VALUE IN HEALTH Volume 17 Number 7 November 2014 ISSN 1098-3015 IN HEALTH VOLUME 17 NUMBER 7 NOVEMBER 2014 ISPOR 17th Annual European Congress Research Abstracts November 8-12, 2014, Amsterdam, The Netherlands ISPOR 6th Asia-Pacific Conference Abstracts September 6-9, 2014, Beijing, China Research Podium Abstracts Research Poster Abstracts PAGES A323–A822 ELSEVIER www.ispor.org EDITORIAL BOARD Co-Editor-in-Chief Michael Drummond, PhD University of York Heslington, York, UK michael.drummond@york.ac.uk Co-Editor-in-Chief C. Daniel Mullins, PhD University of Maryland Baltimore, MD, USA dmullins@rx.umaryland.edu CO-EDITORS Maiwenn Al, PhD iMTA - Erasmus University Rotterdam Rotterdam, The Netherlands al@bmg.eur.nl Ulla S. Skjoldborg, PhD, MA Novo Nordisk A/S Copenhagen, Denmark ulla@skjoldborg.biz Peter Neumann, ScD Tufts University School of Medicine, Boston, MA, USA pneumann@tuftsmedicalcenter.org Kalipso Chalkidou, MD, PhD NICE London, UK kalipso.chalkidou@gmail.com David Veenstra, PhD, PharmD University of Washington Seattle, WA, USA veenstra@u.washington.edu Chris L. Pashos, PhD United BioSource Corporation Lexington, MA, USA chris.pashos@unitedbiosource.com Joshua Cohen, PhD Tufts CSDD Boston, MA, USA joshua.cohen@tufts.edu Allan Wailoo, PhD, BSc, MA University of Sheffield Sheffield, UK A.J.Wailoo@sheffield.ac.uk Shelby Reed, PhD, RPH Duke Clinical Research Institute Durham, NC, USA shelby.reed@duke.edu Jalpa Doshi, PhD University of Pennsylvania Philadelphia, PA, USA jdoshi@mail.med.upenn.edu Danielle Whicher, PhD, MHS PCORI Washington, DC, USA dwhicher@jhsph.edu Uwe Seibert, MD, MPH, MSc, ScD University of Health Sciences, Medical Informatics & Technology Hall i.T., Austria uwe.siebert@umit.at Sheri Fehnel, PhD, MA RTI Health Solutions Research Triangle Park, NC, USA sfehnel@rti.org Alex Z. Fu, PhD Georgetown University Washington, DC, USA zf54@georgetown.edu Benjamin P. Geisler, MD, MPH Health Care Value Strategies Somerville, MA, USA ben.geisler@gmail.com Ron Goeree, MA PATH Research Institute, McMaster University Hamilton, ON, Canada goereer@mcmaster.ca Dan Greenberg, PhD Ben-Gurion University of the Negev Beer-Sheva, Israel dangr@bgu.ac.il Teresa Kauf, PhD University of Florida Gainesville, FL, USA tkandrb@gmail.com Andrew Lloyd, DPhil, BSc ICON Oxford, UK andrew.lloyd@oxfordoutcomes.com Andrea Manca, PhD, MSc University of York York, UK andrea.manca@york.ac.uk Richard Willke, PhD Pfizer Inc. New York, NY, USA Richard.J.Willke@pfizer.com EDITORIAL ADVISORY BOARD Alan Brennan, PhD University of Sheffield Sheffield, UK A.Brennan@sheffield.ac.uk Andrew Briggs, DPhil University of Glasgow Glasgow, UK a.briggs@clinmed.gla.ac.uk John Hornberger, MD, MS Cedar Associates, LLC Menlo Park, CA, USA jhornberger@cedarecon.com Johan L. (Hans) Severens, PhD Erasmus University Rotterdam, The Netherlands severens@bmg.eur.nl Sean Sullivan, PhD University of Washington Seattle, WA, USA sdsull@u.washington.edu Milton C. Weinstein, PhD Harvard School of Public Health Boston, MA, USA mcw@hsph.harvard.edu MANAGEMENT ADVISORY BOARD Maarten J. IJzerman, PhD (Chair) University of Twente Enschede, The Netherlands m.j.ijzerman@utwente.nl Don Husereau, MSc, BSc University of Ottawa Ottawa, ON, Canada don.husereau@gmail.com Josephine Mauskopf, PhD RTI Health Solutions Triangle Park, NC, USA jmauskopf@rti.org Pablo Lapuerta, MD Lexicon Pharmaceuticals Princeton, NJ, USA lapuertp@yahoo.com Jan Busschbach, PhD Erasmus University Rotterdam, Netherlands j.vanbusschbach@erasmusmc.nl Adrian Levy, PhD Dalhousie University Halifax, NS, Canada adrian.levy@dal.ca EDITORIAL OFFICE Managing Editor Paul Scuffham, PhD, BA Griffith University - School of Medicine Queensland, Australia p.scuffham@griffith.edu.au Bryan Luce, PhD, MBA PCORI Washington, DC USA bluce@pcori.org Ya-Chen (Tina) Shih, PhD, MS University of Chicago Chicago, IL, USA tshih@medicine.bsb.uchicago.edu Richard J. Milne, PhD University of Auckland Auckland, New Zealand rj.milne@auckland.ac.nz Stephen L. Priori ISPOR Lawrenceville, NJ, USA spriori@ispor.org Editorial Assistant Angela Buziak ISPOR Lawrenceville, NJ, USA abuziak@ispor.org VOLUME 17 NUMBER 7 NOVEMBER 2014 ISPOR 17TH ANNUAL EUROPEAN CONGRESS RESEARCH ABSTRACTS A323 A324 A324 A325 A326 A326 Research Podium Presentations – Session I Cancer Outcomes Research Studies: CN1–CN4 Conceptual Papers: CP1–CP4 Diagnostic Research Studies: DI1–DI4 Health Care Expenditure Studies: HC1–HC4 Research on Methods – Modeling Studies: MO1–MO4 QALY-Related Studies: QA1–QA4 A327 A328 A329 A329 A330 A331 Research Podium Presentations – Session II Cardiovascular Disease Research Studies: CV1–CV4 Health Services Research Studies: HS1–HS4 Medication Adherence Studies: MA1–MA4 Studies on NICE Assessments: NI1–NI4 Research on Methods – Preference Studies: PR1–PR4 Quality of Life and Utility Studies: QL1–QL4 A332 A337 A353 A357 Research Poster Presentations – Session I Disease-Specific Studies Diabetes/Endocrine Disorders – Clinical Outcomes Studies: PDB1–PDB32 Diabetes/Endocrine Disorders – Cost Studies: PDB33–PDB119 Diabetes/Endocrine Disorders – Patient-Reported Outcomes & Patient Preference Studies: PDB120–PDB143 Diabetes/Endocrine Disorders – Health Care Use & Policy Studies: PDB144–PDB174 A362 A363 A368 A370 Gastrointestinal Disorders – Clinical Outcomes Studies: PGI1–PGI6 Gastrointestinal Disorders – Cost Studies: PGI7–PGI33 Gastrointestinal Disorders – Patient-Reported Outcomes & Patient Preference Studies: PGI34–PGI42 Gastrointestinal Disorders – Health Care Use & Policy Studies: PGI43–PGI56 A372 A375 A383 A388 Muscular-Skeletal Disorders – Clinical Outcomes Studies: PMS1–PMS13 Muscular-Skeletal Disorders – Cost Studies: PMS14–PMS62 Muscular-Skeletal Disorders – Patient-Reported Outcomes & Patient Preference Studies: PMS63–PMS84 Muscular-Skeletal Disorders – Health Care Use & Policy Studies: PMS85–PMS97 A390 A393 A400 A403 Neurological Disorders – Clinical Outcomes Studies: PND1–PND16 Neurological Disorders – Cost Studies: PND17–PND55 Neurological Disorders – Patient-Reported Outcomes & Patient Preference Studies: PND56–PND73 Neurological Disorders – Health Care Use & Policy Studies: PND74–PND80 A405 A405 A406 A406 A417 A418 A418 A432 A436 Research Poster Presentations – Session II Health Care Use & Policy Studies Health Care Use & Policy Studies – Consumer Role In Health Care: PHP1–PHP4 Health Care Use & Policy Studies – Diagnosis Related Group: PHP5–PHP6 Health Care Use & Policy Studies – Disease Management: PHP8–PHP10 Health Care Use & Policy Studies – Drug/Device/Diagnostic Use & Policy: PHP11–PHP74 Health Care Use & Policy Studies – Equity and Access: PHP75–PHP82 Health Care Use & Policy Studies – Formulary Development: PHP83–PHP86 Health Care Use & Policy Studies – Health Care Costs & Management: PHP87–PHP168 Health Care Use & Policy Studies – Health Care Research & Education: PHP169–PHP193 Health Care Use & Policy Studies – Health Technology Assessment Programs: PHP194–PHP247 VALUE IN HEALTH 17 (2014) A445 A446 A447 A447 A448 A448 A449 Health Care Use & Policy Studies – Patient Registries & Post-Marketing Studies: PHP248–PHP249 Health Care Use & Policy Studies – Population Health: PHP250–PHP254 Health Care Use & Policy Studies – Prescribing Behavior & Treatment Guidelines: PHP255–PHP258 Health Care Use & Policy Studies – Quality Of Care: PHP259–PHP260 Health Care Use & Policy Studies – Regulation Of Health Care Sector: PHP261–PHP265 Health Care Use & Policy Studies – Risk Sharing/Performance-Based Agreements: PHP266–PHP271 Health Care Use & Policy Studies – Conceptual Papers: PHP272–PHP297 A453 A456 A460 A463 Disease-Specific Studies Mental Health – Clinical Outcomes Studies: PMH1–PMH15 Mental Health – Cost Studies: PMH16–PMH38 Mental Health – Patient-Reported Outcomes & Patient Preference Studies: PMH39–PMH56 Mental Health – Health Care Use & Policy Studies: PMH57–PMH74 A466 A467 A471 A472 Urinary/Kidney Disorders – Clinical Outcomes Studies: PUK1–PUK4 Urinary/Kidney Disorders – Cost Studies: PUK5–PUK26 Urinary/Kidney Disorders – Patient-Reported Outcomes & Patient Preference Studies: PUK27–PUK34 Urinary/Kidney Disorders – Health Care Use & Policy Studies: PUK35 A472 A478 A492 A497 Research Poster Presentations – Session III Disease–Specific Studies Cardiovascular Disorders – Clinical Outcomes Studies: PCV1–PCV35 Cardiovascular Disorders – Cost Studies: PCV36–PCV113 Cardiovascular Disorders – Patient-Reported Outcomes & Patient Preference Studies: PCV114–PCV140 Cardiovascular Disorders – Health Care Use & Policy Studies: PCV141–PCV183 A505 A506 A512 A518 Individual’s Health – Clinical Outcomes Studies: PIH1–PIH11 Individual’s Health – Cost Studies: PIH13–PIH44 Individual’s Health – Patient-Reported Outcomes & Patient Preference Studies: PIH45–PIH82 Individual’s Health – Health Care Use & Policy Studies: PIH83–PIH99 A512 A525 A534 A539 Systemic Disorders/Conditions – Clinical Outcomes Studies: PSY1–PSY18 Systemic Disorders/Conditions – Cost Studies: PSY20–PSY73 Systemic Disorders/Conditions – Patient-Reported Outcomes & Patient Preference Studies: PSY74–PSY97 Systemic Disorders/Conditions – Health Care Use & Policy Studies: PSY98–PSY121 A543 A547 A553 A555 A567 A576 A580 A583 Research Poster Presentations – Session IV Research on Methods Studies Research on Methods – Clinical Outcomes Methods: PRM1–PRM25 Research on Methods – Cost Methods: PRM26–PRM55 Research on Methods – Databases & Management Methods: PRM56–PRM72 Research on Methods – Modeling Methods: PRM73–PRM134 Research on Methods – Patient-Reported Outcomes Studies: PRM135–PRM184 Research on Methods – Statistical Methods: PRM186–PRM211 Research on Methods – Study Design: PRM212–PRM228 Research on Methods – Conceptual Papers: PRM229–PRM256 A588 A591 A599 A602 Disease-Specific Studies Respiratory-Related Disorders – Clinical Outcomes Studies: PRS1–PRS16 Respiratory-Related Disorders – Cost Studies: PRS17–PRS65 Respiratory-Related Disorders – Patient-Reported Outcomes & Patient Preference Studies: PRS66–PRS81 Respiratory-Related Disorders – Health Care Use & Policy Studies: PRS82–PRS89 A604 A605 A610 A612 Sensory Systems Disorders – Clinical Outcomes Studies: PSS1–PSS9 Sensory Systems Disorders – Cost Studies: PSS10–PSS36 Sensory Systems Disorders – Patient-Reported Outcomes & Patient Preference Studies: PSS37–PSS51 Sensory Systems Disorders – Health Care Use & Policy Studies: PSS52–PSS59 VALUE IN HEALTH 17 (2014) A614 A621 A645 A651 Research Poster Presentations – Session V Disease-Specific Studies Cancer – Clinical Outcomes Studies: PCN1–PCN40 Cancer – Cost Studies: PCN41–PCN176 Cancer – Patient-Reported Outcomes & Patient Preference Studies: PCN177-PCN212 Cancer – Health Care Use & Policy Studies: PCN212–PCN281 A664 A667 A681 A682 Infection – Clinical Outcomes Studies: PIN1–PIN20 Infection – Cost Studies: PIN21–PIN97 Infection – Patient-Reported Outcomes & Patient Preference Studies: PIN98–PIN103 Infection – Health Care Use & Policy Studies: PIN104–PIN125 A687 ISPOR 17TH ANNUAL EUROPEAN CONGRESS DISCLOSURE INFORMATION A700 ISPOR 17TH ANNUAL EUROPEAN CONGRESS RESEARCH ABSTRACTS AUTHOR INDEX ISPOR 6TH ASIA-PACIFIC CONFERENCE ABSTRACTS A719 A719 A720 A721 A722 Research Podium Presentations – Session I Cancer Outcomes Research Studies: CN1–CN4 Diabetes Outcomes Research Studies: DB1–DB4 Drug Use Studies: DU1–DU4 Health Services Research Studies: HS1–HS4 Mental Health Outcomes Research Studies: MH1–MH4 A722 A723 A724 A725 A725 Research Podium Presentations – Session II Cardiovascular Disease Outcomes Research Studies: CV1–CV4 Health Care Reimbursement Studies: HC1–HC4 Infectious Disease Outcomes Research Studies: IN1–IN4 QALY-Related Studies: QA1–QA4 Respiratory-Related Disorders Outcomes Research Studies: RR1–RR4 A726 A727 A728 A729 A730 A731 A731 A732 Research Poster Presentations – Session I Research on Methods Studies Research on Methods – Clinical Outcomes Methods: PRM1–PRM7 Research on Methods – Cost Methods: PRM8–PRM10 Research on Methods – Databases & Management Methods: PRM11–PRM16 Research on Methods – Modeling Methods: PRM17–PRM24 Research on Methods – Patient-Reported Outcomes Studies: PRM25–PRM31 Research on Methods – Statistical Methods: PRM32 Research on Methods – Study Design: PRM33–PRM34 Research on Methods – Conceptual Papers: PRM35–PRM40 A733 A734 A737 A739 Disease-Specific Studies Cancer – Clinical Outcomes Studies: PCN1–PCN5 Cancer – Cost Studies: PCN6–PCN28 Cancer – Patient-Reported Outcomes & Patient Preference Studies: PCN29–PCN36 Cancer – Health Care Use & Policy Studies: PCN37–PCN48 A740 A743 A746 A747 Diabetes/Endocrine Disorders – Clinical Outcomes Studies: PDB1–PDB12 Diabetes/Endocrine Disorders – Cost Studies: PDB13–PDB30 Diabetes/Endocrine Disorders – Patient-Reported Outcomes & Patient Preference Studies: PDB31–PDB40 Diabetes/Endocrine Disorders – Health Care Use & Policy Studies: PDB41–PDB48 A749 A749 A750 Gastrointestinal Disorders – Clinical Outcomes Studies: PGI1–PGI4 Gastrointestinal Disorders – Cost Studies: PGI5 Gastrointestinal Disorders – Health Care Use & Policy Studies: PGI6 VALUE IN HEALTH 17 (2014) A750 A751 A752 A754 Individual’s Health – Clinical Outcomes Studies: PIH1–PIH8 Individual’s Health – Cost Studies: PIH10–PIH13 Individual’s Health – Patient-Reported Outcomes & Patient Preference Studies: PIH15–PIH27 Individual’s Health – Health Care Use & Policy Studies: PIH28–PIH37 A756 A759 A763 A764 Research Poster Presentations – Session II Cardiovascular Disorders – Clinical Outcomes Studies: PCV1–PCV23 Cardiovascular Disorders – Cost Studies: PCV25–PCV44 Cardiovascular Disorders – Patient-Reported Outcomes & Patient Preference Studies: PCV45–PCV51 Cardiovascular Disorders – Health Care Use & Policy Studies: PCV52–PCV58 A765 A767 A769 A770 Mental Health – Clinical Outcomes Studies: PMH1–PMH9 Mental Health – Cost Studies: PMH10–PMH23 Mental Health – Patient-Reported Outcomes & Patient Preference Studies: PMH25–PMH30 Mental Health – Health Care Use & Policy Studies: PMH32–PMH36 A771 A772 A774 A775 Muscular-Skeletal Disorders – Clinical Outcomes Studies: PMS1–PMS7 Muscular-Skeletal Disorders – Cost Studies: PMS8–PMS19 Muscular-Skeletal Disorders – Patient-Reported Outcomes & Patient Preference Studies: PMS20–PMS27 Muscular-Skeletal Disorders – Health Care Use & Policy Studies: PMS28–PMS32 A776 A777 A779 A780 Respiratory-Related Disorders – Clinical Outcomes Studies: PRS1–PRS8 Respiratory-Related Disorders – Cost Studies: PRS9–PRS19 Respiratory-Related Disorders – Patient-Reported Outcomes & Patient Preference Studies: PRS20–PRS22 Respiratory-Related Disorders – Health Care Use & Policy Studies: PRS23–PRS28 A781 A781 A783 A783 Sensory Systems Disorders – Clinical Outcomes Studies: PSS1–PSS2 Sensory Systems Disorders – Cost Studies: PSS3–PSS9 Sensory Systems Disorders – Patient-Reported Outcomes & Patient Preference Studies: PSS10–PSS12 Sensory Systems Disorders – Health Care Use & Policy Studies: PSS13 A783 A784 A784 A785 Systemic Disorders/Conditions – Clinical Outcomes Studies: PSY1–PSY4 Systemic Disorders/Conditions – Cost Studies: PSY5–PSY6 Systemic Disorders/Conditions – Patient-Reported Outcomes & Patient Preference Studies: PSY7–PSY8 Systemic Disorders/Conditions – Health Care Use & Policy: PSY9–PSY11 A785 A785 A786 A786 A789 A790 A790 A796 A797 A798 A799 A799 A800 A800 A801 Research Poster Presentations – Session III Health Care Use & Policy Studies Health Care Use & Policy Studies – Consumer Role in Health Care: PHP1–PHP2 Health Care Use & Policy Studies – Diagnosis Related Group: PHP3 Health Care Use & Policy Studies – Disease Management: PHP4–PHP5 Health Care Use & Policy Studies – Drug/Device/Diagnostic Use & Policy: PHP6–PHP25 Health Care Use & Policy Studies – Equity and Access: PHP26–PHP29 Health Care Use & Policy Studies – Formulary Development: PHP31 Health Care Use & Policy Studies – Health Care Costs & Management: PHP32–PHP70 Health Care Use & Policy Studies – Health Care Research & Education: PHP71–PHP79 Health Care Use & Policy Studies – Health Technology Assessment Programs: PHP81–PHP87 Health Care Use & Policy Studies – Population Health: PHP88 Health Care Use & Policy Studies – Prescribing Behavior & Treatment Guidelines: PHP89–PHP92 Health Care Use & Policy Studies – Quality of Care: PHP93–PHP94 Health Care Use & Policy Studies – Regulation of Health Care Sector: PHP95–PHP98 Health Care Use & Policy Studies – Risk Sharing/Performance-Based Agreements: PHP100–PHP104 Health Care Use & Policy Studies – Conceptual Papers: PHP105–PHP113 A802 A804 A807 A808 InfectIon – Clinical Outcomes Studies: PIN1–PIN12 InfectIon – Cost Studies: PIN14–PIN27 InfectIon – Patient-Reported Outcomes & Patient Preference Studies: PIN28–PIN31 InfectIon – Health Care Use & Policy Studies: PIN32–PIN35 VALUE IN HEALTH 17 (2014) A808 A809 A809 A810 Neurological Disorders – Clinical Outcomes Studies: PND1 Neurological Disorders – Cost Studies: PND2–PND4 Neurological Disorders – Patient-Reported Outcomes & Patient Preference Studies: PND5–PND6 Neurological Disorders – Health Care Use & Policy Studies: PND7–PND11 A810 A811 A812 Urinary/Kidney Disorders – Clinical Outcomes Studies: PUK1–PUK6 Urinary/Kidney Disorders – Cost Studies: PUK7–PUK13 Urinary/Kidney Disorders – Patient-Reported Outcomes & Patient Preference Studies: PUK14–PUK16 A814 ISPOR 6TH ASIA-PACIFIC CONFERENCE ABSTRACTS DISCLOSURE INFORMATION A818 ISPOR 6TH ASIA-PACIFIC CONFERENCE ABSTRACTS AUTHOR INDEX Editorial Office. Value in Health, ISPOR, 505 Lawrence Square Blvd. South, Lawrenceville, NJ 08648. ISPOR Office. Marilyn Dix Smith, RPh, PhD, Executive Director, 505 Lawrence Square Blvd. South, Lawrenceville, NJ 08648. Tel: (609) 586-4981, Fax: (609) 586-4982, E-mail: info@ispor.org, Web site: http://www.ispor.org/valueinhealth_index.asp. CUSTOMER SERVICE (orders, claims, online, change of address): Elsevier Health Sciences Division, Subscription Customer Service, 3251 Riverport Lane, Maryland Heights, MO 63043. Tel: (800) 654-2452 (U.S. and Canada); (314) 447-8871 (outside U.S. and Canada). Fax: (314) 447-8029. E-mail: JournalsCustomerService-usa@elsevier.com (for print support); JournalsOnlineSupport-usa@elsevier.com (for online support). Address changes must be submitted four weeks in advance. YEARLY SUBSCRIPTION RATES: United States and possessions: Individual $298.00. All other countries (prices include airspeed delivery): Individual $298.00. To receive student/resident rate, orders must be accompanied by name of affiliated institution, date of term and the signature of program/residency coordinator on institution letterhead. Orders will be billed at the individual rate until proof of status is received. Current prices are in effect for back volumes and back issues. Further information on this journal is available from the Publisher or from this journal’s website (http://www.elsevier.com/locate/jval). Information on other Elsevier products is available through Elsevier’s website (http://www. elsevier.com). Author inquiries For inquiries relating to the submission of articles (including electronic submission where available), visit http:/www.elsevier.com/authors. The site also provides the facility to track accepted articles and set up e-mail alerts to inform you of when an article’s status has changed, as well as detailed artwork guidelines, copyright information, frequently asked questions, and more. Please see Information for Authors for individual journal. Contact details for questions arising after acceptance of an article, especially those relating to proofs, are provided after registration of an article for publication. English language help service: Upon request, Elsevier will direct authors to an agent who can check and improve the English of their paper (before submission). Please contact authorsupport@elsevier.com for further information. Reprints. For queries about author offprints, e-mail authorsupport@ elsevier.com. To order 100 or more reprints for educational, commercial, or promotional use, contact Derrick Imasa at (212) 633-3874, Elsevier Inc., 360 Park Avenue South, New York, NY 10010-1710. Fax: (212) 462-1935; e-mail reprints@elsevier.com. Reprints of single articles available online may be obtained by purchasing Pay-Per-View access for $31.50 per article on the journal website http://www.elsevier.com/locate/jval. Copyright © 2014, International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc. This journal and the individual contributions contained in it are protected under copyright by International Society for Pharmacoeconomics and Outcomes Research, and the following terms and conditions apply to their use: Photocopying Single photocopies of single articles may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier’s Rights Department in Oxford, UK: phone +44 (0) 1865 843830, fax +44 (0) 1865 853333. Requests may also be completed online via the Elsevier homepage (http://www.elsevier.com/authors/obtaining-permission-to-re-use-elsevier-material). In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (978) 750-8400, fax: (978) 750-4744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (+44) 20 7631 5555; fax: (+44) 20 7631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Subscribers may reproduce tables of contents or prepare lists of articles including abstracts for internal circulation within their institutions. Permission of the Publisher is required for resale or distribution outside the institution. Permission of the Publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this journal, including any article or part of an article. Except as outlined above, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier Rights Department, at the fax and e-mail addresses noted above. Abstracting and Indexing Value in Health is indexed in Index Medicus/MEDLINE, Current Contents/Social & Behavioral Sciences, SciSearch/SCI Expanded, Social Sciences Citation Index, International Pharmaceutical Abstracts, Embase/Excerpta Medica, SciVerse Scopus and PsychInfo/Psychological Abstracts, and Journal Citation Reports/ Science Edition (Thomson ISI). Notice No responsibility is assumed by the Publisher or the International Society for Pharmacoeconomics and Outcomes Research for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. Although all advertising material is expected to conform to ethical (medical) standards, inclusion in this publication does not constitute a guarantee or endorsement of the quality or value of such product or of the claims made of it by its manufacturer. VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Available online at www.sciencedirect.com journal homepage: www.elsevier.com/locate/jval ABSTRACTS ISPOR 17TH ANNUAL EUROPEAN CONGRESS RESEARCH ABSTRACTS intermediate-risk prostate cancer at relatively low cost/QALY thresholds, and RP is expected to be the most cost-effective of available treatments at the prevailing range of cost/QALY thresholds (i.e. ₤20,000-₤30,000) . However, large decision uncertainty exists and acquiring further information is likely cost-effective. Future research on costs and utilities associated with treatment outcome and adverse events is expected to be most valuable. Research Podium Presentations – Session I Cancer Outcomes Research Studies CN1 Long-Term Impact of the Dutch Colorectal Cancer Screening Programme on Cancer Incidence: Exploration of the Serrated Pathway Greuter M J 1, Lew J B 2, Berkhof J 1, Canfell K 2, Dekker E 3, Meijer G A 1, Coupe V M 1 1VU University Medical Center, Amsterdam, The Netherlands, 2University of New South Wales, Sydney, Australia, 3Academic Medical Center, Amsterdam, The Netherlands . . . . . . . . . . . Objectives: The Netherlands has recently started with the stepwise implementation of biennial faecal immunochemical testing for colorectal cancer (CRC). We evaluated the impact of the transition to, and the fully implemented screening programme on the long-term CRC incidence and colonoscopy demand. Methods: The previously reported and calibrated ASCCA model was set up to simulate the Dutch CRC screening programme between 2014 and 2044. We adopted an open-model approach by simulating multiple birth cohorts and combining the results while accounting for the ageing of the population. Besides a no screening scenario, we evaluated the impact of screening under three sets of natural history assumptions which differed in the contribution of the serrated pathway to the CRC incidence (0%, 15% and 30%). Model-predicted outcomes were CRC incidence and colonoscopy demand per year. Results: Due to ageing, the model-predicted CRC incidence in the no screening scenario increased from 77/100,000 in 2014 to 109/100,000 in 2044. Under screening, the predicted CRC incidence in 2014 was between 105/100,000 (assuming all CRCs originate from adenomas) and 109/100,000 (assuming that 30% of CRCs arises from serrated lesions) due to the detection of asymptomatic, prevalent tumours. After this peak, the predicted incidence gradually decreased until in 2039 a new equilibrium was reached, ranging between 65/100,000 and 71/100,000 assuming that 100% versus 70% of CRCs originate from adenomas, respectively. Due to the stepwise implementation, the predicted number of colonoscopies required for the screening programme increased gradually over time from 38,000 (752,199 invitees) in 2014 to 117,000 (2,154,875 invitees) in 2044. Conclusions: The Dutch screening programme will markedly decrease CRC incidence in the next 25 years. The conclusions about the impact of screening were robust to key natural history assumptions. With the results of this study, decision-makers can anticipate the expected change in CRC-related health care use and colonoscopy demand. CN2 Primary Treatments for Intermediate-Risk Prostate Cancer: a CostEffectiveness and Value-of-Information Analysis M 1, IJzerman M J 2, Steuten LMG3 Piena 1PANAXEA bv, Enschede, The Netherlands, 2University of Twente and MIRA institute for Biomedical Technology & Technical Medicine, Enschede, The Netherlands, 3University of Twente, Enschede, The Netherlands . . . . . . Objectives: Intermediate-risk prostate cancer patients are recommended primary treatment with either radical prostatectomy (RP), external beam radiotherapy (EBRT), brachytherapy (BT), or EBRT plus high-dose rate BT boost (EBRT + HDR-BT); or expectant management with active surveillance (AS). The costs of these treatments differ considerably, whilst the amount and quality of evidence for their comparative effectiveness in terms of disease progression, adverse events and health-related quality of life is unbalanced and inconclusive. Therefore, we undertook a cost-effectiveness analysis of RP, EBRT, BT, EBRT + HDR-BT and AS, and performed a value-of-information analysis to direct future research. Methods: We developed a probabilistic Markov model estimating the expected incremental cost/(Quality Adjusted) Life Years from a UK-NHS perspective, with a time horizon of 10 years. Input data were obtained from the best available literature. We explored the uncertainty around the model outcomes by identifying the most influential parameters and estimating the expected value of perfect (parameter) information. Results: AS is most likely to be cost-effective at a cost/QALY threshold (λ ) < ₤3,000/QALY, BT for λ ₤3,000 to ₤12,000/QALY and RP for λ > ₤12,000/QALY. One-way sensitivity analysis shows that utilities and probabilities of adverse events are main effect drivers and initial treatment costs are main cost drivers. Large decision uncertainty exists around λ ₤11,000 with a population EVPI of nearly ₤100 million. The EVPPI suggests that eliminating uncertainty around costs and utilities is most worthwhile. Conclusions: With current information AS and BT are cost-effective treatments for CN3 Early Stage Cost-Effectiveness Analysis of a Brca1-Like Test to Detect Triple Negative Breast Cancers Responsive to High Dose Alkylating Chemotherapy Miquel Cases A 1, Steuten L M G 2, Retèl V P 1, van Harten W H 2 1Netherlands Cancer Institute, Amsterdam, The Netherlands, 2University of Twente, Enschede, The Netherlands . . . . . . . . Objectives: Triple negative breast cancers (TNBC) with a BRCA1-like profile may benefit from high dose alkylating chemotherapy (HDAC). This study examines whether treating TNBC with personalized HDAC based on BRCA1-like testing can be more cost-effective than current clinical practice. Additionally we estimated the minimum required prevalence of BRCA1-likeness and the required positive predictive value (PPV) for a BRCA1-like test to render this strategy cost-effective. Methods: Our markov model compared the outcomes of treating TNBC women with personalized HDAC based on BRCA1-like testing vs. current clinical practice from a societal Dutch perspective and a 20-year time horizon. From our base-case model we assessed: 1) the incremental number of respondents; 2) the incremental number of Quality Adjusted Life Years, 3) the incremental costs, and 4) the incremental cost-effectiveness ratio (ICER). We performed one-way sensitivity analysis (SA) of all model parameters, and two-way SA of prevalence and PPV. Data were obtained from a current trial (NCT01057069), published literature and expert opinions where necessary. Results: Based on our base-case analysis with 68% BRCA1-like prevalence, 100% PPV, and costs of € 164 / test, treating TNBC according to BRCA1-like testing would be cost-effective (€ 16.192/QALY). One-way SA on the prevalence and PPV demonstrated that only the PPV drives the ICER changes. In two-way SA, the lower bound for the two parameters was: prevalence 39.6% and PPV 46.4%. Regardless of prevalence, at PPVs > 46.4% BRCA1-like testing was always cost-effective. Conclusions: Treating TNBC with personalized HDAC based on BRCA1-like testing is expected to be cost-effective at a minimum PPV of 46%. This information can help test developers in decisions on further research and development. CN4 The Cost of Costing Treatments Incorrectly: Errors in the Application of Drug Prices in Economic Models Due to Differing Patient Weights Hatswell A J 1, Porter J 1, Hertel N 2, Lee D 1 1BresMed, Sheffield, UK, 2Bristol Myers Squibb, Uxbridge, UK . . . . . Objectives: Drug costs are generally a key driver of the results of economic models. We tested the impact on drug cost estimates for the following common approaches: using mean patient weight, individual patient weights or fitting a distribution to the observed patient weights. Methods: For the analysis, we utilised patient weight and height data from trial CA184-024 (517 patients) in metastatic melanoma. Based on this dataset, costs of a single administration of drug therapy were calculated using UK list prices. Costs were calculated for four recently licensed treatments with different posologies: ipilimumab (mg/kg, with 2 vial sizes), cabazitaxel (mg/ m2), ustekinumab (doubled dosage over 100kg patient weight) and romiplostim (µg/ kg, with a large, single vial size). Results: Cost estimates using the mean patient weight were £18,750, £3,696, £2,147 and £482 per administration of ipilimumab, cabazitaxel, ustekinumab and romiplostim, respectively. These results increased by 4.9%, 2.3%, 10.3% and 36.6% when costing individual patient weights, and by 5.2%, 2.3%, 11.8% and 36.9% when fitting a distribution to the patient weights. The use of only mean patient weight consistently underestimated costs compared to methods that incorporated the distribution of weight data. Sampling from the observed patient weight distribution provides a more accurate estimation of costs; however, it is subject to over- or under-estimation, depending on enrolment in a trial programme, particularly amongst patients who are substantially over- or under-weight. Conclusions: Accurate estimation of drug costs requires an understanding of the distribution of patient weights. Failing to take this into account can result in cost estimates that are substantially lower than will be seen in practice, which could (in turn) impact treatment (implementation) decisions. These errors would be further compounded should drug wastage not be adequately captured. Modellers should be mindful of these issues when costing therapies or conducting health technology assessment submissions. Copyright © 2014, International Society for Pharmacoeconomics and Outcomes Research (ISPOR). Published by Elsevier Inc. A324 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Conceptual Papers CP1 The Evaluation of Economic Methods to Assess the Social Value of Medical Interventions for Ultra-Rare Disorders (URDS) Schlander M 1, Garattini S 2, Holm S 3, Kolominsky-Rabas P L 4, Nord E 5, Persson U 6, Postma M J 7, Richardson J 8, Simoens S 9, de Sola-Morales O 10, Tolley K 11, Toumi M 12 1Institute for Innovation & Valuation in Health Care (InnoVal-HC), Wiesbaden, Germany, 2Mario Negri Institute for Pharmacological Research, Milano, Italy, 3University of Manchester, Manchester, UK, 4University of Erlangen, Erlangen, Germany, 5Norwegian Institute of Public Health, Oslo, Norway, 6The Swedish Institute for Health Economics (IHE), Lund, Sweden, 7University of Groningen, Groningen, The Netherlands, 8Monash University, Clayton, Victoria, Australia, 9KU Leuven, Leuven, Belgium, 10Sabirmedical, Barcelona, Spain, 11Tolley Health Economics Ltd., Buxton, Derbyshire, UK, 12University Claude Bernard Lyon 1, Lyon, France . . . . . . . . . . . . . . Objectives: To develop a set of criteria to critically appraise the strengths and weaknesses of health economic methods for the systematic valuation of interventions for ultra-rare disorders (URDs). Methods: An international group of clinical and health economic experts met in conjunction with the Annual European ISPOR Congresses in Berlin/Germany and Dublin/Ireland, November 2012 and 2013, to deliberate and agree on a set of criteria to assess the potential of the various methods, which have been used or proposed to estimate the social value of medical interventions for URDs. Results: The group identified a broad set of potential criteria, which may be grouped according to the following dimensions: theoretical foundations (normative premises, i.e., links to moral and economic theories, including - but not limited to – nonutilitarian consequentialist and deontological reasoning, definition and treatment of core concepts of economic thinking such as opportunity costs and efficiency), empirical underpinnings (social preferences related to attributes of the health condition or of the person afflicted with it), and pragmatic aspects (feasibility of implementation and potential for bias and misuse). For each of the dimensions, a set of criteria has been agreed upon, which in turn will need further scrutiny and justification. Conclusions: Previously, a need had been identified for modifications or alternatives to the conventional logic of cost effectiveness applying benchmarks for the maximum allowable cost per qualityadjusted life year (QALY). We propose a framework for the systematic assessment how well different evaluation approaches reflect prevalent social norms and value judgments. As a next step, the framework shall be applied on multi-criteria decision analysis methods and social cost value analysis, either using the person trade-off (PTO) or the relative social willingness-to-pay (RS-WTP) instrument. validity of network meta-analysis; and (iv) the possibility of disagreement between direct evidence and indirect evidence. We illustrate the framework using a network meta-analysis of topical antibiotics without steroids for chronically discharging ears with underlying eardrum perforations. CP4 Amending the Guide to Methods of Technology Appraisal at Nice to Incorporate two New Value Elements: Burden of Illness And Wider Societal Impact Knight H 1, Boysen M 1, Stevens A 2, Longson C 1 Institute for Health and Care Excellence, Manchester, UK, 2University of Birmingham, Birmingham, UK . . . . 1National Background: In July 2013 the Department of Health referred terms of reference for value based assessment of health technologies to NICE. Objective: We present the approach taken over the past 11 months to amending the Guide to Methods of Technology Appraisal to incorporate burden of illness and wider societal impact. Methods: Given the time frame available, NICE built on prior work undertaken by the Department of Health (in the context of value based pricing) on the concepts of burden of illness and wider societal benefits by commissioning the NICE Decision Support Unit to review and critique this existing work. NICE reconvened the working party from the Guide to the Methods of Technology Appraisal review which took place in 2012, which included standing membership drawn from the stakeholder communities such as patient and professional organisations, academia and pharmaceutical industry. The working party considered the prior work undertaken by the Department of Health and the Decision Support Unit’s critique for burden of illness and wider societal benefit, and provided advice on the incorporation of the 2 new value elements into NICE’s current methods at 4 meetings. A consultation paper describing NICE’s proposals and draft of the amended sections of the methods guide was published in March 2014, and consultation ran for 12 weeks. It is anticipated that the final amendment of the methods guide will be considered by the NICE Board in advance of the ISPOR conference. Results: Key points drawn from the discussion at the working party and consultation responses regarding burden of illness and wider societal impact, will be discussed. Discussion: Considering NICE’s ‘position’ in the world of health technology assessment and appraisal, the conclusions from this latest amendment of the Guide to Methods of Technology Appraisal to incorporate value based assessment will be (highly) anticipated. Diagnostic Research Studies CP2 Value in the Making: Harvesting the Value of Complex Medical Innovations in Practice Abrishami P 1, Boer A 2, Horstman K 1 University, Maastricht, The Netherlands, 2National Health Care Institute, Diemen, The Netherlands . . . 1Maastricht Rapid development of medical innovations in the face of rising health care costs have been calling for a more value-conscious adoption and diffusions of innovations. This conceptual paper departs from swift adoption of the da Vinci surgical robot in the Netherlands. It describes three challenges facing health care systems to evaluate promising, yet complex and often expensive medical innovations. Firstly, they are often adopted and diffused prior to their evidence-based superiority being proven. Secondly, formal evaluation frameworks are somehow detached from the dynamics of and incentives for adoption and diffusion of these innovations. Third, the real risks and benefits of these innovations are not easily amenable to an experimental inquiry. Unlike pharmaceuticals, whose impact is intrinsic to its biochemical components and thus can be subject to experiment, the value of complex surgical devices, imaging equipments, or targeted therapy interventions are inseparable from actual patterns of human practices and clinical pathways that utilize them. Multi-stakeholder (early) deliberation has often been proposed to better align value requirements with the adoption and diffusion processes. This article examines the importance of developing a shared value perspective on the implementation of complex innovations through early deliberation. Product developers, (potential) adopters (providers or patients), purchasers, and policy makers may engage in an upfront iterative deliberation on all the particularities and (pre)conditions that account for delivering value of a certain innovation during early adoption in a given care delivery setting. Such deliberation offers a cumulative learning as to how to reduce true-to-life uncertainties and risks ‘along the way’, thereby serving for value ‘fulfillment’ in practice. Implication of such situated deliberative platforms for technology (outcome) assessment and for the role of authorities is discussed. A concrete framework for multi-stakeholder deliberation applied to the case of the da Vinci surgical robot in the Netherlands is also proposed. CP3 Evaluating the Quality of Evidence from a Network Meta-Analysis Higgins J P 1, Del Giovane C 2, Chaimani A 3, Caldwell D M 1, Salanti G 3 1University of Bristol, Bristol, UK, 2University of Modena and Reggio Emilia, Modena, Italy, 3University of Ioannina School of Medicine, Ioannina, Greece . . . . . . . Systematic reviews that collate data about the relative effects of multiple interventions via network meta-analysis are highly informative for decision-making purposes. A network meta-analysis provides two types of findings for a specific outcome: the relative treatment effect for all pairwise comparisons, and a ranking of the treatments. It is important to consider the confidence with which these two types of results can enable clinicians, policy makers and patients to make informed decisions. We propose an approach to determining confidence in the output of a network meta-analysis, based on methodology developed by the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group for pairwise meta-analyses. The suggested framework for evaluating a network meta-analysis acknowledges (i) the key role of indirect comparisons (ii) the contributions of each piece of direct evidence to the network meta-analysis estimates of effect size; (iii) the importance of the transitivity assumption to the DI1 Cost-Effectiveness (CE) of Imaging-Guided Strategies for the Diagnosis of Coronary Artery Disease (CAD): Results From the Evinci Study Lorenzoni V 1, Pierotti F 1, Bellelli S 1, Neglia D 2, Rovai D 2, Turchetti G 1 Superiore Sant’Anna, Pisa, Italy, 2National Research Council, Pisa, Italy . . . . . . 1Scuola Objectives: To evaluate the cost-effectiveness (CE) of imaging-guided strategies for the diagnosis of significant coronary artery disease (CAD) in patients with intermediate pre-test likelihood. Methods: Significant CAD was defined at invasive coronary angiography (ICA) as > 50% stenosis in the left main or > 70% stenosis in a major coronary vessel or 30-70% stenosis with fractional flow reserve ≤ 0.8.Nine diagnostic strategies were compared using a CE analysis. Strategies included the use of one single or two combined non-invasive imaging tests (CTCA as first line test and then stress ECHO, CMR, PET or SPECT) followed by ICA in the case of positivity of the single test or both non-invasive examinations in the case of combinations. ICERs were obtained using per-patient data collected throughout the EVINCI multicentre European study. Strategy costs were calculated using examination countryspecific reimbursements, while effectiveness was defined as the percentage of correct diagnosis. All costs were converted to Euro 2012 and adjusted using PPP. A propensity-score adjustment was used in the analysis and 95%CI were obtained with non-parametric bootstrap. Results: Among the strategies analysed only three resulted cost-effective for the diagnosis of significant CAD. These included stress ECHO and CTCA as single non-invasive test, CTCA first then ECHO, CTCA first and then stress PET, all followed by ICA when required. Stress ECHO approach was the least costly but also the least effective, while CTCA alone [ICER: 2345 (2287-2400)] or in combination with PET [ICER: 5227(5161-5296)] had increasingly higher effectiveness for a willingness to pay (WTP) exceeding 2,000 Euro and 5,000 Euro, respectively. Conclusions: Results from the health-economic analysis of the EVINCI study showed that stress ECHO guided diagnostic strategy could be cost-effective when the WTP is low. Strategies involving CTCA alone or as first line exam followed by stress PET could allow a more accurate diagnostic workflow for higher WTP. DI2 The Value of Risk-Stratified Information in the National Lung Cancer Screening Trial Soeteman D I 1, Cohen J T 1, Neumann P J 2, Wong J B 1, Kent D M 1 1Tufts Medical Center/Tufts University School of Medicine, Boston, MA, USA, 2Tufts Medical Center, Boston, MA, USA . . . . . . . . . . Objectives: Clinical guideline recommendations are generally informed by population-based evidence. However, interventions that are (cost-)effective on average may not be (cost-)effective for many (even for most) patients meeting trial inclusion criteria. This study aims to investigate the value of risk-stratified recommendations for lung cancer screening among current or former smokers between the ages of 55 and 74 years compared to a screen-all policy. Methods: Using data from the National Lung Cancer Screening Trial (NLST), we calculated the costs and QALYs for low-dose computed tomography (CT) versus chest radiography (X-ray) from empirically observed health states and 6 years life expectancy. Based on Kovalchik’s risk of lung cancer death prediction model, we stratified 53,454 NLST trial patients into quintiles. The expected value of individualized care (EVIC) was calculated to A325 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 quantify the value of using stratified information over population-based information. Results: The incremental cost-effectiveness ratio (ICER) of CT versus X-ray was $31,942 per QALY, for the “average” trial patient, indicating that CT would be the preferred option at a cost-effectiveness threshold of $50,000 per QALY. However, when stratified into quintiles, CT is dominated for the lowest risk quintile (i.e., X-ray is the preferred option for quintile 1) and CT is preferred for higher risk groups (quintiles 2 to 5). The EVIC was calculated at around $180 per person for cost-effectiveness thresholds of $50,000 per QALY and higher. Conclusions: Tailoring screening strategies to avoid CT scan in the lowest risk quintile of patients appears to be a superior strategy compared to population-wide CT scan screening, although results were sensitive to the cost-effectiveness threshold and the level of granularity of the analysis. This study shows the value of considering the risk-based heterogeneity of cost-effectiveness in clinical guideline recommendations and policy decisions. DI3 Effect of Self-Monitoring of Blood Glucose on Glycemic Control, Clinical Outcomes, and Health Care Costs in Diabetic Patients Using Insulin: a Retrospective Analysis Degli Esposti L , Saragoni S , Blini V , Buda S CliCon Srl, Ravenna, Italy . . . . Objectives: Self-monitoring of blood glucose (SMBG) may improve diabetes management. We analyzed the effect of SMBG on glycemic control, clinical outcomes and health care costs among insulin-users diabetic patients in a clinical practice setting. Methods: A retrospective analysis using data from the administrative databases, clinical registries containing laboratory results and medical devices databases including SMBG strips data of two Italian Local Health Units was performed. Insulin-users were defined if they had at least one prescription of insulin agents from November, 2009 and April, 2011. The first prescription was selected as index-date. Patients were divided into two groups based on testing frequency of SMBG during the 18 months after the index-date: no test strip claims (no SMBG use) and more than two test strips per day. We calculated incidence rates to estimate the risk for fasting blood glucose levels < 70 mg/dl and for diabetes-related hospitalizations or death occurrence during the 18-months follow-up period. Total annual costs included hypoglycemic therapy and the direct costs due to diabetes-related hospitalizations and outpatient services. Results: We identified 394 insulin-users patients with no SMBG use and 1350 with SMBG performed more than twice per day. Compared with non-SMBG use group, patients using SMBG showed a significant reduced risk of glucose levels < 70 mg/dl (unadjusted rate was 10,6 vs 27,3 per 100 person-years, p<0.001) and of diabetes-related hospitalizations or death (30,0 vs 60,8 per 100 person-years, p<0.001). The higher hospitalization rate resulted in higher hospitalization costs per patient (€ 2.419 vs € 1.512 of those with SMBG use) and consequently higher total annual direct costs per patient (€3.060 vs €2.738 of those using SMBG). Conclusions: Results indicate that patients using SMBG, compared with non-SMBG patients, are associated with better glycemic control and reduced risk of diabetes-related hospitalizations and consequently with lower overall total annual cost per patient. DI4 Diagnosing Anxiety Disorders in Primary Care: a Systematic Review and Meta-Analysis Olariu E 1, Rodrigo M 2, Alvarez Lopez P 3, Castro-Rodriguez J I 3, Martin-Lopez L M 3, Alonso J 4, Garcia Forero C 1 1PRBB - IMIM Instituto Hospital del Mar de Investigaciones Médicas, Barcelona, Spain, 2Pompeu Fabra University, Barcelona, Spain, 3Institut de Neuropsiquiatria i Addiccions, Barcelona, Spain, 4IMIM-Research Institute Hospital del Mar, Barcelona, Spain . . . . . . . . . Objectives: Anxiety and mood disorders are highly prevalent in Primary Care but research shows that general practitioners (GPs) fail to diagnose up to half of cases. In this study we try to determine the diagnostic accuracy of GPs’ diagnoses of anxiety disorders with and without any help from diagnostic (assisted vs unassisted diagnosis). Methods: We searched for articles published from January 1980 to June 2014 in 7 databases. We included studies in English, Spanish, French, and German reporting the ability of GPs to identify any anxiety disorder (DSM III/ IV/ IV-TR diagnostic criteria) in Primary Care community samples. We excluded studies from general population and those addressing specific physical or mental disorders, along with vignette and case-series studies. Two authors independently performed abstract and full-text reviews and data extraction. Study was assessed with the QUADAS-2. Coupled forest plots summarized estimated studies’ sensitivity and specificity and 95% confidence intervals. We fitted random-effects meta-analysis models and undertook a bivariate meta-analysis to construct a summary Receiver Operator Characteristic Curve (sROC). Results: From a total 17.964 detected papers, 443 were included for full text review. So far, we have analyzed 111 papers, out of which 8 studies were included with N= 3608 patients with pooled anxiety prevalence 26% (CI=25-27%). Preliminary results shows an overall ROC curve with lower GP diagnostic accuracy when performing unassisted diagnoses for a total diagnostic accuracy 80% (CI= 79.4-80.1) with overall sensitivity= 49% (CI= 45-53), and Specificity = 92% (CI= 90-94). GP’s accuracy was higher with assisted diagnoses (86.7%, CI=85%-89%) than unassisted diagnoses (45.5%; CI= 43.7%-47.3%). Specificity was lower in assisted (89.15%; CI= 87.9-90.4) than unassisted diagnosis (92.5%; CI= 91.9-93.1). Conclusions: Low diagnostic sensitivity might hinder the adequate detection and management of anxiety in primary care. Results suggest that detection might be improved by using diagnostic tools. Results for all included articles will be presented. Health Care Expenditure Studies HC1 Determinants of Increasing the Likelihood for a Positive Drug Reimbursement Recommendation in Scotland Pantiri K , Baeten S , Majer I M , Heeg B , Charokopou M Pharmerit International, Rotterdam, The Netherlands . . . . . . Objectives: A binary reimbursement prediction model was previously developed based on a dataset of submissions to the Scottish Medicines Consortium (SMC) between 2006 and 2014. The objective of this study is to build on the previous model by identifying factors that influence the different levels of SMC recommendation, defined as “recommend”, “restrict” or “not recommend” pharmaceutical technologies for use in Scotland. Methods: Univariate and multivariate ordered logistic regression analyses were performed to assess the impact by means of odds ratios (OR) of the submitted evidence to the SMC on the decision. The proportional odds assumption underlying the current approach was tested. Results: Out of 463 applications, 115 received positive recommendation (25%), 150 received restricted recommendation (32%) and 198 (43%) were not recommended. Univariate analyses showed that 14 variables significantly affected the SMC decision. The multivariate analyses showed significant associations (p≤ 0.05) between the SMC decision and several variables, including: (1) a product demonstrating cost savings and QALY gains [OR= 6.11], (2) a product not being cost-effective (ICER≥ £30,000/QALY) [OR= 0.50], (3) a non-superior efficacy outcome versus placebo [OR= 0.15], (4) the product’s therapeutic indication (nervous system [OR= 0.51], blood forming organs [OR= 2.29]), (5) whether the product was indicated for non-chronic use [OR= 1.48] and (6) whether the submission was performed by a big company [OR= 1.86]. The proportional odds assumption was not violated, proving the appropriateness of the current model. The present model yielded similar results with the previously developed binary logistic one, further ensuring face validity, yet this approach is considered to better fit the multidimensional nature of SMC’s decision and increase the predictive power of the model. Conclusions: This study identified superior efficacy using an active comparator as well as a beneficial cost-effectiveness outcome to increase the likelihood of receiving a positive recommendation by the SMC. HC2 Biosimilars Versus Brands for Rheumatoid Arthritis: Eu5 Payers and Prescribers Place their Bets Cox J 1, Fletcher-Louis M 1, Kang B 2 1Decision Resources Group, London, UK, 2Decision Resources Group, Burlington, MA, USA . . . Objectives: The entrenched positioning of biologics to treat moderate-to-severe rheumatoid arthritis (RA) has generated a lucrative market. However, amid ongoing economic constraints, the EU5 (France, Germany, Italy, Spain, and the UK) must tighten their health care belts. As biosimilars versions of key brands appear on the horizon, this study explored the expected impact of these cheaper options on reimbursement and prescribing for RA in each country. Methods: Across the EU5, 254 rheumatologists were surveyed regarding their views on biosimilars for RA and on current and expected prescribing patterns. In addition, 15 payers who influence reimbursement at national or regional level were interviewed. Results: Considering 54-week Phase III data, ≥80% of surveyed rheumatologists in each country believe CT-P13 (biosimilar infliximab) has similar efficacy to branded Remicade; however, respondents are less confident in the biosimilar’s safety. Furthermore, > 80% of respondents in most countries are willing to prescribe biosimilars of infliximab, and of etanercept and rituximab, though largely not before branded biologics. Unsurprisingly, given likely price discounts, interviewed payers will somewhat encourage biosimilar uptake. However, excluding those in Germany, consensus is that discounts offered on biosimilars will not significantly impact their budgets. German payers, however, report that additional rebates to statutory insurers are expected; they admit to financial incentives for physicians to prescribe rebated drugs, thus manufacturers may consider robust uptake will compensate for hefty discounts. Conclusions: Available data have inspired prescriber confidence in biosimilar efficacy, although safety concerns, likely stemming from complex biosimilar manufacturing and lack of long-term safety data, will ensure continued brand uptake, at least initially. Furthermore, the expected modesty of biosimilar discounts in most countries will somewhat curb payer policy promoting use of such agents. However, as prescribers become more familiar with biosimilars, and, as the full extent of cost savings are revealed, increasing uptake of biosimilars is probable. HC3 Preferences for Prioritizing Patients with Rare Diseases: a Survey of the General Population in Sweden Wiss J , Levin L Å Linköping University, Linköping, Sweden . . . Objectives: Incentives are offered to pharmaceutical companies in order to increase the number of treatments for patients with rare diseases. As a consequence, a number of new drugs have been introduced on the market—drugs that often fail to meet traditional cost-effectiveness criteria. This study aims to investigate if there are societal preferences for treating patients with rare diseases differently in priority setting situations compared with common diseases. Moreover, psychological mechanisms that potentially could explain such preferences are explored. Methods: A postal questionnaire in three versions was sent out to a representative sample of the general Swedish population. Respondents were asked to choose to give treatment to a patient with a rare or a common disease in eight different scenarios. Rarity of the disease, different alternative costs, and group/ individual level decisions was investigated. Psychological aspects in the presented scenaros that varied between subjects was related to proportion dominance, the identifiability of the patient, pseudo-inefficacy and if the scenario was expressed in priority or rationing terms. Results: Response rate was 41 % (n= 1239). For equal cost scenarios, 42.3 % were indifferent between the rare and the common group, 23.9 % chose to prioritize the rare disease and 33.4 % the common disease. When questions were framed to be on an individul as opposed to a group level repondents were significantly (p< .001) more likely to be indifferent. Proportion dominance increased individuals’ preferences to prioritize rare diseases (p<.001). Identifiability and pseudo-inefficacy had no major effect on respondents’ choices. Conclusions: All else equal we see no strong support that a societal preference for rarity exists. However, we observe psychological effects influencing the judgments individuals make when setting priorities related to rare diseases. Whether or not these should A326 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 be viewed as biases or an expression of true preferences is a matter for further discussion. HC4 Understanding the Payer Dilemma with Biosimilar Mabs: Striking the Right Balance Between Budget Needs and Patient Outcomes Vidal Pinheiro A , Ziai Buetas A , Storer M ICON, London, UK . . . Objectives: The first infliximab biosimilars reached the EU in September 2013, representing the first biosimilar monoclonal antibodies (mAbs) to obtain EMA approval. Although commercialization in the major European markets will only start in February 2015, payers in Nordic and Eastern European countries have already faced the dilemma of striking the balance between potential savings accrued from use of less expensive infliximab biosimilars and demands for robust proof of clinical efficacy and safety. This work identifies payers’ evidence expectations, their reliance on regulators’ decisions and how potential savings can influence access and recommendations to target patient populations. Methods: Exploratory qualitative primary research with payers (N= 12) from France, Italy, Spain, UK, Germany and Netherlands. Collection of data about the current and future attitudes towards biosimilar health technology assessments at the national and, if applicable, local levels will be conducted, as well as perceived price and access trade-offs. Results: (1) Payers will mainly defer to the EMA the decision on acceptability of biosimilar indication extrapolation (indications where biosimilars do not have direct clinical trial data); (2) It is understood that mAb biosimilar clinical development is more onerous and costly than small molecule generics, thus payers do not expect the same magnitude of discounts offered vs. originator; (3) Although eager to obtain savings from broad patient populations, payers will not implement pharmacy-level substitution or enforce biosimilar use in originator-experienced patients; (4) Use in naïve patients will be recommended in most markets. Conclusions: Across the EU5, payers acknowledge physicians’ concerns over long term safety and efficacy of biosimilars. Nonetheless, they will rely on the regulators evaluations and expert panels to justify implementing recommendations, and in some markets, restrict formularies based exclusively on cost. Moreover, they have conservative discount expectations at launch, with the long-term aim of incentivizing further competition from other biosimilar manufacturers. Research On Methods – Modeling Studies MO1 Quasi-Monte Carlo Simulation and Variance Reduction Techniques Substantially Reduce Computational Requirements of Patient-Level Simulation Models: an Application to a Discrete Event Simulation Model Treur M 1, Postma M 2 International, Rotterdam, The Netherlands, 2University of Groningen, Groningen, The Netherlands . . 1Pharmerit Objectives: Patient-level simulation models provide increased flexibility to overcome the limitations of cohort-based approaches in health-economic analysis. However, computational requirements of reaching convergence is a notorious barrier. The objective was to assess the impact of using quasi-monte carlo simulation (QMCS) and variance reduction techniques (VRTs) on computational requirements. Methods: A recently published discrete event simulation model assessing the cost-effectiveness of an adjunctive antipsychotic treatment for depression was used. The following VRTs were implemented: antithetic variables, common random numbers (CRN) and the combination (Anti_CRN). In addition, QMCS was conducted using the Sobol low discrepancy sequence. The minimal number of patients required to reach equal precision as the reference situation of 1,000,000 simple monte carlo simulations (MCS) was recorded. Precision was defined by the standard error (SE) of the incremental net monetary benefit (INMB) at a willingness to pay of € 20,000 per quality adjusted life year gained. VRT simulations were replicated 100 times. INMB estimates were compared with the reference situation using mean squared error (MSE), mean absolute error (MAE) and percentage of under- and overestimations. Results: Reference INMB (SE) was € 1,413 (76). The average number of patients required to reach reference precision were 929,628, 35,692, 41,683 and 36,803 for antithetic variables, CRN, Anti_CRN and Sobol respectively. This implied a computation time reduction ranging between 7% and 96% compared to simple MCS. MSE was 346,036, 16,314, 155,950 and 7,475 respectively. MAE was 588, 105, 387 and 86 respectively. Antithetic variables and Anti_CRN structurally underestimated INMB (99% and 100%). CRN marginally overestimated INMB in 76 replications. Conclusions: QMCS and VRT reduce computational requirements in terms of simulated patients and computational time up to 96%, enhancing the practical feasibility of patient-level simulation models. This particularly applies to Sobol and CRN. Antithetic variables should be used with caution and its structural bias warrants further research. MO2 Transition Probability Estimation Using Repeated Sampling from a Fitted Mixed Model Gupta S 1, Bhattacharyya S 1, Sonathi V 1, Bakuli A 1, Mathur A K 1, Leteneux C 2 1Novartis Healthcare Pvt. Ltd., Hyderabad, India, 2Novartis Pharma AG, Basel, Switzerland . . . . . . . Objectives: Markov model is one of the most used decision analytic models in health care. Transitions between health states in a Markov model is driven by transition probability matrix. When the number of patients and observed transitions are limited, transition probability estimation becomes challenging. The objective of this exercise is to demonstrate how transition probabilities can be estimated by simulating data from a statistical model fitted to patient-level data. Methods: An economic model for ranibizumab in mCNV secondary to pathological myopia (submitted to NICE in June 2013) was adapted for forthcoming Asian reimbursement submissions. BCVA (Best Corrected Visual Acuity) scores were available for limited number of East Asian patients (N= 35) from a phase III, 12-month, randomized, double-masked, multicenter, active-controlled study (RADIANCE). To populate a transition probability matrix with 8 health states based on BCVA scores, a statistical model was proposed to simulate a larger hypothetical patient cohort. A mixedeffect model was fitted on the observed BCVA scores with baseline BCVA score as covariate, patients as random effect and an autoregressive AR(1) error correlation structure amongst the repeated observations. This model was used to simulate a patient cohort of 35,000. Transition probabilities were estimated using traditional division by row sum method. Several simulations were run to confirm consistency of results. Results: From baseline to month 3, percentage of patients with BCVA ≥ 20 letters gain was 22.45% in observed data vs 22.49% in simulated data, and percentage of patients with BCVA ≥ 20 letters loss was 0.008% in observed data vs 0.009% in simulated data. BCVA change from baseline to month 3 in simulated data (mean= 13.3, SD= 8.3) was verified with that of the observed data (mean= 13.3, SD= 8.8). Conclusions: Transition probability estimation by simulation from a fitted statistical model can overcome the challenges posed by small patient cohorts and multiple state transitions. MO3 Extrapolation of Trial-Based Survival Curves Using External Information Guyot P 1, Welton N J 2, Beasley M 3, Ades A E 2 1Mapi, Houten, The Netherlands, 2University of Bristol, Bristol, UK, 3Bristol Haematology and Oncology Centre, Bristol, UK . . . . . . Objectives: In cost-effectiveness analysis (CEA), mean survival difference (QALYadjusted) over a lifetime horizon is required. Parametric models are necessary to extrapolate survival outcomes beyond the Randomized Controlled Trial (RCT) period. However, mean survival is very sensitive to the assumed model and different mean survival times may result from models fitting similarly well to the RCT data. We investigate the idea that other sources of information, external to the trial data, could be used to inform model choice and estimation. Methods: We explored various survival models and we show how external information can be used to put constraints on spline-based survival models. We illustrate with a Technology Appraisal (TA) of head and neck cancer where RCT evidence had 5 year follow up. A US cancer database (SEER), general population data and expert opinion were used to impose constraints on overall survival, conditional survival, and hazard ratio. RCT and external data were fitted simultaneously within a Bayesian framework. Results: Standard survival time distributions were insufficiently flexible to simultaneously fit both the RCT data and general population constraints. Spline models were sufficiently flexible, although there were difficulties choosing initial values. A good fit to all sources of internal and external evidence was achieved within one integrated model using splines on the log hazard. Cetuximab in addition to radiotherapy improves the expected survival by 4.7 months [95% CrL: 0.4; 9.1] compared to radiotherapy alone. Conclusions: The method enabled us to estimate models consistent with all evidence. Clinical knowledge is essential to guide the interpretation of the external data sources. The method could be used to analyze other RCTs on other cancers and with other treatments. Other flexible models than splines could be investigated. MO4 Estimating Survival Data from Published Kaplan-Meier Curves: a Comparison of Methods Perry R 1, Taylor M 2, Lewis L 2, Yellowlees A 1, Fleetwood K 1, Barata T 1 1Quantics Consulting Ltd, Edinburgh, UK, 2York Health Economics Consortium, York, UK . . . . . . Objectives: Health technology assessment of treatments often requires estimates of their survival curves. Individual patient data (IPD) are often unavailable and the survival curves are usually calculated by fitting a nonlinear least squares (NLS) model directly to Kaplan Meier plots provided in the published literature. This method does not account for the uncertainty associated with the Kaplan Meier curve and can lead to biased estimates. Although the IPD are often missing, the Kaplan Meier curve itself can be digitised and used to approximate what the original IPD could have been. Methods: We simulated trial IPD data from different survival distributions in order to assess the accuracy of the IPD reconstruction methods. The assessment of accuracy is made at multiple stages and ultimately the effects on the incremental cost effectiveness ratio (ICER) estimates are compared. To do so, a simple cost-effectiveness model was developed, assuming two health states (alive and dead), and assigning costs (£1,000 per month plus drug costs) and a utility score (0.70) to generate ICERs. Two additional methods to curve fitting are compared against the NLS approach – those suggested by Guyot (G), and by Hoyle & Henley (HH). Results: We find that the methods differ in accuracy at each of the following two stages; (a) model selection via the AIC and secondly (b) survival model parameter estimation. When an underlying Weibull function was assumed, the ‘true’ ICER should be £28,924, compared against £31,182 £33,449 and £31,650 for the NLS, HH and G methods respectively. When an underlying loglogistic function was assumed, the NLS, HH and GG methods produced ICERs of £26,507, £25,559 and £25,857, compared to a ‘true’ ICER of £25,779. Conclusions: These findings suggest that inherent biases may be apparent in each of the approaches, and these may manifest themselves differently, depending upon the ‘true’ shape of the underlying data. Qaly-Related Studies QA1 Economic orphans? the prevalence of child-specific utilities in nice appraisals for paediatric indications Montgomery S , Hassan M , Kusel J Costello Medical Consulting Ltd., Cambridge, UK . . . A327 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Children have been termed “therapeutic orphans” due to the paucity of age-specific therapeutic data. Here we review the extent to which utility data derived from under-18s were used to inform National Institute for Health and Care Excellence (NICE) Technology Appraisals (TAs) providing cost-effectiveness guidance in paediatric indications, in line with the NICE reference case. Methods: All 311 published TAs up to April 2014 were initially sifted to identify therapeutic recommendations for children. Identified TAs were reviewed to determine if a cost-utility analysis (CUA) was performed. For each CUA, the published TA along with the manufacturer’s submission (single TAs) or the assessment report (multiple TAs) were examined to determine the origin of the utilities used. Results: Of 35 published TAs reviewed, 27 analysed cost-per-QALY and made recommendations for treatment of under-18s. Of these, 17 used adult utilities, 1 of which attempted to adjust the adult values for children; 3 considered child and adult populations as one, with child-derived data used within the overall model inputs for the whole population, 1 of which adjusted both child and adult utilities by age. Only 6 studies used child-specific utilities: 1 assumed a specified change from treatment on a generic QoL instrument, 2 used parent-reported utilities on a generic QoL instrument, 1 used parent-reported utilities mapped from a disease-specific scale and 2 used child-reported utilities mapped from a disease-specific scale. One MTA contained diverging submissions, 1 adult-derived and 1 child-reported. No trends over time in the types of utilities used were apparent from visual examination of the results. Conclusions: Despite NICE’s reference case specifying that utilities should be measured in the population in question, children may also be termed “economic orphans” with the majority of cost-utility submissions applying adultderived utilities to paediatric indications and no trend away from this apparent over time. QA2 Cost-Utility of Cancer Therapies – the ‘Cost’ of Different Utility Generation Strategies Meads D M 1, McCabe C 2, Hulme C T 1, Edlin R 3, Kharroubi S A 4, Browne C 5, Ford H 6, Dunn J 7, Marshall A 7 1University of Leeds, Leeds, UK, 2University of Alberta, Edmonton, AB, Canada, 3University of Auckland, Auckland, New Zealand, 4University of York, York, UK, 5Evidera, London, UK, 6Addenbrooke’s Hospital, Cambridge, UK, 7University of Warwick, Coventry, UK . . . . . . . . . . . . Objectives: To explore the impact of different utility measurement strategies on the results of a cost-effectiveness analysis, funding decisions, decision uncertainty and value of information. Methods: Data from a UK trial of two cancer therapies (active versus standard care) were analysed using NICE reference case methods. Within-trial, cost-utility analyses were conducted with utility based on a number of strategies: A) Observed EQ-5D; cancer-specific utility based on the EORTC QLQ-C30 B) the EORTC-8D and C) the QLQ_U; Mapping from QLQ-C30 to EQ-5D using an algorithm generated in D) the same cancer patient group and E) a different cancer group. Incremental cost-effectiveness ratios (ICERS) were calculated. Bootstrapped net benefit estimates allowed generation of cost-effectiveness acceptability curves (CEACs) and population expected value of perfect information (EVPI) was calculated using incremental cost scenarios. Results were compared across utility strategies. Results: There were small but important differences observed in the incremental QALYs which ranged from 0.067 (EQ-5D) to 0.036 (EORTC-8D). Large differences were observed in the ICERs generated; for strategies A to E these were: £57,513; £106,264; £102,785; £90,049; £78,885. Using an incremental cost scenario of £3,000 only strategy A yielded an ICER <£30,000. At a QALY willingness to pay threshold (WTPT) of £20,000 there was little decision uncertainty. However, assuming WTPT=£50,000, the probability the active treatment was cost-effective ranged 0.34 (EQ-5D) to 0.025 (EORTC-8D). Using this threshold, the population EVPI for the strategies were: £3,597,844; £120,621; £155,858; £354,094; £805,847. Conclusions: Different utility sources can lead to very different estimates of cost-effectiveness and value of further research and change funding decisions. Estimates of cost-effectiveness based on mapping (even when the algorithm appears to perform well) can differ substantively from those based on observed scores. The lowest ICERs were obtained with the EQ-5D but this may not capture side-effects picked up by the cancer-specific utility measures. QA3 Do new cancer drugs offer good value for money? the perspective of oncologists, payers, patients, and general population Dilla T 1, Lizán L 2, Paz S 2, Garrido P 3, Avendaño C 4, Cruz J J 5, Espinosa J 6, Sacristan J A 1 1Lilly S.A., Madrid, Spain, 2Outcomes 10, Castellon, Spain, 3Hospital Ramon y Cajal, Madrid, Spain, 4Hospital Puerta de Hierro, Madrid, Spain, 5Hospital Clinico Universitario, Salamanca, Spain, 6Hospital La Paz, Madrid, Spain . . . . . . . . . . Objectives: To analyze oncologists’, payers’, patients’, and general population´ views on the cost and value of new cancer treatments. Methods: An electronic self-administered questionnaire was developed and randomly distributed, to assess participants’ attitudes towards new cancer treatment outcomes and costs during reimbursement decisions. Among the questions asked were two hypothetical scenarios. First, participants were asked to indicate the minimum survival benefit that a new treatment, that cost € 50,000 more than the standard therapy, should have to be funded by the Spanish National Health System (NHS). Second, participants were requested to state the highest costs to be afforded by the NHS for a medication increasing patient’s quality of life (QoL) twofold with no changes in survival. Responses were used to calculate incremental cost-effectiveness ratios (ICER). Results: 53 oncologists, 60 patients, 25 payers, and 50 individuals from general population answered the questionnaire. The minimum improvement median in patient survival that justified the inclusion into the NHS was 5.66 months for oncologists, 8.16 for patients, 9.08 for general population and 10.44 for payers; implying different ICER for oncologists (€ 106,000/QALY), patients (€ 73,520/QALY), general population (€ 66,074/QALY) and payers (€ 57,471/QALY). The cost stated in QoL-enhancing scenario was € 33,167 for patients, € 30,200 for general population, € 26,000 for oncologists and € 17,040 for payers; resulting in ICERs of € 82,917/QALY for patients, €75,500/QALY for general population, €65,000/QALY for oncologists, and €42,600/QALY for payers. Conclusions: All the estimated ICER values were higher than the thresholds usually described in the literature (€ 20,000-30,000/QALY), with relevant differences among the groups. In both scenarios, payers were less prone to pay for therapeutic improvements compared to the rest of the participants. On the other hand, oncologists were the ones that most valued gains in survival for a new treatment while patients assigned a higher value for money to a treatment that enhanced the quality of life. QA4 Reimbursement Decisions for Pharmaceuticals in Sweden: the Impact of Cost-Effectiveness and Disease Severity Nilsson F O L 1, Svensson M 2, Arnberg K 1 1Dental and Pharmaceutical Benefits Agency, Stockholm, Sweden, 2Örebro University, Örebro, Sweden . . . . . Objectives: The purpose of this study is to evaluate the impact of cost-effectiveness and disease severity on the drug reimbursement decisions made by the reimbursement agency TLV in Sweden. Methods: Cost-effectiveness is measured through the continuous variable cost per QALY, while disease severity is measured by a dichotomous variable indicating high- or not high disease severity. We analyze all reimbursement decisions from 2005 through 2011 where there is data available on cost per QALY and disease severity. Logistic regressions are used to evaluate the impact of cost-effectiveness and disease severity on the drug reimbursement decisions. Results: There are 102 decisions with the required data available, 86 where reimbursement was granted and 16 where reimbursement was denied. The median cost per QALY for the drugs that were granted reimbursement was 39 000 euro (9sek/euro), ranging from a negative cost per QALY (better and cheaper) to 136 000 euro. The median cost per QALY for the drugs that were denied reimbursement was 111 000 euro, ranging from 78 000 euro to 1 111 000 euro. The results from the logistic regression analysis show that both the cost per QALY and the level of disease severity are statistically significantly related to the probability of a drug being granted reimbursement. When the cost per QALY exceeds 56 000 euro for non-severe diseases, and 92 000 euro for severe diseases, the probability that reimbursement is denied is higher than the probability that reimbursement is granted. Conclusions: In Sweden, it is sometimes stated as a rule of thumb that 55 000 euro per QALY is a threshold for cost-effective interventions. Our model shows that at this cost-effectiveness ratio, the probability of a new drug becoming reimbursed is 91 % or 98 %, depending on disease severity. Research Podium Presentations – Session II Cardiovascular Disease Research Studies CV1 The Importance of Treatment Classifications that Account for Concomitant Treatments in the Context of a Network MetaAnalysis Comparing Pharmacological Treatments for Chronic Heart Failure Burnett H 1, Cope S 1, Vieira M C 2, Sagkriotis A 3, Senni M 4, Deschaseaux C 3 Pharma, Health Economics and Outcomes Research, USA, East Hanover, NJ, USA, 3Novartis Pharma AG, Basel, Switzerland, 4Scompenso e Trapianti di Cuore, Bergamo, Italy . . . . . . . 1Mapi, Toronto, ON, Canada, 2Novartis Objectives: The aim of the study was to assess the comparative efficacy of recommended treatment for chronic heart failure with reduced ejection fraction in terms of all-cause mortality based on a network met-analysis (NMA) of randomized controlled trials (RCTs) and to explore the impact of alternative treatment classifications depending on concomitant treatments. Methods: A systematic literature search identified 56 relevant RCTs (1980-2013) that reported mortality data that were synthesized using a Bayesian Poisson regression NMA model. Treatments were classified as angiotensin converting enzyme inhibitors (ACEI), beta-blockers (BB), angiotensin II receptor blockers (ARB), mineralocorticoid/aldosterone receptor antagonists (MRA) and the Ifchannel inhibitor (IF) ivabradine. Analysis 1 classified treatments according to the main drugs of interest, whereas Analysis 2 defined treatments according to the main drugs of interest as well as the concomitant treatments belonging to classes of interest if more than 50% of patients were taking concomitant drugs. Results: Six regimens were compared in Analysis 1 and 10 regimens were compared in Analysis 2. Analysis 1 resulted in the following rate ratios (RR) versus placebo: ACEI: 0.81 (95% Credible Interval 0.61, 0.95); BB: 0.71 (0.60, 0.80); ARB: 0.90 (0.75, 1.02); ACEI+BB: 0.48 (0.30, 0.76). Analysis 2 resulted in the following RRs versus placebo: ACEI: 0.81 (0.68, 0.95); BB: 0.57 (0.35, 0.87); ARB: 0.81 (0.61, 1.01); ACEI+BB: 0.61 (0.54, 0.68). The treatments that are expected to be most efficacious depended on the treatment classification: Analysis 1 supported ACEI+BB and BB, whereas Analysis 2 supported ACEI+BB+MRA+IF and ACEI+BB+MRA [RR: 0.44 (0.34, 0.58) and 0.48 (0.38, 0.60), respectively]. Conclusions: Combination treatments were likely to be more efficacious than monotherapy and adding a class to a regimen was likely to make it more efficacious regardless of the approach. However, treatment classifications affect the results and interpretation. The approach that accounts for concomitant treatments is preferred. CV2 Work Productivity Loss and Indirect Costs Associated with New Cardiovascular Events in High-Risk Patients with Hyperlipidemia Estimates from Population-Based Register Data in Sweden Banefelt J 1, Hallberg S 1, Fox K M 2, Mesterton J 1, Paoli C J 3, Johansson G 4, Levin L Å 5, Sobocki P 6, Gandra S R 3 1Quantify Research, Stockholm, Sweden, 2Strategic Healthcare Solutions, LLC, Monkton, MD, USA, 3Amgen, Inc., Thousand Oaks, CA, USA, 4Uppsala University, Uppsala, Sweden, 5Linköping University, Linköping, Sweden, 6IMS Health, Stockholm, Sweden . . . . . . . . . . . . . A328 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To estimate productivity loss and associated indirect costs in working-age high-risk patients treated for hyperlipidemia who experience new cardiovascular (CV) events. Methods: A retrospective population-based cohort study was conducted using Swedish electronic medical records linked to national health registers and the Social Insurance Register. Patients were included based on a prescription of lipid-lowering therapy between January 1, 2006 and December 31, 2011 and followed until December 31, 2012 for identification of CV events and estimation of work productivity loss (e.g. sick leave and disability pension) and indirect cost. Patients were stratified into two cohorts based on CV risk level. Propensity score matching was applied to compare patients with new events (cases) to patients without new events (controls). For all outcomes, the incremental effect estimate of a new CV event was the difference between cases and controls in the differences between the year before and the year after the cases’ new event. Results: The incremental effect estimate on mean indirect costs of sick leave was largest in the CV risk-equivalent (RE) cohort (n= 2,946) at 38,395 SEK. The corresponding figure in the CV event history cohort (n= 4,508) was 23,931 SEK. There was substantial variation in work productivity loss with regard to type of new CV event. Transient ischemic attack and percutaneous transluminal coronary angioplasty had no significant effect on indirect cost. Other types of CV events yielded a substantial incremental cost estimate, such as myocardial infarction (38,002 SEK), unstable angina (27,189 SEK) and most notably ischemic stroke at 61,500 SEK. New CV events did not have a significant impact on disability pension in either cohort. Conclusions: High indirect costs are related to work productivity losses associated with new CV events in high-risk patients treated for hyperlipidemia. The effect of new CV events on indirect costs varied by event type. CV3 Incidence Description and Costs of Acute Heart Failure in the Netherlands Stevanovic J 1, Denee L 2, Koenders J M 2, Postma M J 1 of Groningen, Groningen, The Netherlands, 2Novartis Pharma B.V., Arnhem, The Netherlands . . . . . . 1University Objectives: Acute heart failure (AHF) is frequent, severe and costly, however detailed population-based epidemiological data are currently unavailable for the Netherlands. Our aim was to characterize the incidence, clinical features and outcomes of AHF, and estimate associated hospitalization costs in the Netherlands. Methods: Using the 2010 Dutch Hospital Data (DHD), we identified all patients admitted to hospital with AHF as a primary diagnosis. DHD provide data on patient characteristics, primary diagnosis, date of admission and discharge, surgical procedures, prior location and discharge destination. We applied contemporary estimates of health care activity associated with AHF in order to calculate its cost in 2014. Major components of health care activity included in this estimate were hospital admissions associated with a primary diagnosis of AHF, associated drug utilization generally observed in those patients, major surgeries conducted during the hospital stay and autopsy associated with inhospital mortality. Results: Primary analysis of the data identified 7,717 patients to be admitted at least once into Dutch hospitals in 2010 due to primary diagnosis of AHF. The mean age of patients was 77.1 (±11.5) and 51% were women. The most common comorbid conditions were cardiac dysrhythmias, essential hypertension, old myocardial infarction, other diseases of endocardium and diabetes. The mean hospital length of stay was 8.67 days during the first admission. Inhospital mortality was 11.3% and readmission to hospital was observed in 13.8% of the patients. Finally, the cost of an AHF hospitalization in the Netherlands was estimated to be € 4,623. Conclusions: Our study provided important insights into the clinical characteristics and costs of AHF hospitalizations in the Netherlands. Further analysis including secondary diagnosis will indicate what the exact number of AHF hospitalizations is, and whether this resembles previously published figures from the National Institute for Public Health and the Environment of 29,838 patients hospitalized and diagnosed with congestive HF. CV4 Health Utility in Patients Following Cardiovascular Events Pockett R D 1, McEwan P 2, Beckham C3, Shutler S 4, Martin S 5, Yousef Z 6, Bakhai A 7 University, Cardiff, UK, 2Health Economics and Outcomes Research Ltd, Cardiff, UK, 3Roche Products Ltd, Welwyn Garden City, UK, 4F. Hoffmann-La Roche Ltd, Basel, Switzerland, 5Peterborough and Stamford Hospitals NHS Foundation Trust, Peterborough, UK, 6University Hospital of Wales, Cardiff, UK, 7Barnet and Chase Farm Hospitals NHS Trust, Barnet, UK . . . . . . Health Services Research Studies HS1 Follow-UP Autoantibody Testing and Health Care Utilization Among Patients with (Sustained) Celiac Disease Activity in the Netherlands Kuiper J G 1, Penning-van Beest F J A 1, Naessens D 2, Leon F 3, Herings R M C 1 1PHARMO Institute for Drug Outcomes Research, Utrecht, The Netherlands, 2Janssen BVBA, Beerse, Belgium, 3Janssen Research & Development, Spring House, PA, USA . . . . . . . . . . Objectives: To examine follow-up of autoantibody testing and health care utilization among patients with (sustained) celiac disease. Methods: From the PHARMO Database Network, patients with a positive autoantibody test for deamidated gliadin peptide (anti-DGP), endomysial (EMA) or tissue transglutaminase (anti-tTG) (period 1998-2011) were classified as celiac disease patients. The first positive test served as index date. For patients with ≥ 12 months follow-up, autoantibody tests in the year after index date were assessed. Patients with a second positive test, using the same cut-off, between 6-12 months after index date were classified as ‘sustained celiac disease’ patients and matched 1:1 on gender, birth year and year of index date to celiac disease patients without a second positive test. For these patients, health care utilization in 2012 was determined. Results: 1,815 patients had at least one positive autoantibody test, of which anti-tTG was most common (86%). 1,724 patients had ≥ 12 months follow-up, of which 75% did not have any type of autoantibody test in the year after index date. 183 patients (11%) were classified as ‘sustained celiac disease’ patients. Of these, 63 were active in the database in 2012 and could be matched. The proportion of patients with at least one GP visit and/ or hospitalization in 2012 was lower among sustained celiac disease patients: 76% and 14% vs. 83% and 24% for patients without sustained celiac disease. However, the mean (±SD) number of GP visits was higher among sustained celiac disease patients (4.3 ± 7.3 vs. 3.4 ± 3.5) as was the mean (±SD) number of outpatient drug dispensings (21.6 ± 96.3 vs.10.5 ± 17.5). Conclusions: This study shows limited follow-up autoantibody testing among patients with celiac disease and does not suggest higher health care utilization with sustained celiac disease. These findings are consistent with notion of patients managing the disease without engaging the health care system. HS2 The Impact of Different Levels of Clinical Pharmacist Interventions on the Therapeutic Plan and Cost Saving Alkhalaf M S Dammam Medical Complex, MOH, Dammam, Saudi Arabia . . Objectives: To assess the impact of different levels of clinical pharmacy intervention (CPI) on therapeutic plan and cost saving through monitoring and follow-up of some medications (including 20% Human Albumin, Meropenem and Cefepime) over three separated months. Methods: A prospective observational study. Includes three phases, first phase: NO-CPI. While the second phase: partial clinical pharmacy intervention (PCPI) in which the clinical pharmacist contacted physicians directly or indirectly by telephone and give suggestions without evidence based supported or follow up. And third phase: an effective clinical pharmacy intervention (ECPI) defined by direct contact between the clinical pharmacist and physician through daily round supported by evidence based and follow up for patients and physician’s orders. Results: The percentage of quantity saved of Human Albumin was 46.5% during PCPI and 76.7% in ECPI. Also, there was monthly cost saving by both PCPI and ECPI (4880 $ and 8052 $ respectively). This means that ECPI resulted in more increment in cost saving in comparison to PCPI (39.4%). For Meropenem, 47.7% of the dispensed quantity was saved by PCPI and 90% by ECPI. Likewise there was a significant cost saving per month that is 3052.4 $ for PCPI and 5799.5 $ for ECPI. Thereby 47.4% more cost saving resulted from ECPI. Regarding Cefepime, although both PCPI and ECPI reduced the dispensed quantity, interestingly PCPI saved more than ECPI (25% and 4.58% respectively). This could be due to the clinical pharmacist recommendation according to local antibiogram. Conclusions: CPI generally results in cost saving of therapeutic plan. ECPI can lead to more remarkable cost saving. . 1Swansea Objectives: Cardiovascular (CV) disease is a major contributor to morbidity and mortality in the UK. Health-related quality of life (HRQoL) data is an important requirement of the development process and is used to inform the health state utilities within economic models. Methods: EuroQol-5 dimension (EQ-5D) surveys were sent to patients (age ≥ 18 years) from three centres in the UK (Barnet, Cardiff, Peterborough) 1 month following hospital admission for a myocardial infarction (MI), unstable angina (UA) or stroke. Patient demographics, lifestyle and baseline utility score were collected in the first survey. Follow-up surveys were sent at 6, 12, 18 and 24 months capturing utility and subsequent health events. Descriptive statistics and general linear regression models were used to describe the patients and to identify changes in utility over time. Results: 1350 patients (mean age 68.8 years; SD 12.3) were recruited. Of these, 755 (55.9%) suffered a MI, 571 (42.3%) had UA, and 24 (1.8%) had a stroke; 345 (25.6%) patients also had diabetes. Baseline utilities were 0.690 (SD 0.322) in patients with a MI and 0.623 (SD 0.322) in patients with UA. Using regression, mean utility was 0.767 (MI) and 0.724 (UA) at 1 month, changing to 0.846 (MI) and 0.807 (UA) at 6 months, 0.877 (MI) and 0.845 (UA) at 12 months, 0.855 (MI) and 0.841 (UA) at 18 months, and 0.885 (MI) and 0.836 (UA) at 24 months. Diabetes was associated with a decrement of 0.106, 0.046, 0.074, 0.076 and 0.059 at 1, 6, 12, 18 and 24 months, respectively. Conclusions: In this prospective, robustly conducted study with good follow up, HRQoL associated with CV events appeared to improve in the 6 months post-event. However, over the next 18 months HRQoL plateaued with little to no improvement in this time period. Diabetic patients had lower scores at each timepoint. HS3 Impact of Morbidity in Populations of North London Clinical Commissioning Groups on Patient Admission Rates and Gp Referrals Groom Z C , Burgon J , Eddowes L A , Wilson T , Kusel J Costello Medical Consulting Ltd., Cambridge, UK . . . . . . . Objectives: To ascertain whether disease prevalence alone explains elective and non-elective patient admission rates, general practitioner (GP) referral rates and prescribing spends in 20 North London Clinical Commissioning Groups (CCGs). Methods: Using information provided by National Health Service (NHS) England CCG Information packs and the NHS CCG Outcomes Tool, age and sex standardised elective and non-elective admissions rates, GP referral rates and average prescribing spends were extracted and compared to national averages and to the prevalence of 19 commonly occurring diseases available through the Outcomes Tool (2011). Results: Of the 20 North London CCGs included in this analysis, 4 reported a higher non-elective admission rate (per 1,000 of the population) than the national average. For CCGs reporting a higher than average non-elective admission rate the disease prevalence was, on average, higher in only 6/19 diseases compared to those CCGs reporting lower than average non-elective rates. Four CCGs reported higher than national average elective admission rates (per 1,000 of the population); of these, the disease prevalence was, on average, higher in 13/19 diseases than CCGs reporting lower than average elective rates. Sixteen CCGs reported higher than national average GP referral rates, and in these CCGs 10/19 diseases had a higher prevalence compared to those CCGs reporting lower than average rates. No CCGs reported a higher than national average prescribing spend (£, per person, per 1,000). Conclusions: Disease prevalence appears to explain rates of elective admissions and GP referrals relatively well in North London CCGs compared to non-elective admission rates. Non-elective admissions commonly occur in emergency situations, thus making A329 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 the prediction of these events challenging for primary care management and NHS community provision. This could potentially result in increased hospital admissions, irrespective of morbidity in the population. Alternatively, the discrepancy may be influenced by patient proximity to accident and emergency (A&E). HS4 Information Used in the Decision-Making Process Regarding Influenza Vaccination Policy: Perceptions of Stakeholders in France and the Netherlands Silva M L 1, Perrier L 2, Paget J 3, Mosnier A 4, Buthion V 5, Cohen J M 4, Späth H M 6 Group in Health Economics (GATE, UMR 5824, CNRS); University Lyon 2, Ecully, France, 2Cancer Centre Léon Bérard, Lyon, France, 3Netherlands Institute For Health Services Research (NIVEL), Utrecht, The Netherlands, 4Regional Group for the Surveillance of Influenza - GROG, Open Rome, Paris, France, 5COACTIS EA 4161, University of Lyon, Lyon, France, 6University Claude Bernard, Lyon 1 EAM 4128, Lyon, France . . . . . . . . . . 1Research Objectives: To minimize the medical and societal impact of influenza, most WHO countries recommend seasonal vaccination in targeted populations; however, little is known about the decision-making procedures at a country-level. In Europe, the Netherlands has the highest rate of influenza vaccination and France is not far behind. Our purpose was to analyze differences and similarities in the information used in the decision-making process between these two countries, according to the stakeholders involved. Methods: A preliminary documentary analysis identified all stakeholders, at national level in both countries, as decision-makers (governmental authorities), advisors and information providers (research institutions, groups of experts), and vaccine manufacturers. We undertook a qualitative study including at least one actor from each stakeholder group involved in the process. Thirty-three faceto-face or telephone semi-structured interviews were conducted during summer 2013 in France (n=16), and autumn 2013 in the Netherlands (n=17). Every interview was recorded and transcribed. NVivo10® was used for the qualitative analysis. Results: Stakeholders in France and the Netherlands follow international recommendations. The most relevant information is clinical trials and epidemiological studies. Economic models gained importance after the 2009 influenza pandemic, especially in the Netherlands. In both countries, the advice of experts is crucial. All types of studies are assessed through a standard checklist for public health vaccinations in the Netherlands. In France, the assessment is not standardized, but based on general checklists. Decision-makers are increasingly worried about the quality of studies, due to the lack of standardized methods and influenza uncertainty. When published studies are not generalizable, local studies are required. Conclusions: Information used in the decision-making process is similar in both countries, although economic models have greater importance in the Netherlands. The excellence of the process is challenged by the poor quality of influenza data. Efforts should be made on standardization of study methods, together with harmonization of European policy. Medication Adherence Studies MA1 Cost-Effectiveness of Real-Time Medication Monitoring in Children with Asthma Goossens L M A 1, Vasbinder E C 2, Van den Bemt P M L A 3, Rutten-van Mölken M P M H 1 University, Rotterdam, The Netherlands, 2Groene Hart Hospital, Gouda, The Netherlands, 3Erasmus University Medical Center, Rotterdam, The Netherlands . . . . . . . . . . . . survey from a representative sample of adults from France, Germany, Italy, Spain and UK stratified by age and gender. Out of 62,000 respondents, 8,462 (11%) reported a diagnosis of depression and 3,937 (6%) having a prescription medication for depression (Rx). Respondents classified as adherent according to the Morisky Medication Adherence Scale (MMAS) were compared to the non-adherent on severity (PHQ-9), sociodemographics, health characteristics, health-related quality of life (SF-36), work productivity and activity impairment (WPAI) and health care resource use (physician, hospital and emergency visits). Results: Compared to adherent respondents (54%), the non-adherent (46%) were more severe (37% vs. 31% with PHQ-9 score ≥ 15); had lower Mental Component Summary (MCS: 32 vs. 33); higher Absenteeism (18 vs. 16); more emergency visits in the previous 6 months (0.49 vs. 0.42); and their satisfaction with medication was lower (4.97 vs. 5.27) (All p< 0.05). About 75% of both groups were participating in psychotherapy at the time of survey and showed no significant difference in Physical Component Summary (PCS); Presenteeism; and number of hospitalizations. Conclusions: While efficacy measured during clinical trials is one of the most influential measures in treatment assessment, its ecological validity may be jeopardized by non-adherence to medication in real life. The current study shows that low adherence is associated with more severe depression, lower treatment satisfaction and lower mental quality of life (MCS) in respondents taking antidepressants. These results point at the importance of combining adherence and efficacy in the assessment of treatments for depression. MA3 Assessing the Relationship Between Patient Compliance to Blood Glucose Monitoring and Health Related Quality of Life Dierick K 1, Mcbride M 2, Pike I 3 Disease Atlas, Brussels, Belgium, 2GfK USA, New York, NY, USA, 3GfK NOP, London, UK . . . 1GfK Objectives: The objective of our research was to evaluate whether there is a relationship between patient compliance to blood glucose monitoring (BGM) and HRQoL. Moreover we wanted to understand what drives patients not to be compliant to BGM. Methods: Data were taken from the ROPER Diabetes program which captures information direct from diabetes patients across 27 countries on a regular basis. For this specific study data from a sample of 1480 diabetes patients living in the USA were collected from June to August 2013. Each patient completed a questionnaire comprising some 2000+ variables, which included the EQ-5D-5L instrument and accompanying VAS. Patients were asked to state their recommended frequency of BGM and what their actual BGM frequency was during the past month. Moreover patients were asked to explain why they were not compliant to the recommended BGM frequency. Results: Not being compliant explained 52% (R² = 0,52) of the variations in HRQoL. The main reasons for not being compliant to the recommended BGM frequency were: no coverage of strips by the insurance, pain and discomfort related to blood testing, not willing to know the test result. Other less important drivers of non-compliance were: inconvenience, issues with food intake and meter malfunction. Conclusions: Meter manufacturers are right when they reckon that BGM is crucial to diabetes patients’ disease management. Manufacturers have been innovating to make the blood testing as convenient as possible. Yet an important driver for not complying to the recommended BGM frequency remains to be pain and discomfort during blood testing. Manufacturers should also continue their efforts to ensure coverage of strips by the different insurance providers in the USA. . 1Erasmus Objectives: Poor asthma control in children is partly caused by poor adherence to medication. The aim of this study was to investigate the effectiveness and costeffectiveness of a Real-Time Medication Monitoring system (RTMM) to improve adherence to inhalation corticosteroids. Methods: We performed a multi-center, randomized controlled trial. Included were 209 children (< 12 years) with moderate to severe asthma, who had used inhaled corticosteroids (ICS) for at least 3 months. Patients were followed for 12 months. All children received an RTMM device, but only in the intervention group text messages were sent to the parents whose child appeared to forget an inhalation. The effectiveness measures were adherence (percentage of inhalations taken within the correct timeframe), clinically relevant improvement in asthma control score (ACT, ≥ 3) and asthma-related quality of life (PAQLQ, ≥ 0.5) at the end of the study. Costs were calculated from a health care perspective (including GP, hospital, medication and RTMM device costs) and from a societal perspective (additionally including costs of parents’ absence from work). Uncertainty around the point estimates was assessed using bootstrapping. Results: Adherence was 73% in the treatment group and 58% in the control group (difference 15%-pt, 95%-CIL 8.6%-22.0%). Of the RTMM patients, 33% showed clinically relevant improvement in ACT and 22% in PAQLQ. In the control group, these numbers were 37% and 38% respectively (differences not statistically significant). Costs were higher in the intervention group: € 825 versus € 713, a difference of € 112 (95%-CI:-€ 139- € 338) from the health care perspective and € 1084 versus € 845 from a societal perspective (difference 239, 95%-CI: -€ 84 – € 565).The incremental costs per 10% improvement from each perspective were € 74 and € 157 respectively. Conclusions: RTMM increases inhalation adherence, but there is no evidence of better health outcomes in this patient population within the first year. In these circumstances, this is not a cost-effective intervention. MA2 The Burden Associated with Non-Adherence in European Patients with Depression MA4 Adherence to Antiretroviral Therapy (Art) Among Adult Hiv Positive Patients in Volta Regional Hospital, Ghana Okotah A N 1, Korbuvi J 2 1Volta Regional Health Directorate, Ghana Health Service, Ho, Ghana, 2Volta Regional Hospital, Ho, Ghana . . . Objectives: Adherence to antiretroviral therapy (ART) is a critical element towards reducing the emergence and spread of drug resistant strains of the virus. To achieve a sustained virological suppression, at least a 95% optimal adherence is necessary. This study sought to explore the level of adherence and also identify the factors contributing to non adherence to therapy among people living with the HIV/ AIDS. Methods: The study was a descriptive cross sectional type. A systematic sampling method was used to recruit 146 adult ARV users who have been on therapy for at least 3 months and attended the ART clinic between March to May, 2014. Using a structured and pretested questionnaire, data on medication adherence were collected by adopting a one month visual analogue scale (VAS) recall, a 4 days self reported adherence, and a pill identification test technique (PIT). A multivariate logistic regression was then used to determine key factors that were associated with adherence. Results: Of the three methods used, the optimal adherence ( ≥ 95%) for the pill identification test (PIT) was 76%, followed by the visual analogue scale (71.2%). The 4 days self report recorded the least adherence rate (65.1%). The overall rate of high optimal adherence was found to be 51.4%. Respondents aged 46years or more were highly adherent (61%) than their counterparts who are less than 25years (60.0%). Those between 25 to 45years of age were the least adherent (45.0%). However, the association between the level of adherence and the sociodemographic variables (sex, age, employment, and marital status) were not statistically significant. Conclusions: The overall optimal adherence was found to be relatively low and fell below the minimum expected adherence of ≥ 95%. Adherence to ART should aim at improving the pre treatment counselling and ensure the availability of ARV’S at all times. Pedersini R , Kuehl M Kantar Health, Epsom, UK Studies on Nice Assessments Objectives: Adherence to medication is regarded as an important factor for predicting clinical outcomes in mental disorders such as depression, bipolar disorder or schizophrenia. The current study investigates the relation between adherence and the burden of depression on society and individuals. Methods: Data were from the 2013 EU National Health and Wellness Survey (NHWS), an internet-based NI1 Exploring the Flaws in Clinical Data that Lead to Rejection of Nice Submissions . . Hendrich J , Griffiths E A PAREXEL, London, UK . . . A330 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: New health technologies are required to demonstrate clinical- and cost-effectiveness before being recommended by NICE for use in the National Health Service. A large proportion of submissions are rejected, at least in part, due to poor trial design or flaws in the presented clinical data. Published NICE guidance includes a comprehensive critique of submitted clinical evidence. Therefore, the flaws in clinical data leading to rejection by NICE were examined, as a means of providing guidance for future submissions. Methods: All single technology appraisals from January 2006 to May 2014 from NICE were included in the analysis. Multiple technology appraisals, resubmissions, vaccination programmes, and requests for advice were excluded. The recommendation and reasoning behind each decision were assessed, focusing on the critique of the clinical evidence. Results: 121 NICE submissions met the inclusion criteria, of which 28 (19.8%) were rejected. Notable flaws in the presented clinical data were reported in 22 (78.6%) of the submissions. Major drivers of rejection due to clinical flaws included: uncertainty in the clinical evidence (32.1%), statistical flaws in trial design (28.6%), flaws in the choice of comparator (17.9%), and a perceived lack of transparency in trial design (14.3%). These factors contributed to a perceived failure to demonstrate clinical superiority over the comparator in 42.9% of the rejected submissions. Conclusions: A failure to convincingly demonstrate clinical superiority is a major driver in the rejection of submissions to NICE, often due to clinical trial design weaknesses or uncertainty surrounding the data presented. Manufacturers should not underestimate the need for effective planning and review of trial design early in the clinical development process, in order to avoid rejection by HTA agencies in later development stages. NI2 Predicting the Impact of Value-Based Assessment on Future Nice Appraisals Beale R C , Maruszczak M , Kusel J Costello Medical Consulting Ltd., Cambridge, UK . . . . Objectives: Value-based assessment will be introduced into National Institute for Health and Care Excellence (NICE) appraisals where the willingness-to-pay above £20,000 will depend upon the wider societal impact (absolute shortfall of qualityadjusted life years [QALYs]) and burden of illness (proportional QALY shortfall). The objective of this research was to re-evaluate past appraisals and determine whether there was a relationship between the absolute and proportional QALY shortfalls and the NICE decision. Methods: Final incremental cost-effectiveness ratios (ICERs) were identified from NICE single technology appraisals published between January 2013 and May 2014. The age that treatment commenced was taken from manufacturer submissions and combined with the average life expectancy in the UK to calculate the discounted QALYs accrued by a healthy person (X). Discounted QALYs accrued by current treatment were extracted from manufacturer submissions (Y). Consequently, absolute (X-Y) and proportional (absolute/X) QALY shortfalls were calculated for each condition. Logistic regression (LR) was performed on the data. Appraisals were excluded if the manufacturer submission was missing. Results: Of the appraisals assessed, 23 appraisals met the selection criteria; 13 were recommended and 10 not recommended by NICE. The LR confirmed that the probability of an intervention being recommended was reduced significantly if the ICER was > £30,000 (p< 0.05). Taking into account the ICER, LR analysis demonstrated that neither the proportional nor absolute QALY shortfall had a significant effect on recommendation by NICE (p> 0.05). Conclusions: These results suggest that in recent appraisals, proportional and absolute QALY shortfalls have had no significant effect on the NICE committee’s decisions, and that the introduction of value-based assessment may therefore cause a substantial change in the future outcomes of HTA processes. It is recognised that further appraisals before 2013 should be evaluated to confirm these results, as the sample size was small. NI3 Nice’s Proposed Value-Based Assessment of Health Technologies: Concerns of Inconsistent Consideration of Social Values Paulden M 1, O’Mahony J F 2, Culyer A J 3, McCabe C 1 of Alberta, Edmonton, AB, Canada, 2Trinity College Dublin, Dublin, Ireland, 3University of Toronto, Toronto, ON, Canada . . . . . . 1University Background: The UK’s National Institute for Health and Care Excellence (NICE) recently proposed amendments to its health technology appraisal methods. Previous amendments in 2009 and 2011 placed a greater value on the health of patients at the “end of life” and in cases where “treatment effects are both substantial in restoring health and sustained over a very long period”. The recent proposals repeal “end of life” considerations but add consideration of the “proportional” and “absolute” QALY loss from illness. NICE’s cost-effectiveness threshold may increase from £20,000 to £50,000 per QALY based upon these and four other considerations: “certainty of the ICER”; if health related quality of life is “inadequately captured”; the “innovative nature” of the technology; and “non-health objectives of the NHS”. Analysis: We demonstrate that NICE’s previous amendments are flawed for two reasons: they contain logical inconsistencies which can result in different values being placed on health gains for identical patients, and they do not apply value weights to patients bearing the opportunity cost of NICE’s recommendations. The newly proposed amendments also suffer from a third flaw that the suggested weightings are poorly justified. The second of the three flaws is particularly significant, as the failure to apply special considerations to displaced services may result in NICE’s cost-effectiveness threshold being too high to adequately represent opportunity cost. Furthermore, the baseline threshold of £20,000 per QALY is greater than current estimates of the opportunity cost. Overall, these flaws likely lead to several undesirable consequences including age discrimination, a systematic bias in favour of new interventions over existing services, and the displacement of effective interventions by less effective interventions. Conclusion: NICE’s proposed threshold range is too high, for empirical and methodological reasons. NICE’s proposals will harm the health of unidentified patients, whilst privileging the identified beneficiaries of new health technologies. NI4 Do health technology agencies accept methods for dealing with treatment switching? Maervoet J 1, Skaltsa K 2, Ivanescu C 1, Van Engen A 1 1Quintiles Consulting, Hoofddorp, The Netherlands, 2Quintiles Consulting, Barcelona, Spain . . . . Objectives: Upon disease progression or early termination of a trial, cancer patients in the control arm of clinical studies commonly switch to the experimental drug or receive other post-study treatments. The objective of this study was to assess the willingness of HTA agencies to accept the statistical methods that are available to adjust overall survival (OS) estimates for resulting bias. Methods: PubMed and the ADIS R&D Insights database were searched to identify pivotal trials involving treatment switching. Related HTA appraisals, published between January 2011 and February 2014, were then selected from the HTA agencies websites for further analysis. Reports from 10 agencies were considered for review. Results: Sixteen pivotal trials and 45 related HTA appraisals were selected for in-depth analysis. It was widely acknowledged by all HTA agencies that treatment switching could confound the OS estimates when an inferior drug is used as a comparator. Our analysis suggests that statistical adjustment methods are discussed more often and more thoroughly by agencies that base their decisions on cost-effectiveness outcomes. On one end, NICE accepts and reviews the selection and implementation of such methods in detail, and has recently developed its own guidelines on the appropriate handling of treatment switching. Various examples have been identified and accepted in SMC and PBAC submissions. TLV, ZiNL, pCODR and INESSS have accepted statistical adjustments in at least one case. No evidence of such adjustments was identified in the selected HAS reports. On the other end, recent case examples among IQWiG submissions suggest that they do not accept the use of these methods. Conclusions: The rationale to adjust for treatment switching and the methods used need to be justified towards agencies accepting the correction techniques. Other payers will rather want to base their decision on results of the last “unbiased” dataset. Research on Methods - Preference Studies PR1 Survival or Mortality: Framing of the Risk Attribute in a Discrete Choice Experiment Veldwijk J 1, Essers B A B 2, Dirksen C D 3, Smit H A 4, Lambooij M S 1, de Wit G A 1 Institute for Public Health and the Environment, Bilthoven, The Netherlands, 2Clinical and Medical Technology Assessment, Maastricht University Medical Centre; CAPHRI, Maastricht University, Maastricht, The Netherlands, 3Department of Clinical Epidemiology and Medical Technology Assessment, Maastricht University Medical Centre, Maastricht, The Netherlands, 4University Medical Center Utrecht, Utrecht, The Netherlands . . . . . . . . . . . . 1National Objectives: To empirically test whether and how framing of a risk attribute in a Discrete Choice Experiment (DCE) affects study results with respect to the relative importance of the attributes, trading behavior and potential uptake rates. Methods: By means of ongoing data collection, two versions of a DCEquestionnaire containing nine D-efficiently designed choice tasks were distributed among a representative sample of the Dutch population aged 55-65years. The DCE consisted of four attributes related to the decision whether to participate in genetic screening for colorectal cancer (CRC). Three fixed attributes were; risk of being genetically predisposed, risk of developing CRC, and frequency of follow-up colonoscopies. The included risk attribute was framed positively as survival rate and negatively as mortality rate. Mixed logit models were conducted to estimate the relative importance of the attributes. Dominant decision behavior was determined and potential uptake rates were calculated. Results: Overall, risk attribute framing significantly interacted with most of the attribute level estimates. Based on the positive frame, the frequency of follow-up colonoscopies was most important followed by survival rate, while based on the negative fame, mortality rate was most important. Twice as many respondents dominated on the survival attribute compared to the mortality attribute. Potential uptake rates were calculated for multiple hypothetical scenarios, in all cases they were lower based on the data of the negative frame. Conclusions: The use of a positive frame leads to significantly increased frequency of dominant choices. Negative framing of the risk attribute resulted in a different relative importance of the attributes and a lower willingness to participate in genetic screening for CRC compared to positive framing. These results call for greater attention and more research with regard to the impact of framing of risk attributes in DCEs aiming to elicit preferences within the health care or public health context. PR2 Mapping from SF-6D to EQ-5D: Changes in Estimates Based on the Choice of Algorithm Hatswell A J 1, Ito T 2, Ganguly R 3, Nassens D 4, Almond C 1 1BresMed, Sheffield, UK, 2Janssen-Cilag Ltd, High Wycombe, UK, 3Janssen R&D, LLC, Spring House, PA, USA, 4Janssen BVBA, Beerse, Belgium . . . . . . Objectives: Where data collected using a preferred utility instrument are not available, mappings may be available to link either disease specific or generic quality of life instruments to the preferred measure. The objective of this study was to investigate the similarity of a widely used older mapping from SF-36 to EQ-5D and a more recently published mapping linking the same two instruments. Methods: Patient level SF-36 data from the PSUMMIT trials of ustekinumab versus placebo in psoriatic arthritis (PsA) were used to generate EQ-5D utilities based upon the older algorithm (Gray) and the newer algorithm (Rowen). A regression linking disease specific instruments to the resulting EQ-5D utilities, used in a published economic evaluation of anti-TNFα treatments for PsA, was also replicated to investigate the difference the choice of mapping would make if the results were used in economic analysis. Results: The mapping algorithms showed similar mean values (Gray = 0.572, Rowen = 0.568) and a large degree of agreement in estimates (R2 = 0.95). A331 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Replicating a published regression predicting EQ-5D based on Health Assessment Questionnaire (HAQ) and Psoriasis Area and Severity Index (PASI) scores, however, led to subtle yet statistically significant differences in coefficients (Gray EQ-5D = 0.782 -0.196xHAQ -0.00227xPASI vs. Rowen EQ-5D = 0.816 -0.220xHAQ –0.0231xPASI). In this analysis the Rowen algorithm was more sensitive to changes in disease status, which is potentially important in a disease area with multiple available therapies. Conclusions: The choice of mapping algorithm has the potential to affect the results of analyses performed. Although mean values obtained from different mapping algorithms may be similar, regression coefficients were shown to differ by mapping algorithm. Analyses using all available mappings should be presented to avoid accusations of ‘cherry picking’ the most favourable estimate. When suggesting the most appropriate analysis, clear justification should be provided. PR3 Investigating the Impact of Perspective on Weighting Qalys: a Discrete Choice Experiment Butt T 1, Longworth L 2, Rubin G 1, Orr S 1 1University College London, London, UK, 2Brunel University, Uxbridge, UK . . . . Objectives: Discrete choice experiments (DCEs) are increasingly used to elicit preferences for weighting QALYs by other characteristics. Recent studies for weighting QALYs have asked respondents to take the role of decision maker and prioritize treatment options for groups of patients (‘someone else’). The procedure normally used to obtain preferences for health states is for respondents to consider oneself in that state from behind a veil of ignorance. The aim of this study was to elicit public preferences for weighting QALYs by other characteristics, and to test if the decisionmaker perspective or naming the condition is equivalent to choosing for oneself from behind a veil of ignorance. Methods: A DCE was developed with attributes for health gain, severity, unmet need and process of care. Four questionnaires were designed to reflect three different perspectives and one alternative framing: a) ‘oneself’ b) ‘someone like oneself’ c) ‘someone else’ d) ‘someone else’ with a disease label (age-related macular degeneration). 800 members of the public were recruited via an online survey panel, with 200 completing each of the four questionnaires. Results were analysed using a conditional logit model and utility weights derived using Hicksian compensating variation. Results: Attributes of health gain, severity and process of care were significant for all perspectives (p< 0.05). Respondents preferred treatments that offered greater health gain, for a higher severity (lower starting level of health), and more convenient process. Unmet need was significant only in the ‘oneself’ perspective with respondents preferring treatments that addressed an unmet need. Conclusions: Members of the public are willing to trade health gain for other characteristics. These include severity, process of care, and in some cases unmet need. There is evidence that preferences for weighting QALYs vary by perspective. The study may help inform criteria for decision-making when prioritizing health care resources. PR4 Valuing EQ-5D-5L: Does the Ordering of the Health Dimensions Impact on Health State Valuations? Mulhern B 1, Shah K 2 1University of Sheffield, Sheffield, UK, 2Office of Health Economics, London, UK . QL2 The Impact of Discrimination, Victimization and Social Standing on Health-Related Quality of Life Samnaliev M Boston Childrens Hospital, Boston, MA, USA . Objectives: To examine the link between discrimination, victimization and social standing on health-related quality of life (HRQL). This information is currently not known, but is crucial as it can aid decision makers in the design of cost-effective policies to reduce such victimization among specific minority populations. Methods: We used data from 6,194 females and males from a prospective cohort of youth that has been tracked for over 15 years. We specifically examined the impact of discrimination, victimization and social standing of sexual orientation minorities on their HRQL (using EQ-5D-5L). In 2013, participants reported sexual orientation (Completely Heterosexual [CH], Mostly Heterosexual [MH], Bisexual [B], and Lesbian/ Gay [LG])). Participants previously reported past-year bullying victimization and discrimination (range: 1 [never] to 5 [several times/week]) and self-perceived social standing (range: 1 [top] to 10 [bottom]) in adolescence. Longitudinal, multivariable linear regression models were used to assess the effect of bullying and social standing in adolescence on sexual orientation HRQL score disparities in young adulthood, controlling for gender and age. Results: Compared to CHs, sexual minorities reported more bullying and lower social standing in adolescence and lower HRQL in young adulthood: MH HRQL score= 0.880; B 0.845; LG 0.870 vs. CH 0.916 (all differences had p-values< 0.05). When discrimination, victimization and social standing were added to the multivariable models, orientation-group effect estimates were attenuated ~10%, suggesting adolescent bullying and lower social standing partly explained HRQL disparities in young adulthood. Conclusions: Our findings are a proof of concept that victimization is not only a social justice issue but imposes an economic burden on society, as evidenced by its impact on HRQL.Our findings can also be used in cost-effectiveness analyses of interventions to reduce stigma against sexual orientation minorities. . Objectives: Health states defined by multi-attribute instruments such as the EQ-5D-5L can be valued using Time Trade Off (TTO) or Discrete Choice Experiment (DCE) methods. A key feature of the tasks is the order in which the health state dimensions are presented to respondents. This is because respondents may use a variety of heuristics to answer the questions (for example focusing on the first dimension presented), and therefore the order of the dimensions may impact on the importance assigned to particular health state descriptions. The objective of this study is to assess the impact of different EQ-5D-5L dimension orderings on health state valuations. Methods: Preferences for EQ-5D-5L health states were elicited from a broadly representative sample of members of the UK general public. Respondents were allocated to one of three EQ-5D-5L dimension orderings, and completed 10 TTO and 9 DCE tasks via a computer assisted personal interview carried out in their home. Differences in mean values and the relative importance of the coefficients across the arms were compared using difference testing and regression analyses. Results: Descriptive analysis suggests that there are minimal differences between the mean TTO health state values across the different dimension orderings. Regression analysis suggests that the magnitude of the dimension coefficients differs across the different dimension orders (for both TTO and DCE), but there is no clear pattern. Conclusions: There is some evidence that the order in which the dimensions are presented impacts on the coefficients, which may impact on the health state values provided. The order of dimensions is an important consideration in the design of health state valuation studies. Quality of Life and Utility Studies QL1 Elicitation of Health State Utilities Associated with Varying Severities of Flare in Systemic Lupus Erythematosus Hartz S 1, Liu-Leage S 2, Paget M A 2, Pollard C 3, Cook J 3, Enstone A 3 Research Centre, Surrey, UK, 2Lilly France, Cedex, France, 3Adelphi Values, Bollington, UK . during face-to-face interviews with a minimum representative sample (n= 100) of the general population, per-country, in Australia, Canada, France, Japan, Spain and the UK. Visual Analog Scale (VAS) scores were obtained to validate TTO scores. TTO scores were converted into utility values. Results: The highest mean TTO utility scores (range 0.66-0.82) were observed for the anchor HS (minimal disease activity) across all countries. All flare HS were associated with a disutility compared with the anchor HS: mild flare HS (0.55-0.71), moderate flare HS (0.38-0.53), severe renal flare HS (0.33-0.45), severe Central Nervous System (CNS) flare HS (0.30-0.45) and severe generalised flare HS (0.19-0.33). Significant differences were reported between the anchor state and each flare state across all countries (p< 0.05). Mean VAS scores followed the same trend. The severe generalised flare HS received the lowest mean TTO utility score across all six countries suggesting that the perceived day-to-day impact of a severe generalised flare was greater than a severe CNS or severe renal flare. Conclusions: These results show that a decrease in utility, representing a detrimental impact on HRQoL, was observed with increasing severity of flare. These results could be applied in cost-utility analyses for interventions for SLE. . . . . . QL3 Quality of Life Decrements after Stroke Dewilde S 1, Thijs V 2, Annemans L 3, Peeters A 4, Belgian Stroke Council N P 5 1SHE, Brussels, Belgium, 2KULeuven, Leuven, Belgium, 3Ghent University & Brussels University, Ghent, Belgium, 4Cliniques universitaires Saint-Luc, Bruxelles, Belgium, 5Virga Jesseziekenhuis, Hasselt, Belgium . . . . . . Objectives: To quantify which factors affect the quality of life valuations (QoL) of patients after stroke. Methods: 569 ischemic stroke patients were recruited into a retrospective, observational study in 10 teaching and regional hospitals across Belgium. Patients were stratified according to their modified Rankin Score (mRS) ranging from 0 (full health) to 5 (severely dependent). QoL as measured by the EQ5D and the VAS were collected for their health status before and after the index stroke. Utility decrements were calculated and using a general linear model the relationship between a set of variables and quality of life evaluation and its decrement associated with the stroke experience was investigated. Different distributions and link functions were tested against each other using the AICC and BIC criterion. Results: Stroke had a significant effect on patient’s QoL: a positive relationship with the degree of disability was found: EQ5D decrements were 0.08, 0.17, 0.25, 0.40, 0.64 and 0.75 for mRS categories 0 to 5; and equivalent decrements on the VAS scale were 7, 11, 20, 27, 33 and 39. The EQ5D and VAS decrements were significantly related to each other (r= 0.49, p< 0.001) and the decrements were also correlated with their baseline values (r= -0.74 for EQ5D baseline measurement and its decrement; r= -0.55 for VAS baseline value and decrement, p< 0.001). The decrements were of higher magnitude when measured by the EQ5D compared to the VAS. Furthermore the utility decrements were significantly associated with being female, being diabetic, having atrial fibrillation, being in a wheelchair, experiencing a change in one’s living situation and the number of days spent in an inpatient care facility during the first 3 months after stroke. Conclusions: Stroke results in a significant decrease in patient’s QoL, the magnitude of which is related to patient characteristics, disease severity measures, clinical risk factors and living circumstances. . 1Lilly Objectives: Systemic Lupus Erythematosus (SLE) is characterised by fluctuating periods of minimal disease activity and ‘flare’. Flare is an important outcome variable impacting the economic and humanistic burden of SLE. The objective of this study was to obtain population-based utility values for varying severities of flare to measure the impact on health-related quality of life (HRQoL). Methods: Six health states (HS) for varying severities of flare were developed based on literature, patient blogs, and interviews with patients (n= 12), rheumatologists (n= 7) and nurses (n= 2). HS were verified by independent clinical experts (n= 6) and pilot interviews (n= 10, UK). HS were evaluated using the time-trade-off (TTO) method Ql4 Differences Between Hypothetical and Experience-Based Value Sets for Eq-5d: Implications for Decision Makers Aronsson M , Husberg M , Kalkan A , Eckard N , Alwin J Linköping University, Linköping, Sweden . . . . . Objectives: The aim of the study was to analyse the differences between hypothetical and experience-based EQ-5D value sets. Furthermore, the aim was to evaluate the health economic implications of such differences for policy decisions in health care. Methods: We studied the differences in hypothetical and experiencebased value sets using three different methods: numerical comparison, empirical A332 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 comparison, and health economic examples. In the numerical comparison, the valuations of all 243 different EQ-5D states and all pure improvements were compared. In the empirical study, a database of 23,925 individuals was used to identify patient groups that could be influenced by the implementation of experience-based value sets. Two hypothetical health economic examples were used to examine the implication of the choice of value set for decision makers. Results: The numerical comparison showed that only three health states were assigned a lower QALYweight in the experience-based value set. The empirical comparison showed that severe conditions were assigned higher values in case of experience-based value set. Furthermore, the health economic examples showed the choice of a value set has an effect on the health economic result. Conclusions: Shifting to experience-based QALY-weights would increase the estimated level of quality of life in virtually all health conditions. In extension, quality of life enhancing interventions are given higher priority in decision making situations where hypothetical values are used to construct QALY-weights. On the other hand, in situations where experience-based QALY-weights are used, life-prolonging interventions will be prioritised. RESEARCH POSTER PRESENTATIONS – SESSION I Disease-Specific Studies Diabetes/Endocrine Disorders – Clinical Outcomes Studies PDB1 Hipos-ER (Hypoglycemia in Portugal Observational Study – Emergency Room): Outcomes with Different Anti-Hyperglycemic Agents J 1, Laires P 1, Dores J 2, Araújo F 3, Carr R 4, Brodovicz K 5, Radican L 6, Vicente V 7, Conceição Nogueira A M 1 1Merck Sharp & Dohme, Oeiras, Portugal, 2Hospital de Santo António, Porto, Portugal, 3Hospital Beatriz Ângelo, Loures, Portugal, 4Merck Sharp & Dohme, Ballerup, Denmark, 5Merck & Co., North Wales, PA, USA, 6Merck & Co., Inc., Whitehouse Station, NJ, USA, 7Eurotrials, Lisbon, Portugal . . . . . . . . . . Objectives: HIPOS-ER is an observational study to describe the patient population of Type 2 Diabetics treated with an anti-hyperglycemic agent (AHA) and admitted to the emergency room (ER) with a hypoglycemic event. In this analysis we aim to describe the hypoglycemia events by AHA class. Methods: The study enrolled patients from 7 centers in mainland Portugal from Jan 2013 – Jan 2014. Sociodemographic and clinical data were collected at the emergency room and patients who required hospital admission were followed up. Episodes were enrolled consecutively within the sampling period. AHA therapy classes were Insulin - Group 1, secretagogue based - Group 2, oral AHA excluding secretagogue - Group 3 and Insulin+secretagogue – Group 4. Results: A total of 238 patients were admitted to the ER with severe hypoglycemia and 105 (44%) were hospitalized. The distribution based on AHA therapy: 55% (131) Group 1, 32% (75) Group 2, 7% (16) Group 3 and 7% (16) Group 4. Previous severe hypoglycemia in the last 12 months was more frequent in Group 1 vs. Group 2 (p= 0.009). Group 2 patients were more often followed up in Primary Care vs. Group 1 (84% vs. 48%; p<0.001) and Group 4 (44%, p=0.002). Group 2 patients were more often hospitalized vs. Group 1 (71% vs. 29%; p<0.001) and Group 4 (31%; p=0.003). There was no difference in terms of the length of stay or hospitalization outcome. 9 deaths occurred: 5 in Group 1, 3 in Group 2 and 1 in Group 3. Conclusions: In the first national study on severe hypoglycemia in Type 2 diabetics, patients treated with Group 2 drugs were followed mainly in Primary care setting. More patients on Group 2 drugs were hospitalized. Within our sample there is little clinical difference in the hypoglycemia events with different therapys besides the need for hospitalization. PDB3 Regional Assessment of Severe Hypoglycemic Coma Events in Finland Vattulainen P 1, Mushnikov V 1, Hoti F 1, Saukkonen T 2, Korhonen P 1 1EPID Research, Espoo, Finland,, 2Novo Nordisk Farma, Espoo, Finland . . . . . Objectives: In a previous study about severe hypoglycemic coma events (SH) among diabetes mellitus (DM) patients we saw regional variation in SH occurrence. The purpose was to compare the incidence of SH between hospital districts in Finland. Methods: This study among type 1 (T1) and 2 (T2) diabetic patients was based on nationwide health care registers. The study cohort contained insulin naïve diabetic patients who had filled at least one prescription of insulin during follow-up in 2006-2009. SH was defined as a hospitalization or a secondary health care visit due to hypoglycemic coma (ICD-10: E10.00 and E11.00). Stratified incidence rates and adjusted hazard ratio (HR) estimates with 95% confidence intervals (CI) were calculated. Analyses were performed for the first and recurrent SHs. Results: The population comprised 5271 (17.6%) patients with T1 and 24602 (82.4%) with T2. Altogether 3.1% patients experienced at least one SH during the follow-up. In different hospital districts the rate of first SHs varied from 5.6 (Varsinais-Suomi) to 47.8 (Länsi-Pohja) for T1, and from 4.8 (Pohjois-Karjala) to 50.3 (Länsi-Pohja) for T2. Compared to the capital area, the risks of first SHs varied differently for T1 and T2, but were commonly higher in Länsi-Pohja (T1: HR 3.46, CI 1.41-8.46; T2: HR 1.60, CI 1.13-2.29) and lower in Pohjois-Karjala (T1: no events; T2: HR 0.275, CI 0.14-0.56). Taking all SH events into account did not change these differences remarkably. Conclusions: We found differences between regions in risk of hospital-treated SH in both DM types. Further analyses will be performed for e. g. ambulance density. PDB4 Hipos-ER (Hypoglycemia in Portugal Observational Study – Emergency Room): Clinical Outcomes in Admitted Patients Conceição J 1, Dores J 2, Araújo F 3, Laires P 1, Carr R 4, Brodovicz K 5, Radican L 6, Nogueira AM1 1Merck Sharp & Dohme, Oeiras, Portugal, 2Hospital de Santo António, Porto, Portugal, 3Hospital Beatriz Ângelo, Loures, Portugal, 4Merck Sharp & Dohme, Ballerup, Denmark, 5Merck & Co., North Wales, PA, USA, 6Merck & Co., Inc., Whitehouse Station, NJ, USA . . . . . . . . . PDB2 Hipos-ER (Hypoglycemia in Portugal Observational Study – Emergency Room): Clinical Outcomes in the Emergency Room Objectives: HIPOS-ER is an observational, cross-sectional, multicenter study to describe the patient population of Type 2 diabetics treated with an anti-hyperglycemic agent (AHA) and admitted to the emergency room (ER) with a hypoglycemic event. This is the first national hypoglycemia study in Portugal and the first study collecting hypoglycemia specific resource data directly in this setting. Here we aim to describe the clinical features of hospitalized patients. Methods: The study enrolled patients from 7 centers in mainland Portugal for a period of 12 months (Jan 2013 – Jan 2014). Sociodemographic and clinical data were collected at the emergency room and patients who required hospital admission were followed up. Episodes were enrolled consecutively within the sampling period. Results: A total of 238 patients were enrolled, and 105 (44%) were hospitalized and 2 (1%) were transferred outside the hospital center for likely need of hospitalization. Mean age was 78 years, average disease duration was 19 years, 51% were female, 36% were on insulin, 51% on a secretagogue, 9% on an oral AHA excluding secretagogue and 5% on insulin+secretagogue. 26% had complications diagnosed in the ER: Trauma (37%) and Cardiovascular (22%) and Infection/Sepsis (22%) were the most frequent. Mean and median hospitalization time was 9 and 5 days, respectively. Most (95%) patients were admitted to a medical department: Internal medicine (80%) and the ER observation/short stay unit (15%) were the most frequent. 6% of patients were admitted to an intensive care unit. 8% (9) of hospitalized patients died. Conclusions: Hospitalized diabetic patients following an ER episode due to hypoglycemia were treated mainly with secretagogue type drugs. Internal medicine was key in the hospital approach of these patients. The length of stay exceeded the 48-72h typical surveillance period for secretagogue-induced hypoglycemia. Severe hypoglycemia in Portugal is associated with several complications which also include death. Conceição J 1, Dores J 2, Araújo F 3, Laires P 1, Carr R 4, Brodovicz K 5, Radican L 6, Nogueira A M 1 1Merck Sharp & Dohme, Oeiras, Portugal, 2Hospital de Santo António, Porto, Portugal, 3Hospital Beatriz Ângelo, Loures, Portugal, 4Merck Sharp & Dohme, Ballerup, Denmark, 5Merck & Co., North Wales, PA, USA, 6Merck & Co., Inc., Whitehouse Station, NJ, USA PDB5 Comparative Efficacy and Safety of Empagliflozin with Other Anti-Diabetic Drugs for the Treatment of Patients with Type 2 Diabetes Mellitus Who are Failing Insulin . . . . . . . . . Objectives: HIPOS-ER is an observational, cross-sectional, multicenter study to characterize the population of Type 2 Diabetics treated with an anti-hyperglycemic agent (AHA) who were admitted to the emergency room (ER) with a hypoglycemic event. This is the first national ER hypoglycemia study and the first study collecting primary resource data in this setting. Here we aim to describe this population and the hospital management. Methods: The study enrolled patients from 7 centers in mainland Portugal from Jan 2013 – Jan 2014. Sociodemographic and clinical data were collected within the emergency room and patients who required hospital admission were followed-up. Results: A total of 238 events were recorded. Mean age: 76 years; average disease duration: 19 years; 58% female; 80.9% were not living alone; 83% no formal education or schooling < 4 years; 25% had a previous episode of severe hypoglycemia in the preceding 12 months and 61% were treated in a primary care setting. Regarding drug treatment: 55% on insulin; 32% on a secretagogue; 7% on other oral AHA and 7% on insulin+secretagogue. Missing a meal was the most frequent immediate cause of hypoglycemia (56%). All patients received lab evaluations. 71% underwent radiological procedures. Time spent in the ER was 11 hours (mean) and the average medical and nurse time utilized was 85 and 71 min respectively. AHA therapy was changed in 65% of cases: insulin adjustment (56%) being the most frequent modification. 56% (132) of patients were discharged. Conclusions: Diabetic patients admitted to the ER have several markers of frailty and low educational status. Strategies to mitigate hypoglycemia underline the need to avoid missed meals. ER episodes are lengthy and consume significant physician and nurse time as well as laboratory and other diagnostic procedures. Primary care stakeholders should be involved in actions to mitigate hypoglycemia in type 2 diabetes. Thorlund K 1, Siliman G 1, Eapen S 1, Lund S 2, Palencia R 3 Outcomes, Vancouver, BC, Canada, 2Boehringer Ingelheim Pharma GmbH & Co. KG, Ingelheim am Rhein, Germany, 3Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany . . . . . 1Redwood Objectives: The aim of the present network meta-analysis is to compare the efficacy and safety of empagliflozin versus other anti-diabetic drugs, in patients with type 2 diabetes mellitus (T2DM) who are inadequately controlled with any insulin treatment and with or without other anti-diabetic drugs (ADs). Methods: We performed a systematic review of randomized controlled trials (RCTs) and Bayesian network meta-analysis to establish the comparative efficacy and safety of SGLT-2s, DPP-4s, GLP-1s and TZDs in T2DM patients. The principal outcome of this analysis was the effect of these drugs on HbA1c, weight, systolic blood pressure (SBP), incidence of hypoglycaemia and urinary tract infections (UTIs) at 24 weeks. Results: Sixteen RCTs were included. All patients received insulin treatment (any type) with or without other additional ADs. Compared with placebo, mean changes in HbA1c were -0.51 % [95% confidence interval (CI) -0.93 to -0.11%] and -0.60% [95%CI -1.03 to -0.19%] for empagliflozin 10mg and 25mg, respectively. HbA1c reduction was similar for all other interventions, and no significant differences were detected. Compared with placebo, Empagliflozin 10mg and 25mg significantly reduced patients weight by approximately -1.5kg. Other SLGT-2s and all GLP-1s were associated with a similar weight loss, not change was observed for DPP-4s, and non-significant weight gains were observed for TZDs. Empagliflozin was also associated with significant reductions in SBP, and no significant differences between treatments were detected. For hypoglycaemic events, empagliflozin 10mg and 25mg were associated with nonsignificant relative risks of 1.03 and 1.23. All interventions had non-significant relative risks of UTIs. Conclusions: Compared with other treatments, empagliflozin A333 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 offers similar or reduced HbA1c reduction, had comparable significant weight loss to other SLGT-2s and GLP-1s, and appeared to have a superior weight loss profile compared with DPP-4s and TZDs.. No increased risk of adverse events were observed for empagliflozin compared with placebo and other ADs. PDB6 Comparative Efficacy and Safety of Empagliflozin with Other Antidiabetic Drugs for the Third Line Treatment of Type 2 Diabetes Mellitus Thorlund K 1, Siliman G 1, Eapen S 1, Lund S 2, Palencia R 3 Outcomes, Vancouver, BC, Canada, 2Boehringer Ingelheim Pharma GmbH & Co. KG, Ingelheim am Rhein, Germany, 3Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany . . . . . 1Redwood Objectives: The aim of the present network meta-analysis is to compare the efficacy and safety of empagliflozin versus other antidiabetic drugs used in third line for the treatment of patients with type 2 diabetes mellitus (T2DM). Methods: We conducted a systematic review randomized controlled trials (RCTs) and Bayesian network meta-analysis to establish the comparative efficacy and safety of SGLT-2s, DPP-4s, GLP-1s, and TZDs. RCTs enrolling subjects with T2DM inadequately controlled on metformin plus sulfonylurea were included. The principal outcome of this analysis was the effect of these drugs on HbA1c, weight, systolic blood pressure (SBP), incidence of hypoglycaemia and urinary tract infections (UTIs) at 24 weeks. Results: From 6969 abstracts, 13 were included in the analysis. No RCTs involving TZDs were identified. Compared with placebo, mean changes in HbA1c were -0.65% [95% confidence interval (CI) -1.59 to -0.08%] and -0.60% [95%CI -1.14 to -0.14%] for empagliflozin 10mg and 25mg. No significant differences were detected between interventions. Mean changes in weight with empagliflozin 10mg and 25mg were -1.77 [95%CI -2.19 to -1.35) and -2.00 (95%CI -2.44 to -1.57), respectively. Mean weight losses were fairly similar across SLGT-2s and GLP-1s ranging between -1.26 to -2.12. All DPP-4s were associated weight gains, ranging form 0.33 to 0.98, of which most were statistically significant. SBP data were only available for SGLT-2s and DPP-4s. Empagliflozin 10mg and 25mg compared with placebo had statistically significant reductions of -2.70 and -2.09. All interventions (except exenatide) yielded relative risk of hypoglycaemia greater than 1.00. For UTIs, no differences were found between SLGT-2s or DPP-4s and placebo. Conclusions: Compared with other SGLT-2s, DPP-4s, and GLP-1s, empagliflozin generally offers similar HbA1c control at week 24, an advantageous profile in weight loss and reduction of SBP, as well as similar safety profile. PDB7 Comparative Efficacy and Safety of Empagliflozin with Other Oral Antidiabetic Drugs for the Second Line Treatment of Type 2 Diabetes Mellitus Thorlund K 1, Siliman G 1, Eapen S 1, Lund S 2, Palencia R 3 1Redwood Outcomes, Vancouver, BC, Canada, 2Boehringer Ingelheim Pharma GmbH & Co. KG, Ingelheim am Rhein, Germany, 3Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany . . . . . Objectives: To compare the efficacy and safety of empagliflozin versus other second-line treatment for patients with type 2 diabetes mellitus (T2DM). Methods: A systematic review and Bayesian network meta-analysis were performed to identify randomized controlled trials (RCTs) assessing the efficacy and safety of SGLT-2s, DPP-4s, GLP-1s, TZDs, and SUs in patients with T2DM. RCTs enrolling subjects with T2DM inadequately controlled with metformin monotherapy were included. The principal outcomes were HbA1c, weight, systolic blood pressure (SBP), hypoglycaemia, and urinary tract infections (UTIs) at 24 weeks. Results: Forty-eight RCTs were included. The mean differences (MD) in HbA1c were -0.56% [95% confidence interval (CI) -1.06% to -0.08%] and -0.64% [95%CI -1.14 to -0.14%] for Empagliflozin 10mg and 25mg vs placebo. All other interventions yielded similar reductions and no significant differences were detected between interventions. Empagliflozin 10mg and 25mg significantly changed patients’ weight by -1.63kg (95%CI -2.66 to -0.64) and -2.01kg (95%CI -3.02 to -1.02), versus placebo. Other SLGT-2s had similar MDs, all DPP-4s had no change, and GLP-1s fell in between. All SUs and TZDs were associated with significant weight gain versus placebo. For SBP, the MDs for Empagliflozin 10mg and 25mg versus placebo were -4.09mmHg (95%CI -6.97 to -1.18) and -4.81mmHg (95%CI -7.69 to -2.00). No significant differences between Empagliflozin and other interventions were detected. Incidence of hypoglycaemia for empagliflozin 10 and 25 mg relative to placebo was 2.71 [95%CI 0.46 to 11.44] and 1.97 [95%CI 0.23 to 9.57], respectively. No significant differences were detected between emplagliflozin and other interventions. For UTIs, all yielded relative risks close to 1.00 when compared with placebo. Conclusions: Compared with other SGLT-2s, DPP-4s, GLP-1s, TZDs, and SUs, empagliflozin offers similar HbA1c control at 24 weeks, a marked reduction in weight compared with DPP-4s, TZDs, and SUs, and a similar safety profile as other interventions. PDB8 Long-Term Modeling of Using Manually Coded and Autocoded Blood Glucose Meters in Diabetes Treatment Yagudina R , Kulikov A , Babiy V I. M. Sechenov First Moscow State Medical University, Moscow, Russia . . 26 years period was associated with 18,59 LYG. At the same time use of autocoded blood glucose meters was associated with 18,92 LYG. In case of using autocoded meters instead of using manually coding meters patients obtained 0,33 LYG more (120 days). Conclusions: Obtained results showed that difference in glucose measurement errors between manually coded and autocoded blood glucose meters can lead to the difference in long-term outcomes in diabetes treatment. PDB9 Assessing the Relationship between Improved Life Expectancy Due to Better Cardiovascular Risk Factor Management and the Likelihood of Microvascular Complications in Type 2 Diabetes Mellitus McEwan P 1, Grant D 2, Foos V 3 1Health Economics and Outcomes Research Ltd, Cardiff, UK, 2IMS Health, London, UK, 3IMS Health, Basel, Switzerland . . . Objectives: Type 2 diabetes mellitus (T2DM) is a chronic disease associated with increased risk of cardiovascular (CV) and microvascular complications. Improvements in blood pressure and cholesterol control have resulted in a reduction in CV event rates in clinical practice. The objective of this study was to assess the relationship between increased life expectancy, due to reduction in CV event rates, and the risk of microvascular disease for a range of glycemic control levels. Methods: A lifetime analysis was conducted using the CORE diabetes model (CDM). Newly diagnosed T2DM simulated patients aged 52 years at baseline with HbA1c 7.1%, SBP 135.1 mmHg, total cholesterol: HDL 5.2 mmol/l were modelled. The impact of HbA1c on microvascular complications was assessed by running the CDM with baseline HbA1c ±1% for scenario 1: 100% of patient receiving CV risk factor management; and scenario 2: no CV risk factor management. Results: Improved CV risk factor management reduced the predicted cumulative incidence of fatal myocardial infarction (MI) from 27% to 18%, increasing life expectancy by an average of 2 years. For scenario 1, baseline HbA1c +1% versus -1% was associated with a 20%, 11% and 4% increase in microalbuminuria (MA), gross proteinuria (GRP) and end stage renal disease (ESRD), respectively; for scenario 2, the increase was 15.5% for MA, 7.6% for GRP and 2.5% for ESRD. Cumulative incidence of neuropathy ranged from 68.4% (baseline HbA1c +1%) to 42.1% (baseline HbA1c -1%) for scenario 1 and from 65.2% (baseline HbA1c +1%) to 39.7% (baseline HbA1 c-1%) for scenario 2. Cumulative retinopathy rates were similar across both scenarios: 56.7% versus 56.0% for scenarios 1 and 2, respectively. Conclusions: This modeling study suggests that improvements in blood pressure and cholesterol management may result in increased rates of microvascular complications, in particular renal disease, over the long term as patient survival increases. PDB11 A Systematic Review and Network Meta-Analysis Assessing the Effectiveness and Tolerability of Gliptins and Sulfonylureas as Monotherapy in Patients with Type 2 Diabetes Mellitus If Metformin is not Considered Appropriate Bartmus T , Mansmann U Institut für medizinische Informationsverarbeitung, Biometrie und Epidemiologie, LudwigMaximilians-Universität, Munich, Germany . . Objectives: A significant proportion of patients with type-2-diabetes mellitus (T2DM) are unable to take Metformin as recommended first-line therapy due to gastrointestinal intolerance or contraindications such as chronic kidney disease. In contrast to combination therapy no network meta-analysis (NMA) has been undertaken for oral anti-diabetic drugs (OAD) as monotherapy in this population particularly with respect to gliptins and sulfonylureas as second-line options. The purpose of this study is to assess the comparative effectiveness and tolerability of gliptins versus sulfonylureas in terms of glycated hemoglobin (HbA1c), body weight and hypoglycemia. Methods: A systematic review was conducted searching bibliographic databases, reports of regulatory authorities and clinical trial registries through July 2012 to identify randomized controlled trials in adult T2DM patients receiving at least 12 weeks of OAD monotherapy or placebo. A Bayesian NMA was performed to yield mixed treatment comparisons. Consistency was examined by the node split method. Results: A total of 62 studies enrolling 21,302 patients informed the entire network. Due to their improved model fit estimates from random effect models are reported to account for heterogeneity across the set of studies. After a mean follow-up of 32 weeks, the difference in mean HbA1c was 0.26, 0. 95 credible interval (CrI0.95): [0.1; 0.42], in favour of sulfonylureas. However, gliptins induced weight loss (difference in means: -1.21 kg; CrI0.95: [-1.57; -0.84] ) and were associated with a considerably lower incidence of any hypoglycemia compared to sulfonylureas (odds ratio: 0.22; CrI0.95: [0.15; 0.31] ). All effect estimates were statistically significant and consistent in terms of combining direct and indirect evidence. Conclusions: This is the first network assessing OAD monotherapy that can readily be extended to emerging therapies. With regard to glycemic control gliptins were slightly inferior to sulfonylureas, whereas they positively affected body weight and risk of hypoglycemia, confirming their role in second-line monotherapy. . Objectives: To obtain long-term clinical outcomes of using manually coded and autocoded blood glucose meters in diabetes treatment in the Russian Federation. Methods: The model used in this study analyzed the influence of errors in blood glucose measurements (due to using manually or autocoded glucose meters) on the treatment of patients with Type 1 and Type 2 diabetes during the 26 years period (the life-time period). Life years gained (LYG) was chosen as an outcome measure in assessment of health intervention. Calculation of LYG was based on prior clinical studies that evaluated glucose meters’ errors in glucose level measurements and risk of complications associated with blood glucose level. Data for patients with diabetes was obtained from prior epidemiological studies that had been provided in Russian Federation. Results: Use of manually coded blood glucose meters in the analyzed population with median age of 53 years during PDB12 Glycemic, Lipid, and Blood Pressure Control Among Individuals with Type 2 Diabetes Mellitus in Saudi Arabia Osenenko K M 1, Szabo S M 1, Donato B M K 2, Korol E E 1, Qatami L 3, Al Jaser S 4, Al Saggabi A 4, El Seid M E 4, Maclean R 5, Levy A R 1 1ICON Epidemiology, Vancouver, BC, Canada, 2Bristol-Myers Squibb Company, Wallingford, CT, USA, 3Bristol-Myers Squibb Company, Dubai, United Arab Emirates, 4King Fahad National Guard Hospital, Riyadh, Saudi Arabia, 5Bristol-Myers Squibb Company, Plainsboro, NJ, USA . . . . . . . . . . . . . . . . . Objectives: Inadequate glycemic, blood pressure (BP), and low-density lipoprotein (LDL) control among persons with type 2 diabetes mellitus (T2DM) increases the risk of T2DM-related complications, which require more intensive and costly therapy. Extending on a recently-conducted study in Dubai, UAE, we assessed levels of glycemic, LDL, and BP control, and estimated the proportion of those meeting A334 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 guideline targets, among a cohort of individuals treated for T2DM in Riyadh, Saudi Arabia. Methods: Charts from 455 adults with T2DM who visited the King Fahad National Guard Hospital from October 2009 to March 2010 (enrolment period) were systematically sampled until the target (n= 250) was reached. Haemoglobin A1c (HbA1c), LDL, and BP test results from enrolment to September 2011 were abstracted. The most recent test values were compared to guideline targets. The proportion of well-controlled (target met on all tests) and never-controlled (target not met on any test) subjects over the study period was calculated. Analyses were stratified by T2DM duration. Results: Forty-four percent of the cohort was male; at enrolment, mean (SD) age was 61 (13) years and mean T2DM duration was 11 (8) years. At the most recent assessment, 36 subjects (14%) had HbA1c < 7%, 91 (36%) had HbA1c ≥ 9%, and 177 (72%) had LDL < 100mg/dL. Although 109 subjects (44%) met BP targets (< 130/80mmHg), 30% had BP ≥ 140/90mmHg. HbA1c, LDL and BP were well-controlled in 5.2%, 45.9%, and 8.4% of subjects, respectively, while 71.2%, 13.4% and 22.9% were never-controlled, respectively. The proportion of the cohort that was never-controlled for HbA1c increased with T2DM duration. Conclusions: While rates of HbA1c control were low among subjects with T2DM, nearly half met BP targets and nearly three quarters met LDL targets. Given the increased risk of complications associated with poor control, achieving higher rates of control could reduce the burden of T2DM in Saudi Arabia. PDB13 Budget Impect Analysis Sun T chian pharmaceyical university, nanjing, China . Objectives: To quantify the number and costs of relapses avoided over 2 years in the first-line treatment of RRMS based on the findings of the Cochrane report. Methods: An Excel-based financial model estimated the relapses and costs incurred by a hypothetical cohort of 1000 RRMS patients treated with first-line disease-modifying drugs (DMDs). The modelled cohort evaluated the consequences of treatment with subcutaneous (SC) interferon beta-1a versus intramuscular (IM) interferon beta-1a, as this was the only comparison whose data quality was assessed as ‘high’ by the Cochrane Review (Filippini et al., 2013). Risk of relapse was based on the 2-year data from the Cochrane Review network meta-analysis. The analysis was performed from a US payer perspective. The cost of a relapse was sourced from Panitch et al., 2005, and adjusted to 2012 US dollars. Net annual cost of therapy was based on wholesale acquisition cost. Given the model’s short time horizon, disability-related costs were not included as these tend to be an important economic driver only over the long-term progression of the disease. In order to test how variability in the model’s inputs might impact the analysis’ results, two-way sensitivity analyses were performed based on the reported 95% risk of relapse credible intervals for SC interferon beta-1a and IM interferon beta-1a. Results: In a hypothetical cohort of 1000 RRMS patients, treatment with SC interferon beta-1a is expected to result in the avoidance of 173 (sensitivity analysis range: -20 to 399) relapses versus IM interferon beta-1a over 2 years. Assuming a direct cost of relapse of $5141, this represents a savings of $890,212 (sensitivity analysis range: -$102,138 to $2,052,934) versus IM interferon beta-1a. Conclusions: Subcutaneous interferon beta-1a is likely to result in fewer relapses and lower direct costs of relapse versus IM interferon beta-1a over a 2-year period treatment. PDB14 A Decision-Focused Mixed Treatment Comparison (MTC) of Alternative Dpp-4 Inhibitors (Dpp-4i’s) Used in Combination With Metformin or a Sulfonylurea for the Treatment of Type 2 Diabetes Mellitus (T2DM) Tolley K 1, Strickson A 2, Kay S 2, Benson E 2, Selby R 2 1Tolley Health Economics Ltd., Buxton, Derbyshire, UK,, 2Hardwick House, Buxton, Derbyshire, UK . . . . . Objectives: To conduct decision-focused mixed treatment comparisons (MTCs) of the relative efficacy and safety of a new DPP-4 inhibitor (DPP-4i), alogliptin 25mg daily compared to current DPP-4i’s used in UK clinical practice (sitagliptin, saxagliptin, linagliptin, vildagliptin) in dual therapy for the treatment of type 2 diabetes mellitus (T2DM) in combination with metformin or sulfonylurea (SU). Methods: A decision-focused systematic review was conducted to identify RCTs comparing the DPP-4i’s in combination with metformin or SU compared to metformin or SU alone or against each other in the target patient population. Separate Bayesian MTCs were conducted for each DPP-4i combination. Outcomes of interest were change in HbA1c from baseline (primary), weight change, proportion of patients HbA 1c < 7%, and proportion of patients ≥ 1 hypoglycaemic episode. Fixed and random effects models were run with sensitivity analysis conducted to account for confounding factors (study length, baseline HbA1c), study heterogeneity and inconsistency. Results: Twenty-five RCTs met inclusion criteria for the MTCs. For the primary outcome, fixed effects models suffered from heterogeneity problems whilst random effect models struggled to estimate between trial heterogeneity. However, the base case and all sensitivity analyses showed the same results - alogliptin in combination with metformin or SU had a high probability of non-inferiority to comparator DPP-4’s in HbA1c change (61%-100% at a 0.3% HbA1cmargin). Deletion of trial outliers via leverage plots vastly improved model fit within fixed effects models whilst not changing underlying results. For all other outcomes alogliptin was shown to be comparable to alternative DPP-4i’s. Conclusions: Alogliptin 25mg has comparable efficacy and safety as other DPP-4i’s at their recommended doses in UK clinical practice. This is in line with expectations based on prior meta-analyses of DPP-4i’s. The use of a decision-focused approach to the MTC enables a focus on the data of direct interest for clinical and HTA based decision making. PDB15 Identifying Consistent Inconsistency in Network Meta-Analyses An Illustration In Type 2 Diabetes Hawkins N 1, Scott D A 2 PLC, Oxford, UK, 2ICON Health Economics, Oxford, UK . 1ICON . . Objectives: Network meta-analyses (NMA) provide estimates of comparative efficacy for multiple treatments based on an analysis of connected networks of trial comparisons. A key concern is the comparability of treatment effect estimates from different trials. Where there is both indirect and direct evidence for one or more comparisons (‘loops’ in the network) it is possible to evaluate empirically the ‘consistency’ of the network. Methods: A variety of methods have been proposed to examine inconsistency including: (i) node-splitting where the direct and indirect estimates are compared across the network (ii) comparison to an ‘inconsistency’ model where estimates for each treatment comparison are allowed to be independent, (iii) inclusion of treatment by design interaction terms, (iv) investigation of residual deviance estimates for individual trial arms, and (v) investigation of mixed predictive p-values. We compare the implementation and, most importantly, the interpretation of these methods using a previously published NMA in type 2 diabetes. In this analysis HbA1c was compared across six treatments in a network of 22 studies with multiple ‘loops’. Results: The methods agreed in showing the presence of inconsistency with the network. For example, the inconsistency model showed an improved fit (DIC -62.35) compared to the consistency model (DIC -60.25). The node splitting method identified statistically significant inconsistency in two treatment arcs (liraglutide 1.8mg vs placebo and liraglutide 1.8mg vs exenatide QW). Conclusions: The alternative methods vary in their ability to provide an omnibus ‘test’ of inconsistency across the network and their ability to identify which parts of the network contain inconsistencies. We highlight that none of the methods alone can identify individual studies as being the cause of inconsistencies and argue that we need to consider the whole structure of the network and the characteristics of the studies (in terms of treatments, subjects and design) within the network. PDB17 The Efficacy and Effectiveness in HBA1C-Lowering is Dependent on Baseline Body Mass Index (BMI) for Sitagliptin but not Canagliflozin in the Treatment of Type 2 Diabetes Mellitus (T2DM) Diels J 1, Angermund R 2, Schroeder M 3, Worbes-Cerezo M 3, Thompson G 3 1Janssen Research & Development, Beerse, Belgium, 2Janssen-Cilag Germany, Mainz, Germany, 3Janssen-Cilag UK, High Wycombe, UK . . . . . Objectives: To investigate HbA1c-reduction by baseline BMI in patients treated with canagliflozin or sitagliptin, using clinical trial and electronic medical record (EMR) data. Methods: Patient-level data from two randomised controlled trials (RCTs) were used to explore HbA1c-reduction from baseline after 52 weeks treatment with canagliflozin (100/300mg) or sitagliptin (100mg) by baseline BMI. Ordinary least squares (OLS) regression was performed with HbA1c, BMI, eGFR and demographics as covariates, in patients with metformin (MET) or metformin+glimepiride (MET+SU) background therapy. EMR-data (UK General Practitioner data from CPRD) on HbA1c over time in patients treated with sitagliptin were analysed by background therapy using repeated measures analysis, with baseline BMI, HbA1c and demographics as covariates. Results: In both RCTs sitagliptin showed a decreasing HbA1c-reduction by increasing baseline BMI, while efficacy of canagliflozin was independent of BMI. The estimated HbA1c-reduction (%) from baseline for sitagliptin in patients with baseline BMI of 25 vs. 40 varied between -0.87 to -0.58 (MET; Δ = 0.29, p= 0.01) and -0.87 to -0.54 (MET+SU; Δ = 0.33, p= 0.0014), while a non-significant increase in HbA1c-reduction was observed in high BMI-patients in all canagliflozin-arms (Δ between -0.04 and -0.07). EMRdata showed similar decreasing effectiveness of sitagliptin in high BMI-patients. Estimated HbA1c-reduction (month 10; baseline HbA1c 9%) was significantly less (MET: Δ = 0.30, MET+SU: Δ = 0.35, p< 0.0001) in patients with BMI 40 vs. 25. No data for canagliflozin were yet available. Lower efficacy of sitagliptin in obese patients has been previously reported in the literature. Conclusions: RCT and EMRdata consistently show that the relative efficacy of anti-diabetic treatments may depend on baseline BMI. Reduced efficacy of sitagliptin and DPP-4 inhibitors in general in obese patients may be explained by a higher degree of insulin-resistance. Efficacy of canagliflozin is independent of BMI, due to its insulin-independent mechanism of action. Patients’ BMI should be taken into account to select effective therapeutic options for patients with T2DM. PDB18 Treatment Maintenance Duration of Dual Therapy with Metformin and Sitagliptin in Type 2 Diabetes – Real-World Data From Odyssee Study Leproust S 1, Dallongeville J 2, Valensi P 3, Boutmy E 4, Moisan C 1, chanut-Vogel C 5, de Pouvourville G 6 1Merck (MSD France), Paris, France, 2INSERM U744, Lille, France, 3Service d’EndocrinologieDiabétologie-Nutrition, Bondy, France, 4Cegedim Strategic Data, Boulogne-Billancourt, France, 5Laboratoires MSD France, Courbevoie, France, 6ESSEC, Cergy-Pontoise, France . . . . . . . Objectives: comparative effectiveness of new oral anti-diabetic drugs in primary care practice remains poorly characterized. This presentation focus on the relevant statistical methods used to handle common potential bias related to real-world observational designs. Methods: multicenter, longitudinal, observational study, conducted in primary care in France. Participating physicians were to include adult type 2 diabetes patients initiating a treatment with metformin and sitagliptin dual therapy (M-Sit group) or metformin and sulfonylurea dual therapy (M-SU group). Planned follow-up period was three years. The primary endpoint was the treatment maintenance duration, from initiation of dual therapy to a strict change, defined as the addition, replacement or withdrawal of an agent used for initial dual therapy. Survival Kaplan-Meier analysis, multivariate Cox model adjusted on the propensity score in order to limit bias due to baseline imbalance between the two groups and sensitivity analyses including multiple imputations to deal with missing data have been performed. Results: a total of 3 453 patients have been analyzed: 1 874 in the M-Sit group and 733 in the M-SU group. In the principal analysis, the median treatment maintenance duration was 20.2 months in the M-SU group and 43.2 months in the M-Sit group (p < 0.0001). A335 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Multivariate Cox model adjusted for propensity score and sensitivity analysis confirmed a reduced risk of treatment change for M-Sit versus M-SU with a relative risk of 0.70 [0.62; 0.78] (p< 0.0001). Conclusions: We observed an approximate two-fold increase in the median treatment maintenance duration in the met-Sit group versus the met-SU group. Given the observational design, confounding factors on the primary outcome cannot be excluded. However, multivariate and sensitivity analyses showed no qualitative change in the principal finding, providing some confidence in the observed difference between the two dual therapies. PDB19 Assessing Consistency in a Network Meta-Analysis to Compare Once Weekly Dulaglutide Versus Other Glp-1 Receptor Agonists in Patients with Type 2 Diabetes Hawkins N 1, Padhiar A 1, Thompson J 1, Scott D A 1, Eaton J N 1, Varol N 2, Norrbacka K 2, Boye K S 3, Nicolay C 4 1ICON Clinical Research UK Ltd., Oxford, UK, 2Eli Lilly and Company, Windlesham, UK, 3Eli Lilly and Company, Indianapolis, IN, USA, 4Eli Lilly and Company, Bad Homburg, Germany . . . . . . . . . . . . Objectives: To demonstrate the use and interpretation of alternative approaches to evaluate and understand inconsistency within network meta-analyses (NMA). Methods: A network meta-analysis was performed to compare the efficacy of once weekly dulaglutide 1.5mg to licensed doses of liraglutide, exenatide, lixisenatide, and albiglutide for the treatment of type 2 diabetes. The primary endpoint was reduction in HbA1c. To allow for potential heterogeneity, networks were stratified by background therapy: add-on to metformin (MET) and add-on to metformin in combination with sulfonylurea and/or thiazolidinediones (MET±SU±TZD). Networks were further stratified by time of evaluation (16-36 weeks and 37-56 weeks). The ‘node-splitting’ approach and models including treatment-by-design interaction effects (where ‘design’ refers to the study comparator set), were used to examine inconsistency. Results: For the 37-56 week add-on to MET network there were no trials linking to dulaglutide 1.5mg; therefore, a NMA could not be conducted. For the 37-56 week combination network there were no comparisons for which there were both direct and indirect evidence. For the 16-36 week add-on to MET network, there were no comparisons with statistically significant inconsistency in HbA1c reduction in the node splitting model and no significant design by treatment interactions. For the 16-36 week add-on to Met±SU±TZD network there was statistically significant variation between the direct and indirect estimates for the comparison between glargine and placebo and the comparison between glargine with liraglutide 1.80mg. Further sensitivity analyses were conducted removing the LEAD 5 trial (comparing placebo, glargine and liraglutide 1.80mg) from the network. Conclusions: The node-splitting and treatment-by-design interaction models showed no evidence of inconsistency in the 16-36 week monotherapy network. Following removal of the LEAD 5 trial, the 16-36 week combination network also showed no evidence of inconsistency. The analyses were useful in investigating potential inconsistency across the network. PDB20 Management of Type 2 Diabetes Mellitus Among Patients Attending a Primary Health Care Setting in Qatar: A Study on Medication Use Pattern and Clinical Outcomes Mohamed Ibrahim M I 1, Awaisu A 1, Berzou S 1, Abu Samaha R 1, Elshami S 1, Elhafiz M 2, Babiker I 2 1Qatar University, Doha, Qatar, 2Primary Health Care, Doha, Qatar . . . . . . . . Objectives: To evaluate the pattern and clinical outcomes of insulin therapies compared to oral antidiabetic therapies in the management of T2DM. Methods: A retrospective cohort study of patients with T2DM attending the Mesaimeer Primary HealthCare Center in Doha, Qatar. Inclusion criteria included were adult patients with at least with T2DM for 12 months period. Exclusion criteria included were age below 18 yrs, T2DM history less than 1 year, pregnant women and patients with kidney/hepatic disease. Patients’ medical records were reviewed for a period of one year. Data collected included socio-demographic profiles, clinical and laboratory data, medication regimens, and regimen cost per year. Data were analyzed descriptively and using Chi-Square test with priori alpha level of 0.05. Results: A total of 295 patients with T2DM attending the primary health center were included in the study. Majority of the patients were male (n= 210, 71.2%), age between 25-64 years old (n= 241, 82.3%) and obese (n= 163, 55.3%). Metformin was the most frequently used oral anti-diabetic medication (89.8%), followed by gliclazide (30.6%) and sitagliptin (26.2%). 32.3% of the patients were on insulin at the end of the 12 month follow-up period. The most common treatment regimen at both time points was oral dual therapy (28.3%), followed by oral monotherapy (21.3%) and triple oral therapy (20.2%). The mean HbA1c was 8.0+1.6% at both the beginning and endpoint of the 12 month follow-up period, indicating uncontrolled DM. Those patients receiving insulin-containing therapy had a significantly higher proportion of uncontrolled DM than those who did not receive insulin (98% vs. 81%; p< 0.001) at the endpoint of 12-month follow-up period. Conclusions: The findings of this study showed that T2DM patients attending primary health care clinics in Qatar were not achieving glycemic control. PDB21 Preventing the Progression to Type 2 Diabetes Mellitus in Adults at High Risk: A Systematic Review And Network Meta-Analysis of Lifestyle, Pharmacological and Surgical Interventions Stevens J W 1, Harvey R C 1, Johnson M 1, Khunti K 2 of Sheffield, Sheffield, UK, 2University of Leicester, Leicester, UK . . . . . . 1University Objectives: Individuals with impaired fasting glucose (IFG) or impaired glucose tolerance (IGT) have an increased risk of progression to Type 2 diabetes mellitus. The objective of this study was to quantify the effectiveness of lifestyle, pharmacological and surgical interventions in reducing the progression to Type 2 diabetes mellitus in people with IFG or IGT. Methods: A systematic review was carried out and a Bayesian network meta-analysis of log-hazard ratios was performed. The primary outcome of the network meta-analysis was the time to progression to Type 2 diabetes mellitus. Results are presented as hazard ratios and the probabilities of treatment rankings. Results: 30 studies were included in the network meta-analysis. There was a reduced hazard of progression to Type 2 diabetes mellitus associated with all interventions versus standard care. The most effective interventions compared to standard care were diet plus pioglitazone (HR 0.17, 95% CrI [0.09, 0. 33] ), glipizide (HR 0.16, 95% CrI [0.02, 1.62] ), diet plus exercise plus metformin plus rosiglitazone (HR 0.20, 95% CrI [0.11, 0.39] ), diet plus exercise plus orlistat (HR 0.31, 95% CrI [0.16, 0.61] ) and diet plus exercise plus pedometer (HR 0.35 95% CrI [0.11, 1.14] ). The least effective intervention was ramipril (HR 0.91, 95% CrI [0.72, 1.14] ). Conclusions: Pharmacological and lifestyle interventions are beneficial in reducing the risk of progression to Type 2 diabetes mellitus. Lifestyle interventions require significant behaviour changes and this may be achieved through incentives such as the use of pedometers. Lifestyle interventions alone, whilst beneficial, are unlikely to be as effective as pharmacological interventions alone or in combination with lifestyle interventions. Adverse events and costs of pharmacological interventions should be taken into account when considering potential risks and benefits, and their cost-effectiveness relative to lifestyle interventions. PDB22 New Meta-Analysis of Patient-Level Data on Efficacy And Hypoglycaemia with Insulin Glargine or Nph Insulin in Type 2 Diabetes Mellitus (T2DM) According to Concomitant Oral Therapy Owens D R 1, Trayor L 2, Landgraf W 3, Mullins P 4 of Life Sciences, Swansea University, Swansea, UK, 2Sanofi US, Inc., Bridgewater, NJ, USA, 3Sanofi, Frankfurt, Germany, 4Department of Statistics, University of Auckland, Auckland, New Zealand . . . . . 1Institute Objectives: Previous meta-analyses show inconsistent results regarding benefits of insulin glargine (GLA) over NPH insulin (NPH) in terms of nocturnal hypoglycaemia. We analysed standardized efficacy and safety outcomes in uncontrolled insulin-naïve subjects with T2DM treated with GLA or NPH according to the oral antidiabetic drug (OAD) to which insulin was added (sulfonylurea [SU] ± metformin [MET] ). Methods: Patient-level data from 4 Treat-To-Target (TTT) RCTs (FPG < 100 mg/dL) of ≥ 24 weeks duration were pooled. HbA1c, weight, dose, and hypoglycaemia (overall and nocturnal, plasma glucose < 56 mg/dL) were assessed. Results: 2,091 subjects were analysed; 49% male, mean age 57.8 years. SU-treated subjects had longer T2DM duration, higher baseline HbA1c, and were less obese than MET+SUtreated subjects. Endpoint HbA1c values were similar for GLA and NPH overall (7.4 vs 7.5%). Subjects adding GLA or NPH to MET+SU had numerically lower weightadjusted endpoint insulin doses (GLA 0.40 vs 0.44 U/kg; NPH 0.36 vs 0.42 U/kg), with less weight gain (GLA: +1.9 vs +3.7 kg; NPH: +1.8 vs +3.0 kg) versus subjects adding insulin to SU. Hypoglycaemia incidence (overall: odds ratio [OR] 0.79 [95% CI 0.66−0.95]; P= 0.011, nocturnal: OR 0.64 [0.51−0.81]; P< 0.001) and event rates (overall: rate ratio [RR] 0.78 [0.65−0.93]; P= 0.006, nocturnal: RR 0.54 [0.42−0.69]; P< 0.001) were significantly lower in GLA-treated subjects versus NPH, irrespective of concomitant OAD. In general, higher hypoglycaemia incidences/rates were observed in the MET+SU groups versus the SU only groups. MET+SU-treated subjects were more likely to achieve HbA1c < 7.0% without overall hypoglycemia versus SU-treated subjects (GLA: 24.6 vs 16.1%; NPH: 24.2 vs 13.0%). Conclusions: In insulin-naive T2DM patients from 4 TTT RCTs, GLA significantly reduced overall and nocturnal hypoglycemia risk versus NPH, irrespective of concomitant OAD. Subjects adding GLA or NPH to MET+SU had better efficacy outcomes than those adding to SU only, with slightly increased hypoglycaemia risk. PDB23 Efficacy and Safety of Dipeptidyl Peptidase-4 Inhibitors: Systematic Review and Meta-Analysis Pérez A 1, Franch J 2, Fuster E 3, Paz S 4, Prades M 4, Granell M 3 1Hospital de la Santa Creu I Sant Pau, Barcelona, Spain, 2EAP Raval Sud- Institut Català de la Salut - USR Barcelona ciutat - IDIAP Jordi Gol, Barcelona, Spain, 3Novartis Farmaceutica, Barcelona, Spain, 4Outcomes’10, Castellon, Spain . . . . . . Objectives: To determine the efficacy and safety of dipeptidyl peptidase-4 (DPP4) inhibitors in type 2 diabetes patients according to published data. Methods: A systematic review of randomized clinical trials (RCT) in MEDLINE, Cochrane, ISI WOK, SCOPUS and clinicaltrials. gov databases was performed. Eligible studies were RCT with a treatment duration of at least 24 weeks evaluating efficacy (HbA1c, fasting plasma glucose-FPG- and weight variation from baseline) and/or safety (hypoglycemia rate) of DPP4 inhibitors (linagliptin, saxagliptin, sitagliptin, vildagliptin) compared with placebo or non-insulin monotherapy or combination, published in English or Spanish until June 2013. A meta-analysis was conducted using a random effects model. Standardized mean difference (SMD) for efficacy variables and relative risk (RR) for safety variable with 95% confidence intervals (CI) were calculated. Results: Of the 4,582 publications retrieved, 3,807 were deleted by duplicate review, 581 by title/abstract review and 139 by criteria compliance. Finally, 55 RCT were selected. DPP-4 inhibitor monotherapy was associated to greater reductions in HbA1c and FPG compared with placebo [SMD= -0.60, 95% CI= -0.75; -0.46 and SMD= -0.51, 95% CI= -0.62; -0.39, respectively] while compared with metformin the reductions were lower [SMD= 0.28, 95% CI= 0.20; 0.26 and SMD= 0.36, 95% CI= 0.27; 0.44, respectively]. DPP-4 inhibitors added to metformin lowered HbA1c and FPG significantly more than metformin monotherapy [SMD= -0.52, 95% CI= -0.62; -0.41 and SMD= -0.41, 95% CI= -0.51; -0.30, respectively], and achieved a greater decrease in weight and hypoglycemia risk versus sulfonylurea plus metformin [SMD= -0.55, 95% CI= -0.64; -0.45 y RR= 0.16, 95% CI= 0.11; 0.21, respectively]. Moreover, the addition of DPP-4 inhibitors to sulfonylurea showed a greater reduction in HbA1c compared with sulfonylurea monotherapy [SMD= -0.54, 95% CI= -0.70; -0.37]. Conclusions: DPP-4 inhibitors added to metformin achieved a better glycemic control compared with metformin mono- A336 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 therapy, with a lower risk of hypoglycemia and without affecting weight versus sulfonilurea and metformin combination. PDB24 Recombinant Growth Hormone Therapy in Children with GH Deficiency: First Interventional Study in Armenia Markosyan R , Perikhanyan A Yerevan State Medical University, Yerevan, Armenia . . Objectives: The purpose of this study was to evaluate the effectiveness and safety of treatment with recombinant growth hormone (RGH) in children with GH deficiency. Methods: This was an interventional study with 6 and 12 months followup. Treatment was received by 15 children. The patients were receiving the RGH in 0.033 mg/kg (0.1 Unit/kg) at the same time each day (9-10 pm) for period of 1 year. The effectiveness of treatment was evaluated based on change in growing speed, growth SDS and bone age maturation. Results: The mean age of children was 9.5±3.6 years. In the given sample 6 children had MHPD, other 9 children had IGHD. There was a great improvement in absolute growth at 6 and 12 months period of treatment (p-0.001; p-0.001). The same was found for growth SDS (p-0.001; p-0.001). Effectiveness of RGH therapy on bone age maturation also showed great improvement (p-0.001; p-0.001). The level of IGF-1 was increased (p-0.001; p-0.001); at 12 months the level of IGF-1 reached to 248.72±70.7 ng/mL and remained consistently high. The same improvement was in IGF-pb3 levels (p-0.001; p-0.001). The lipidemic analysis showed that the blood cholesterol levels were from 3.21 to 12.39 mmol/L (norm 5.68±1.55 mmol/L) and the level of LDL - 1.3 to 10.86 mmol/L (norm 3.83±1.44 mmol/L). During the treatment period we observed the significant improvement in cholesterol levels (p-0.001; p-0.001). High density lipoprotein and triglyceride levels did not change significantly (p. 0.05). Conclusions: It can be concluded that the treatment with RGH in patient with GH deficiency is beneficial as it normalized the levels of cholesterol and LDL. During the treatment there were no any changes in indicators of kidney’s function, indicators of liver’s function as well as the indicators of carbohydrate metabolism. PDB25 Effectiveness, Safety and Patients’ Subjective Feelings of Insulin Pen-Needle: A Systematic Review Sun L , Sun J , Chen B , Sun S , Wang R , Ren X Shenyang Pharmaceutical University, Shenyang, China . . . . . . Objectives: To compare the differences of effectiveness, safety and patients’ subjective feelings for using different lengths of insulin pen-needle in diabetic patients. Methods: A retrospective analysis of relevant publications that were identified via electronic searches of databases using multiple search terms related to insulin pen-needle. Results: Totally, 21 literatures were included. Firstly, for the effectiveness, 85.71% of the studies suggested that there was no difference between longer and shorter needle in controlling HbA1c, 14.29% thought the shorter needle was better than the longer. No changes were observed with respect to fructosamine, glycated albumin and body mass index. Secondly, about the safety, all of the studies proved that the shorter needle was better in intramuscular injections, adverse device effects, subcutaneous lipodystrophy and barb phenomenon. 33.33% reported less hypoglycemic events, bleeding, bruising and needle bending with the shorter needle, the others showed no difference. All of the studies considered the shorter needle was undifferentiated with the longer in the needle break, hyperglycemia and lipohypertrophy. 6.25% have pointed out that the shorter needle was better than the longer in leakage, while 81.25% showed no difference in the length. Thirdly, in terms of subjective feelings, for convenience and acceptance, all studies agreed that shorter needle was superior to the longer. For fear and pain, half of studies suggested that shorter needle was superior to the longer one; the other half thought that there was no difference. In all the studies, 69.23% suggested patients prefer the shorter, 23.08% suggested the patients not prefer a particular needle length. Conclusions: Overall, the effectiveness of insulin pens with longer and shorter needle are comparable in treating diabetes, but the shorter needle is little better in parts of the safety indexes. As for patients’ subjective feeling, our findings show that patients are generally willing to accept shorter needle. PDB26 Economic Impact of Combining Metformin with Dipeptidyl Peptidase Inhibitors In Diabetic Patients With Renal Failure PDB27 Uses of Electronic Patient Information Systems and National Registers – Implementation of the Clinical Practice Guideline and Evaluation of Costs and Use of Resources in Patients with Incident Type 2 Diabetes in Finland Prami T 1, Sulamaa A 2, Sipilä R 3, Linna M 4, Hahl J 5, Miettinen T 5, Leppä E 6, Haukka J 1, Tuomilehto J 7, Enlund H 8, Niskanen L 8, Korhonen P 1 1EPID Research, Espoo, Finland, 2Pharma Industry Finland, Helsinki, Finland, 3Finnish Medical Society Duodecim, Helsinki, Finland, 4Aalto University, HEMA Institute, Espoo, Finland, 5AT Medical Affairs Consulting, Espoo, Finland, 6Pharmaceutical Information Centre, Helsinki, Finland, 7University of Helsinki, Helsinki, Finland, 8Finnish Medicines Agency Fimea, Helsinki, Finland . . . . . . . . . . Objectives: Effective management of diabetes is the cornerstone for prevention of diabetic complications. However, how well the Finnish Current Care guideline for diabetes is implemented in practice is unknown. Combining local and nationwide patient registers provide a valuable resource for evaluating risks, benefits and costs. The purpose of this study was to identify how the Finnish electronic patient information systems and national registers can be used to explore the treatment for patients with incident type 2 diabetes. Methods: Selected primary and specialty care organizations representing different geographical areas and patient information system providers were invited to participate in the study. Study permits were obtained from several local and nationwide register holders. The study protocol was reviewed by the Ethical Review Board of Hospital District of Helsinki and Uusimaa. Results: Register linkage is accomplished using unique personal identification numbers. We collect nationwide data on prescriptions, hospital and primary care, reimbursed dental care, and the causes of death. Cost data are based on hospital benchmarking database, sickness allowances and rehabilitations. We use local registers as a source of information on diagnoses, medical procedures, prescriptions and contact types. High quality laboratory data are also included from several local providers. Conclusions: Register linkages enable longitudinal follow-up of patients for research purposes in Finland. In our study a unique combined register database of diabetic patient cohort is created that improves the evaluation of prognosis and care of diabetic patients. This is a promising and versatile source for research in pharmacoepidemiology. PDB28 Epidemiology and Unmet Medical Need in Diabetes Mellitus Type 2 in Germany –Results of a Literature Search Greiner R A 1, Paulus G 1, Driedger J 1, Schröder C 1, Timtschenko V 1, Batscheider A 1, Eheberg D 2, Vosgerau S 1 1IMS Health, Munich, Germany, 2IMS HEALTH, Munich, Germany . . . . . . . Objectives: Diabetes mellitus Typ2 (T2DM) is a metabolic disease characterized by hyperglycemia with a high risk-potential of microvascular and macrovascular complications. In addition to glycemic control important therapy targets are the prevention of hypoglycemia and weight gain as well as blood pressure control due to national guidelines (German Medical Association 2013). To describe the current state of T2DM epidemiology and therapeutic needs in Germany which is mandatory when submitting AMNOG dossiers. Methods: To describe epidemiology of diabetes a targeted literature research was conducted in PubMed in 2014 using the search terms (epidemiology OR incidence OR prevalence). To identify relevant comorbidity information the following terms were used (metabolic syndrome OR glycemic control OR hypoglycemia OR obesity OR blood pressure) and combined with AND diabetes AND Germany. PubMed research was supplemented by additional searches in guidelines in German/English. Results: The screening of the epidemiologic results identified nine relevant publications: two specified a T2DM-prevalence of 15.3% and 14.7% (Wittchen et al 2007, Huppertz et al 2009) and two studies estimated a T2DM incidence of 15.8 per 1000 patient years (KORA, MONICA). Treatment prevalence increased from 5.9% in 1998 to 8.9% in 2007 related to the total population (Hauner 2013). Arterial hypertension was the most frequent comorbity (83%) of T2DM (Hagen et al. 2010). In 2010, a disease management program in North Rhine showed that only 15% of participants with T2DM achieved a BMI < 25 (Hagen et al. 2010). Long-term trials investigating the efficacy of antidiabetics on the prevention of macrovascular complications are limited (Drug Commission of German Medical Association 2009; Matthaei et al. 2009). Conclusions: While treatment prevalence is increasing and glycemic control seems to be sufficiently achieved a substantial unmet medical need is identified for antidiabetics with a significant effect on weight reduction and blood pressure control in patients with T2DM in Germany. PDB29 Progression of Physiological Parameters Over Time in Type 1 Diabetes Mellitus Patients in France Objectives: To evaluate resource use and health costs due to the combination of metformin and dipeptidyl peptidase-4 (DPP-4) inhibitors in patients with diabetes and renal failure (RF) in routine clinical practice. Methods: An observational, retrospective study was performed. Patients aged ≥ 30 years treated with metformin who initiated a second oral antidiabetic treatment in 2008-2009 were included. Two groups of patients were analysed: a) metformin + DPP-4 inhibitors and b) other oral antidiabetics. The main measures were: compliance, persistence, metabolic control (glycosylated haemoglobin < 7%) and complications (hypoglycemia, cardiovascular events) and total costs. Patients were followed up for two years. Results: We included 395 patients, mean age 70.2 years, 56.5% male: 135 patients received metformin + DPP-4 inhibitors and 260 patients received metformin + other oral diabetics. Patients receiving DPP-4 inhibitors showed better compliance (66.0% vs. 60.1%), persistence (57.6% vs. 50.0%) and metabolic control (63.9% vs. 57.3%), respectively, compared with those receiving other oral diabetics (p < 0.05), and also had a lower rate of hypoglycemia (20.0% vs. 47.7%) and lower total costs (€ 2,486 vs. € 3,002), p = 0.001. Conclusions: Despite the limitations of the study, patients with renal failure treated with DPP-4 inhibitors had better metabolic control, lower rates of hypoglycaemia, and lower health costs for the Spanish national health system. Beaudet A , Ong R C IMS Health, Basel, Switzerland . . . . Sicras-Mainar A 1, Navarro-Artieda R 2 Serveis Assistencials, Badalona. Barcelona, Spain, 2Hospital Universitari Germans Trias i Pujol, Badalona, Spain 1Badalona . . . . . Objectives: The objective of this study was to understand the progression over time of physiological parameters, including HbA1c, body mass index (BMI), systolic blood pressure (SBP), total cholesterol, LDL-cholesterol, HDL-cholesterol and triglycerides, in type 1 diabetes mellitus (T1DM) patients to inform disease modeling. Methods: This was a cross-sectional analysis of T1DM patients based on the IMS LifeLink Diabetes Cohort in France, which prospectively collects clinical, biological and treatment information from general practitioners. Patient age, gender, year of diagnosis, BMI, HbA1c, cardiovascular risk factors, renal function and lab test results were collected at baseline and subsequent visits. Data were analyzed using R Studio. T1DM patients who visited their general physician between May 2011 and May 2014 and have received at least one insulin prescription were included in the analysis. Results: A cohort of 605 T1DM patients was included in this analysis. Forty-three percent of patients were male. Average patient age at first visit was 58 years of age. Mean HbA1c was 7.8%, mean SBP was 132 mmHg, and mean BMI was 27.6 kg/m2. Linear regression showed that BMI increased by 0.092 kg/m2 (p< 0.001) for each additional year of age. SBP was projected to increase by 0.248 mmHg (p< 0.001) per additional year of age, LDL-cholesterol decreased by A337 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 0.624 mg/dL for each additional (p= 0.017) year of age and triglycerides increased by 1.417 mg/dL for each additional (p= 0.041) year of age. Changes in HbA1c, total cholesterol and HDL-cholesterol over time were not significantly correlated with patient age or time since diagnosis. Therefore, health economic modelers may assume that these parameters remain stable over time. Sensitivity analyses were performed to address the potential mislabeling of T2DM as T1DM patients. Conclusions: These results provide relevant inputs for the progression of physiological parameters to model the economic and clinical impacts of T1DM therapies over time. PDB30 Trend in Prevalence and Distribution of Diabetes Mellitus Type I and Type II in the Netherlands Overbeek J A 1, Penning F J A 1, van Dongen E 2, Herings R M C 1 Institute for Drug Outcomes Research, Utrecht, The Netherlands, 2Novo Nordisk B.V., Alphen a/d Rijn, The Netherlands . . . . . . . . . medications and obstetric medical history. Set-up was non-random, convenience sampling with 40 women’s data with mean age 33.7± 3.6, between August 2013 – February 2014. Friedmann ANOVA and t-test was applied for analysis with software Statistics for Windows. Results: Progresses the gestation period significant reduction of TSH values was observed in hypothyroid patients (p= 0,0075). Comparing successive TSH values of the individual patients showed the same significant difference. Significant negative correlation was founded between TSH value and thyroxin dose in group of hypothyroid patients (r= -0,35; p< 0,05). Premature birth and other obstetric complications occured more frequently in the thyreotoxical group, especially among older women giving birth. Conclusions: In case of hypothyroid pregnant with increasing dose of thyroxin the TSH levels are well balanced, and obstetric complications did not occur, while in hyperthyroid patients can be reported obstetric complications in addition to proper care. 1PHARMO Objectives: To quantify the trend in prevalence and distribution of diabetes mellitus (DM) type I (T1DM) and type II (T2DM) in the Netherlands. Methods: Using the General Practitioner Database and the Out-patient Pharmacy Database of the PHARMO Database Network, the trend in prevalence of DM and distribution of T1DM and T2DM from 2005 to 2012 was assessed. Per year, patients with ≥ 2 antidiabetic drug dispensings within 6 months were selected as DM patient. Patient numbers were extrapolated to the Netherlands to determine prevalence of DM. For all patients, diabetes treatment at September 30 of that year was assessed. For patients with a GP recorded diagnosis for T1DM or T2DM, distribution of T1DM/T2DM was stratified by treatment. This distribution of DM type by treatment was applied to the treatment of patients with no GP recorded DM type to assess the distribution of T1DM/T2DM. Results: The prevalence of DM in the Netherlands increased from 38 per 1000 males and 40 per 1000 females in 2005 to 54 per 1000 males and 52 per 1000 females in 2012. The distribution of T1DM versus T2DM among patients with DM changed from 15% versus 85% in 2005 to 8% versus 92% in 2012. Among patients with T1DM mean (±SD) age decreased from 48 (±22) years in 2005 to 44 (±22) years in 2012. Among patients with T2DM mean age increased from 63 (±12) years in 2005 to 67 (±12) years in 2012. Conclusions: This study describes the epidemiology of DM in the Netherlands over 2005-2012. Prevalence of DM increased and relatively more patients were diagnosed with T2DM. These changes can be explained by the ageing Dutch population, better survival, more obesity and early detection of T2DM. Furthermore, introduction of the T2DM care program in 2005 probably has led to a better registration of T2DM patients. PDB31 Factors Associated with Hospitalization of Type 2 Diabetic Patients with Hypoglycemic Episodes Assisted at Emergency Departments Conceição J 1, Laires P 1, Araújo F 2, Dores J 3, Vicente V 4, Silva C 4, Carr R 5, Brodovicz K 6, Radican L 7, Nogueira A M 1 Sharp & Dohme, Oeiras, Portugal, 2Hospital Beatriz Ângelo, Loures, Portugal, 3Hospital de Santo António, Porto, Portugal, 4Eurotrials, Lisbon, Portugal, 5Merck Sharp & Dohme, Ballerup, Denmark, 6Merck & Co., North Wales, PA, USA, 7Merck & Co., Inc., Whitehouse Station, NJ, USA . . . . . . . . . . . Diabetes/Endocrine Disorders – Cost Studies PDB33 Budget Impact Analysis of Adding Dapaglifozin To The Therapy of Diabetes Mellitus Type 2 In Bulgaria Manova M 1, Petkova E 1, Yordanova S 1, Petkova V 1, Petrova G 2 1Medical University of Sofia, Faculty of Pharmacy, Sofia, Bulgaria, 2Medical University Sofia, Faculty of Pharmacy, Sofia, Bulgaria . . . . . Objectives: Dapaglifozin is a highly potent, selective and reversible inhibitor of sodium-glucose co-transporter 2 (SGLT2) and is approved for the treatment of T2DM in adults. Diabetes type 2 is one of the most prevalent chronic diseases that can lead to serious complications and disability. The largest costs are those associated with hospitalizations due to the complications, the prevention of which requires a good glycemic control. The objective of the study is to estimate the budget impact of adding dapaglifozin to the therapy of type 2 diabetes in Bulgaria. Methods: The budget impact model was used from the payer perspective for population 7 284 552 people, and out of them 450000 are type 2 diabetics. The retail pharmacy prices were used from the Positive Drug List. Official IMS data for antidiabetic medicines were incorporated in the model. Net budget impact is presented as costs per-member per-month (PMPM) and costs per-patient per-year (PPPY). Results: An increase in the estimated net budget impact from 70 592 € first year to 1 290 716 € for the fifth year was observed after adding dapaglifozin to T2DM therapy, with a cumulative net budget impact of 3 258 047 € . PMPM and PPPY costs show minimal growth with respective cumulative values of 0.06 € and of 65.63 € . The cost for dapaglifozin therapy is comparable to that of DPP-4 inhibitors and is lower than the cost of treatment with a GLP- 1. Conclusions: The results show that adding dapaglifozin to standard therapy will lead to minimal increase in the diabetes type 2 budget in Bulgaria. This increase is considered acceptable in terms of better glycemic control with safe and effective therapy for diabetes type 2. 1Merck Objectives: HIPOS-ER is an observational, cross-sectional, multicenter study to describe the patient population of Type 2 diabetics treated with an anti-hyperglycemic agent (AHA) and admitted to the emergency room (ER) with a hypoglycemic event. In this analysis we aim to identify factors associated with hospitalization following admission in the ER. Methods: The study enrolled patients from 7 centers in mainland Portugal for a period of 12 months (Jan 2013 – Jan 2014). Sociodemographic and clinical data were collected at the ER and patients who required hospital admission were followed up. Multiple logistic regression was used to identify factors associated with hospitalization. Results: A total of 238 patients were enrolled of whom 105 (44%) were hospitalized. These patients were older than non-hospitalized (mean: 78 years vs. 75 years; p= 0.020). About half of the hospitalized patients were on secretagogue based regimen. Glasgow coma scale score was not statistically different between hospitalized and non-hospitalized patients (p= 0.270), however hospital admissions showed lower values of the lowest recorded plasma/capillary glycemia (mean: 35 mg/dL vs. 41 mg/dL; p= 0.004) and a higher prevalence of hypoglycemia complications (26% vs. 9%; p< 0.001). Obesity and higher plasma capillary glycaemia level are associated with a lower risk of hospitalization (OR= 0.29 and OR= 0.96, respectively; p< 0.010) while treatment with secretagogue based regimen, presence of hypoglycemia complications and other atherosclerotic disease are associated with a higher risk (OR= 5.71, OR= 3.89 and OR= 2.87, respectively; p< 0.010). Conclusions: Almost half of the patients with diabetes suffering a hypoglycemic event who required medical assistance at the ER were hospitalized. Presence of some factors may increase the risk of these expensive cases which are also surrogates for the potential severity of the hypoglycemic episodes. Identification of these predictors may help assist physicians at the emergency room to proactively act upon patients at higher risk of hospitalization and generate substantial health and economic gains to the hospitals. PDB32 The Impact of Treatment of Thyreoid Disease in Pregnant Women to the Outcome of Giving Birth Ferenczy M 1, Póhr K 1, Lőcsei Z 2, Oláh A 3, Boncz I 4, Karácsony I 1, Salamonné Toldy E 2 of Pécs, Szombathely, Hungary, 2Markusovszky University Teaching Hospital, Szombathely, Hungary, 3University of Pécs, Pécs, Hungary, 4Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . PDB34 Assessment of The Economic Value Of Dpp-4 Inhibitor Alogliptin Compared With Sitagliptin, Saxagliptin, And Linagliptin Pedrazzoli M 1, Pasquini F 1, Minda K 2 1LSC Lifesciences Consultants, Milan, Italy, 2Takeda Pharmaceuticals, Glattpark-Opfikon (Zurich), Switzerland . . . Objectives: Objective of this study is to provide additional evidence for decision making to payers assessing health care resource utilization, economic impact, and cost-effectiveness of DPP-4 inhibitor alogliptin compared with sitagliptin, saxagliptin and linagliptin, for the treatment of type-2 diabetes mellitus (T2DM). Methods: 29 comparable, randomized clinical trials were selected out of a panel of 58 studies. 6 different clinical endpoints (efficacy and safety) were compared across 5 different combinations: DPP-4 Monotherapy, +metformin (MET), +sulfonylurea (SU), +thiazolidinedione (TZD), +insulin (INS). For each endpoint and combination, alogliptin clinical endpoints were compared with respective average endpoints of other DPP-4s. Differentials were calculated after adjustment for baseline characteristics. Each endpoint was associated with the impact on patient outcomes and related health care costs (T2DM-related complications, treatment escalation, costs associated with adverse events: hypoglycemia, cardiovascular mortality, hospitalization due to heart failure, lipids profile) obtained from published data. Economic value saving of alogliptin was calculated and compared to the other DPP-4s. Results: The proportion of patients at target (HbA1c< 7%) as well as the reduced need for treatment escalation with alogliptin could generate annual savings for a health care system of € 69.62 and € 22.97 per patient-year, respectively. Improved lipids profile and proven CV safety of alogliptin can generate savings of € 40.86 and € 21.47 per patient-year, respectively. Impact of lower hypoglycemia and increased adherence with fixed dose combinations with TZD may generate additional savings (€ 1.53 and € 1.60/ patient-year, respectively). Conclusions: This study suggests that alogliptin could generate significant savings for a Healthcare System, even at price parity with other DPP-4s, thanks to its efficacy and safety profile, particularly in the widely used DPP-4+MET combination. Total savings of up to € 158 per patient-year compare favorably with an overall cost of treatment with a DPP-4i ranging from € 350 to € 481 per patient-year. . 1University PDB35 Cost-Effectiveness Analysis of Autocoded and Manually Coded Blood Glucose Meters In Diabetes Treatment Objectives: The most common endocrine clinical symptom is thyroid disease which has impact to pregnant women and fetus. Leading international references about its treatments are well known, there is no relevant experiences in the Hungarian context. Aim of this study is to inspect the thyroid disease and impact of its treatment to outcome of pregnancy. Methods: Survey was carried out at Markusovszky Hospital in Vas County, Hungary. Retrospective study made by data analysis of pregnant patient with thyroid disease, including hormone parameters, Yagudina R , Kulikov A , Babiy V I. M. Sechenov First Moscow State Medical University, Moscow, Russia . . . Objectives: To conduct a cost-effectiveness assessment of glucose meters (manually coded and autocoded) in diabetes treatment in the Russian Federation. Methods: Clinical effectiveness assessment was based on the results of modeling of the treatment of patients with diabetes that use manually coded and autocoded blood glucose meters. Cost analysis included assessment of direct and indirect costs that can be A338 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 associated with manually coded meters (model 1 and model 2) and autocoded meters (model 3). Cost data was based on median prices for medicines, devices, medical services in National health care system in the Russian Federation. Results: Annual direct costs per patient in the group of manually coded glucose meters were 1533 euro (model 1) and 1574 euro (model 2), and in the group of autocoded meters were 1557 euro (model 3). Annual total costs per patient in the group of manually coded glucose meters were 2992 euro (model 1) and 2921 euro (model 2), and in the group of autocoded meters were 3034 euro (model 3). Cost-effectiveness ratio for autocoded meters was 1875 euro (model 3) and for manually coded meters was 1954 euro (model 1) and 1982 euro (model 2) per 1 LYG, respectively (discounted at 3%). Budget impact analysis showed that use of autocoded meters (model 3) instead of manually coded (model 1 and model 2) leads to the annually cost savings of 33 euro and 53 euro per patient, respectively (discounted at 3%). Conclusions: Obtained results approve the use of autocoded blood glucose meters instead of manually coded blood glucose meters to administrate blood glucose level as part of intensive glucose-lowering therapy from a pharmacoeconomic point of view. PDB36 Health Economic Impact of Bariatric Surgery Revisted: Structured Review of Literature and Health Technology Assessments Chawla A S 1, Tao C 2, Faulkner E C 3, Hsiao C W 4, Patkar A D 5, Romney M 6 1Quintiles Consulting, Durham, NC, USA, 2Quintiles Consulting, Cambridge, MA, USA, 3Institute for Pharmacogenomics and Individualized Therapy, Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, NC, USA, 4Johnson and Johnson Medical Companies, Markham, ON, Canada, 5Ethicon, Inc, Somerville, NJ, USA, 6Jefferson School of Population Health, Philadelphia, PA, USA . . . . . . . . . . Objectives: The costs of obesity are staggering, accounting for 2-6% of global health care costs. The health economic benefits of bariatric surgery while notable, are multi-faceted, resulting in heterogeneous reporting in the literature. To that end, this study seeks to 1) highlight available evidence of the health economic impact of bariatric surgery, and to 2) identify key gaps in current evidence that may influence uptake by health care systems. Methods: Evidence of the health economic impact was collated from 107 scientific articles, of which 19 were systematic reviews, published between 2010-March 2014 and archived in MEDLINE and PubMed. Additionally, HTAWatch identified 10 HTAs, largely from North America and the EU that evaluated economic benefits. Results: In all countries where evaluated, bariatric surgery was cost-effective compared to non-surgical therapy. The surgery was likely more cost-effective with higher patient BMI and with comorbidities, especially for BMI> 40kg/m2, although estimates varied. Furthermore, resolution of underlying comorbidities resulted in reduced utilization of health care services, pharmaceutical utilization, and improved work productivity, among others, resulting in cost saving of 60-70% relative to standard care over a 3-year period. Importantly, time to break even was typically in the 4-7 years range, and was shorter for patients with Type 2 Diabetes Mellitus (T2DM) and with BMI> 40kg/ m2, typically in the 1.25-5 year range. Three key evidence gaps were identified: few studies computed long-term cost-effectiveness; no head-to-head trials have directly compared different surgical procedures; and heterogeneity across populations and health system impede meta-analyses of patient outcomes such as QoL and longterm health benefits. Conclusions: Although heterogeneous, reports of health economic benefits of bariatric surgery indicate an overall positive trend, largely via reduction of health resource utilization. To firmly establish its impact, future studies need to conduct head-to-head comparisons, determine optimal patient populations, and employ standard clinical endpoints to demonstrate real world, long-term benefits. PDB37 Association of Changes In Body Weight With Health Care Costs Among Patients With Newly-Diagnosed Type-2 Diabetes In Sweden Sabale U 1, Bodegård J 1, Sundström J 2, Svennblad B 3, Östgren C J 4, Nilsson P 5, Johansson G 3, Henriksson M 1 1AstraZeneca Nordic-Baltic, Södertälje, Sweden, 2Uppsala University, Uppsala, Australia, 3Uppsala University, Uppsala, Sweden, 4Linköping University, Linköping, Sweden, 5Lund University, Malmö, Sweden . . . . . . . . . Objectives: Type 2-diabetes and excess weight incur large costs to health care systems, but the association between weight progression in diabetes and health care costs is unknown. We investigated those relations using real world data in a sample of newly diagnosed diabetes patients in Sweden by using repeated body mass index (BMI) measurement and health care resource utilization data Methods: Patients with a BMI (kg/m2) measure at diagnosis and subsequent BMI measures at 12,24,36,48, and 60 months were identified from a previously conducted register study. Individuals were classified into three groups based on their BMI change over 5 years: increase (> 1 BMI unit increase), decrease (> 1 BMI unit decrease), and stable (≤ 1 BMI unit change). Each group was stratified by BMI at diagnosis (BMI 18-25; 25-30; ≥ 30). Health care costs for each group were estimated by applying Swedish unit costs to the health care resource data extracted from electronic patient journals and a national patient register. Results: The study included 903 T2D patients (women, 43%; mean age, 62; mean HbA1c, 6.78%; mean BMI, 30.9). The BMI increased, decreased, or remained stable in 178 (20%), 387 (43%), and 338 (37%) patients, respectively. Among patients with baseline BMI 18-25 (n= 104), the five year cumulative health care costs were € 13,695, € 9,059, € 8,936, in the increase, decrease, and stable group, respectively. Corresponding figures were € 11,470, € 7,950, € 8,683 for patients with baseline BMI 25-30 (n= 321), € 14,387, € 9,465, and € 9,302 for patients with BMI ≥ 30 (n= 478). Conclusions: In newly diagnosed diabetes, an increase in BMI lead to increased health care costs, irrespective of baseline BMI. Linking registry data on repeated BMI measurements and health care utilization is a valuable approach to investigate the association between weight changes and costs. Costs of interventions that maintain weight in patients with diabetes should be considered in the context of costs associated with weight gain. PDB38 Effect of Smoking Status on Health Care Costs In Patients With Type 2 Diabetes: A Retrospective Nested Case-Control Economic Study In Routine Clinical Practice Sicras Mainar A 1, Rejas Gutiérrez J 2, Navarro Artieda R 3, Ibánez Nolla J 4, De Lossada Juste A 5 1Badalona Serveis Assistencials SA, Badalona (Barcelona), Spain, 2Pfizer S.L.U., Alcobendas/ Madrid, Spain, 3Hospital Germans Trias i Pujol, Badalona (Barcelona), Spain, 4Badalona Serveis Assistencials, Badalona, Spain, 5Pfizer, S.L.U, Alcobendas/Madrid, Spain . . . . . Objectives: Smoking in diabetics is associated with a worse prognosis and vascular complications. The available evidence on health care resources utilization and associated costs in diabetics who smoke is limited or nonexistent. Thus, the objective was to compare health care resource utilization and costs according to smoking status in patients with type 2 diabetes in clinical practice. Methods: A retrospective cohort nested case-control study was designed. Cases were current smokers, while two types of controls (former smokers and never smokers) were matched, two controls per case, for age, sex, duration of diabetes, and burden of comorbidity using data from electronic medical records. Non-institutionalized diabetics, both genders, age > 18 years, seen consecutively over a 5-year period before the index date were enrolled. Perspective of both the National Health System and the Society were chosen and costs of health care resource utilization and loss of productivity due to sick leaves were compared among groups using a linear general model with covariates. Results: A total of 2,490 records were analyzed: 498 from cases, 996 from former smokers, and 996 from never smokers. Mean age was 63.4 years (64.9% male). Smokers had higher HbA1c (7.4% vs. 7.2% and 7.2%, respectively, p = 0.013) and a lower degree of metabolic control (49.2 % vs. 54.7% and 55.8%, p = 0.036). Smokers had higher average annual costs (€3,583) than former smokers (€2,885) and never smokers (€2,183), p<0.001. Mean annual health care cost saving per patient was associated with elapsed time of quitting smoking from €950 in subjects quitting smoking 2-year to € 1173 5-year or more, p<0.05. Conclusions: Diabetic smoker patients had lower metabolic control, higher health resource utilization, and more sick leave, resulting in higher health care costs and lost productivity compared with both former and never diabetic smokers. Health care cost-savings was associated with elapsed time from quitting smoking. PDB39 Evaluation of Potential Waste Of Growth Hormone Across Available Growth Hormone Pen Devices And An Electronic Growth Hormone Delivery Device Locklear J 1, Edwards N 2, Phillips A L 1 Serono, Inc., Rockland, MA, USA, 2Health Services Consulting Corporation, Boxborough, MA, USA . . . . 1EMD Objectives: The aim of this analysis was to estimate the potential GH waste per patient with pen devices and the easypod® device, and to quantify the potential economic impact of expected GH waste from patient and health care organization perspectives. Methods: A Waste Calculator Model was developed to examine GH waste. All somatropin products available in pen or electronic devices were included. The user may define distribution across cartridges sizes. The mechanical/priming loss applied to each product was based upon each product’s prescribing information and/or instructions for use. The base case model utilizes a US patient daily dose of 1.4 mg. The model assumes that the easypod® dose adjustment feature is activated by the clinician (±25%). Model assumes that 42.6% of caregivers discard the remaining amount left in the cartridge (eg waste) if less than a full dose. Annual amount of GH waste (mg, cartridges, dollars) per patient and per population (based on US national market shares) for each pen/device is reported. Results: The expected annual amount of waste per patient was lowest for easypod®. The expected annual amount of waste per patient was highest for Omnitrope®. The expected annual amount of waste ranged from 0 to 38.9mg per patient per year, which is equal to 0 to 8 cartridges per patient per year and/or 0 to $2,935 per patient per year. For a patient population of 100 GH-treated patients, the annual amount of waste is estimated at 2,009 mg, which can be translated into approximately 342 cartridges or about $162,000 per year. The results in GH waste fluctuated depending upon daily dose, cartridge size, and dose spread assumptions. Conclusions: The expected annual amount of GH waste evaluated in this Waste Calculator was lowest with easypod®. Cost of GH waste can be an important consideration when evaluating GH delivery devices. PDB40 Benefit Of Positive Airway Pressure (Pap) Therapy In Sleep Apnoea (Sa) Patients With Type Ii Diabetes Mellitus (T2dm) In Germany: A Retrospective Comparative Cohort Analysis Based On A Statutory Health Insurance Database Doess A 1, Zucca F 2, Woehrle H 3, Brueggenjuergen B 4 1ResMed Germany Inc., Martinsried, Germany, 2HGC GesundheitsConsult, Duesseldorf, Germany, 3ResMed Science Center, Martinsried, Germany, 4Institut für Sozialmedizin, Epidemiologie und Gesundheitsökonomie, Charité - Universitätsmedizin Berlin, Berlin, Germany . . . . Objectives: It is estimated that the prevalence of moderate-to-severe SA (apnoeahypopnoea index > 15/h) is 10%. Patients with T2DM have a particularly high incidence of SA. T2DM and SA influence the development and progression of each other. This study investigated the effects of PAP therapy in SA patients with T2DM on all-cause mortality and cost of illness (COI) in Germany from a statutory health insurance (SHI) perspective. Methods: A total of > 4 million individuals covered by the SHI database were analysed (≈5% of the German SHI population). PAP therapy was initiated in 4,068 patients with SA (PAP group). Propensity score matching was used to define a control group (CG) of 4,068 SA patients matched for age, sex, risk factors/aetiology, region and medication who received usual care (no PAP). Of these, 1,280 patients in the PAP group and 1,186 patients in the CG had comorbid T2DM. This subgroup of patients was followed for 3 years after initiation of PAP therapy. Results: Total COI was higher in the PAP group versus CG in year 1 (€8,105 vs € 7,037, p< 0.0001). After 2 years’ follow-up, COI in the PAP group decreased but remained higher versus CG (€ 6,842 vs € 6,625, p< 0.001). After 3 years, PAP group COI A339 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 was significantly reduced versus CG (€ 6,798 vs € 6,954, p< 0.001). PAP recipients had a significantly lower 3-year mortality rate versus CG (7.2% vs 10.9%, p< 0.008; relative risk reduction 33.8%). Conclusions: SA patients with T2DM treated with PAP showed significantly reduced mortality and morbidity. Total COI was higher in PAP recipients versus CG in the first two years of follow-up, but during year 3 COI was significantly lower in the PAP group versus CG. In addition to the known advantages of PAP therapy in SA patients with T2DM (e. g. improved glycaemic control), PAP therapy may be beneficial from an economic perspective. PDB41 ARE TOTAL HEALTH CARE EXPENDITURES IMPACTED BY A NEW DIABETES DIAGNOSTIC FACTOR: HBA1C? Bhounsule P 1, Peterson A M 2 of the Sciences in Philadelphia, Philadelphia, PA, USA, 2University of the Sciences, Philadelphia, PA, USA . . . 22700 diabetic patients in Madrid was used to estimate transition probabilities in the model. Time horizon was 10 years. Prices were taken from the health care tariffs published by the Regional Health Service in Madrid and from Spanish literature. They were updated to 2013 price levels and adjusted to subsequent years by a 3% inflation rate. We used the perspective of the Regional Health Service in Madrid (public payer). Results: The model included an acute event and 3 health states. The increase in total costs in 10 years was € 1,328 per DM patient (133 € per year). In patients with hypertension the costs increased by 1,519€ in 10 years (152 € ); whereas the total increase in costs in patients with obesity was 1,535 € (153 €). Conclusions: IS has a relevant impact on the costs of management of DM. Hypertension and obesity increase these costs even further. Prevention of DM complications and an adequate control of risk factors can lead to cost savings for the management of DM. Understanding future costs of DM might be valuable for economic evaluations. 1University Objectives: To determine differences in total health care expenditures among general diabetics and newly diagnosed diabetics before and after HbA1c was implemented as the standard diagnostic factor. Methods: Medical Expenditure Panel Survey-Household component 2009 and 2011 databases were used. Annual health care expenditures formed the dependent variable. Demographic factors, comorbidities, prescription drug costs, medical events and utilization variables for health care services like visits to office-based, outpatient and inpatient facilities, emergency rooms, home health care for 2009 and 2011 formed the independent variables. Patients diagnosed within the 2009 and 2011 years, were categorized into a dichotomous new diagnosis variable. General linear regression was conducted to compare predictors of total diabetes health care expenditures in 2009 and 2011. Results: The mean total health care expenditure decreased in 2011 compared to 2009 among general diabetics {$10,901 (95%CI= $9,013-$11,836) vs $11,458 (95%CI= $10,715-$12,201)} and the newly diagnosed ones {$9,462 (95%CI= $6201$12,721) vs $8,429 (95%CI= $6793- $10,064)}. Among general diabetics in 2009, total expenditures showed significant relationships with prescription drugs, emergency room, home health care, outpatient facility and physician services expenses (all values p< 0.0001). The 2011 model was similar, but included associations between office-based provider visits (p< 0.0001) and previous incidences of heart attacks (p= 0.008). In 2009, among newly diagnosed patients, total expenditures were significantly associated with home health facilities expenditure (p< 0.0001) as opposed to no association in 2011. In both years, prescription medication, emergency room and hospital inpatient facilities expenses contributed to spending. Visits to office-based and outpatient department providers, number of homehealth provider days (all values p< 0.0001) and hospital discharges (p= 0.0011) impacted total expenditures in 2009 and 2011 respectively. Conclusions: Due to the many variables significantly affecting diabetes health expenditures, reduction in costs could not be solely attributed to the implementation of the HbA1c diagnostic criteria. Further research on cost-effectiveness of the HbA1c factor is warranted to establish any possible association. PDB42 Direct Costs Of Diabetes Mellitus In Poland Kawalec P Jagiellonian University Medical College, Krakow, Poland . . . . . Background: High prevalence and chronic complications are key determinants of management costs of Diabetes Mellitus (DM). Given that DM affects more than 4 million people in Spain, the costs associated with this disease require an efficient management. Cardiovascular diseases are complications associated with DM with a major impact in health care budgets. Objectives: Develop a model to predict future costs of DM in Spain. Estimate the increase in health care costs in DM patients associated with potential cardiovascular diseases. We used the perspective of the Regional Health Service in Madrid (public payer). Methods: Foro Gerendia is an initiative constituted by health care professional, experts in DM management. A Markov model for DM was developed based on the opinion of Foro Gerendia and on previous works in this area. A 5-year primary care data registry of 22700 diabetic patients in Madrid was used to estimate transition probabilities in the model. Time horizon was 10 years. Prices were taken from the health care tariffs published by the Regional Health Service in Madrid and from Spanish literature. They were updated to 2013 price levels and adjusted to subsequent years by a 3% inflation rate. Results: Cardiovascular complications included in the model were ischemic heart diseases (acute myocardial infarction and angina) and heart failure with a total of 5 health states. The estimated increase in costs was € 1,080 per patient in 10 years (€ 108 per year). Considering an 8% prevalence of DM in Madrid, the increase in costs of DM due to cardiovascular diseases would be € 52 million per year. Conclusions: Cardiovascular diseases in DM patients have a relevant impact on health care expenditure. Prevention of cardiovascular complications can lead to significant cost savings for the management of DM. Understanding future costs of DM might be valuable in terms of budget allocation and economic evaluation. PDB45 Cost and Burden Of Hypercholesterolemia In Portugal Gouveia M 1, Borges M 2, Augusto M 2, Caldeira D 2, Alarcão J 2, Pinheiro L 2, Sousa R 2, Fareleira F 2, Ascenção R 2, Costa J 3, Laires P 4, Fiuza M 5, Dias N C 5, Martins S 5, Belo A 5, Vaz-Carneiro A 2 1Católica Lisbon School of Business and Economics, Lisbon, Portugal, 2Center for Evidence Based Medicine, Faculty of Medicine, University of Lisbon, Lisbon, Portugal, 3Institute of Molecular Medicine, Lisbon, Portugal, 4Merck Sharp & Dohme, Oeiras, Portugal, 5Sociedade Portuguesa de Cardiologia, Lisbon, Portugal . . . . . . . . . . . . . . . . PDB43 Estimation of Costs Associated With Stroke In Diabetic Patients in Madrid (Spain) Using A New Simulation Model . García-Pérez S , Sánchez-Piedra C , Sarría-Santamera A Instituto de Salud Carlos III, Madrid, Spain . Objectives: The aim of this study was to assess the direct costs of ambulatory treatment associated with types 1 and 2 diabetes mellitus (DM) from the public payer perspective in Poland, and to explore the relative contribution of different cost components in the total direct cost of DM. Methods: The estimates were based on reimbursement data from the years 2012 and 2013 provided by the public payer in Poland. Reimbursement costs of insulin, oral anti-diabetes drugs and monitoring strips were taken into account. Costs were presented in Polish zloty (PLN). Results: Total reimbursement costs for the public payer of anti-diabetes treatment were 1.3 billion PLN in 2012 and 1.5 billion PLN in 2013, representing 19% and 21% of the total Polish reimbursement expenditures in 2012 and 2013, respectively. The highest component of anti-diabetes treatment costs was monitoring strips (645 million PLN in 2012 and 793 million PLN in 2013, 49% and 52%, respectively). Insulin costs (518 million PLN in 2012 and 583 million PLN in 2013) and anti-diabetes oral drug costs (144 million PLN in 2012 and 155 million PLN in 2013) constitute 38-40% and 10-11% of total reimbursement costs, respectively. Regarding insulin, human insulin had a larger share of the costs (60-62%) than insulin analogs (38-40%). Premixed human insulin (65-66% reimbursement costs) and premixed insulin analogs (47-52% reimbursement costs) represent the largest share of the reimbursement costs of human insulin and insulin analogs, respectively. Reimbursement costs of biguanides and sulphonylureas represent 49-54% and 40-44% of total reimbursement costs of oral anti-diabetes drugs, respectively. Conclusions: Total direct costs of diabetes mellitus treatment constitute a major part of reimbursement expenditures in Poland and are still growing. The main cost driver was reimbursement of monitoring strips, and the costs of insulin reimbursement were slightly lower. Sánchez-Piedra C , García-Pérez S , Sarría-Santamera A Instituto de Salud Carlos III, Madrid, Spain PDB44 Estimation of Costs Associated With Cardiovascular Complications of Diabetic Patients In Madrid (Spain) Using A New Simulation Model . Background: Diabetes (DM) has a relevant impact on health care budgets mainly due to its high prevalence and its chronic and acute complications. Isquemic stroke (IS) is one of the macrovascular complications associated with DM and its incidence increases with the presence of some risk factors such as hypertension and obesity. Objectives: Develop a model to predict future costs of DM in Spain. Estimate the increase in health care costs associated with potential IS in DM patients. Methods: Foro Gerendia is an initiative constituted by health care professional, experts in DM management. A Markov model for DM was developed based on the opinion of Foro Gerendia and on previous works in this area. A 5-year primary care data registry of Objectives: Hypercholesterolemia is a risk factor for ciculatory diseases. This study estimates the impact of hypercholesterolemia on populations’ health levels and its economic impact in Portugal. Methods: The impact on health status is measured by the loss of life years adjusted for disability (DALYs - Disability Adjusted Life Years). The economic impact analysis includes two components. The first estimates the direct costs generated by hypercholesterolemia including consumption of health care, social support, and other associated costs (e. g. transport). The second estimates indirect costs, those related to the losses of productivity. The disease burden and the costs of illness attributable to hypercholesterolemia include also the complications attributable to this risk factor, namely ischemic stroke and ischemic heart disease. In order to estimate the attributable fractions to hypercholesterolemia (total cholesterol ≥ 200mg/dL or statin use) of the diseases considered, a microsimulation approach was employed by using Framingham equations on the national database (VALSIM). In a counterfactual scenario hypercholesterolemia was eliminated from these observations and the resulting proportional change in the probability of CV events was taken as the hypercholesterolemia attributable fractions. These fractions were used to estimate the contribution of hypercholesterolemia to the burden and the annual costs of the aforementioned circulatory diseases. Results: 1,689 deaths can be attributed to hypercholesterolemia, which corresponds to 1.6% of the total deaths in Portugal in 2010. The DALYs resulting from disability and premature deaths caused by hypercholesterolemia in 2010 totaled 12,174. The estimated direct cost attributable to hypercholesterolemia in 2013 prices is € 320 million (€ 32 million for in-patient care and €288 million for ambulatory care). Indirect costs generated by disability attributable to hypercholesterolemia add up to € 198 million. The overall costs of disease are therefore estimated at € 518 million, about 0.3% of Portuguese GDP. Conclusions: In Portugal, we observed substantial burden and costs associated with Hypercholesterolemia. PDB46 Metformin in Combination With Dipeptidyl Peptidase-4 Inhibitors or Sulfonylureas in The Treatment of Type 2 Diabetes: Clinical And Economic Impact Sicras-Mainar A 1, Navarro-Artieda R 2 Serveis Assistencials, Badalona. Barcelona, Spain, 2Hospital Universitari Germans Trias i Pujol, Badalona, Spain . 1Badalona . A340 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To determine the clinical (compliance, metabolic control, hypoglycaemia and cardiovascular events [CVE]) and economic (resources and costs) impact of metformin in combination with dipeptidyl peptidase-4 inhibitors (DPP4-I) or sulfonylureas in patients with type 2 diabetes. Methods: Design: Multicentre, observational retrospective study. Patients aged ≥ 30 years under treatment with metformin who initiated a second oral antidiabetic treatment in 2008-2009 were evaluated. Two study groups were established: a) metformin + DPP4-I, b) metformin + sulfonylureas. Main measures: comorbidity, metabolic control (HbA1c < 7%), compliance and complications (hypoglycaemia, CVE). Patients were followed for 2 years. The cost model considered direct (primary/specialised care) and indirect (productivity) health care costs. Statistical analysis: logistic regression models and ANCOVA, p < 0.05. Results: We recruited 1,405 patients (men age 67.1 years, 56.2% male) of whom 37% initiated a second treatment with DPP4-I and 63% with sulfonylureas. At 2 years follow-up, patients treated with DPP4-I showed better compliance (70.3% vs 60.6%), better metabolic control (64.3% vs 60.6%), and a lower proportion of hypoglycaemia (13.9% vs 40.4%) (p < 0.05). The mean unit costs of the total adjusted costs were € 2,341 vs € 2,512, respectively; p = 0.038. Rates of CVE and renal failure were 3.7% vs 6.4%; p = 0.027. Vildagliptin was the most commonly used DPP4-I. Conclusions: Sulfonylureas were the most frequently used drugs for the treatment of diabetes. Patients treated with DPP4-I had better compliance and control of diabetes, with lower rates of hypoglycaemia and CVE, resulting in reduced health care costs. PDB47 Patient-Level Estimates of Diabetic Complications on Direct Medical Cost Jiao FF, Lam C L K , Wong C K H , Fung C S C , McGhee S The University of Hong Kong, Hong Kong Island, Hong Kong . . . . . . . . . . Objectives: To estimate the impact of diabetic complications on immediate and long-term direct medical costs in Hong Kong. Methods: A retrospective cohort study was conducted among 137,634 diabetic subjects from a territory-wide administrative database over six years (2008-2013). The trends of annual direct medical costs were analyzed by three groups: group 1) subjects without complications over study period; group 2) subjects with existing complications at baseline; and group 3) subjects developing new complications during follow-up. We employed panel data regression to investigate the impact of each diabetic complication on direct medical costs in the event year and subsequent years, adjusting for age and Charlson Comorbidity Score. Results: We found 10,322 subjects with existing diabetic complications at baseline (January 1st, 2009), and 14,349 newly developed diabetic complications over 5 years. The annual direct medical costs increased from $US4,629 to $US15,585 in the new complications group, which is substantially higher than the modest rise in the no complication group (from $US1,157 to $US1,984). The annual direct medical cost of baseline case was $US1,062 (62 year-old, no complication, Charlson Comorbidity Score= 0). After adjusting for age and comorbidity, compared to the baseline case, the multipliers of annual direct medical costs in the event year were: acute myocardial infarction (AMI) 10.61; other ischemic heart disease (IHD) 3.67; congestive heart failure (CHF) 8.46; stroke 10.17; sight threatening diabetic retinopathy (STDR) 3.41; blindness 2.78; end stage renal disease (ESRD) 12.16; peripheral vascular disease (PVD) 3.00; amputation: 3.48. The multipliers in subsequent years were: AMI 1.45; other IHD 1.32; CHF 1.86; stroke 1.37; STDR 2.16; blind 1.32; ESRD 1.86; PVD 1.42; amputation 1.32. Conclusions: There were wide variations in direct medical cost in event year and subsequent years across different major complications. These data would be useful for economic evaluations of diabetes prevention or treatment programs. PDB48 Cost Of Diabetes Related Chronic Complication In South Korea 2011 Yu J 1, Park S 2 . . 1Sungkyunkwan University, Suwon, South Korea, 2Handok Inc., Seoul, South Korea Objectives: This study aimed to analyze the direct medical cost of diabetes related micro- and macrovascular complications in Korean people with diabetes mellitus. Methods: Using the Health Insurance Review & Assessment Service-National Patients Sample (about 1.3 million patients), which was a stratified sampling from the entire population (about 46 million persons) under the Korean national health security system (2011), estimation of direct medical cost complication for patients who have diabetes related chronic complications were performed. We used the 6th revision of Korean Standard Classification of Diseases (KCD-6) which had been developed with the reference of the 10th version of International Classification (ICD10). All statistical analyses were performed using the Statistical Analysis System (SAS, version 9.3). Results: The mean age of the subjects was 61.4 years, 52.1% were male and 47.9% were female. Of the 91,463 patients with DM, 20,584 patients (22%) had at least one microvascular or macrovascular complication, 70,879 patients (78%) had no complication. The average annual direct medical cost in patients who had no complications was 351,660 won in 2011. The average annual direct medical cost in patients who had microvascular complications such as retinopathy, nephropathy, neuropathy was 506,160 won, 1.4 times higher than without DM complications, who had macrovascular complications such as MI, CVA, ESRD was 1,362,928 won, 3.8 times higher than without DM complications. Conclusions: Diabetic complications have a substantial impact on the direct medical costs of DM patients. As the number of people with diabetes continues to rise, early detection of the disease and implementation of timely and appropriate therapeutic strategies could decrease the burden of diabetes chronic complications and also huge related expenditures. PDB49 Medical Expenditure for People With Diabetes in Urban Employee Basic Medical Insureance in Fujian Yu L 1, Zhang C 2 Office, China Labour and Social Security News, Fuzhou, China, 2Shenyang Pharmaceutical University, Shenyang, China . 1Fujian . Objectives: To estimate medical expenditure for people with diabetes covered by Urban Employee Basic Medical Insurance (UEBMI) in Fujian. Methods: Claims data were extracted from UEBMIF database during 2012.1. 1 to 2012.12.31. People with diabetes were identified by keyword – diabetes. To estimate expenditures, we used Sum_All Medical method. Descriptive analyses were conducted by using EXCEL 2010. Results: In 2012, one of the Fujian UEBMI database recorded claims data of 240079 patients, including 5953 with diabetes (2.48%), among which 68.81% were male and 91.03% were age over 50. The number of General Outpatient (GO)/Catastrophic Outpatient (CO)/inpatient with diabetes was 5196/4964/2325. GO/CO with diabetes visited physicians averagely 24.13/46.02 times, almost 60% of the visits happened in tertiary hospitals. People with diabetes each used inpatient services 1.70 times; the mean number of LOS was 16.52 days. Total Direct Medical Expenditure for people with Diabetes (TDMED) was CNY 132 million (16.33% of total direct medical expenditure); out-of-pocket spending accounted for 27.20%. GO/CO/inpatient medical expenditures were CNY 3824.16/7960.88/31365.25 per person and CNY 158.49/172.97/18396.62 per visit. Expenses for medicine was CNY 89 million (67.01% of TDMED), only CNY 17 million (13.03% of TDMED) was spent on anti-glycaemic medicine. 94.56% of people with diabetes took prescribed anti-glycaemic medicine, including OAD (88.34%), insulin (41.21%) and traditional Chinese medicine (23.13%). For CO, 47.26% of the visits were due to diabetes and the medical expenditure accounted for 43.40%. Among hospitalized people with diabetes, only 10.68% of their primary diagnoses were diabetes. Inpatient with diabetes whose primary diagnoses were diabetes consumed less health care resources (average LOS/medical expenditure per person/medical expenditure per visit) than those whose primary diagnoses were diabetic complications or other diseases (15.47days/CNY 14567.57/CNY 12481.90 vs. 16.65days/CNY 27666.67/CNY 19161.66). Conclusions: Diabetes, as one of the major chronic diseases, consumed a large amount of medical resources in Fujian. Majority of direct medical expenditure were spent on treating diabetic complications and other diseases. PDB50 Indirect Costs of Diabetes Mellitus (Dm) From the Perspective of The Social Insurance Institution (Zus) In Poland Kawalec P Jagiellonian University Medical College, Krakow, Poland . Objectives: The aim of this study was to assess the indirect costs associated with types 1 and 2 diabetes mellitus (DM) from the perspective of the Social Insurance Institution (ZUS) in Poland. Methods: The estimates were based on data from the year 2012 concerning absence from work due to the illness (sick leave) the amount of short term disability, the sufferers of which claim rehabilitation benefit, and the amount of permanent (or long term) disability, the sufferers of which claim disability pension. Costs were presented in Polish zloty (PLN). Results: Total indirect costs of DM types 1 and type 2 in the year 2012 in Poland were 59 013 912 PLN and 66 597 701 PLN, respectively. The highest component of indirect costs of DM type 1 was sick leave (61%). Long and short term disability costs constitute 28% and 11% of total indirect costs of DM type 1, respectively. A slightly different situation was reported in the case of type 2 diabetes: long term disability costs were only 1% of total indirect costs, and short term disability had a slightly larger part of indirect costs (less than 5%). The highest component of indirect costs of DM type 2 was sick leave (95%). One sick leave of a person with types 1 or 2 diabetes generated a cost of lost productivity equal to 1 771 PLN or 1 585 PLN, respectively. The cost of disability pension per one person was higher than rehabilitation benefit, and equalled 41 398 PLN compared with 17 249 PLN (the average value of one short and long term benefit payment was the same for both types of diabetes). Conclusions: DM in Poland generated high indirect costs. The main component was sick leave; rehabilitation benefit and disability pension generated much lower costs for Social Insurance Institution. PDB51 Key Cost Drivers Of Type 2 Diabetes Mellitus: an International Literature Review Villoro R 1, Zozaya N 1, Hidalgo A 2, Oliva J 2, Rubio M 3 Economía y Salud, Madrid, Spain, 2University of Castilla La Mancha, Toledo, Spain, 3Sanofi, Barcelona, Spain . . . . . 1Weber Objectives: Type 2 Diabetes Mellitus (T2DM) is a worldwide prevalent chronic disease, related to high morbidity and mortality, and to significant socioeconomic costs. However, the magnitude of cost varies significantly among empirical studies. The objective of this literature review is to identify the main drivers that influence the costs of T2DM. Methods: We searched relevant databases for studies estimating T2DM costs, published in English and Spanish in the USA and Europe (1995-2014). Search terms included “diabetes”, “costs”, “burden”, and “economic impact”. No exclusion was made based on study design. Cost drivers and differences in methodologies across studies were identified. Results: A total of 25 papers were included out of 618 relevant titles identified. Costing methodologies and cost drivers vary significantly causing important variations in results. The main difference in methodology refers to estimating T2DM costs versus the costs of people living with T2DM, which include all direct sanitary costs -whether directly related to T2DM or not- and may increase total costs by up to 74%. Direct sanitary costs range between 40% and 75% of total costs depending partially on whether productivity losses and caregiving are considered. The relative weight of hospitalizations range between 20% and 60% of sanitary costs depending on the categories included. Pharmacologic costs range between 13% and 46%, which include antidiabetic drugs and treatments for complications derived from poor control. Optimal glycemic control reduces costs but is not always considered in the studies. Microvascular and/ or macrovascular complications and hypoglycemic events increase total cost per patient. Conclusions: Differences in costing methodology, the type of cost categories included in the analysis and the source of data greatly influence the results of the studies and impede reaching accurate conclusions regarding the cost of T2DM. Designing standardized costing methodology guidelines would help future studies estimate the real burden of T2DM. A341 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PDB52 How to Estimate The Cost Of Diabetes Based on Information From The French Health Insurance Database (Sniiram)? Aguadé A S 1, Gastaldi-Menager C 1, Denis P 1, Fagot-Campagna A 1, Gissot C 2, Polton D 2 (National Health Insurance), Paris Cedex 20, France, 2CNAMTS (National Health Insurance), paris cedex 20, France . . . . . . . 1CNAMTS Objectives: The aim of this study is to assess for 2012 the cost of diabetes from a payer perspective, based on the French health insurance database and using two different approaches (top down versus bottom-up). Methods: Using information about 60 millions of individuals from the general scheme insurance database (85% of the French population), we developed algorithms to identify all people who received care for each of 56 groups of diseases or medical events or treatments, which are frequent, severe and/or costly. Algorithms have been applied to each patient. For diabetes, we used ICD-10 diagnoses for long-term chronic diseases, reimbursement for anti-diabetic drugs. Costs of all reimbursed expenditures (outpatient/inpatient care, disability/sickness benefits) were extracted per individual. The top-down method allocated expenditure to each of the 56 diseases based on the average expenditure by disease calculated for individuals with only one disease. All expenditures were thereafter extrapolated to the whole population to fit national health account aggregates. For the bottom-up approach, diabetes expenditures were estimated by identifying finely in our database expenditure items which are partly or wholly directly related to diabetes according to expert judgment. Results: Based on the top-down approach, among the 146 billion euros of expenditures reimbursed by national health insurance (all insurance schemes) in 2012, 7. 5 billion (5%) are attributable directly to diabetes. Expenditures for chronic renal insufficiency and cardio-vascular disease, frequent diabetes complications, have been assessed separately. Drugs and medical devices (3.8 billion € ) represents more than 50% of the diabetes expenditures, other outpatient care 34% (2.5 billion), inpatient care 9% (700 millions) and disability/sickness benefits 7% (500 millions). Based on our bottom-up, around 8 billion would be directly attributable to diabetes. Conclusions: Our study provides estimation of the cost of diabetes from a payer perspective, according to two different approaches but with concordant results. PDB53 Direct Cost of Diffuse Toxic Goiter and Its Complications In Ukraine Vadziuk I Ivan Horbachevsky Ternopil State Medical University, Ternopil, Ukraine . Objectives: To determine the direct cost of health care technologies used for treatment of diffuse toxic goiter in Ukraine. Methods: We made the retrospective analysis of 52 patients’s medical records with DTG that were hospitalized to the endocrinology department of Ternopil University Hospital (Jan-Dec 2012). The method of pharmacoeconomic analysis “cost of illness” was used to estimate the cost of health care for patients with DTG. We have made calculations of costs: the cost of laboratory analysis, the cost of instrumental analysis, the cost of drug treatment, doctor’s consultations, costs of patient’s stay in hospital. While determining the direct costs of medical services in monetary terms we used the rates for medical services that were in Ternopil University Hospital. Results: The study found that among the patients there were 11 (21.15%) men and 41 (78.85%) women aged 20 to 65 years (46.88±8.9). All patients have got to the hospital in stage of medication subcompensation. The average duration of stay in hospital for patients was 10.67±2.07 days. The total cost of laboratory tests was EUR 1,555.53. The costs of instrumental methods of patients examination amounted EUR 204.25. The cost of consultations by specialists was EUR 285.97. In determining the amount of direct costs for medicines we found that the cost of drug treatment of the underlying disease is EUR 211.83, the cost of drug therapy of DTG complications is EUR 1,163.64. Total cost of patients stay in hospital was EUR 3,513.45. After calculation of all direct costs we determined that the total cost of DTG per patient is EUR 133.36 per course of treatment. Conclusions: In the cost structure of health care provision for patients with diffuse toxic goiter the most significant costs were spent for laboratory tests and for patient’s stay in hospital. The presence of underlying disease complications significantly increases cost of drugs. PDB54 Comparison of the Economic Burden and Health care Utilizations of u.S. Veteran Patients Diagnosed with Type 2 Diabetes Mellitus Xie L 1, Kariburyo M F 1, Wang Y 1, Baser O 2 1STATinMED Research, Ann Arbor, MI, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . . Objectives: To evaluate the economic burden and health care utilizations of Type 2 diabetes mellitus (T2DM) among U.S. veteran patients. Methods: T2DM patients (International Classification of Disease 9th Revision Clinical Modification [ICD-9-CM] diagnosis codes 250.x0, 250.x2) were identified using the U.S. Veterans Health Administration Medical SAS datasets (01OCT2008-31SEP2011). The first diagnosis date was defined as the index date. A comparison cohort of patients without a T2DM diagnosis but of the same age, region, gender and index year were identified and matched according to baseline Charlson Comorbidity Index scores, with a randomly chosen index date to minimize selection bias. Patients in both cohorts were required to be at least age 18 years, with 1-year continuous health plan enrollment pre- and post-index date. Baseline body mass index (BMI) and glycated hemoglobin (HbA1c) values and follow-up health care costs and utilizations were compared using 1: 1 propensity score matching (PSM). Results: A total of 1,211,748 T2DM patients were identified for study. T2DM patients had significantly higher HbA1c (7.25 vs. 5.78, p< 0.0001) and BMI (31.75 vs. 29.16, p< 0.0001) results during the baseline period. After 1: 1 PSM, each cohort included 323,962 patients, with well-balanced baseline demographic and clinical characteristics. A higher percentage of T2DM patients had inpatient admissions (8.65% vs. 1.58%, p< 0.0001), emergency room (ER) (13.12% vs. 4.44%, p< 0.0001) and physician office visits (99.54% vs. 45.22%, p< 0.0001) and prescription fills (84.55% vs. 45.43%, p< 0.0001). The T2DM cohort also incurred higher inpatient ($3,051 vs. $424, p< 0.0001), ER ($127 vs. $39, p< 0.0001), physician office ($2,934 vs. $875, p< 0.0001), outpatient ($3,299 vs. $971, p< 0.0001), pharmacy ($610 vs. $224, p< 0.0001) and total costs ($6,958 vs. $1,618, p< 0.0001) than the comparison cohort. Conclusions: Study results suggest that patients diagnosed with T2DM utilized more health resources and incurred four times higher costs compared to those without a T2DM diagnosis. PDB55 Medical Expenditures Associated With Type 2 Diabetes Mellitus in Japan: A Large Claims Database Study Fukuda H 1, Ikeda S 2, Shiroiwa T 3, Igarashi A 4, Fukuda T 3 1Kyushu University, Graduate School of Medical Sciences, Fukuoka, Japan, 2International University of Health and Welfare, Otawara City, Tochigi, Japan, 3National Institute of Public Health, Saitama, Japan, 4University of Tokyo, Graduate School of Pharmaceutical Sciences, Tokyo, Japan . . . . . Objectives: The objective of the study was to estimate the excess costs associated with type 2 diabetes mellitus (T2DM) and diabetes-related complications in Japan through the use of a large claims database. Methods: We performed a retrospective cohort analysis using a large commercial claims database obtained from the Japan Medical Data Center Co., Ltd. (Tokyo, Japan). Data from the period between January 2005 and June 2012 were analyzed. Patients diagnosed with T2DM were identified using the International Classification of Diseases 10th revision (ICD-10) diagnosis codes E11–E14, with the month of initial diagnosis designated as the index month. Cost and health care utilization data from claims for outpatient, inpatient, and dispensing services during the study period were summarized to per-patient-per-month (PPPM) levels. Costs were calculated from the perspective of a public health care payer. Diabetes-related complications were identified through the occurrence of the following diseases after the index month and/or their associated treatment: retinopathy, nephropathy, neuropathy/extremity disease, ischemic heart disease, and cerebrovascular disease. Regression-adjusted medical costs associated with each health state of T2DM were estimated using a fixed-effects model. Results: A total of 8,063,139 PPPM records from 152,791 T2DM patients were identified and examined. The average follow-up duration per patient was 52.8 months. The average incremental cost for T2DM was US$123 PPPM. The average incremental costs for diabetes-related complications were US$94 (retinopathy without surgery), US$1,933 (retinopathy with surgery), US$17 (renal proteinuria), US$319 (renal failure), US$3,677 (dialysis), US$131 (neuropathy and/or extremity disease without surgery), US$4,498 (neuropathy and/or extremity disease with surgery), US$93 (ischemic heart disease without surgery), US$13,280 (ischemic heart disease with surgery), US$56 (cerebrovascular disease with surgery), and US$2,605 (cerebrovascular disease with hospitalization). Conclusions: These estimates of incremental medical expenditure in relation to health state may contribute to economic evaluations of various aspects of health care in Japan. PDB56 Health Care Costs in Patients With Type 2 Diabetes in Flanders Based on A Combination of Clinical And Health Insurance Data Odnoletkova I 1, Annemans L 2, Ceuppens A 3, Aertgeerts B 1, Ramaekers D 1 1University of Leuven, Leuven, Belgium, 2Ghent University & Brussels University, Ghent, Belgium, 3Independant Health Insurance Fund, Brussels, Belgium . . . . . Objectives: To analyse the annual health care costs of patients with type 2 diabetes in Flanders from the perspective of the health care system and to assess the impact of comorbidities on total cost. Methods: Study participants were selected from the Independent Sickness Fund database on consumption of hypoglycemic agents and invited to participate in a clinical trial on diabetes education. Information on resource utilization during 12 months preceding inclusion was extracted from the database and combined with baseline clinical assessment data. Total health care costs consisted of costs paid by the public health insurance and patient co-payments and was analyzed as function of reported comorbidities through stepwise multiple regression. Results: 574 patients agreed to participate. Mean age was 64 years (35 75) and type 2 diabetes was diagnosed since 7 years on average. 62% were men, 86% were treated with oral antidiabetics. 38% reported to have at least one comorbidity. 30% were hospitalized in the study period. Hospitalizations accounted for 41% of the total cost, followed by outpatient prescription medications (22%) and physician consults (8%). The mean (95% CI) health care cost was € 4,522 (€ 3,799 to € 5,245). The mean annual cost of patients with no comorbidities was € 3,357 (€ 2,599 to € 4,116) and increased till € 4,750 (€ 3,511 to € 5,990) for those with one comorbidity (p= .001) and till € 7,303 (€ 3,909 to € 10,697) and € 9,868 (€ 4,685 to € 15,051) for those with two and three comorbidities respectively (p= .039 and. 057). Coronary heart disease was the main contributor to the cost variability (R2= 0.063, p= .000). Conclusions: Our study updates the knowledge on the actual annual cost of diabetes treatment in Flanders. Combination of clinical information and health insurance claims data allowed analysis based on the patient medical condition. The high marginal cost of people with comorbidities emphasises the importance of prevention for people diagnosed with type 2 diabetes. PDB57 Multimorbidity Pharmaceutical Cost of Diabetes Mellitus Vivas-Consuelo D 1, Alvis-Estrada L 2, Uso-Talamantes R 3, Caballer-Tarazona V 1, BuiguesPastor L 3, Sancho-Mestre C 1 1Universitat Politecnica de Valencia, Valencia, Spain, 2Universidad de Cartagena, Cartgena, Colombia, 3Conselleria de Sanidad. Generalitat de Valencia, Valencia, Spain . . . . . . Objectives: To estimate the multimorbity associated with diabetes mellitus type 2 and its relation to pharmaceutical cost in a primary health care setting. Methods: Cross-sectional study during 2012. A health region of 5,150,540 population was analysed to determine the diabetic individuals. 350,015 diabetic individuals were identified through clinical codes using the ICD-9-MC classification and the 3M Clinical Risk Groups software. We analyzed the consumption of A342 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 pharmaceutical and blood glucose reagent strips. All measurements were obtained at individual level. Results: The raw prevalence of diabetes was 6.7%. All patients were stratified into four morbitity groups. The first group corresponded to the initial state (CRG 1-4); the second group included the core multimorbidity patients in the intermediate and advanced stages (CRG 5-7); the third group included patients with diabetes and malignancies; the last group was of patients with catastrophic statuses, manly ERSD. The most common comorbitities in the patient group for CRG 5 to 7 were hypertension (47% - 97%), dislipemia (50.9% - 59%) and cardiovascular disease (18% - 39.9%). The average cost of insulin was € 117.1. The average cost of oral antidiabetic agents secretagogues was € 15.2 and was highest in the CRG 5 to 7 group at € 16.3. The average cost of non-secretagogue oral antidiabetic agents was € 212.9 and was highest in the CRG 5 to 7 group at € 232.8. The total average cost was € 476.5 in CRG 1-4 group, € 1350.5 in CRG 5-7 group, € 2174.1 in CRG 8 group and € 2840.5 in CRG 9 group. Conclusions: Diabetes is characterised by a strong presence of other chronic conditions and events related to its development throughout life. This multimorbidity has a great impact on the pharmaceutical cost. PDB58 Impact of Early Microbiologic Culture in the Economic Burden of Care for Patients With Infected Diabetic Foot Ulcers At A Model Of Secondary Level Imss’ Hospitals in Mexico Balderas-Peña L M A 1, Sat-Muñoz D 2, Alvarado-Iñiguez M R 3, Salcedo-Rocha A L 4, Ramírez-Conchas R E 1, Sánchez-Sandoval F H 5, Macías-López G G 6, Ruíz-Quezada S L 6, Chagollán-Ramírez J M 3, Prieto-Miranda S E 5, García de Alba-García J E 4 1UMAE Hospital de Especialidades Centro Médico Nacional de Occidente IMSS, Guadalajara, Jalisco, Mexico, 2UMAE Hospital de Gineco-Obstetricia Centro Médico Nacional de Occidente IMSS, Guadalajara, Jalisco, Mexico, 3Centro Universitario de Ciencias Económico Administrativas. Universidad de Guadalajara, Zapopan, Jalisco, Mexico, 4Delegación Jalisco IMSS, Guadalajara, Jalisco, Mexico, 5Instituto Mexicano del Seguro Social, Hospital General Regional N° 46, Guadalajara,, Mexico, 6Universidad de Guadalajara, Centro Universitario de Ciencias Exactas e Ingenierias, Guadalajara, Mexico . . . . . . . . . . . . . . . . . . . . . . Objectives: Estimate the economic impact of the early microbiological culture in the cost-reduction associated to the attendance process of the infected ulcers in diabetic foot through an economic model based on a Monte Carlo simulation. Methods: Economic model of Monte Carlo simulation in two steps: The first step was a descriptive and longitudinal study. Were included diabetic subjects with infected ulcers in diabetic foot, patients were initially treated in Emergency Room of Mexican IMSS’ secondary level hospital in a period from January 1st to April 30th2010. Results: Micro-Costing Analysis Results: Total costs for the sample were $502,438.04 USD, the mean cost per patient was $7,177.69 (±$5,043.51) USD, the median cost was $6,422.99 (p25 $3,502.93; p75 $9,298.33). 72.75% ($365,527.45) of the total cost was related to hospital stay. 10.6% ($53,240.86) of the expenditure during treatment was associated to wound healing care, 9.98% ($50,132.94) surgeries, 1.87% ($9,389.27), anti-diabetic drugs and concomitant chronic diseases. 1.79% ($9,008.47) to clinical chemistry tests, 1% ($5,405.98) antibiotics, similar percentage to the imaging studies ($5,000.08), 0.66% ($3,309.88) acute phase reactants, 0.19% ($965.65) microbiologic cultures, and the lowest percentage was destined to intravenous fluids. Monte Carlo Modelling Results: Antibiotic cost analysis showed: If the physician takes a secretion sample in the first 48 hours of the patient admission in emergency room and begin the empirical antibiotic treatment, and modifies the criteria according microbiological results from the culture, the cost will be reduced (through antibiotic expenditure) 9 to 15%. The reduction could be 20 to 32% associated to reduce the hospital stay days. The total cost for the attendance process could be reduced 10 to 25%. Conclusions: According Monte Carlo Modelling results the early microbiological culture as the base of antibiotic selection can reduce treatment cost in more than 30%. PDB59 Direct Costs of Diabetic Foot Ulcers in Russia Ignatyeva V 1, Avxentyeva . M 1, Galstyan . G R 2, Bregovskiy V 3, Udovichenko . . . O4 . 1The PDB60 Health Economics Evaluation for Insulin Injection Pen-Needle with Different Lengths in Patients With Diabetes in China Sun L , Ren X , Zhu S , Liu Y , Chen Y , Sun J , Chen B , Sun S , Wang R Shenyang Pharmaceutical University, Shenyang, China . . . . . . . . . Objectives: To perform a health economics evaluation for insulin injection penneedle with different lengths in diabetic patients in China. Methods: With the relationship between HbA1C level and diabetes-related complications obtained from literature, cost-benefit analysis was applied to evaluate the net cost saving (including the cost of needles and cost of complications and adverse events treatment) of shorter pen-needle compared to the longer one, from the perspective of society. The unit costs were retrieved from published sources and outcomes of the effectiveness were derived from meta-analysis of relevant literatures related to different lengths insulin pen-needle. One-way sensitivity analysis was conducted to demonstrate the robustness of results. Results: The meta-analysis showed that compared to the longer needle, the shorter one decreased HbA1C by an average of 0.11% and reduced the risk of hypoglycaemic events by 17%. A net saving of $27 per patient based on shorter needle resulted from decreased adverse events, while the costs of needles were increased by $9 during the trial time. If the retinopathy was included, shorter needle produced a net saving of $320 per patient due to decreased complications, and the economic evaluation revealed shorter needle can totally save $338 per patient compared to the longer one. If without retinopathy, the net saving was $129 per patient due to decreased complications, and the totally saving was $147 per patient. Sensitivity analysis showed that the results were stable. Conclusions: Current findings indicate that shorter insulin injection pen-needle lead to a decreasing of complications and adverse reaction compared to longer needle, and subsequently could lead to a reduction in health costs. PDB61 Burden of Non-Adherence To Type 1 Diabetes Mellitus Therapeutic Guidelines In France Beaudet A , Ong R C IMS Health, Basel, Switzerland . . . Objectives: To investigate the burden of type 1 diabetes mellitus (T1DM) in France associated with non-adherence to clinical guidelines using the IMS CORE Diabetes Model (CDM) and clinical data from the IMS Lifelink Diabetes Cohort. Methods: The CDM is a health economic model based on 17 inter-dependent sub-models that simulate micro- and macrovascular complications associated with diabetes. Physiological parameter inputs such as HbA1c, blood lipids, body mass index (BMI), and systolic blood pressure (SBP) were taken from real-life data and compared with guidelines from the “Haute Autorité de Santé”. T1DM patients (age ≥ 18) who visited a general physician between May 2011 and May 2014 and have received at least one insulin prescription were included in the analysis. Costs and outcomes were discounted at 4% per annum. Costs of complications were taken from published sources. Results: A cohort of 605 T1DM patients (43% male) was analyzed. Mean age at first visit was 58 years, HbA1c was 7.8%, SBP was 132 mmHg, and BMI was 28 kg/m2. Smokers represented 23% of the cohort. Results from economic modeling using the CDM suggest that for prevalent T1DM patients in France, potential savings for the health care system associated with meeting the therapeutic guidelines would be on average of 1,661 EUR per patient from a lifetime perspective. Most of the savings would come from the prevention of renal complications. Any new complication would be delayed by 2.4 months on average. Conclusions: The “Haute Autorité de Santé” recommends that T1DM patients maintain their HbA1c lower than 7.5%, SBP lower than 130 mmHg, BMI lower than 25 kg/m2 and quit smoking. Our results showed that these recommendations were not strictly followed among T1DM patients in France. Better adherence to the T1DM guidelines would lead to cost savings in the French health care system and improved patient outcomes. PDB62 Health Economic Evaluation of Canagliflozin in the Treatment of Type 2 Diabetes Mellitus in Czech Republic Troelsgaard A 1, Pitcher A 2, Veselá Š 3, Lovato E 2, Hemels M 1 Health, London, UK, 3Janssen-Cilag Czech Republic, Prague, Czech Republic Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia, 2The Endocrinological Scientific Center, Moscow, Russia, 3Federal Almazov Medical Research Center, St. Petersburg, Russia, 4Moscow Municipal outpatient clinic #22, Moscow, Russia 1Janssen A/S, Birkerød, Denmark, 2IMS Objectives: To evaluate the outcomes and current costs of 4 different scenarios of diabetic foot ulcers (DFU) treatment in Russian current medical practice. Methods: We developed a decision tree model estimating annual number of outcomes (major and minor amputations) and costs for 4 simplified possible scenarios of medical care: 1) outpatient diabetic foot clinic, 2) non-specialized outpatient care, 3) cessation of the outpatient treatment by patient after the second visit and 4) care provided only at hospital (without previous visits to any outpatient clinic). The distribution of the patient cohort (1000 patients) among the scenarios and rates of possible events were based on published Russian data and experts’ survey. Costs were calculated from the overall governmental budget point of view and included inpatient and outpatient care for DFU treatment, total contact casting, medications and dressings provided in the outpatient care, prosthetic devices and services. Cost data was derived from published retrospective study comparing provision of care to DFU patients in specialized and non-specialized clinics in Russia and reimbursement rates in public medical insurance and social care. Results: The lowest rate of amputations per patient (0.07 for minor and 0.01 for major) is expected for those receiving treatment in the outpatient diabetic foot clinic with mean annual cost € 279.68 per patient. The highest costs (€ 1141.17) and amputation rates (0.29 for minor and 0.19 for major) per patient are observed in the group treated at hospital only. The minor amputation rate and annual costs are higher for patients following non-specialized outpatient care than for those ceasing the outpatient treatment – 0.11 vs 0.09 and € 449.82 vs € 381.30, and the reverse was observed for the rate of major amputations - 0.03 vs 0.06. Conclusions: Referring DFU patients to the outpatient diabetic foot clinics as early as possible appears to be the most cost-effective way of treating them. Objectives: Canagliflozin is a novel drug for treatment of diabetes belonging to the drug class known as sodium glucose co-transporter-2 (SGLT2) inhibitors which method of action is insulin independent. Canagliflozin has been shown to not only reduce glucose levels, but also weight and blood pressure. The objective is to evaluate the cost-effectiveness of canagliflozin 100mg in dual therapy (add-on to metformin) compared to sitagliptin and glimepiride, in triple therapy (add-on to metformin and sulfonylurea) compared to sitagliptin and as an insulin add-on compared to dapagliflozin in the Czech Republic setting from a payer perspective. Methods: The IMS CORE Diabetes Model was used to evaluate the costeffectiveness of canagliflozin using clinical trial data and network meta-analysis data, combined with Czech Republic specific data, where available. Results: The cost-effectiveness analyses indicate that in dual therapy when compared with sitagliptin and glimepiride, canagliflozin 100mg is found to be cost-effective with an incremental cost-effectiveness ratio (ICER) of 242,783 Kč per QALY gained and 626,939 Kč per QALY gained, respectively. In triple therapy when compared to sitagliptin, canagliflozin 100mg is found to be cost-effective with an ICER of 335,759 Kč per QALY gained. As an add-on to insulin (with or without metformin), canagliflozin compared to dapagliflozin appears to be cost-effective with an ICER of 375 Kč per QALY gained. A wide range of deterministic sensitivity analyses revealed that in the majority of scenarios, canagliflozin remained cost-effective with ICERs well below the willingness-to-pay threshold of 1,100,000 Kč per QALY. Conclusions: Canagliflozin 100mg represents a cost-effective option for the treatment of type 2 diabetes in the Czech Republic. Canagliflozin 100mg offers greater health benefits than the included alternatives in this study, at a cost that provides good value for money. . . . . . A343 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PDB63 Short-Term Cost-Effectiveness Analysis of Insulin Detemir Versus Insulin Neutral Protamine Hagedorn (Nph) In Patients With Type 2 Diabetes Mellitus in Spain Ramírez de Arellano A 1, Morales C 2, De Luis D 3, Ferrario M G 4, Lizán L 5 Nordisk Pharma SA, Madrid, Spain, 2Hospital Virgen de la Macarena, Sevilla, Spain, 3Hospital Rio Hortega, Valladolid, Spain, 4Outcomes’10, Castellon, Spain, 5Outcomes 10, Castellon, Spain . . . . . . 1Novo Objectives: To estimate the short-term cost-effectiveness of insulin detemir compared with Neutral Protamine Hagedorn (NPH) insulin when initiating insulin treatment in patients with Type 2 Diabetes Mellitus (T2DM) in Spain. Methods: A short-term (1 year) cost-effectiveness model was adapted to the Spanish public health care system. Based on a head-to-head randomized controlled trial (NCT00104182) that showed similar efficacy in glycemic control for both insulin types, weight gain (Δ=0.9Kg) and the rate of non-severe hypoglycemia (between-arms RR= 0.52; IC95% 0.44-0.61) of detemir vs. NPH were selected as clinical outcomes. Costs (Euros 2014) were estimated from the perspective of the Spanish national health system and derived from national public sources. Only insulin treatment and management costs associated with non-severe hypoglycemic episodes were included in the analysis. According to a published study, non-severe hypoglycemia (a self-managed event) was assumed to imply the use of 5.3 glucometer strips and a visit to a general practitioner for 25% of patients (Orozco-Beltrán et al., 2014). The disutility value associated to weight gain was -0.0100 per BMI unit (Lee et al., 2005). The disutility associated to daytime and nocturnal non-severe hypoglycemia was -0.0041 and -0.0067 per event, respectively (Evans et al., 2013). Results: The yearly gain in quality-adjusted life years (QALY) associated to insulin detemir versus NPH was 0.015. The estimated incremental cost of treating patients with insulin detemir versus NPH was € 244.03. The incremental cost-effectiveness ratio (ICER) of insulin detemir versus NPH in insulin-naïve T2DM patients was estimated to be €16,381.18/QALY in Spain. This value is lower than those published for other European countries (€21,768-28,349/ QALY) and is beneath the ICER threshold commonly accepted for Spain (€30,000/ QALY). Conclusions: Insulin detemir is a cost-effective alternative to NPH insulin in the first and subsequent years of treatment of insulin-naïve T2DM patients in Spain. PDB64 Cost-Effectiveness Analysis of Insulin Detemir Versus Insulin Neutral Protamine Hagedorn (Nph) In Patients With Type 1 Diabetes Mellitus In Spain Ramírez de Arellano A 1, Lizán L 2, Prades M 2, Morales C 3, De Luis D 4 1Novo Nordisk Pharma SA, Madrid, Spain, 2Outcomes 10, Castellon, Spain, 3Hospital Virgen de la Macarena, Sevilla, Spain, 4Hospital Rio Hortega, Valladolid, Spain . . . . . Objectives: To estimate the short-term cost-effectiveness of insulin detemir compared with Neutral Protamine Hagedorn (NPH) insulin when initiating insulin treatment in patients with Type 1 Diabetes Mellitus (T1DM) in Spain. Methods: A short-term (1 year) cost-effectiveness model was adapted to the Spanish public health care system. Based on the Update of CADTH Technology Report No. 92 (2008) that showed similar efficacy in glycemic control for both insulin types, the rate of hypoglycemia with detemir vs. NPH (RR=0.84; IC95% 0.74-0.97) was the considered clinical outcome. Costs, expressed in Euros 2014, were estimated from the perspective of the Spanish national health system and derived from national health care cost databases and publications. Only insulin treatment and management costs associated with non-severe hypoglycemic episodes were included in the analysis. Non-severe hypoglycemia, defined as a self-managed event, was assumed to imply the use of extra 5.3 glucometer strips during the following week and a visit to a general practitioner for 25% of patients (Orozco-Beltrán et al., 2014). The disutility associated to daytime and nocturnal non-severe hypoglycemia was -0.0041 and -0.0067 per event, respectively (Evans et al., 2013). Results: The gain in quality-adjusted life years (QALY) associated to determir versus NPH was 0.108. The estimated incremental cost of determir versus NPH was €247.40. The incremental cost-effectiveness ratio (ICER) of detemir vs. NPH in patients with T1DM was estimated to be € 2,286.67/QALY in Spain. This value is significantly lower than those reported for other European countries (€10,938-13,310/ QALY) and are much lower than the ICER threshold commonly accepted for Spain (€30,000/QALY). Conclusions: Detemir is a cost-effective alternative to NPH insulin in the first year of treatment of insulin-naive T1DM patients in Spain. PDB65 Cost-Effectiveness-Analysis of The New-Born Screening in Austria Walter E 1, Kasper D C 2 1Institute for Pharmaeconomic Research, Vienna, Austria, 2Department of Pediatrics and Adolescent Medicine, Vienna, Austria . . infant in comparison to no-screening strategy. Transferred to the entire birth cohort newborn screening is able to reduce total costs by 14 million € from the Austrian health care systems perspective each year. Conclusions: Funding the new-born screening saves money and is cost-effective for the Austrian health care system. . Objectives: Since more than 45 years, a preventive program for the detection of congenital metabolic and endocrine diseases is carried out successfully in Austria. The goal is to investigate every new-born a few days after birth to initiate a quality assured therapy as quickly as possible. Since 1966, this program is carried out by the Federal Ministry of Health at the University Clinic for Child and Adolescent Medicine, Medical University of Vienna. The aim of this study was to determine cost-effectiveness of the new-born screening. Methods: We developed a decision-analytic model, which include specific Markov processes for the core disorders: Cystic Fibrosis (CF), phenylketonuria (PKU), medium-chain acyl-CoA (MCAD), congenital hypothyroidism (CH), galactosemia (GAL) and Maple syrup urine disease (MSUD). Costs and health benefits were estimated for a cohort of new-borns in Austria in 1 year. The analysis focused on lifetime consequences. This encompassed direct costs (including screening costs and cost of illness), quality-adjusted-life-years (QALYs) and reduced expectation of life. Costs were presented per child and for the Austrian birth cohort. Costs from published sources were used (2014 Euro) from the health care systems perspective. QALYs, life-years (LYs) and costs were projected over a life-time horizon and discounted at 3% p. a. Results: We found ten-times higher lifetime costs per child without screening compared to screening. The incremental costs of screening ranged from 12.308 € (MCAD) to 291.332 € (PKU). Screening saved 181 € and 0.09 QALYs per PDB66 The Impact of Long-Term Clinical Evidence on Cost-Effectiveness of Exenatide Once Weekly (Bydureon®) Versus Insulin Glargine for Patients With Type 2 Diabetes Mellitus (T2dm) From A Uk Nhs Perspective Charokopou M 1, Vioix H 2, Verheggen B G 1, Bratt T 2, Franks D 2 1Pharmerit International, Rotterdam, The Netherlands, 2AstraZeneca UK Ltd., Luton, UK . . . . . . Objectives: When patients start their first injectable therapy, clinicians can choose between glucagon-like peptide-1 (GLP-1) agonists and basal insulins. This study investigates the cost-effectiveness of exenatide once weekly (Bydureon®), a GLP-1 agonist, compared with insulin glargine in patients inadequately controlled with metformin (±sulfonylureas) based on long-term clinical evidence. Methods: The validated CARDIFF model was used to conduct the analyses. Clinical inputs were derived from a randomized clinical trial and a 3-year follow-up study of it comparing exenatide once weekly (ExQW) versus insulin glargine once daily. Based on these clinical inputs and the United Kingdom Prospective Diabetes Study (UKPDS) equations, the model predicts disease progression and the number of micro- and macro-vascular complications, along with diabetes-specific and all-cause mortality. The perspective of the National Health Service in the UK was adopted over a lifetime horizon. Local unit costs and utility data were assigned to the appropriate model parameters to calculate total Quality-Adjusted-Life-Years (QALYs) and total costs. Deterministic and probabilistic sensitivity analyses (PSA) were conducted. Results: Long-term treatment with ExQW was well tolerated and associated with sustained glycaemic control and sustained weight loss over at least 3 years. Compared to glargine, ExQW in combination with metformin was associated with an incremental benefit of 0.123 QALYs (95%CI: 0.057; 0.178) at an additional cost of £1,722 (95%CI: £1,396; £2,089), resulting in an incremental cost-effectiveness ratio of £13,967 per QALY gained. The PSA showed that at a willingness-to-pay threshold of £20,000 per QALY gained, ExQW treatment had an 83% probability to be cost-effective compared to the strategy including glargine. Sensitivity analyses showed that results were robust to variation in model parameters that carry uncertainty. Conclusions: Exenatide once weekly in combination with metformin was shown to be a cost-effective treatment option as first injectable therapy in patients inadequately controlled with metformin within established UK cost-effectiveness thresholds. PDB67 Dapagliflozin (Forxiga®) Versus Glipizide As Add-On Therapies In Type 2 Diabetes Mellitus (T2dm); An Update of The Cost-Effectiveness Based On Long-Term Clinical Evidence From Uk Nhs Perspective Charokopou M 1, Vioix H 2, Verheggen B G 1, Dillon S 2, Franks D 2 1Pharmerit International, Rotterdam, The Netherlands, 2AstraZeneca UK Ltd., Luton, UK . . . . . . Objectives: To update the cost-effectiveness of dapagliflozin (Forxiga®), a selective sodium-glucose co-transporter-2 (SGLT-2) inhibitor, compared with a sulphonylurea (SU) when added to metformin in patients inadequately controlled with metformin mono-therapy based on long-term clinical evidence. Methods: The published and validated CARDIFF diabetes model was used to conduct the analyses. Clinical inputs were derived from a 4-year follow-up study of a randomized clinical trial comparing dapagliflozin and glipizide in combination with metformin. Based on these clinical inputs and the United Kingdom Prospective Diabetes Study (UKPDS) equations, the model predicts disease progression and the number of micro- and macro-vascular complications, along with diabetes-specific and all-cause mortality. The perspective of the National Health Service in England and Wales was adopted over a lifetime horizon. Local unit costs and utility data were assigned to the appropriate model parameters to calculate total Quality-Adjusted-Life-Years (QALYs) and total costs. Deterministic and probabilistic sensitivity analyses (PSA) were conducted. Results: Dapagliflozin showed greater durability of HbA1c reduction compared with SU and sustained weight loss over 4 years. Compared to SU added on top of metformin, dapagliflozin add-on to metformin was associated with an incremental benefit of 0.181 QALYs (95%CI: 0.088; 0.268) at an additional cost of £819 (95%CI: £415; £1,259), resulting in an ICER point estimate of £4,521 per QALY gained. The univariate analyses showed that no input parameter change inflated the ICER above £15,000 per QALY. The PSA showed that at a willingness-to-pay threshold of £20,000 per QALY gained, dapagliflozin treatment had an estimated 100% probability to be cost-effective compared to an SU treatment strategy. These findings were shown to be robust with all sensitivity analyses. Conclusions: Dapagliflozin in combination with metformin was shown to be a cost-effective treatment option for patients who are inadequately controlled with metformin mono-therapy within established UK cost-effectiveness thresholds. PDB68 Health Economic Evaluation of Canagliflozin in the Treatment of Type 2 Diabetes Mellitus in Portugal Troelsgaard A 1, Knudsen M 2, Maia-Lopes S 3, Luz M 3, Hemels M 1 Health, Hellerup, Denmark, 3Janssen-Cilag Farmacêutica, Barcarena, Portugal . . . . . 1Janssen A/S, Birkerød, Denmark, 2IMS Objectives: To evaluate the cost-effectiveness of canagliflozin in dual therapy as add-on to metformin compared to sitagliptin and in triple therapy as add-on to metformin (MET) and sulfonylurea (SU) compared to sitagliptin. Methods: The IMS CORE Diabetes Model was used to evaluate the cost-effectiveness of canagliflozin 100 mg and 300 mg versus sitagliptin 100 mg using data from both clinical trials and network meta-analysis, combined with Portuguese-specific data when available. The perspective of the analysis is societal in accordance with Guidelines for Economic Drug Evaluation Studies from INFARMED. Results: The cost-effectiveness analyses indicate that canagliflozin (100 mg and 300 mg weighted average 65: 35) is cost-saving A344 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 and results in higher QALYs in comparison with sitagliptin 100 mg in dual therapy as add-on to metformin and in triple therapy as add-on to MET plus SU. In dual therapy as add-on to MET, canagliflozin (100 mg and 300 mg weighted average 65: 35) has an average cost saving of 24 € and an average QALY gain of 0.036. In triple therapy as add-on to MET+SU, canagliflozin (100 mg and 300 mg weighted average 65: 35) has an average cost saving of 171 € and an average QALY gain of 0.033. Probabilistic analysis suggests a likelihood of 59% of canagliflozin being cost-effective compared to sitagliptin in dual therapy and 59% in triple therapy at a willingness-to-pay of 30,000 € per gained QALY. Sensitivity analyses showed that canagliflozin is cost-effective also from a payer perspective and even when the time horizon, which is 30 years in the base case, is reduced to 10 years. Conclusions: Canagliflozin 100 mg and 300 mg will be a cost-effective alternative to sitagliptin in both dual and triple therapies, as add-on to MET or as add-on to MET+SU, respectively. PDB69 Health Economic Evaluation of Canagliflozin in the Treatment of Type 2 Diabetes Mellitus in France Granados D 1, Maurel F 2, Knudsen M 3, Troelsgaard A 4, Hemels M 4 1Janssen, Paris, France, 2IMS Health, Paris, France, 3IMS Health, Hellerup, Denmark, 4Janssen A/S, Birkerød, Denmark . . . . . Objectives: Canagliflozin is a sodium glucose co-transporter 2 (SGLT2) inhibitor developed for the treatment of adult patients with type 2 diabetes mellitus (T2DM). To evaluate the cost-effectiveness of canagliflozin in dual therapy as add-on to metformin (MET) compared to sitagliptin, and in triple therapy (add on to MET and sulphonylurea (SU)) compared to sitagliptin, liraglutide and a mixed strategy with both drugs. Methods: The IMS CORE Diabetes Model was used to evaluate the cost-effectiveness of canagliflozin versus the aforementioned comparators using French-specific data, where available. Results: In dual therapy, as addon to metformin versus sitagliptin, canagliflozin (100 mg and 300 mg weighted average 50: 50) resulted in an incremental cost of 392 € and an average QALY gain of 0.013, which leads to an ICER of 30,154 € . In triple therapy canagliflozin (100 mg and 300 mg weighted average 50: 50) dominates sitagliptin with average cost savings of 286 € and an average QALY gain of 0.050. Canagliflozin was estimated to be cost-saving (1,280 € ) compared to liraglutide (distribution liraglutide 1.2 mg / 1.8 mg= 0.71/0.29 based on French market research data) with incremental QALYs of -0.015. Canagliflozin (100 mg and 300 mg weighted average 50: 50) dominates a mixed strategy with liraglutide (12.5%) and sitagliptin (87.5%) average cost saving of 410 € and average QALY gain of 0.041. Sensitivity analyses showed that HbA1cand SBP treatment effects were key drivers of the cost effectiveness results. Conclusions: Canagliflozin 100 mg and 300 mg will be a cost-effective alternative to sitagliptin in dual therapy as add-on to metformin. In triple therapy as add-on to metformin and SU, canagliflozin dominates in comparison with a mix of patients treated with sitagliptin or liraglutide. Canagliflozin 100 mg or 300 mg is expected to be considered good value for money for the treatment of T2DM in France. PDB70 The Cost-Effectiveness of Exenatide Bid Versus Insulin Lispro Tid As Add-On Therapy to Titrated Insulin Glargine in Patients With Type 2 Diabetes – An Analysis From The Swedish Health Care Perspective Gordon J 1, McEwan P 1, Sabale U 2, Kartman B 2 Economics and Outcomes Research Ltd, Cardiff, UK, 2AstraZeneca Nordic-Baltic, Södertälje, Sweden . . . . 1Health Objectives: In patients with type 2 diabetes (T2D) receiving basal insulin therapy, the addition of bolus insulin is a common therapeutic strategy once glycated hemoglobin (HbA1c) becomes uncontrolled. However, this strategy is associated with weight gain and increased risk of hypoglycemia. The objective was to assess the cost-effectiveness of exenatide twice daily (BID) (‘ExBID’) versus insulin lispro three times daily (TID) (‘LisTID’) as add-on therapy to titrated insulin glargine in patients with T2D, from the Swedish health care perspective. Methods: This analysis utilized the Cardiff Model, a previously published T2D disease model. Treatment effects were taken from the 4B Study, where ExBID and LisTID resulted in similar glycemic control with less hypoglycemia with ExBID. Furthermore, ExBID was associated with a weight loss of 2.6kg and LisTID with a weight gain of 1.9kg. An assumption made in this analysis is that once the HbA1c level returned to its baseline value, patients escalated to intensified basal-bolus therapy based on published treatment effects. Quality-adjusted life years (QALYs) were calculated, with health-state utilities applied to weight changes, hypoglycemia, and T2D-related complications. Costs (Swedish kronor, SEK) included medication and T2D-related complications. The model was run over a 40-year time horizon. Costs and QALYs were discounted at 3% annually. Results: Cost/patient with ExBID and LisTID was SEK 381,177 and SEK 364,381, respectively. ExBID was associated with a QALY gain of 0.64 per patient versus LisTID (11.51 and 10.86 QALYs, respectively). This result was driven by the difference in weight between ExBID and LisTID. The cost/QALY gained with ExBID versus LisTID was SEK 26,292. Conclusions: With a cost/QALY gained of SEK 26,292 (3,039 Euros, 2,582 British pounds, 4,033 US dollars), evaluated against commonly used cost-effectiveness threshold values, ExBID represents a cost-effective treatment alternative to LisTID as add-on therapy to titrated insulin glargine in patients with T2D in Sweden. PDB71 The Cost-Effectiveness OF Canagliflozin (Cana) Versus Dapagliflozin (Dapa) In Patients With Type 2 Diabetes Mellitus (T2dm) With Inadequate Control On Metformin (Met) Monotherapy In The United Kingdom Schroeder M 1, Johansen P 2, Thompson G 1, Willis M 3, Neslusan C 4 UK, High Wycombe, UK, 2The Swedish Institute for Health Economics (IHE), Lund, Sweden, 3The Swedish Institute for Health Economics, Lund, Sweden, 4Janssen Global Services, LLC, Raritan, NJ, USA . 1Janssen-Cilag . . . . Objectives: Two SGLT-2 inhibitors, CANA and DAPA, are recommended in the UK for combination therapy in T2DM. Through an insulin-independent mechanism of action, SGLT-2 inhibitors improve glucose levels, blood pressure, and weight with a low inherent risk of hypoglycaemia. The cost-effectiveness of using CANA or DAPA in combination with MET was evaluated in patients inadequately controlled with MET monotherapy, from the perspective of the UK NHS. Methods: The ECHOT2DM model was used to estimate 40-year outcomes and costs associated with using CANA (100mg or 300mg) versus DAPA 10mg in dual therapy. HbA1c efficacy estimates were obtained from a Network Meta-Analysis (NMA). Analyses of pooled data from trials investigating CANA dual therapy (with MET) were used for parameters unavailable in the NMA (i. e., SBP, LDL, HDL and AEs). A broad set of sensitivity analyses were performed. Results: Both doses of CANA were associated with more QALYs (0.01,0. 03 for 100mg and 300mg, respectively) and higher costs (£101, £594, respectively). The associated incremental cost-effectiveness ratios (ICERs) were £7,423 and £17,734, respectively; both below the willingness-to-pay for QALY threshold in the UK. The key driver of the result for CANA 300mg was the greater HbA1c efficacy versus DAPA 10mg, and for CANA 100mg, a lower need for insulin rescue since CANA (but not DAPA) can be used in persons with moderate renal impairment. The ICERs were robust under all scenarios tested. Only use of UK-specific and not clinical trial patient characteristics substantively impacted the results; none reversed the interpretation of CANA as cost-effective versus DAPA. Conclusions: SGLT-2 inhibitors reduce HbA1C, body weight, and blood pressure, and thus the risk of micro- and macrovascular complications. Economic simulations suggest that both doses of CANA are cost-effective versus DAPA in dual therapy treatment of T2DM (with MET) in the UK. PDB72 Health-Economic Comparison of Sensor-Augmented Pump With Low Glucose Suspend Versus Insulin Pump Alone For The Treatment of Hypo-Prone Type 1 Diabetes In Hungary Roze S 1, Lynch P 2, Boncz I 3, Dunne N 2, Varga C 4, Klots M 2, Karamalis M 2, Felszeghy E 5 1HEVA HEOR, Lyon, France, 2Medtronic, Tolochenaz, Switzerland, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary, 4Medtronic Hungary, Budapest, Hungary, 5University of Debrecen, Medical School and Health Science Centre, Pediatric Dept., Debrecen, Hungary . . . . . . . . Objectives: To project the long-term costs and outcomes of sensor-augmented pump (SAP) with low glucose suspend (LGS) versus insulin pump (CSII) alone for the treatment of hypo-prone Type 1 diabetes in Hungary. Methods: The CORE Diabetes Model is a peer-reviewed, validated model, which employs standard Markov/Monte Carlo simulation techniques to describe the long-term incidence and progression of diabetes-related complications. It was used to simulate disease progression in a cohort of patients with baseline characteristics (mean age 18.6 years, duration of diabetes 12 years, mean HbA1c 7.5%) and clinical outcomes (severe hypoglycaemic event rates; Quality of Life; HbA1c) taken from a recent randomised controlled trial (Ly et al, 2013). Local treatment and complication cost data was used. The main scenario considered in this cost-effectiveness analysis was the comparison of sensor-augmented insulin pump (SAP) with low glucose suspend (LGS) versus pump alone (CSII). The target population was hypo-prone type I diabetes patients with the analysis based on a deterministic microsimulation of 1,000 patients, using a 1 to 5 year time horizon. Direct costs were calculated from a third-party payer perspective. Discount rates of 3.7% per annum were applied to both costs and clinical outcomes. Results: The Incremental-Cost-Effectiveness-Ratio (ICER) for SAP+LGS vs CSII was HUF 6,196,086 (€ 20,298) per Quality-Adjusted-Life-Year gained over a 1 year time horizon. Results were similar using a 5 year time horizon (HUF 6,125,823 [€ 20,068] per QALY gained). Extensive sensitivity analyses showed the robustness of the results. Conclusions: Using a payer’s perspective, our analysis showed that SAP (w LGS) is cost-effective over a short term (1-5 year) time horizon in hypo prone patients with Type 1 Diabetes in Hungary (using a WTP threshold of 3x Hungary GDP). PDB73 Economic Assessment of Delaying Insulin Treatment Through The Use of Newer Anti-Diabetic Agents, Dapagliflozin (Forxiga®) And Exenatide (Bydureon®), Both As Add-On To Metformin; A CostEffectiveness Analysis From A Uk Nhs Perspective Charokopou M 1, Vioix H 2, Verheggen B G 1, Maddocks D 2, Bratt T 2, Franks D 2 International, Rotterdam, The Netherlands, 2AstraZeneca UK Ltd., Luton, UK . . . . . . . 1Pharmerit Objectives: New classes of anti-diabetes drugs may delay the onset of Insulin treatment. This study investigates the cost-effectiveness of a treatment pathway starting with dapagliflozin (Forxiga®), followed by exenatide once weekly (Bydureon®) - both as add-on to metformin - and insulin treatment, compared with a treatment pathway recommended in clinical guidelines that commences with SU add-on to metformin, followed by the addition of insulin regimens in patients inadequately controlled with metformin alone. Methods: The validated CARDIFF model was used for the analyses. Clinical inputs for dapagliflozin versus SU, both as add-on to metformin, and exenatide once weekly were derived from relevant head-to-head clinical trials and long-term follow up studies. Based on these and the United Kingdom Prospective Diabetes Study (UKPDS) equations, the model predicts disease progression and number of micro- and macro-vascular complications, along with diabetes-specific and all-cause mortality. The perspective of the National Health Service in UK was adopted over a lifetime horizon. Local unit costs and utility data were assigned to the appropriate model parameters to calculate total Quality-Adjusted-Life-Years (QALYs) and costs. Deterministic and probabilistic sensitivity analyses (PSA) were conducted. Results: Long-term evidence showed that the durability of the treatment effects of dapagliflozin and exenatide can delay the onset of insulin treatment by 5-6 years. Compared to the traditional clinical practice, treatment with dapagliflozin+metformin followed by exenatide+metformin, was associated with an incremental benefit of 0.343 QALYs (95%CI: 0.239; 0.450) at an additional cost of £2,827 (95%CI: £2,352; £3,267), resulting in an incremental cost-effectiveness ratio of £8,233 per QALY gained. The PSA A345 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 showed that at a willingness-to-pay threshold of £20,000 per QALY, the proposed treatment pathway had a 100% probability to be cost-effective. Conclusions: The proposed alternative treatment sequence was shown to be a cost-effective treatment option in patients inadequately controlled with metformin alone within established UK cost-effectiveness thresholds. PDB74 Cost-Effectiveness Analysis of Liraglutide Versus Sitagliptin or Exenatide in Patients With Inadequately Controlled Type 2 Diabetes On Oral Antidiabetic Drugs In Greece Tzanetakos C 1, Melidonis A 2, Verras C 2, Kourlaba G 3, Maniadakis N 1 1National School of Public Health, Athens, Greece, 2Tzanio General Hospital, Piraeus, Greece, 3National and Kapodistrian University of Athens School of Medicine, Athens, Greece . . . . . Objectives: To evaluate the long-term cost-effectiveness of liraglutide versus sitagliptin or exenatide, added to oral antidiabetic drug mono- or combination therapy respectively, in patients with Type 2 diabetes in Greece. Methods: The CORE Diabetes Model, a validated computer simulation model developed to determine the long-term health and economic outcomes of interventions in Type 2 diabetes, was adapted to the Greek health care setting. Patient and intervention effects data were gathered from a clinical trial comparing liraglutide 1.2mg once daily vs. sitagliptin 100mg once daily, both combined with metformin, and a clinical trial comparing liraglutide 1.8mg once daily vs. exenatide 10μ g twice daily, both as add-on to metformin, glimepiride or both. Direct costs were reported in 2013 Euros and calculated based on published and local sources. All future outcomes were discounted at 3.5% per annum, and the analysis was conducted from the perspective of a third-party payer in Greece. Results: Over a patient’s lifetime, treatment with liraglutide 1.2mg vs. sitagliptin drove a mean increase in discounted life expectancy of 0.13 (SD 0.23) years and in discounted quality-adjusted life expectancy of 0.19 (0.16) quality-adjusted life years (QALYs), whereas therapy with liraglutide 1.8mg vs. exenatide yielded increases of 0.14 (0.23) years and 0.19 (0.16) QALYs respectively. As regards lifetime direct costs, liraglutide 1.2mg resulted in greater costs of € 2797 (€ 1468) versus sitagliptin, and so did liraglutide 1.8mg compared with exenatide (€ 1302 [€ 1492]). Liraglutide 1.2 and 1.8mg doses were associated with incremental cost effectiveness ratios of € 15101 and € 6818 per QALY gained, respectively. Conclusions: Liraglutide is likely to be a cost-effective option for the treatment of Type 2 diabetes in a Greek setting. PDB75 Health Economic Evaluation of Canagliflozin In The Treatment of Type 2 Diabetes Mellitus In Slovakia Troelsgaard A 1, Pitcher A 2, Binder R 3, Stetka R 4, Ondrusova M 5, Lovato E 2, Hemels M 1 Health, London, UK, 3Janssen, Bratislava, Slovak Republic, 4Pharm-In, Ltd, Bratislava, Slovak Republic, 5Pharm-In Ltd, Bratislava, Slovak Republic . . . . . . . 1Janssen A/S, Birkerød, Denmark, 2IMS Objectives: Canagliflozin is a novel drug for treatment of diabetes belonging to the drug class known as sodium glucose co-transporter–2 (SGLT-2) inhibitors. To evaluate the cost-effectiveness of canagliflozin 100 mg in the Slovakian setting from a payer perspective when compared to sitagliptin in dual therapy (add-on to metformin), sitagliptin in triple therapy (add-on to metformin plus sulfonylurea) and dapagliflozin in combination with insulin (with or without metformin). Methods: The IMS CORE Diabetes Model was used to evaluate the cost-effectiveness of canagliflozin versus comparators using Slovakia specific data, where available. Results: The cost-effectiveness analyses indicated that canagliflozin 100 mg in dual therapy when compared with sitagliptin, was found to be cost-effective with an incremental cost-effectiveness ratio (ICER) of 14,930 € per QALY gained. In triple therapy when compared to sitagliptin, canagliflozin 100 mg was found to be cost-effective with an ICER of 5,251 € per QALY gained and in combination with insulin, canagliflozin 100 mg was a dominant alternative compared to dapagliflozin with a cost saving of 118 € per patient and higher QALYs. One-way sensitivity analyses revealed that in the majority of scenarios considered canagliflozin remained cost-effective in the dual therapy, triple therapy and add-on to insulin comparisons. Conclusions: Based on calculations performed using the CORE Diabetes model, canagliflozin 100 mg appear to represent a cost-effective option for the treatment of type 2 diabetes in Slovakia. Canagliflozin 100 mg was found to offer greater health benefits than currently available alternatives and to be a cost-effective treatment option when used in dual and triple therapy instead of sitagliptin or as an add-on to insulin instead of dapagliflozin. PDB76 Is Canagliflozin Cost-Effective Compared to Sitagliptin Across Multiple Lines of Type 2 Diabetes Mellitus (T2dm) Therapy In Ireland? Bacon T 1, Willis M 2, Johansen P 2, Neslusan C 3, Nuhoho S 4, Worbes-Cerezo M 5 1Janssen-Cilag Ltd, Dublin, Ireland, 2The Swedish Institute for Health Economics, Lund, Sweden, 3Janssen Global Services, LLC, Raritan, NJ, USA, 4Janssen-Cilag A/S, Birkerød, Denmark, 5Janssen-Cilag UK, High Wycombe, UK . . . . were sourced from the literature. Both costs and outcomes were discounted at 5%. Results: Incremental costs, QALYs and ICERs for canagliflozin vs. sitagliptin were € 1,360, 0.059 QALYs and € 23,118 per QALY, respectively, in dual therapy; € 108, 0.093 QALYs and € 1,172 per QALY, respectively, in triple therapy; and € 550, 0.068 QALYs and € 8,047 per QALY, respectively, in add-on to insulin. In all three scenarios, canagliflozin was cost-effective using the acceptable willingness-to-pay threshold in Ireland. Sensitivity analyses suggest that these results are robust. Conclusions: These simulations suggest that the use of canagliflozin in patients in need of additional glycaemic control in dual, triple and add-on to insulin lines of therapy is a more efficient use of health care funds than the use of sitagliptin in the Irish setting. . . Objectives: Canagliflozin is a new oral agent for the treatment of T2DM that inhibits sodium-glucose co-transporter 2 (SGLT2), thereby leading to inhibition of glucose reabsorption and urinary glucose excretion which results in reductions in blood glucose, weight, and blood pressure. The purpose of this analysis was to evaluate the cost-effectiveness of canagliflozin vs. sitagliptin – a recommended and widely used dipeptidyl peptidase-4 inhibitor (DPP4) – in dual (with metformin), triple (with metformin + sulphonylurea) and add-on to basal insulin (with or without other anti-hyperglycaemic agents) therapy lines in Ireland. Methods: The Economic and Health Outcomes Model of T2DM (ECHO-T2DM) [using updated UKPDS 82 mortality and risk equations] was used to simulate lifetime outcomes and costs of patients on either canagliflozin (100mg, titrated to 300mg as needed to maintain glycaemic control) or sitagliptin 100mg. Patient characteristics and treatment effects were sourced from head to head randomized clinical trials for dual and triple therapy. For the add-on to insulin therapy simulations, treatment effects were sourced from a network meta-analysis. Costs were localised and inflated to 2013 euros. Utilities PDB77 The Cost-Effectiveness Of Canagliflozin VersUS Liraglutide In Patients With Type 2 Diabetes (T2dm) Failing To Achieve Glycaemic Control On Metformin Monotherapy In Ireland Bacon T 1, Willis M 2, Johansen P 2, Neslusan C 3, Nuhoho S 4, Worbes-Cerezo M 5 1Janssen-Cilag Ltd, Dublin, Ireland, 2The Swedish Institute for Health Economics, Lund, Sweden, 3Janssen Global Services, LLC, Raritan, NJ, USA, 4Janssen-Cilag A/S, Birkerød, Denmark, 5JanssenCilag UK, High Wycombe, UK . . . . . . Objectives: Canagliflozin is a novel oral agent for the treatment of T2DM that inhibits sodium-glucose co-transporter 2 (SGLT2), a mechanism that is complementary to other anti-hyperglycaemic drug classes, including insulin. SGLT2 inhibition leads to inhibition of glucose reabsorption and urinary glucose excretion, thereby reducing blood glucose, weight, and blood pressure. An economic (cost-effectiveness) evaluation of new technologies versus routine care is required prior to uptake in Ireland to ensure good value-for-money. This study evaluates the cost-effectiveness of canagliflozin compared to liraglutide – a currently reimbursed and routinely used glucagon-like peptide 1 (GLP-1) agonist – as dual therapy in combination with metformin from the payer perspective in the Irish health care setting. Methods: The Economic and Health Outcomes Model of T2DM (ECHO-T2DM) [using updated UKPDS 82 mortality and risk equations] was used to simulate 40-year costs and outcomes associated with canagliflozin (100mg titrated to 300mg in patients requiring tighter glycaemic control) compared to liraglutide 1.2mg. Patient characteristics were sourced from canagliflozin RCTs in patients uncontrolled on metformin monotherapy and treatment effects were obtained from a network meta-analysis. The costs of treatments and outcomes were localised and inflated to 2013 values where possible. Utilities were sourced from the literature. Costs and outcomes were discounted at 5% annually. Results: In the base case, canagliflozin was associated with incremental cost savings of € 3,382 compared to liraglutide 1.2mg. It was also associated with a 0.022 more life years and 0.020 more quality adjusted life years (QALYs), suggesting that canagliflozin dominates liraglutide. Results were driven by lower acquisition costs for canagliflozin. Sensitivity analyses indicated that the dominance observed was robust. Conclusions: In these simulations, canagliflozin dominated liraglutide 1.2mg in dual therapy (add-on to metformin) in the Irish setting. These results suggest that canagliflozin represents good ‘value for money’ in treating these patients, compared to a routinely used GLP-1 agonist. PDB78 Health Economic Evaluation of Canagliflozin In The Treatment of Type 2 Diabetes Mellitus In Norway Troelsgaard A 1, Huetson P 2, Kjellberg J 3, Hemels M 1, Knudsen M 4 1Janssen A/S, Birkerød, Denmark, 2IMS Health, Stockholm, Sweden, 3Zefferin Farma AB, Stockholm, Sweden, 4IMS Health, Hellerup, Denmark . . . . . Objectives: Canagliflozin is a novel drug for treatment of diabetes belonging to the drug class known as sodium glucose co-transporter 2 (SGLT2) inhibitors. To evaluate the cost-effectiveness of canagliflozin in dual therapy (add-on to metformin) compared to sitagliptin and sulfonylurea (SU), in triple therapy (add on to metformin and SU) compared to sitagliptin and as add-on to insulin versus placebo. Methods: The IMS CORE Diabetes Model was used to evaluate the cost-effectiveness of canagliflozin (using a weighted average of 80/20 for the 100 mg and 300 mg dosage, respectively) versus the aforementioned comparators using Norwegian-specific data, where available. Results: In dual therapy, as add-on to metformin versus sitagliptin, canagliflozin appears to dominate sitagliptin with average cost savings of 606 NOK and an average QALY gain of 0.030 and as add on to metformin canagliflozin is cost-effective versus SU, with an incremental cost-effectiveness ratio (ICER) of 79,309 NOK and an incremental cost of 5,757.80 NOK and an average QALY gain of 0.0726. As add on to insulin canagliflozin appears to dominate placebo with an incremental cost saving of 13,506 NOK and an incremental QALY of 0.080. In triple therapy as add on to metformin and SU canagliflozin appears to dominate sitagliptin with average cost savings of 556 NOK and an average QALY gain of 0.021. Conclusions: Canagliflozin is associated with cost savings and QALY gain compared to sitagliptin in dual therapy as add-on to metformin, and in triple therapy as add on to metformin and SU. Canagliflozin will be a cost-effective alternative to SU in dual therapy as add on to metformin. Adding canagliflozin to insulin will be cost-effective compared with placebo i. e. it is cost-effective to add canagliflozin treatment rather than not. PDB79 Health-Economic Comparison of Sensor-Augmented Pump With Low Glucose Suspend Versus Insulin Pump Alone For The Treatment Of Hypo-Prone Type 1 Diabetes In Slovakia Roze S 1, Lynch P 2, Machova R 3, Micieta V 3, Dunne N 2, Klots M 2, Karamalis M 2, Martinka E 4 1HEVA HEOR, Lyon, France, 2Medtronic, Tolochenaz, Switzerland, 3Medtronic Slovakia, Bratislava, Slovak Republic, 4National Endo and Diabetes Center, Bratislava, Slovak Republic . . . . . . . . Objectives: To project the long-term costs and outcomes of sensor-augmented pump (SAP) with low glucose suspend (LGS) versus insulin pump alone (CSII) for the treatment of hypo-prone Type 1 diabetes in Slovakia. Methods: The CORE Diabetes Model is a peer-reviewed, validated model, which employs standard A346 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Markov/Monte Carlo simulation techniques to describe the long-term incidence and progression of diabetes-related complications. It was used to simulate disease progression in a cohort of patients with baseline characteristics (mean age 18.6 years, duration of diabetes 12 years, mean HbA1c 7.5%) and clinical outcomes (severe hypoglycaemic event rates; Quality of Life; HbA1c) taken from a recent randomised controlled trial (Ly et al, 2013). Local treatment and complication cost data was used. The main scenario considered in this cost-effectiveness analysis was the comparison of sensor-augmented insulin pump (SAP) with low glucose suspend (LGS) versus insulin pump alone (CSII). The target population was type I hypo-prone diabetes patients with the analysis based on a deterministic microsimulation of 1,000 patients, using a 5 year time horizon. Direct costs were calculated from a third-party payer perspective. Discount rates of 3% per annum were applied to both costs and clinical outcomes. Results: The Incremental-CostEffectiveness-Ratio (ICER) for SAP+LGS (vs CSII) was € 17,893 per Quality-AdjustedLife-Year gained over a 5 year time horizon. Results were similar across a 1 to 10 year time horizon. Other extensive sensitivity analyses showed the robustness of the results. Conclusions: Using a payer’s perspective, our analysis showed that SAP (w LGS) is cost-effective over a short term (5 year) time horizon in hypoprone Type 1 Diabetes patients in Slovakia (using a WTP threshold of 1x [€ 18,000] or 3x [€ 54,000] Slovakia GDP). PDB80 Is a Home Based Video Teleconcultation Setup Cost Effective For Lowering Hba1c For Patients With Type-2 Diabetes Over A Six-Month Period? Jensen M S 1, Rasmussen O W 2 1Region of Southern Denmark, Middelfart, Denmark, 2Kolding Hospital, Kolding, Denmark . . . . Objectives: A RCT assessed the effectiveness and costs of a home based video teleconsultation (HVT) setup to lower HbA1c in patients with type-2 diabetes against usual out-patient treatment on the hospital. The HVT equipment was delivered to the patients by the hospital. This analysis shows the potential incremental cost-effectiveness ratio (ICER) of using a HVT setup on six-months health care effects and costs. Methods: The study effectiveness outcome was HbA1c level in %. The economic analysis was performed with a spreadsheet decision tree model with a Danish hospital payer’s direct cost perspective. Cost data were based on study measured time consumption pr. HVT, consultations at out-patient clinic, HVT-equipment, -subscription, -support costs, and hospital operating cost. Medicine costs weren’t included in the model. Model output included the cost of a 1 % point reduction of HbA1c, ICER, with a probabilistic sensitivity analysis (PSA). Two scenario analyses (SA) were made to capture costs of patient transport to the hospital and a future online platform, were patients can use their own computer/tablet to the video teleconsultations. Results: A total of 39 patients (mean age 62, HbA1c 8.5%) were randomized to either usual care (UC) or HVT. At 6 months follow up the HVT group showed greater improvements from baseline HbA1c levels (-1.38% vs. -0.92%) and less costly (€ 199.9 vs. € 208.2) against UC. The base case ICER showed a potential € –17.58 saving per reduction of 1% HbA1c point. A PSA confirmed the ICER trends despite data uncertainties. Both SA showed further savings (ICER: € -67.85 and € -69.13). Compliance was 100% for HVT group were several planned visits were cancelled in the UC group. Conclusions: The present analysis shows the potential benefits of a HVT setup on 6-months health care cost and effects against UC. Further savings could include cost associated with lost work days. PDB81 The Cost-Effectiveness of Canagliflozin VersUS InsulinSecretagogues (Sulphonylureas) or Insulin In Patients With Type 2 Diabetes Mellitus (T2dm) As An Add-On To Metformin In Ireland Bacon T 1, Willis M 2, Johansen P 2, Neslusan C 3, Nuhoho S 4, Worbes-Cerezo M 5 1Janssen-Cilag Ltd, Dublin, Ireland, 2The Swedish Institute for Health Economics, Lund, Sweden, 3Janssen Global Services, LLC, Raritan, NJ, USA, 4Janssen-Cilag A/S, Birkerød, Denmark, 5Janssen-Cilag UK, High Wycombe, UK . . . . . . Objectives: Sulphonylureas (SU) and insulin are used routinely in the management of T2DM but are associated with weight gain and increased risk of hypoglycaemia. Canagliflozin is a new insulin-independent oral glucose lowering agent with added benefits of weight loss, blood pressure reduction and no increased risk of hypoglycaemia. This analysis estimated the cost-effectiveness of canagliflozin compared to either SU or insulin in patients failing to achieve glycaemic control on metformin monotherapy in Ireland. Methods: The Economic and Health Outcomes Model of T2DM (ECHO-T2DM) was used to simulate the lifetime outcomes and costs associated with canagliflozin (100mg, titrated to 300mg as needed to maintain glycaemic control) versus SU and versus insulin glargine. Patient characteristics and treatment effects for the SU comparison were sourced from a head-to-head randomized clinical trial vs. glimepiride. Hypoglycaemia rates were halved to reflect gliclazide MR (the preferred SU in Ireland). Patient characteristics for the insulin glargine comparison were obtained from the pooled canagliflozin add-on to metformin RCTs; treatment effects were sourced from a network meta-analysis. Costs were localised and inflated to 2013 euros. Utilities were sourced from the literature. Costs and outcomes were discounted at 5% annually. Results: The incremental costs, QALY gains and ICERs associated with canagliflozin were € 2,404, 0.215 QALYs and € 11,191 per QALY gained, respectively, versus SU and € 2,352, 0.228 QALYs and € 10,305 per QALY gained, respectively, versus insulin glargine. Key drivers were decreased hypoglycaemia and lower weight-related disutility versus both comparators, as well as better HbA 1c durability versus SU. In both cases, using an acceptable Irish willingness-to-pay threshold, the probability of being cost-effective was in excess of 97%. Sensitivity analyses support the robustness of these results. Conclusions: These simulations suggest that canagliflozin is a cost-effective treatment choice versus both gliclazide MR and insulin glargine in patients failing to control glycaemia on metformin alone. PDB82 Cost Effectiveness Evaluation of Canagliflozin In Combination With Metformin in the Treatment Of Type 2 Diabetes Mellitus In Poland Szmurlo D 1, Drzal R 1, Plisko R 1, Schubert A 2, Skrzekowska-Baran I 3 1HTA Consulting, Krakow, Poland, 2Janssen Cilag Poland, Warszawa, Poland, 3Janssen-Cilag Polska, Warszawa, Poland . . . . . Objectives: To evaluate the cost-effectiveness of canagliflozin, an active inhibitor of sodium glucose co-transporter – 2 (SGLT2) in dual therapy as add-on to metformin compared to sitagliptin and glimepiride. Canagliflozin in clinical trial results showed effective glucose reduction, along with other benefits in diabetes treatment including weight loss and SBP reduction. Cost effectiveness analyses were conducted in the Polish setting from a public perspective in accordance with guidelines of Polish HTA Agency (PolAHTA). Methods: The IMS CORE Diabetes Model was used to evaluate the cost-effectiveness of canagliflozin versus the aforementioned comparators using Polish-specific data, where available. Direct costs were reported in Polish zloty and an annual discount rate of 5% and 3.5% were applied on costs and effects respectively. Results: In dual therapy as add-on to metformin, canagliflozin 100 mg dominates sitagliptin with average cost savings of 2 811 zł and an average QALY gain of 0.06, canagliflozin 300 mg is cost effective option in comparison to sitagliptin with an incremental cost effectiveness ratio (ICER) of 45 008 zł per QALY and QALY gain of 0.09. As add-on to metformin canagliflozin is a cost effectiveness option in comparison with glimepiride with ICER of 28 454 zł and 73 102 zł, QALY gain 0,112 QALY and 0,140 QALY for canagliflozin 100 mg and 300 mg respectively. All results are below defined in Polish reimbursement act cost-effectiveness threshold. Conclusions: These results suggest that adding Canagliflozin to metformin versus sitagliptin or glimepiride in patients inadequately controlled with metformin would be a more efficient use of health care resources in the Polish setting. PDB84 Cost-Effectiveness of Interventions Aimed at Decreasing The Number of Amputations Among Patients With Diabetes Mellitus Ignatyeva V 1, Avxentyeva M 1, Galstyan G R 2, Bregovskiy V 3, Udovichenko O 4 1The Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia, 2The Endocrinological Scientific Center, Moscow, Russia, 3Federal Almazov Medical Research Center, St. Petersburg, Russia, 4Moscow Municipal outpatient clinic #22, Moscow, Russia . . . . . . Objectives: To evaluate the cost-effectiveness of interventions aimed at decreasing the number of amputations among patients with diabetic foot ulcers (DFU) in Russia. Methods: We have modeled the changes in the annual outcomes (minor and major amputations) and costs (services provided in outpatient clinics and hospitals, medications, orthopedic shoes and prosthetic devices and services provided in case of amputation) from the perspective of public health and social care. Two interventions were assessed: preventive services for patients with the very high risk of DFU (additional outpatient visits for foot care and orthopedic shoes) and provision of care for DFU patients at hospital by multidisciplinary foot care team (MDT). The current number of amputations and costs among DFU patients in Russia was assessed on the basis of published Russian data and experts’ survey. The expected effectiveness of interventions was derived from the international publications. Costs were estimated on the basis on reimbursement rates in public medical insurance and social care. Results: The implementation of hospital care by MDT for cohort of 1000 DFU patients at the current rate of hospitalizations will require additional annual spending of € 532,520, and the expected annual number of major amputations will decrease by 41. The ICER for this intervention is € 12,988 per prevented amputation, which is almost 2 times higher than the costs associated with major amputation at the current moment. For the preventive services, if all patients are compliant, additional costs per prevented amputation are slightly lower - € 10,216, but also well above the costs of major amputation. Conclusions: Both interventions require considerable additional budget spending. Preventive measures, if all the patients follow the recommendations, are more cost effective than introduction of hospital MDT. PDB85 The Cost-Effectiveness of Canagliflozin Compared With Liraglutide in Patients With Type 2 Diabetes Inadequately Controlled With Metformin and Sulfonylurea In France Troelsgaard A 1, Pitcher A 2, Granados D 3, Hemels M 1, Lloyd A 2 1Janssen A/S, Birkerød, Denmark, 2IMS Health, London, UK, 3Janssen, Paris, France . . . . . Objectives: Canagliflozin is a sodium-glucose co-transporter 2 (SGLT2) inhibitor used in treatment of patients with type 2 diabetes mellitus (T2DM). The objective is to estimate the cost-effectiveness of canagliflozin (100mg once daily and 300mg once daily) compared with liraglutide in combination with metformin and sulfonylurea (SU) for the treatment of T2DM inadequately controlled with metformin and SU in France. Methods: The IMS CORE Diabetes Model was used to project clinical and economic outcomes for patients with T2DM treated with canagliflozin or liraglutide, each in combination with metformin and SU. Since direct trial data were not available, the relative treatment effects on HbA1c, SBP and BMI for liraglutide 1.8mg in combination with metformin and SU were derived from a network meta-analysis (NMA) of treatment effects at 26 weeks. This study is limited by the absence of direct or indirect data on the effect of liraglutide 1.2mg in combination with metformin and SU, therefore the relative treatment effects on HbA1cand BMI at 26 weeks for liraglutide 1.2mg were estimated using the dose-response relationship from a NMA based on treatments in combination with metformin only. French market share data were used to weight the results of liraglutide 1.8mg and 1.2mg. Results: Canagliflozin 100mg showed cost savings when compared to treatment with liraglutide (1,388 € ); incremental QALYs were estimated as -0,035. Canagliflozin 300 mg was dominant, with cost savings of 1,411 € and relatively small incremental QALY gain of 0.003. Conclusions: The analyses found that treatment of T2DM with canagliflozin 100mg or 300mg instead of liraglutide as add on to metformin A347 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 and SU would generate substantial cost savings in France. Canagliflozin 300mg showed similar outcomes to liraglutide, suggesting that the treatments are similar in effectiveness, and is thus likely to be a highly cost-effective treatment option. PDB86 Layering Interventions for Type-2 Diabetes Prevention Using The Sphr Diabetes Model Thomas C 1, Watson P 1, Greaves C 2, Squires H 1, Chilcott J 1, Brennan A 1 of Sheffield, Sheffield, UK, 2University of Exeter, Exeter, UK . . . . . . 1University Objectives: We have developed a model to evaluate type-2 diabetes prevention interventions. The model allows flexible layering of multiple interventions in order to determine the optimal combination of strategies for maximal cost-effectiveness. Our objective was to demonstrate the utility of the model for analysis of multiple interventions in different population sub-groups. Methods: A model of type-2 diabetes prevention was developed using a micro-simulation framework. Individual patients have simulated trajectories of metabolic factors such as BMI, HbA1c, systolic blood pressure and total cholesterol, and can be diagnosed with diabetes, cardiovascular disease, cancer, osteoarthritis or microvascular complications of diabetes over the course of a lifetime. Interventions targeting metabolic factors influence the likelihood of patients developing ill health and dying prematurely. Five interventions were chosen for layering analyses: soft drinks taxation; retail policy; a workplace intervention; an educational intervention aimed at deprived individuals; and a diabetes screening programme followed by intensive intervention for high risk individuals. Eight different intervention combinations were modelled, using the assumption of either additive, synergistic or antagonistic effects in individuals subject to multiple interventions. Results: All interventions generated cost-savings and QALY gains, with the screening intervention performing particularly well, followed by the soft drinks tax. Combining interventions results in roughly additive effects; this holds whether the layering is additive, synergistic or antagonistic due to the relatively low proportions of individuals subject to multiple interventions. Certain pairs of interventions were found to be almost as cost-effective as combining all five interventions, in particular, a soft drinks tax combined with a screening programme is likely to be one of the most cost-effective options. Conclusions: The SPHR Diabetes Model is a useful tool for analysing the cost-effectiveness of different diabetes prevention interventions, either singly or in combination. This will enable development of intervention strategies tailored to the needs of the target population. PDB87 Cost-Effectiveness of Dapagliflozin Versus Dpp-4 Inhibitors as Monotherapy in the Treatment of Type 2 Diabetes Mellitus From A Uk Health Care Perspective Charokopou M 1, Vioix H 2, Verheggen B 1, Eddowes L A 3, Griffiths M 3, Gabriel Z 2, Tolley K 4, Franks D 2 1Pharmerit International, Rotterdam, The Netherlands, 2AstraZeneca UK Ltd., Luton, UK, 3Costello Medical Consulting Ltd., Cambridge, UK, 4Tolley Health Economics Ltd., Buxton, Derbyshire, UK . . . . . . . . . Objectives: To explore the cost-effectiveness of dapagliflozin, the first-in-class sodium-glucose co-transporter-2 (SGLT-2) inhibitor, compared to the dipeptidyl peptidase-4 inhibitor (DPP4i) class, as monotherapy for the treatment of type 2 diabetes mellitus (T2DM) that is inadequately controlled by diet and exercise alone in patients who are intolerant to metformin. Methods: The validated CARDIFF diabetes model was used. Clinical inputs for the model were sourced from a systematic review and network meta-analysis (NMA) in monotherapy. A UK health care perspective was used for costs, and quality-adjusted life years (QALYs) were calculated from utility data in the published literature. A range of assumptions to reflect potential clinical practice were utilised. Alterations to the HbA1c switching threshold and assumptions on weight extrapolation were considered, in order to investigate the impact of different profiles of weight change and longer-term evidence for the maintenance of weight loss. The cost-effectiveness was reported as an incremental cost-effectiveness ratio (ICER). Results: When the model assumptions were varied around the HbA1c switching threshold and around the profile of maintenance of weight loss to better reflect clinical practice, the ICER point estimates ranged from £1,847 to £30,795. The lowest ICER point estimate was associated with an assumption of maintaining the weight differential that is created at treatment initiation, which is generated by the better weight profile of dapagliflozin compared to DPP4i, over the whole time horizon. The highest estimate arose when the weight differential was maintained for 2 years and there was a higher switching threshold for initiating insulin. Conclusions: Dapagliflozin monotherapy has potential as a cost-effective treatment option from a UK health care perspective for patients with T2DM who are intolerant to metformin and are inadequately controlled by diet and exercise alone, based upon a range of explored scenarios. This study also demonstrates the value of performing scenario analysis to inform decision-making. PDB88 Cost-Effectiveness of Dapagliflozin Compared To Dpp-4 Inhibitors as Triple Therapy In Combination With Metformin and A Sulphonylurea In The Treatment Of Type 2 Diabetes Mellitus From A Uk Health Care Perspective Charokopou M 1, Vioix H 2, Eddowes L A 3, Griffiths M 3, Verheggen B G 1, Gabriel Z 2, Tolley K 4 International, Rotterdam, The Netherlands, 2AstraZeneca UK Ltd., Luton, UK, 3Costello Medical Consulting Ltd., Cambridge, UK, 4Tolley Health Economics Ltd., Buxton, Derbyshire, UK . . . . . . . . . 1Pharmerit Objectives: To assess the cost-effectiveness of dapagliflozin, the first-in-class sodium-glucose co-transporter-2 (SGLT-2) inhibitor, compared to the dipeptidyl peptidase-4 inhibitor (DPP4i) class, when used as triple oral therapy in combination with metformin and a sulphonylurea (met+SU) for the treatment of patients with type 2 diabetes mellitus (T2DM) who are inadequately controlled on met+SU alone. Methods: The validated CARDIFF diabetes model was used. Clinical inputs for the model were sourced from a systematic review and network metaanalysis (NMA) in triple therapy that found clinically relevant differences between dapagliflozin and DPP4i’s. A UK health care perspective was used for costs, and quality-adjusted life years (QALYs) were calculated from utility data sourced from the published literature. The cost-effectiveness was reported as an incremental cost-effectiveness ratio (ICER). To assess uncertainty, univariate deterministic and multivariate probabilistic sensitivity analyses (PSA) were performed. Results: Compared to the DPP4i class, dapagliflozin was associated with 0.023 incremental QALYs (95% CI: -0.058, 0.105), at an additional cost of £253 (95% CI: -£302, £741). This resulted in an ICER of £10,995 per QALY gained. The incremental cost associated with dapagliflozin was primarily due to the additional drug acquisition cost, whereas the QALY gain estimated was associated with superior weight reduction and its impact on health-related quality of life for dapagliflozin relative to the DPP4i’s. Univariate analyses demonstrated that the ICER was most sensitive to varying the weight change parameter for the comparator DPP4i’s according to the 95% credible intervals in the NMA, giving ICER estimates ranging from £3,937 to £16,727. Dapagliflozin had a 59% probability of being cost-effective versus DPP4i’s at a willingness-topay threshold of £20,000 per QALY gained. Conclusions: Dapagliflozin as triple therapy in combination with met+SU was shown to be a cost-effective treatment option from a UK health care perspective for patients with T2DM who are inadequately controlled on met+SU. PDB89 Cost-Effectiveness of Saxagliptin Compared To Glp-1 Analogues As An Add-On To Insulin in the Treatment of Type 2 Diabetes Mellitus From A Uk Health Care Perspective Charokopou M 1, Vioix H 2, Verheggen B 1, Eddowes L A 3, Griffiths M 4, Gabriel Z 2, Tolley K 5, Sibartie M 2 1Pharmerit International, Rotterdam, The Netherlands, 2AstraZeneca UK Ltd., Luton, UK, 3Costello Medical Consulting Ltd., Cambridge, UK, 4Costello Medical Consulting, Cambridge, UK, 5Tolley Health Economics Ltd., Buxton, Derbyshire, UK . . . . . . . . . Objectives: To assess the cost-effectiveness of saxagliptin, a dipeptidyl peptidase-4 inhibitor (DPP4i), compared to the glucagon-like peptide-1 (GLP-1) analogues, exenatide twice-daily or lixisenatide, when added to insulin (±metformin) for the treatment of patients with type 2 diabetes mellitus (T2DM) who are inadequately controlled on insulin. Methods: The validated CARDIFF diabetes model was used to conduct the analyses. Clinical inputs for the model were sourced from a systematic review and network meta-analysis (NMA), which found that the treatments were similarly effective at lowering HbA1c and that the GLP-1 analogues resulted in greater and clinically relevant weight loss. A UK health care perspective was used for costs, and quality-adjusted life years (QALYs) were calculated from utility data sourced from the published literature. To assess uncertainty, univariate deterministic and multivariate probabilistic sensitivity analyses (PSA) were performed. Results: Saxagliptin was found to dominate lixisenatide in the base case, being associated with lower costs (-£472 [95%CI: -1,378, -223]) and slightly more QALYs (0.010 [95%CI: -0.040, 0.076]). Saxagliptin was less costly (-£1,402 [95%CI: -£1,932, -£1,218]) and slightly less effective (-0.012 QALYs [95%CI: -0.054, 0.035]) than exenatide twice-daily. The lower costs associated with saxagliptin were primarily due to the lower drug acquisition costs. The estimated QALY differences were small and associated with the impact of weight reduction on health-related quality of life. Univariate sensitivity analyses demonstrated that the results were most sensitive to varying the HbA1c treatment effect and the effect of saxagliptin on weight change. The PSA estimated that saxagliptin had a 70.4% and 34.3% probability of being dominant, and a 99.4% and 100% probability of being cost-saving compared to lixisenatide and exenatide twice-daily, respectively. Conclusions: Saxagliptin as add-on to insulin was shown to be a cost-saving treatment option from a UK health care perspective for patients with T2DM who are inadequately controlled on insulin. PDB90 The Place of Dpp-4 Inhibitors in the Treatment Algorithm of Diabetes Type 2: A Systematic Review of Cost-Effectiveness Studies Baptista A 1, Teixeira I 1, Romano S 1, Vaz Carneiro A 2, Perelman J 3 of Health Research & Evaluation (CEFAR), National Association of Pharmacies (ANF), Lisbon, Portugal, 2Center for Evidence Based Medicine, Faculty of Medicine, University of Lisbon, Lisbon, Portugal, 3Escola Nacional de Saúde Pública, Universidade Nova de Lisboa, Lisboa, Portugal . . . . . 1Centre Objectives: To conduct a systematic review of cost-effectiveness, cost-utility and cost-benefit studies of new inhibitors of DPP-4 for diabetes treatment versus other antidiabetics. Methods: We searched the CRD York, NICE Health Technology Assessment, Tufts CEA Registry, and MEDLINE (PubMed) databases, and grey literature through 2014 to identify cost-effectiveness, cost-utility and cost-benefit studies of new inhibitors of DPP-4 versus other antidiabetics for diabetes treatment. Three investigators independently reviewed all potentially relevant titles and abstracts (1st screening), and subsequently screened full-text articles (2nd screening), according to pre-established inclusion criteria. We restricted our sample to studies with a lifetime or near-lifetime horizon, and adopting either a societal or a health care perspective. The studies should be available as a full-text publication and published in English, French, Spanish, or Portuguese language. A critical appraisal of the methodology and reporting was performed using the 35item version of the BMJ checklist. Results: A total of 57 studies were identified. From these, 17 studies were accepted after the 1st screening, of which 11 were accepted after the 2nd screening. All selected studies consisted in cost-utility analyses. Most studies were based on a single randomized trial per therapy. Saxagliptin was assessed in 6 studies, Sitagliptin in 4, and vildagliptin in 1. Liraglutide was cost-effective versus sitagliptin as second line therapy (ranged from 12,164 to 27,289€ /QALY across countries). Saxagliptin (ranged from 771 to 10,065 € /QALY across countries), Sitagliptin (ranged from 5,949 € /QALY to 20,350 € /QALY across countries) and vildagliptin (ICER of 9,072 € / QALY in Portugal) were cost-effective versus sulfonylureas as second line therapy. Conclusions: According to commonly accepted thresholds, there is consistent evidence about the cost-effectiveness of DDP-4 inhibitors as second line therapy for diabetes type 2. Though, more evidence (including head-to-head) is necessary to define which DDP-4 inhibitor is the most cost-effective. A348 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PDB91 Systematic Review of Economic Evaluations of Dipeptidyl Peptidase-4 Inhibitors for the Treatment of Type 2 Diabetes Mellitus Lizán L 1, Aceituno S 1, Franch J 2, Pérez A 3, Granell M 4, Fuster E 4 Raval Sud- Institut Català de la Salut - USR Barcelona ciutat - IDIAP Jordi Gol, Barcelona, Spain, 3Hospital de la Santa Creu I Sant Pau, Barcelona, Spain, 4Novartis Farmaceutica, Barcelona, Spain . . . . . . 1Outcomes’10, Castellon, Spain, 2EAP Objectives: To synthesize and analyze the available information on the therapeutic value of dipeptidyl peptidase-4 inhibitors (DPP-4) for the treatment of type 2 diabetes mellitus (T2DM) from the point of view of their efficiency or cost-effectiveness. Methods: A systematic review of national (MEDES, IBECS) and international (MedLine/PubMed, Cochrane Library, ISI WOK, SCOPUS) databases was performed. Eligible studies were economic evaluations (cost-effectiveness, cost-utility and costbenefit analysis) comparing costs and clinical benefits of two alternatives for the T2DM treatment including DPP-4 inhibitors, published in English or Spanish until June 2013. Studies providing data concerning costs and/or disease burden only were excluded. Results: Of 1,634 publications initially selected, 284 were excluded by duplicate review, 1,330 by title/abstract review and 8 by criteria accomplishment. A total of 12 publications [1 conducted in 6 countries including Spain, 4 in Europe, 4 in USA and 3 in others] were selected. All studies were cost-effectiveness or costutility analysis [discrete-event simulation (n= 8), Markov (n= 3) and others (n= 1)] Time horizon ranged from one year (n= 1) to lifetime (n= 7), considering 20 years (n= 1), 35 years (n= 2), and 40 years (n= 1). National Health System perspective was adopted in all publications (n= 12). One study did not include a sensitivity analysis. Results showed that T2DM therapy with DPP-4 inhibitors and metformin resulted cost-effective compared with sulfonylureas plus metformin in all countries which were evaluated (Sweden, UK, Germany, Portugal, Austria, Finland, Spain, USA and Argentina). Although DPP-4 inhibitors cost was higher compared with sulfonylureas, DPP-4 inhibitors plus metformin were associated to higher clinical benefits versus sulfonylureas plus metformin in terms of decreasing hypoglycemia incidence and T2DM complications. Conclusions: DPP-4 inhibitors added to metformin are a cost-effective alternative compared with sulfonylureas plus metformin in T2DM patients, mainly due to a lower hypoglycemia incidence and T2DM complications. PDB92 Projection of Long Term Health-Economic Benefits of Sensor Augmented Pump (Sap) Versus Pump Therapy Alone (Csii) In Type 1 Diabetes, A Uk Perspective Roze S 1, Cook M 2, Jethwa M 2, de Portu S 3 1HEVA HEOR, Lyon, France, 2Medtronic UK, Watford, UK, 3Medtronic International Sàrl, Tolochenaz, Switzerland . . . . Objectives: The main objective of this study was to estimate the health-economic impact of sensor augmented pump (SAP) compared to pump therapy alone (CSII) of type 1 diabetes (T1DM) in the UK. Methods: The Core Diabetes Model was used to project the incidence of diabetes-related complications over a lifetime horizon, based on A) a published meta-analysis comparing SAP versus CSII, and B) a real life observational study. The meta-analysis showed that for the analysed cohort of T1DM with average baseline HbA1c of 10%, every day use of SAP led to a reduction of -1.49% versus -0.62% HbA1c, for CGM and CSII respectively. The mean baseline age of the simulated cohort was 27 years, with a mean duration of diabetes of 13 years. SAP effects also included a decrease in the annual rate of major hypoglycaemic events from 2.2 events per 100 patients’ month for CSII to 0. The quality of life was adjusted for a reduced fear of hypoglycaemic event in the SAP arm. Sensitivity analyses were carried out on several key parameters. Results: The incremental cost-effectiveness ratio (ICER) was 16,986 GBP (£) per Quality Adjusted Life Year gained (QALY). The improvement in discounted QALY was 3.1 years in favour of SAP. Additional SAP related costs were partially offset by the savings due to the reduction in diabetes related complications and the lower frequency of SMBG tests. Remaining extra costs due to SAP were on average 1,143£ per year. When a societal perspective was considered, ICER was reduced to 8,462 £ per QALY. Conclusions: Projection of the improvement in HbA1c of SAP versus CSII translated into cost-effective ratio, generally considered as very good value for money in the UK. Extensive sensitivity analysis on key drivers confirmed the robustness of results under a wide range of assumptions. PDB93 Projection of Long Term Health Economic Benefits of Sensor Augmented Pump (Sap) Versus Pump Therapy Alone (Csii) In Uncontrolled Type 1 Diabetes In France Roze S 1, Payet V 2, Debroucker F 2, de Portu S 3, Cucherat M 4 HEOR, Lyon, France, 2Medtronic France SAS, Boulogne-Billancourt, France, 3Medtronic International Sàrl, Tolochenaz, Switzerland, 4UMR CNRS 5558, Faculté de Médecine Laennec, Lyon, France . . . . . reduction in diabetes related complications and the lower frequency of SMBG tests. Remaining extra annual costs for SAP were 1,258 € per patient. When indirect costs were considered, the ICER was reduced to 23,300 € per QALY. Conclusions: Using a well-accepted simulation model in an uncontrolled T1DM, projection of the improvement in HbA1c of SAP versus CSII translated into cost-effective ratio, generally considered as very good value for money in France. Extensive sensitivity analysis on key drivers confirmed the robustness of results under a wide range of assumptions. PDB94 Cost-Effectiveness of A Short Message Service (Sms) Intervention To Prevent Type 2 Diabetes Among Adults With Impaired Glucose Tolerance Wong C K H 1, Jiao F F 1, Siu S C 2, Fung C S C 1, Wong K W 2, Lam C L K 1 1The University of Hong Kong, Hong Kong Island, Hong Kong, 2Tung Wah Eastern Hospital, Hong Kong Island, Hong Kong . . . . . . . . . . . . . . . Objectives: To investigate the costs and cost-effectiveness of a short message service (SMS) intervention to prevent the onset of type 2 diabetes mellitus (T2DM) in subjects with impaired glucose tolerance (IGT). Methods: A Markov model was developed to simulate the cost and effectiveness outcomes of the SMS intervention and usual clinical practice from the health service provider’s perspective. The annual transition probabilities between health states were taken from several data sources, including Diabetes Prevention Program and Diabetes Prevention Program Outcome Study. The direct program costs and the two-year SMS intervention costs were evaluated in subjects with IGT. All costs were expressed in year 2011 U.S. dollars. The incremental cost-effectiveness ratio were calculated as cost per T2DM onset prevented, cost per life year gained and cost per quality adjusted life year (QALY) gained. Both base-case scenario and sensitivity analysis were conducted. Results: Within the two-year trial period, the net intervention cost of the SMS group was $42.03 per subject. The SMS intervention managed to reduce 5.05% onset of diabetes, resulting in saving $118.39 per subject over two years. In the lifetime model, the SMS intervention dominated the control by gaining an additional 0.071 QALY and saving $1020.35 per person. The SMS intervention remained dominance in all sensitivity analyses. Conclusions: This study revealed that the SMS intervention for subjects with IGT had the superiority of lower monetary cost and a considerable improvement in preventing or delaying the T2DM onset. PDB95 Cost-Minimisation Analysis of Dapagliflozin Compared To Lixisenatide As An Add-On To Insulin In The Treatment of Type 2 Diabetes Mellitus From A Uk Health Care Perspective Vioix H 1, Eddowes L A 2, Griffiths M 2, Gabriel Z 1 1AstraZeneca UK Ltd., Luton, UK, 2Costello Medical Consulting Ltd., Cambridge, UK . . . . . Objectives: To evaluate the cost of using the first-in-class sodium-glucose cotransporter-2 (SGLT-2) inhibitor dapagliflozin compared to the glucagon-like peptide-1 (GLP-1) analogue lixisenatide, when added to insulin for the treatment of patients with type 2 diabetes mellitus (T2DM) who are inadequately controlled on insulin alone. Methods: Bucher indirect comparisons and a network meta-analysis were performed in the add-on to insulin population with regards to the key T2DM outcomes of HbA1c, weight and hypoglycaemia. Dapagliflozin had similar treatment effects to lixisenatide, with no statistically significant differences between dapagliflozin and lixisenatide in these outcomes. Therefore, a cost-minimisation analysis over a 1-year time horizon was developed from a UK health care perspective. Drug and administration costs, including a nurse consultation, were considered in the model. Drug costs were those of the British National Formulary (September 2013), device costs were taken from the Information Services Division Drug Tariff 2013 and the cost of a nurse consultation was estimated using the Unit Costs of Health and Social Care report compiled by the Personal Social Services Research Unit. The application of an annual discount rate of 3.5% and use of a longer time horizon (up to 5 years) were explored in scenario analyses. Results: Dapagliflozin was associated with a yearly cost of £476.96 per patient. The cost per patient for lixisenatide was £765.76 in the first year and £752.07 in subsequent years. Therefore, dapagliflozin was cost-saving compared to lixisenatide, driven by the difference in drug and administration costs. Applying the annual discount rate and using a longer time horizon, dapagliflozin was associated with savings of £1,255.30 over 5 years compared to lixisenatide. Conclusions: Dapagliflozin in combination with insulin was shown to be a cost-saving treatment option compared to lixisenatide from a UK health care perspective for patients with T2DM who are inadequately controlled on insulin alone. 1HEVA Objectives: The main objective of this study was to estimate the health-economic impact of sensor augmented pump (SAP) compared to pump therapy alone (CSII) in uncontrolled type 1 diabetes (T1DM) in France. Methods: The Core Diabetes Model was used to project the incidence of diabetes-related complications over a lifetime horizon, based on a recently performed meta-analysis comparing SAP versus CSII. The meta-analysis showed that for the analysed cohort of T1DM using exclusively Medtronic devices with average baseline HbA1c of 9%, 70% use of SAP led to a reduction of -0.88% versus -0.47% HbA1c, for SAP and CSII respectively. The mean baseline age of the simulated cohort was, according to a French typical TD1M population, 36 years, with a mean duration of diabetes of 17 years. The quality of life was adjusted for a reduced fear of hypoglycaemic event in the SAP arm. Sensitivity analyses were carried out on several key parameters. Results: The incremental cost-effectiveness ratio (ICER) was 27,796 € per Quality Adjusted Life Year gained (QALY). The improvement in discounted QALY was 1.27 years in favour of SAP. Undiscounted life expectancy was increased by 1 year for SAP versus CSII. Additional SAP related costs were partially offset by the savings due to the PDB96 Economic Evaluation of Lanreotide Autogel in the Management of Acromegaly in Greece Kousoulakou H 1, Panayiotou-Pazaitou K 2, Pantazi E 3, Tsentidis C 4, Tzanela M 5, Vryonidou A 6, Markussis V 7, Eleftheriou C 7, Geitona M 1 of Peloponnese, Corinth, Greece, 2Theageneion Anticancer Hospital, Thessaloniki, Greece, 3Alexandra Hospital, Athens, Greece, 4Nikaia Hospital, Nikaia, Greece, 5Evangelismos Hospital, Athens, Greece, 6Red Cross Hospital, Athens, Greece, 7Ipsen, Athens, Greece . . . . . . . . . 1University Objectives: Acromegaly is a chronic and debilitating disease, which can lead to significant increases in health care costs. The objective of the present study was to undertake an economic evaluation of Lanreotide Autogel compared with Octreotide LAR in the management of acromegaly in the Greek health care setting. Methods: A cost minimization model was developed for the analysis which only considers the costs of treatment, as the two drugs have equal efficacy. The time horizon was 30 years; future costs were discounted at 3.5%. In addition, a budget impact analysis was conducted, with a time horizon of 5 years. The perspective for both analyses was that of Social Insurance. Data on resource use (administration setting, drug wastage and time required for the administration of A349 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 each drug), as well as lab and imaging tests, were collected with the Delphi technique via an expert panel with 11 experienced endocrinologists, from University and Public Hospital specialized units. Local unit cost data were collected from officially published sources (Ministry of Health and Social Insurance Funds). One way sensitivity analyses were performed to test the results. Results: Lanreotide Autogel reduced total costs of acromegaly treatment by € 29,896 per patient over the 30-year time horizon. 93% of the savings were attributed to the reduction in drug acquisition and administration costs. Discount rate was the most influential parameter in the sensitivity analysis. The total cost of managing acromegaly in Greece, including lab and imaging tests, over a 5-year time horizon was estimated to range between € 22.9 and € 22.2 million, with a 30% and 60% market share for Lanreotide Autogel, respectively. Therefore, doubling Lanreotide Autogel’s share would lead to total savings of € 781,604. Conclusions: Lanreotide Autogel in comparison with Octreotide LAR may result in a reduction of the total cost in the management of acromegaly in Greece. PDB97 The Opportunity of Treating Type Ii Diabetes With Dpp4i: An Economic Evaluation VersUS Conventional Treatment in the Italian Setting Lorenzoni V , Pierotti F , Turchetti G Scuola Superiore Sant’Anna, Pisa, Italy . . . Objectives: To compare dipeptdyl-peptidase 4 inhibitors (DPP4i) and sulfonylurea (SU) for the treatment of type II diabetes mellitus in terms of economic impact and considering both the Italian National Health System (NHS) and the societal perspective. Methods: The economic evaluation was performed as a model-based costminimization analysis for the comparison DPP4i and SU as second line therapy, in add-on to metformin, over 1-year period. Clinical events to be included in the model were selected from literature review and the opinion of a panel of clinical experts. Resources used were quantified and valued adopting costs and tariffs related to drugs used, glycaemic auto-monitoring, established control visits, incidence of hypoglicaemic events, macrovascular complications and the switch to insulin therapy. One-way sensitivity analyses for model inputs were conducted. Results: Due to the higher cost for drug acquisition in the base case analysis total direct costs for the Italian NHS were about 728 Euro per patient/year in the case of DPP4i and on average 702 Euro for SU. The overall yearly cost for the society was estimated to be about 728 Euro per patient in the case of DPP4i while it was on average 770 Euro when considering SU because DPP4i induced lower direct non-health costs related to stroke and an overall saving of 20.88 Euro per patient/year due to lower costs of productivity loss for hypoglicaemic events and stroke. Conclusions: The use of DPP4i was cost-saving from the societal perspective and just the high cost for drug acquistion made the adoption of DPP4i more costly than SU for the Italian NHS. This result outlined that DPP4i represents a valuable alternative for the management of diabetes both from a clinical and economic perspective and costs will be lowered overall just intervening on cost for drug acquisition. PDB98 Cost-Minimisation Analysis of Saxagliptin Compared to Sitagliptin And Linagliptin As Triple Therapy In Combination With Metformin And A Sulphonylurea In The Treatment Of Type 2 Diabetes Mellitus From A Uk Health Care Perspective Vioix H 1, Eddowes L A 2, Griffiths M 2, Gabriel Z 1 1AstraZeneca UK Ltd., Luton, UK, 2Costello Medical Consulting Ltd., Cambridge, UK . . . . . Objectives: To evaluate the cost of using the dipeptidyl peptidase-4 inhibitor (DPP4i), saxagliptin, compared to sitagliptin and linagliptin, when used as triple therapy in combination with metformin and sulphonylurea (met+SU) for the treatment of patients with type 2 diabetes mellitus (T2DM) who are inadequately controlled on met+SU alone. Methods: Bucher adjusted indirect treatment comparisons (ITCs) were performed with regards to the key T2DM outcomes of HbA1c, weight and hypoglycaemia compared to sitagliptin and linagliptin. The ITCs found no statistically significant differences between saxagliptin compared to either sitagliptin or linagliptin in terms of effectiveness (as measured by Hb1Ac change from baseline), and saxagliptin was found to be at least as safe as the other therapies. Therefore, a cost-minimisation analysis over a 1-year time horizon was developed from a UK health care perspective. Drug costs were considered in the model, sourced from the British National Formulary (BNF; September 2013). The application of an annual discount rate of 3.5% and use of a longer time horizon (up to 5 years) were explored in a scenario analysis. Results: Saxagliptin was associated with a yearly cost of £410.80 per patient. The yearly cost per patient for sitagliptin was £432.38, and the yearly cost per patient for linagliptin was also £432.38, based on drug costs. Therefore, saxagliptin has similar costs compared to the other DPP4i’s. Applying the annual discount rate and using a longer time horizon, saxagliptin was associated with cost-savings of £97.43 per patient over 5 years compared to both sitagliptin and linagliptin. Conclusions: Saxagliptin as triple therapy in combination with met+SU was shown to be a cost-saving treatment option from a UK health care perspective for patients with T2DM who are inadequately controlled on met+SU alone. The cost-saving per patient over 5 years was modest, although this may be important in a large patient population. PDB99 Cost-Effectiveness of Sitagliptin Versus Sulfonylurea As An Add-On Therapy To Metformin In Patients With Type 2 Diabetes In A Belgium Setting Chen J 1, Radican L 2, Shankar R 2, Hiver M 3, Qiu Y 2 1Merck Research Laboratories, North Wales, PA, USA, 2Merck Sharp & Dohme Corp., Whitehouse Station, NJ, USA, 3MSD Belgium BVBA, Brussels, Belgium . . . . . Objectives: Assess the cost-effectiveness of sitagliptin versus sulfonylurea as an add-on therapy to metformin among type 2 diabetes patients currently on metformin but not achieving HbA1c goal in Belgium. Methods: We employed a previously published individual-level simulation model that incorporated risk equations/algorithms from the UKPDS Outcomes Model (68) to predict the long term risks of type 2 diabetes-related complications. The impact of treatments on risk factors and side effects was based on clinical trials, observational studies, systematic reviews and meta-analyses of relevant RCTs, as well as the most recent findings on the potential benefit of DPP4 on other-cause mortality and cardiovascular diseases and the possible detrimental effect of sulfonylurea on myocardial infarction (MI). European patient profiles and Belgium-specific data on drug prices, diabetes-related complication treatment costs, treatment patterns and guidelines were used. Results: A sitagliptin-based treatment strategy was projected to cost € 1,102 more than a sulfonylurea-based treatment strategy per patient lifetime, with the majority of excess costs from prescription drugs. Life expectancy was 0.077 years greater per patient on a sitagliptin-based strategy compared to a sulfonylurea-based strategy. The discounted gain in QALY was 0.082 years with the sitagliptin-based strategy, driven by better hypoglycemia, weight, and MI risk profile. The estimated ICER was € 13,460/QALY. Sensitivity analyses demonstrated that the ICER was somewhat sensitive to the price of sulfonylureas and the weight utility decrement, and most sensitive to assumptions on relative risk parameters. When no relative risk reduction on MI or other-cause mortality was assumed, the ICER was € 17,543/QALY and € 17,053/QALY, respectively. When no relative risk reduction on either MI or other-cause mortality was assumed, the ICER increased to € 23,691/QALY. Conclusions: Using a threshold of € 15,000 per QALY gained, compared to a sulfonylurea-based treatment strategy, a sitagliptinbased treatment strategy was cost-effective in metformin-failed patients with type 2 diabetes in Belgium. PDB100 Cost-Effectiveness of Exenatide Twice Daily (Bid) Added To Basal Insulin Compared To A Bolus Insulin Add-On In Turkey Malhan S 1, Güler S 2, Yetkin I 3, Baeten S 4, Verheggen B 4 University, Ankara, Turkey, 2Hitit University, Faculty of Medicine, Çorum, Turkey, 3Gazi University, Faculty of Medicine, Ankara, Turkey, 4Pharmerit International, Rotterdam, The Netherlands . . . . . 1Baskent Objectives: Type 2 diabetes (T2D) patients on basal insulin with uncontrolled HbA1c levels often receive an add-on with bolus insulin to lower HbA1c levels. The aim of this analysis is to estimate long-term cost-effectiveness of treating T2D patients with – a recently introduced – twice daily (BID) exenatide add-on to basal insulin strategy from a Turkish health care perspective. Methods: Clinical inputs for both treatment strategies were taken from the GWDM clinical trial. The strategies were assessed using a micro-simulation disease model (CARDIFF). The model predicted micro-and macro-vascular complications based on the UKPDS equations. The incidence of adverse events, diabetes related complications and changes in body weight yielded estimates of health care costs and health utilities. The direct (T2D and complication) costs in the model reflect the Social Security Institution costs. Discounting both costs and effects at 3% over the 40 year followup of the model, resulted in life time estimates of costs and quality-adjusted life years (QALYs) on both treatment strategies. Deterministic and probabilistic sensitivity analyses, as well as elaborate scenario analyses were performed. Results: Results showed that exenatide treatment significantly improves QALYs by 0.60 (95% CI: 0.24 to 0.97). Health effects were reached at an additional cost of 217 $ (95% CI: -356$ to 976$), resulting in an incremental cost-effectiveness ratio of 362 $ per QALY gained. Scenario analyses showed that these results were robust to changes in input parameters. At a willingness-to-pay threshold of 30,000 $/QALY the exenatide strategy had a near 100% probability of being cost-effective compared to bolus insulin. Conclusions: A twice daily exenatide add-on to basal insulin treatment for T2DM patients with uncontrolled HbA1c levels is considered a highly cost-effective strategy from the Turkish public health care perspective. PDB101 Comparative Cost-Effectiveness Analysis of Adding Twice-Daily Exenatide To Insulin Glargine Versus Adding Insulin Lispro To Treat Type 2 Diabetes In Spain Sánchez-Covisa J 1, Capel M 1, Baeten S 2, Verheggen B G 2 International, Rotterdam, The Netherlands . . . . . 1AstraZeneca, Madrid, Spain, 2Pharmerit Objectives: When HbA1c is elevated above target in Type 2 diabetes (T2D) patients treated with basal insulin, the widespread strategy consists of adding mealtime insulin. An alternative option is adding twice-daily exenatide (BID), a glucagon-like peptide-1 receptor agonist. The objective was to estimate the cost-effectiveness in Spain of exenatide BID compared to mealtime bolus insulin lispro, both added to insulin glargine and metformin. Methods: The published and validated CARDIFF long-term diabetes model was used to estimate the direct medical costs and qualityadjusted life years (QALYs) associated with each strategy. Patient characteristics at baseline, efficacy and safety inputs were all derived from a head-to-head, doubleblind, randomized controlled trial (NCT00960661), comparing both strategies for 30 weeks. Based on the United Kingdom Prospective Diabetes Study-68 equations, the model predicted long-term disease progression and occurrence of microand macro-vascular complications, including mortality. Costs and utilities were assigned to complications, hypoglycaemias, adverse events and body mass index changes. The analyses were performed from the perspective of the Spanish health care payer, over a lifetime horizon, at a discount rate of 3% (costs and health outcomes). Univariate and probabilistic sensitivity analyses were conducted. Results: Treatment with exenatide BID was projected to produce an incremental benefit of 0.61 QALYs (95 % CI: 0.26 to 0.99) compared to treatment with insulin lispro, at an additional cost of € 146 (95% CI: -€ 1,114 to € 1,679) resulting in an incremental costeffectiveness ratio of € 239 per QALY gained. The exenatide BID strategy reached a probability near 100% of being cost-effective at a willingness-to-pay threshold of €2,500 per QALY gained. Sensitivity analyses showed that results were robust to variation in range of model parameters. Conclusions: Exenatide BID was predicted to be a cost-effective treatment alternative to mealtime bolus insulin in Spain for T2D patients not at target with insulin glargine. A350 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PDB102 Retrospective analysis of the economic burden among cushing’s disease patients in the u.S. Medicaid program Li L 1, Shrestha S 1, Baser O 2, Wang L 1 Research, Plano, TX, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . 1STATinMED Objectives: To evaluate the economic burden among patients diagnosed with Cushing’s disease (CD) in the U.S. Medicaid program. Methods: Patients diagnosed with CD (International Classification of Disease, 9thRevision, Clinical Modification (ICD-9-CM) diagnosis code 255.0) were identified using U.S. Medicaid data from 01 January 2008 through 31 December 2010. The initial diagnosis date was designated as the index date. A matching comparator cohort was created including patients of the same age, race and gender but without a CD diagnosis, and a randomly-chosen index date to minimize selection bias. Patients in both cohorts were required to be age ≥18 years, with continuous medical and pharmacy benefits for 1 year pre- and 1 year post-index date. One-to-one propensity score matching (PSM) was used to compare health care costs and utilizations during the follow-up period between the diseased and comparison cohorts, and was adjusted for baseline demographic and clinical characteristics. Results: After risk adjustment by PSM, a total of 340 patients in each cohort were matched. CD patients had significantly higher health care utilization, including inpatient admissions (36.18% vs. 12.53%, p< 0.0001) and long-term care (5.29% vs. 2.06%, p<0.05), other service (100% vs. 94.12%, p<0.0001) and pharmacy visits (84.41% vs. 78.24%, p<0.05), compared to those without the disease. CD patients incurred significantly higher inpatient ($4,688 vs. $1,139, p<0.05) and pharmacy costs ($4,054 vs. $2,100, p<0.001) compared to those without CD. Long-term care and other service costs incurred were higher for CD patients, compared to comparison patients, but were not statistically significant. Conclusions: In the current study, CD patients in the U.S. Medicaid program had a higher burden of illness in terms of health care resource utilization and costs, compared to those without a CD diagnosis. PDB103 The Potential Value of Ongoing Support In Type-1 Diabetes Mellitus With Dafneplus: Exploratory Pre-Trial Cost-Effectiveness Analysis On Proposed Trial End-Point Target For 12-Month Hba1c Improvement Basarir H , Pollard D , Brennan A , Elliott J , Heller S , Campbell M J University of Sheffield, Sheffield, UK . . . . . . . Objectives: The Dose Adjustment For Normal Eating (DAFNE) structured education programme is shown to be effective both in terms of clinical outcomes and costeffectiveness outcomes in the treatment of T1DM. DAFNEPlus aims to revise the DAFNE 5-day curriculum based on psychological and sociological findings in DAFNE, input from DAFNE graduates and emerging knowledge around behavioural science and technological developments. The current suggested primary endpoint is for the DAFNEplus programme to have an additional 20% DAFNE participants (70% in total) achieve either, (a) a reduction of at least 0.5% in HbA1c, or (b) to have an HbA1c below 7.5% (58.5 mmol/mol), at 12 months. This paper undertakes pre-trial what-if costeffectiveness analyses concerning the DAFNEPlus programme, which aim to be useful both in the design of the intervention itself and of the proposed trial. Methods: The Sheffield Type 1 Diabetes Policy Model is an individual patient-level simulation model of T1DM. It includes long-term microvascular (retinopathy, neuropathy and nephropathy) and macrovascular (myocardial infarction, stroke, revascularization and angina) diabetes-related complications and acute adverse events (severe hypoglycaemia and diabetic ketoacidosis). Econometric methods were used to obtain the target level of HbA1c responders in the DAFNEPlus arm. Results: DAFNEplus would be considered as cost-effective if the additional spending on the intervention would be limited to £455-£751 per patient per year, depending on the assumptions on the length of maintenance period for the HbA1c benefit and the target HbA1c responder endpoint (70% in total) being achieved in the future trial. To achieve a more favourable cost-effectiveness probability of 80%, for example, the additional per patient per year cost should be restricted to £393-£574 range. Conclusions: Pre-trial modelling has enabled a clear understanding of the threshold range for the annual cost of DAFNEplus, which is still being designed, in order to be considered as cost-effective at the £20,000/QALY threshold. PDB104 The Cost-Effectiveness Of Saxagliptin When Added To Metformin And Sulphonylurea In The Treatment Of Type 2 Diabetes Mellitus In Spain Sánchez-Covisa J 1, Franch J 2, Mauricio D 3, López-Martínez N 4, Chuang L H 5, Capel M 1 Raval Sud- Institut Català de la Salut - USR Barcelona ciutat IDIAP Jordi Gol, Barcelona, Spain, 3Germans Trias i Pujol University Hospital, Barcelona, Spain, 4Oblikue Consulting, Barcelona, Spain, 5Pharmerit International, Rotterdam, The Netherlands . . . . . . . 1AstraZeneca, Madrid, Spain, 2EAP Objectives: In patients with type 2 diabetes mellitus (T2DM), when blood glucose is not adequately controlled by the combination of metformin (MET) and sulphonylurea (SU), the clinician has to choose between adding a third oral drug or starting insulin therapy. The objective of this study was to assess the cost-effectiveness in the Spanish setting of adding saxagliptin (SAXA) to MET and SU, compared to adding basal insulin (INS). Additionally, the SAXA strategy was compared with a thiazolidinedione (TZD), also added on top of MET and SU. Methods: The published and validated CARDIFF long-term diabetes model was used to estimate the direct medical costs and quality-adjusted life years (QALYs) associated with each strategy. Clinical inputs were obtained from a network meta-analysis. Based on the United Kingdom Prospective Diabetes Study equations, the model predicted disease progression and occurrence of micro- and macro-vascular complications, including mortality. Costs and utilities were applied to complications, hypoglycaemias and body mass index changes. The perspective of the Spanish Healthcare System was adopted over a lifetime horizon, at a discount rate of 3% (costs and health outcomes). Univariate and probabilistic sensitivity analyses were conducted. Results: SAXA add-on to MET plus SU resulted in a dominant strategy compared to INS add-on to MET plus SU, providing a gain of 0.377 QALYs (95% CI: -0.227 to 0.754) and cost savings of € 264 (95% CI: -€ 1,879 to € 2,768).. At a willingness-to-pay threshold of €30,000 per QALY gained, SAXA strategy had an 82% probability to be cost-effective. Compared to TZD add-on to MET plus SU, the triple therapy with SAXA reached an incremental cost-effectiveness ratio of € 2,610 per QALY gained. Conclusions: Saxagliptin was predicted to be a cost-effective option in Spain when a new drug needs to be added in T2DM patients inadequately controlled with metformin and sulphonylurea alone. PDB105 The Cost-Effectiveness of Dapagliflozin In Combination With Insulin for the Treatment of Type 2 Diabetes Mellitus (T2dm) In Spain Sánchez-Covisa J 1, Capel M 1, Schmidt R 2, Charokopou M 2, Verheggen B G 2 International, Rotterdam, The Netherlands . . . . . . 1AstraZeneca, Madrid, Spain, 2Pharmerit Objectives: To assess the cost-effectiveness of dapagliflozin, a sodium-glucose co-transporter-2 (SGLT-2) inhibitor versus dipeptidyl peptidase-4 inhibitor (DPP4i) both added on top of insulin, or compared to insulin alone (±oral anti-diabetes agents) for patients who are inadequately controlled on insulin strategy. Methods: The CARDIFF diabetes model was used. Clinical inputs were derived from a randomized clinical trial comparing dapagliflozin add-on to insulin with insulin alone, and network-meta-analysis for the comparison with DPP4i. Together with United Kingdom Prospective Diabetes Study (UKPDS) equations, the model predicts disease progression and the number of micro- and macro-vascular complications, along with diabetes-specific and all-cause mortality. The perspective of the Spanish health care payer was adopted over a lifetime horizon. Costs and utilities were assigned to the appropriate model parameters to calculate total Quality-Adjusted-Life-Years (QALYs) and total costs. Deterministic and probabilistic sensitivity analyses were conducted. Results: Compared to insulin alone, dapagliflozin added to insulin was associated with 0.698 incremental QALYs (95%CI: 0.442; 1.211) at an additional cost of € 1,508 (95%CI: € 611; € 1,517), resulting in an incremental cost-effectiveness ratio (ICER) point estimate of € 2,159/QALY. Dapagliflozin was found to dominate DPP4i add-on to insulin, being associated with slightly less costs (-€51; 95%CI: -€ 913; € 553) and higher QALYs (0.168; 95%CI: -0.007; 0.417). At a willingness-to-pay threshold of € 20,000/QALY, the dapagliflozin strategy was estimated to have a 100% probability of being cost-effective when compared to the insulin alone, and a 98% probability when compared to the DPP4i strategy. These findings were shown to be robust to variation in range of model parameters. Conclusions: Dapagliflozin added on top of Insulin was predicted to be a cost-effective (vs. insulin alone) and cost saving (vs DPP4i) alternative in Spain in combination with insulin for patients who are inadequately controlled with insulin treatment regimens. PDB106 The Cost-Effectiveness of Tolvaptan for the Treatment of Hyponatraemia Secondary To Syndrome Of Inappropriate Antidiuretic Hormone Secretion in Sweden Trueman D 1, Robinson P 2, Dale P 2, O’Reilly K 2, Lundberg J 3, Jamookeeah C 2 1Decision Resources Group, London, UK, 2Otsuka Pharmaceutical Europe Ltd, Wexham, UK, 3Otsuka Pharma Scandinavia, Stockholm, Sweden . . . . . . Objectives: Tolvaptan is a selective vasopressin V2-receptor antagonist indicated for the treatment of adult patients with hyponatraemia (HN) secondary to syndrome of inappropriate antidiuretic hormone secretion (SIADH). To date there have been no published economic evaluations assessing the cost effectiveness of tolvaptan in this indication. The aim of this study was to evaluate the cost effectiveness of tolvaptan versus no active treatment (NAT) from a Swedish societal perspective. Methods: The economic evaluation considers a hypothetical population of individuals with HN secondary to SIADH who have either failed to respond to fluid restriction or for whom the use of fluid restriction is not suitable. The analysis considers three clinically relevant patient populations within the SIADH indication: ‘all SIADH’, small-cell lung cancer (SCLC) and pneumonia. A discrete event simulation was developed to model the progression of individuals through multiple inpatient admissions over a 30 day time horizon (180 days in the SCLC scenario). Key sources of evidence included randomised controlled trials (SALT I & II) and observational data sources. Unit costs were collected from publicly available sources. Utility values were obtained from mapping the SF-12 scores from the SALT I & II trials to EQ-5D. The primary outcome of the analysis was the incremental cost-effectiveness ratio (ICER) expressed as a cost per quality-adjusted life-year (QALY). Results: In the ‘all SIADH’ population tolvaptan was associated with reduced costs (SEK 5,778) and increased QALYs (0.0019) versus NAT and was therefore dominant. In the SCLC and pneumonia subgroups tolvaptan was also associated with reduced costs and QALY improvements. The results were most sensitive to the duration of tolvaptan treatment and the assumptions around duration of hospitalisation. Conclusions: In all populations considered (all SIADH, SCLC and pneumonia) tolvaptan was dominant compared to NAT being associated with reduced costs and increased QALYs. PDB107 Cost-Effectiveness of Empagliflozin (Jardiance®) 10 Mg And 25 Mg Administered As An Add-on To Metformin Compared To Other Sodium-Glucose Co-Transporter 2 Inhibitors (Sglt2is) for Patients With Type 2 Diabetes Mellitus (T2dm) In The UK Aguiar-Ibáñez R 1, Palencia R 2, Kandaswamy P 3, Li L 1 1Amaris Consulting UK, London, UK, 2Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany, 3Boehringer Ingelheim UK, Bracknell, UK . . . . Objectives: To assess the cost-effectiveness of the novel SGLT2is empagliflozin 10mg and 25mg compared to other SGLT2is (canagliflozin 100mg, canagliflozin 300mg, and dapagliflozin 10mg) when administered as an add-on to metformin for the treatment of patients with T2DM in the UK. Methods: A micro-simulation model was developed, based on the United Kingdom Prospective Diabetes Study (UKPDS68) and the Januvia Diabetes Economic (JADE) model, to estimate long-term diabetes-related complications, QALYs and costs in a cohort of T2DM patients initiating dual therapy. The model was populated with the results of a network meta-analysis that estimated A351 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 the comparative efficacy and safety across SGLT2is. Data gaps were completed with information derived from published sources, including previous cost-effectiveness analyses. The UK National Health Service (NHS) perspective was considered to estimate costs and QALYs over a patients’ lifetime. Results: There were small differences in efficacy and safety across SGLT2is, which resulted in minor QALY and cost differences across treatment combinations. On average, empagliflozin 25mg obtained incremental QALYs of 0.029 versus dapagliflozin 10mg and 0.019 versus canagliflozin 100mg, and incremental costs of £178 and £86, respectively, whereas both canagliflozin 300mg and empagliflozin 10mg were dominated by empagliflozin 25mg. This resulted in an incremental cost-effectiveness ratio (ICER) of £4,858 per QALY gained with empagliflozin 25mg vs. canagliflozin 100mg. However, the differences across treatments were not significant when 95% percentile confidence intervals were considered. These results were robust to a number of sensitivity analyses including a 10-year time horizon, BMI impact, discount rates and parameter values related to utilities, disutilities, adverse events, and discontinuation rates. Conclusions: Overall, differences in QALYs and costs were minor between SGLT2is used as add-on to metformin in UK T2DM patients. On average, empagliflozin 25mg was the most cost-effective strategy, with an ICER of £4,858 per QALY gained vs. canagliflozin 100mg. PDB108 Cost Effectiveness Analysis of Flash Glucose Monitoring for Type 2 Diabetes Patients Receiving Insulin Treatment In The Uk Li H 1, Bilir S P 2, Donga P 3, Samiian A 4, Munakata J 2 Health, Alexandria, CA, USA, 2IMS Health, San Francisco, CA, USA, 3IMS Health, Plymouth Meeting, PA, USA, 4Abbott Diabetes Care, Alameda, CA, USA . . . . . . 1IMS Objectives: A small, minimally-invasive flash glucose monitor (FGM) has recently been developed. Arm sensors worn up to 14 days interact with a hand-held reader to convey 8 hours of continuous glucose level data. The reader stores data, communicating glucose control via trend charts. Economic evaluation of FGM vs. conventional blood glucose monitoring (BGM) has not been conducted. This analysis estimates potential cost-effectiveness of using FGM in UK insulin-treated type 2 diabetes mellitus (T2DM) patients. Methods: The IMS Core Diabetes Model (CDM) was used for analyses, assuming a lifetime horizon (40 years). Patient characteristics were based on early FGM feasibility trial data. Effectiveness was measured in life years (LY) and quality-adjusted life years (QALY), with assumptions around FGM effectiveness based on expected benefits of use. These include: a) lower HbA1c by 0.35%-0.5% compared to BGM over the horizon; b) utility improvement due to fewer finger pricks of 0-0.03; c) minor hypoglycaemic event rate reduction of 0% or 50% compared to BGM due to potential improved glycaemic control. Cost data (direct costs only) were extracted from published literature and government sources, and inflated to 2013 GBP. Incremental cost-effectiveness ratios (ICERs) were estimated, and threshold analysis was performed to estimate potential total FGM sensor costs for each scenario. Results: Based on assumptions above, the ICER for FGM vs. BGM ranges from £10,034-£29,068/QALY. With 0.5% HbA1c improvement, 0.01 utility benefit, and no difference in hypoglycaemic events, the ICER is £17,808/QALY. Assumptions around utility improvement have a larger ICER impact than HbA1c benefit or change in minor hypoglycaemic events. Threshold analysis shows that with a conventional ICER threshold (£30,000/QALY), £14,606-£27,956 can be spent on sensors over a lifetime across scenarios. Conclusions: Using an alternate glucose monitoring method could be cost-effective across a variety of clinical benefit and cost assumptions in T2DM (T1DM analysis forthcoming). PDB109 Cost Effectiveness Evaluation of Canagliflozin in Combination with Metformin and Sulfonylurea in Comparison To Nph Insulin in the Treatment of Type 2 Diabetes Mellitus In Poland Szmurlo D 1, Drzal R 1, Plisko R 1, Schubert A 2, Skrzekowska-Baran I 3 1HTA Consulting, Krakow, Poland, 2Janssen Cilag Poland, Warszawa, Poland, 3Janssen-Cilag Polska, Warszawa, Poland . . . . . Objectives: To evaluate the cost-effectiveness of canagliflozin, an active inhibitor of sodium glucose co-transporter – 2 (SGLT2), in triple therapy of diabetes as addon to metformin and sulfonylurea compared to NPH insulin in combination with oral antidiabetics. Canagliflozin in clinical trial results showed effective glucose reduction, along with other benefits in diabetes treatment including weight loss and SBP reduction. Cost effectiveness analyses were conducted in the Polish setting from a public perspective in accordance with guidelines of Polish HTA Agency (PolAHTA). Methods: The IMS CORE Diabetes Model was used to evaluate the cost-effectiveness of canagliflozin in triple therapy versus NPH insulin using Polishspecific data, where available. Clinical data were derived from mixed treatment comparison analysis of published studies, as there is no head to head trial comparing canagliflozin with NPH insulin. Direct costs were reported in Polish zloty and an annual discount rate of 5% and 3.5% were applied on costs and effects respectively. Results: In a triple therapy as add-on to metformin and sulfonylurea canagliflozin is a cost- effective treatment option in comparison with NPH insulin with ICERs of 4 477 z³ and 69 081 z³ for canagliflozin 100 mg and 300 mg respectively. Associated QALY gains were 0,084 and 0,106. Both results are below defined in Polish reimbursement act cost-effectiveness threshold. Conclusions: These results suggest that adding Canagliflozin to dual therapy versus insulin intensification in patients inadequately controlled with MET+ SU would be a more efficient use of health care resources in the Polish setting. PDB110 Cost-Effectiveness of Empagliflozin (Jardiance®) 10 Mg And 25 Mg Administered As An Add-On To Metformin And Sulfonilurea (Met+Su) Compared To Other Sodium-Glucose Co-Transporter 2 Inhibitors (Sglt2is) in Patients with Type 2 Diabetes Mellitus (T2dm) In The Uk Aguiar-Ibáñez R 1, Palencia R 2, Kandaswamy P 3, Li L 1 1Amaris Consulting UK, London, UK, 2Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany, 3Boehringer Ingelheim UK, Bracknell, UK . . . . Objectives: To assess the cost-effectiveness of the SGLT2is empagliflozin 10mg and 25mg compared to other SGLT2is (canagliflozin 100mg and canagliflozin 300mg) when administered as an add-on to MET+SU in patients with T2DM in the UK. Methods: Long-term diabetes-related complications, QALYs, and costs were estimated for T2DM patients failing MET+SU. A micro-simulation model was developed based on the United Kingdom Prospective Diabetes Study (UKPDS68) and the Januvia Diabetes Economic (JADE) model. A network meta-analysis comparing efficacy and safety across SGLT2is was used to populate the model. Data gaps were completed with information derived from published sources, including previous cost-effectiveness models. Costs and QALYs were estimated over a patients’ lifetime from the UK National Health Service perspective. Results: Empagliflozin 10mg attained the highest QALYs (6.991, compared to 6.98 for canagliflozin 100mg, 6.978 for empagliflozin 25mg and 6.976 for canagliflozin 300mg) due to slightly better HbA1c, SBP and weight control, and a small number of non-severe hypoglycaemias, compared to higher doses. Canagliflozin 300mg was the most costly strategy (£32,087, vs. £31,217 for canagliflozin 100mg, £31,409 for empagliflozin 10mg and £31,557 for empagliflozin 25mg). Therefore, empagliflozin 10mg dominated both canagliflozin 300mg and empagliflozin 25mg, and resulted in an incremental costeffectiveness ratio of £17,445 per QALY gained vs. canagliflozin 100mg. However, incremental QALY and cost differences were not significant based on 95% percentile confidence intervals. These results remained robust when sensitivity analyses were conducted, including utilities, adverse events, discontinuation, modelling of weight, impact of BMI, duration of effect, time horizon and discount rates. Conclusions: Differences in QALYs and costs between SGLT2is as add-ons to MET+SU were minor. On average, empagliflozin 10mg resulted to be the most cost-effective option for T2DM patients failing MET+SU when commonly accepted thresholds in the UK were considered, with an incremental cost per QALY of £17,445 compared to canagliflozin 100mg. PDB111 Absenteeism and Presenteeism in A Population of Patients with Diabetic Foot Ulcers in Poland Zalewska U 1, Macioch T 1, Sobol E 2, Mrozikiewicz-Rakowska B 1, Krakowiecki A 3, Hermanowski T 1 1Medical University of Warsaw, Warsaw, Poland, 2Medical University of Warsaw Central Clinical Hospital, Warsaw, Poland, 3PODOS, Warsaw, Poland . . . . . . Objectives: Diabetic Foot Syndrome (DFS) is a serious and common complication of diabetes, often leading to limb amputation and disability. Disability and productivity loss in patients with DFS can generate significant indirect costs and potentially significant economic consequences. The purpose of the study is to estimate productivity loss and indirect costs associated with foot ulceration in patients with DFS. Methods: We conducted a prospective survey in a population of DFS patients with foot ulceration. Loss of productivity was measured with a modified WPAI questionnaire. Indirect costs of both absenteeism and presenteeism were estimated using the human capital approach on the basis of the measure of gross value added per employee. Results: Nearly one third of respondents (32%) declared that foot ulceration was the direct reason why they abandoned their professional activity. 40% and 34% of respondents, respectively, were forced to limit or change their professional activity at some point in the past because of the foot ulceration. More than 40% of respondents who changed or limited their professional activity because of the foot ulceration experienced reduction in earnings by 22.9% on average. Mean absenteeism was estimated at 32.63% of the nominal working time, while presenteeism was estimated at 23.48% of real working time. Total annual indirect costs associated with productivity loss amounted to EUR 170.8 million, including EUR 117.3 million of the costs of sickness absence and EUR 53.5 million of the costs of presenteeism. Conclusions: Foot ulceration in patients with DFS is a common cause why patients are forced to give up or change their professional activity, which usually leads to a reduction in earnings. Indirect costs associated with foot ulceration in DFS impose a significant burden on the Polish economy. There is no rationale that would clearly link productivity loss associated with ulceration in DFS and the ulceration severity. PDB112 Examining The Role Of Insulin Pen Devices In Acute Care Settings: A Review And Analysis of Health Resource Utilization Smallwood C A 1, Lamarche D 2, Chevrier A 2 Dickinson, Mississauga, ON, Canada, 2McGill University Health Centre, Montreal, QC, Canada . . . . 1Becton Objectives: Insulin administration in the acute care setting is an integral component of inpatient diabetes management. The current method of administration in acute care settings is by vial and syringe. The aim of this study was to evaluate the impact of insulin pen implementation in the acute care setting on patient and health care worker safety, and health resource utilization (HRU). Methods: A review of published literature was conducted to identify how insulin pen devices in the acute care setting may impact inpatient diabetes management. Additionally, nurse researchers from the McGill University Health Centre conducted a pilot study in a 52-bed unit to quantify this impact in a local context. Together, the results of the literature search and the pilot served as the inputs to an economic model, developed in Excel v14. Costs for the volume of insulin dispensed, injection supplies, needlestick injury management, and nursing labour were assessed. Results: Previous published studies have revealed that insulin pen devices have the potential to improve inpatient management through better glycemic control, increased adherence and improved self-management education. The combined results from the literature and pilot indicate that moving from vial and non-safety syringe to a passive safety pen in acute care results in total estimated annual cost savings of $43,339.66, and 191.42 hours of nursing time saved (site with 52 beds dedicated to patients with diabetes). Cost savings from the adoption of a passive safety insulin pen were predicted based on reductions in insulin volume and needlestick injuries. For an institution of similar size using safety syringes, the move to a A352 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 passive safety pen device would result in total estimated cost savings of $17,865.40 annually. Conclusions: The implementation of insulin pen devices in acute care results in cost savings, as well as time savings for nurses that may be re-directed to increased time at the patient bedside. PDB113 Epidemiology and Direct Health Care Costs of Diabetic Retinopathy: Results From a Population-Based Study Ciampichini R 1, Cortesi P A 1, Cozzolino P 2, Fornari C 1, Madotto F 1, Chiodini V 1, Mantovani L G 3, Cesana G 1 1University of Milano - Bicocca, Monza, Italy, 2Charta Foundation, Milan, Italy, 3Federico II University of Naples, Naples, Italy . . . . . . . . . . Objectives: The aim of this study was to assess the epidemiologic and economic burden of diabetic retinopathy (DR) in terms of incidence, treatment patterns and cost by a population-based study. Methods: Eligible patients were identified through a data warehouse (DENALI), which matches demographic, clinical and economic data of about 9.9 million individuals of Lombardy region. The study population consists of all individuals with a diagnosis of diabetes who, during the period 1-1-2000 to 31-12-2010 received one of the following health care services: hospital admission (HA) for diabetes with ophthalmic manifestations or retinal disorders, fluorescein angiography or angioscopy of eye, destruction of chorioretinal lesion, repair of retinal tear, injection of vitreous substitute, and repeated ophthalmic examinations. The study population was followed for a minimum of 1 to a maximum of 10 years. We evaluated demographic characteristics of the study population and costs from the National Health Service’s perspective. Results: The 2000-2010 DR population was estimated to be around 127,000 (52% male). The average incidence per year and the 2010-prevalence were 6.1 and 24.2 per 100 diabetic patients. Median age (min-max) at the index event was 68.6 (0.3-104.4) with 37% younger than 65-years. Around 15% of the population had Charlson Comorbidity Index> = 1 and the overall mortality was 41.8 deaths/100 patient-years. The 10-year mean cost for DR patients were 17,361€ (95%C.I. 12,673-22,050) compared with 4,771€ in a control diabetic population without complications. HA costs represented the driver of total costs, ranging from 45% (index year) to 36% (last year of follow-up), followed by drug (30-34.4%) and outpatient (25-29.7%). During the index year 13% of HA and outpatient costs were attributed to eye-specialist departments. Approximately 1% received intravitreal injections in 2010. Conclusions: This study attempted to describe the burden of DR in Italy revealing socio-economic aspects relevant in terms of incidence and costs. PDB114 Evaluation of the Burden of Illness of u.S. Medicare Patients Diagnosed with Hyperpotassemia Xie L 1, Dysinger A H 1, Kariburyo M F 1, Du J 1, Baser O 2 Research, Ann Arbor, MI, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . . . . 1STATinMED Objectives: To evaluate the economic burden and health care utilization of patients diagnosed with hyperpotassemia in the U.S. Medicare population. Methods: Hyperpotassemia patients (International Classification of Disease 9 thRevision Clinical Modification ICD-9-CM diagnosis codes 276.7) were identified from the U.S. national Medicare claims from 01JAN2009 through 31DEC2011. The first diagnosis date was designated as the index date for the hyperpotassemia cohort. Patients without a hyperpotassemia diagnosis but of the same age, region, gender, index year and matched baseline Charlson Comorbidity Index score were identified for the comparison cohort, with a randomly-chosen index date to minimize selection bias. Patients were required to have continuous medical and pharmacy benefits 1 year before and after the index date. Study outcomes (health care costs and utilizations) were compared between the disease and comparator cohorts using 1: 1 propensity score matching (PSM). Results: A total of 90,528 patients were included in the hyperpotassemia and comparison cohorts. After 1: 1 PSM, 28,929 patients were matched from each cohort, and baseline characteristics were proportionate. Patients diagnosed with hyperpotassemia were more likely to utilize health care resources, including Medicare carrier (98.4% vs. 77.8%), Durable Medical Equipment (DME; 36.7% vs. 23.6%) and Home Health Agency (HHA; 18.4% vs. 7.3%) claims, outpatient visits (75.5% vs. 47.7%), inpatient (37.4% vs. 3.4%), skilled nursing facility (SNF, 13.3% vs. 3.4%) and hospice stays (1.1% vs. 0.8%) and prescription drug claims (54.1% vs. 50.7%). Patients diagnosed with hyperpotassemia also incurred higher Medicare carrier ($3,447 vs. $1,636), DME ($343 vs. $167), HHA ($938 vs. $412), outpatient ($11,006 vs. $4,534), inpatient ($7,156 vs. $1,463), SNF ($2,688 vs. $612), hospice ($244 vs. $198), pharmacy ($1,104 vs. $812) and total costs ($26,926 vs. $9,834) (p<0.05). Conclusions: The economic burden and health care resource utilization were significantly higher for patients diagnosed with hyperpotassemia. PDB115 Costs of Hospitalization of Type 2 Diabetic Patients Associated with Severe Hypoglycemia Laires P 1, Conceição J 1, Araújo F 2, Dores J 3, Silva C 4, Radican L 5, Nogueira A M 1 Sharp & Dohme, Oeiras, Portugal, 2Hospital Beatriz Ângelo, Loures, Portugal, 3Hospital de Santo António, Porto, Portugal, 4Eurotrials, Lisbon, Portugal, 5Merck & Co., Inc., Whitehouse Station, NJ, USA . . . . . . . . 1Merck Objectives: HIPOS-ER is an observational, cross-sectional, multicenter study aimed (a) to describe type 2 diabetes patients treated with an anti-hyperglycemic agent (AHA) and admitted to the emergency room (ER) due to a hypoglycemic event and (b) to estimate health care resources use and its costs related with the ER hypoglycemic episode. In this analysis, costs within hospitalized patients following severe hypoglycemia assisted at the ER were specifically calculated. Methods: The study was conducted in 7 centers in mainland Portugal for a period of 12 months (Jan 2013 – Jan 2014). Costs related with these hospitalized patients were calculated considering the hospital perspective. Unit costs for 2014 were extracted from official sources and reported in euros. Patient level data were used to calculate average costs. Regarding ER attendance, costs were calculated multiplying resource use by corresponding unit costs. Regarding hospitalization, length of stay was multiplied by daily cost obtained through hospitals accountancy. Both costs were summed and descriptive statistics were calculated. Productivity loss cost within ER admission and hospitalization was calculated for employed patients using the Human Capital Approach. Results: 238 patients were enrolled and 105 (44%) were hospitalized. These patients had a median length of stay of 5.4 days and were most commonly hospitalized at the Internal Medicine department (n= 80; 76%). Nine patients (8.6%) died during hospitalization. Hospitalized patients had the following average (range) costs: pre-hospital care and transport € 34 (€ 0-€ 92); emergency room € 218 (€ 58-€ 1,348); hospitalization € 2,880 (€ 140-€ 26,486); productivity loss costs due to ER and hospitalization € 31 (€ 0-€ 1,579). Thus, mean total cost per hypoglycemic event leading to hospitalization was € 3,163 (€ 230-€ 26,818). Conclusions: We conclude that severe hypoglycemia represent a substantial cost to Society and to the public hospitals of the National Health System in particular for those cases requiring hospitalization. PDB116 The Health Service and Economic Impact of Glucagon Rescue Administration During Severe Hypoglycemic Events Leinwand B , Hughes K E , Inocencio T Avalere Health, Washington, DC, DC, USA . . . . Objectives: Hypoglycemia, which if left untreated, can be severe and result in seizures, unconsciousness, and coma, during which another person’s help is required to administer a rescue dose of glucagon. Injectable glucagon kits are difficult to use and require training to administer. This study aimed to quantify the economic impact of using glucagon kits on resource use and costs, and identify evidentiary gaps requiring future research. Methods: A conceptual model was developed illustrating the series of events resulting from a SHE: successful administration of glucagon, ambulance calls, transport to the ED, inpatient admission, and outpatient follow-up. A literature search was conducted to assess service use and costs associated with severe hypoglycemia events (SHE). English language articles were reviewed in PubMed, EMBASE and Cochrane databases. Results: Resource use associated with SHE, as a function of successful administration of glucagon, has not been systematically evaluated in the literature. Uncertainty exists for the probability of receiving a glucagon prescription for diabetics and successful use of glucagon kits. Furthermore, based on successful administration of glucagon, the probability of ambulance calls, transports to the ED, inpatient admissions, and frequency of outpatient follow-up are lacking in the literature. Conclusions: Diabetes is a costly condition for payers, and a common complication is hypoglycemia. Glucagon kits are effective in stabilizing diabetics’ blood glucose levels during SHE; however, oftentimes physicians do not prescribe, patients do not fill, or caregivers do not successfully administer glucagon due to the complex administration procedures. As less complicated glucagon products are developed, their value propositions must be informed by the economic implications resulting from the complex administration requirements of current kits. However, the current literature does not systematically evaluate these implications. Consequently, future research is needed to quantify the impact of non-successful administration of glucagon rescue kits, as well as the extent of under-utilization. PDB117 Hipos-Er (Hypoglycemia in Portugal Observational Study – Emergency Room): Costs And Health Care Resource Consumption Data Laires P 1, Conceição J 1, Araújo F 2, Dores J 3, Silva C 4, Radican L 5, Nogueira A M 1 Sharp & Dohme, Oeiras, Portugal, 2Hospital Beatriz Ângelo, Loures, Portugal, 3Hospital de Santo António, Porto, Portugal, 4Eurotrials, Lisbon, Portugal, 5Merck & Co., Inc., Whitehouse Station, NJ, USA . . . . . . . . 1Merck Objectives: HIPOS-ER is an observational, cross-sectional, multicenter national study to describe the patient population of type 2 diabetics with hypoglycemic episodes that enter the emergency department caused by an anti-hyperglycemic agent (AHA). A key secondary objective is to estimate health care resource consumption and costs associated with this type of hypoglycemias. Methods: The study was conducted in 7 centers in mainland Portugal for a period of 12 months (Jan2013-Jan2014). Patient level data and resource utilization were collected. Average costs were assessed by multiplying 2014 unit costs (available from public sources) with all relevant health care resource consumption items registered in the emergency room following the hypoglycemia events and through hospital accountancy for the hospitalized patients (length of stay x daily mean cost of hospitalization). Results: The study enrolled 238 patients and the calculated proportion of hypoglycemic episodes among all emergency events in the same period was 0.075% (95%CI: 0.067%, 0.083%). In this population, 55.0% of the patients were using insulin, 31.5% were treated with a secretagogue, 6.7% were on a combination of insulin and a secretagogue oral agent and 6.7% were on oral non-secretagogue based AHA therapy. Mean patient age was 76 years and 57.6% were females. Estimated mean (range) of direct costs assessed in the emergency room were: emergency transportation €33 (€0-€92), drugs €4 (€0-€45), laboratory workup 56€ (€8-€212), other exams €72 (€0-€944), physician and nurse time €30 (€4-€211) and €14 (€1-€90), respectively. Mean hospitalization cost was €1,271 (€0-€26,486). Mean indirect cost related with productivity loss within employed patients was €15 (€0-€1,579). Total cost, including direct and indirect costs, was € 1,493 (€34-€26,818) per hypoglycemic event. Hospitalization was the main cost driver (85% of total costs). Conclusions: We conclude that hypoglycemia represent a substantial cost for the Society and in particular for the public hospitals of the National Health System. PDB118 Cost Analysis of Severe Hypoglycemia in Treated Type 2 Diabetic Patients According to Anti-Hyperglycemic Agent Therapy Laires P 1, Conceição J 1, Dores J 2, Araújo F 3, Silva C 4, Radican L 5, Nogueira A M 1 Sharp & Dohme, Oeiras, Portugal, 2Hospital de Santo António, Porto, Portugal, 3Hospital Beatriz Ângelo, Loures, Portugal, 4Eurotrials, Lisbon, Portugal, 5Merck & Co., Inc., Whitehouse Station, NJ, USA . 1Merck . . . . . . . A353 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: HIPOS-ER is the first national Hypoglycemia study in Portugal collecting specific resource data directly in the hospitals. Here we aim to describe the average cost of severe hypoglycemic event by anti-hyperglycemic agent (AHA) class. Methods: The study was conducted in 7 centers in mainland Portugal for a period of 12 months (Jan13-Jan14). Patient level data and resource utilization were collected. Unit costs for 2014 were extracted from official sources. Regarding emergency room (ER) attendance, costs were calculated multiplying resource use by corresponding unit costs. For hospitalization, length of stay was multiplied by daily cost obtained through hospital account record. AHA therapy classes were: Group 1 (insulin), Group 2 (secretagogue), Group 3 (oral AHA excluding secretagogue), and Group 4 (at least one insulin and one secretagogue). Results: 238 patients were enrolled and 105 (44%) were hospitalized. The distribution based on AHA therapy: 55% (131) Group 1, 32% (75) Group 2, 7% (16) Group 3 and 7% (16) Group 4. After ER episode, Group 2 patients were more often hospitalized versus Group 1 (71% vs. 29%; p< 0.001) and Group 4 (31%; p= 0.003). The global cost was 1,493€ (34€ -26,818€ ) and hospitalization was the main cost driver accounting for 85% of costs. The total cost per AHA class was: Group 1: 1,309€ ; Group 2: 1,880€ ; Group 3: 1,350€ ; Group 4: 1,330€ . When comparing ER-only vs. hospitalized: Group 1: 167€ vs. 4,105€ ; Group 2: 185€ vs. 2,583€ ; Group 3: 156€ vs. 2,278€ ; Group 4: 237€ vs. 3,734€ . Group 1 accounted for 48% of overall costs, Group 2 40%, Groups 3 and 4, 6% each. Group 2 had the highest hospitalization rate (70.7%). Conclusions: AHA classes may contribute differently for costs associated with severe hypoglycemia. Insulin based therapy had the greatest overall cost followed closely by secretagogue type drugs, which were associated with more hospitalizations. PDB119 The Burden of Severe Hypoglicaemias and Diabetes Ketoacidosis: A Population-Based Study Ciampichini R 1, Cozzolino P 2, Cortesi P A 1, Fornari C 1, Madotto F 1, Chiodini V 1, Mantovani L G 3, Cesana G 1 1University of Milano - Bicocca, Monza, Italy, 2Charta Foundation, Milan, Italy, 3Federico II University of Naples, Naples, Italy . . . . . . . . . . Objectives: The aim of this study was to assess the burden of severe hypoglycemia (HYPO) and diabetes ketoacidosis (DKA) in terms of incidence, treatment patterns and cost in Italy. Methods: Eligible patients were identified through a data warehouse (DENALI), which matches with a probabilistic linkage demographic, clinical and economic data of about 9.9 million individuals of Lombardy region. The study population was made of all individuals with a diagnosis of type-1 diabetes (T1DM) who, during the period 1-1-2000 to 31-12-2010 developed a first episode of severe HYPO (ICD-9-CM: 251.0-251.2) or DKA (ICD-9-CM: 250.10-250.13) leading to hospitalization (index event). The identified individuals were followed-up from the index event to a maximum of 10 years. We evaluated demographic characteristics of the study population and costs (hospitalizations, drugs and outpatient examinations/visits) from the National Health Service’s perspective. Results: During the observational period 1,321 T1DM subjects had at least one hospital admission for HYPO (27.5%) or DKA (72.5%) event. Median age (min-max) at the first HYPO or DKA event was 66.9 (1.3-96.7) and 39.4 (1.0-95.7) years, respectively. The overall mortality was 14.3 deaths/100 patient-years among HYPO subjects and 7.0 among DKA ones: the difference in mortality rates between groups disappeared once controlling for age. The mean costs with 95% C.I. (€ /patient-year) for HYPO and DKA patients were, respectively 10,442 (8,755-12,129) and 9,720 (8,659-10,782) in the event year, 10,296 (8,446-12,145) and 5,805 (4,539-7,071) in the year before the event, and 5,619 (3,841-7,398) and 4,974 (3,912-6,035) in the year after the event. Hospitalizations represented the driver of total costs: in the year of event and in the year after it varied from 70 to 56% and 76 to 60% for HYPO and DKA patients, respectively. Conclusions: This study attempted to address the burden of severe T1DM HYPO and DKA events in Italy in an unselected population. The burden showed to be relevant in terms of incidence, mortality and costs. p= 0.03), use of 3rd generation SU (OR= 0.86 [0.78 - 0.95]; p= 0.002), and chronic renal disease (OR= 1.34 [1.07 - 1.67]; p= 0.01). Conclusions: In conclusion, multiple factors, including efficacy and hypoglycemia, are associated with discontinuation of SU treatment after insulin initiation. PDB121 Impact of Hypoglycemia on Discontinuing or Down-Titrating Sulfonylurea Among Type 2 Diabetes Patients Without Insulin Laires P 1, Iglay K 2, Fan C P S 3, Li Z 3, Tang J 3, Qiu Y 2 Sharp & Dohme, Oeiras, Portugal, 2Merck Sharp & Dohme Corp., Whitehouse Station, NJ, USA, 3AscelpiusAnalytics Ltd., Hong Kong, Hong Kong . . . . . . . . 1Merck Objectives: Sulfonylurea (SU) may be discontinued or down-titrated due to hypoglycemia. Hypoglycemia may be more concerning for patients not receiving aggressive efficacy-driven treatment such as the dual-therapy of SU and insulin. A retrospective cohort study using the MarketScan database was conducted to assess the association between hypoglycemia and therapy changes (discontinuation or down-titration) among adults receiving SU therapy without insulin. Methods: Patients with the first SU prescription (Rx) (index date) in 2009-2011, ≥ 18 years of age on the index date, and with ≥ 1 year continuous enrollment pre- and postindex were included. Patients were excluded if they received insulin within the 1-year pre- or post-index, had ≥ 2 SUs on the index date, or had type 1, gestational or secondary diabetes. Therapy changes were determined during the 1-year postindex period. Discontinuation occurred when consecutive SU fills were ≥ 90 days apart. Down-titration occurred when an SU fill had a lower equivalent dose than the index dose. Hypoglycemic events were identified using ICD-9 code between the index date and the therapy change or the end of the 1-year post-index period. Cox regression was used to evaluate the association between hypoglycemic events and therapy changes. Results: 97,570 patients were included in the study, of which 50,854 (52.1%) experienced therapy changes within 1-year post-index. Patients with hypoglycemic events were at significantly higher risk for therapy changes (HR= 1.86 [1.75, 1.97]; p< . 01). Specifically, they were 197% more likely to down-titrate (HR= 2.97 [2.53, 3.46]; p<. 01) and 80% more likely to discontinue (HR=1.80 [1.69, 1.92]; p< . 01). Conclusions: Post-index hypoglycemic events are significantly associated with therapy changes among patients receiving SU without insulin, especially down-titration. PDB122 Guideline Adherence and Control Of Diabetes Mellitus With Co-Morbidities in a Tertiary-Care Hospital in Malaysia Iqbal M Z 1, Iqbal M S 1, Khan A H 2, Sulaiman S A 2, Iqbal M W 3 of Clinical Pharmacy, Faculty of Pharmacy, AIMST University, Kedah, Malaysia, 2Department of Clinical Pharmacy, School of Pharmaceutical Sciences, Universiti Sains Malaysia, Pulau Pinang,,,, Malaysia, 3Faculty of Law, Universiti Malaya,, Kualalumpur, Malaysia . . . . . . . . . . 1Department DIABETES/ENDOCRINE DISORDERS – Patient-Reported Outcomes & Patient Preference Studies Objectives: To evaluate doctors’ adherence to Malaysian Clinical Practice Guideline (CPG) 2009 in the management of diabetes mellitus with co-morbidities in Malaysia. Methods: Cross-sectional study was done at a tertiary-care hospital in Malaysia. Total 51 physicians and 1020 patients’ prescriptions written by same physicians (20 prescriptions per physician) were analyzed. All patients had diabetes mellitus with co-morbidities. Depending on the recommendations of CPG 2009, the prescriptions were clustered as adherent and non-adherent prescriptions. All obtained data were analyzed using descriptive and inferential statistics. Results: A statistically significant negative association (Ф= 0.094, p-value= 0.003) was observed between diabetes mellitus control and co-morbidities. CPG adherent had statistically weak negative association (Ф= - 0.081, p-value= 0.010) with patients having co-morbidities (41.6%). No statistically significant association was observed between CPG adherence and any other co-morbidity. Majority of the patients received guidelines-compliant pharmacotherapy. The overall good level of physician adherence with CPG 2009 was observed in the management of diabetes mellitus with comorbidities. Conclusions: The study explored several features of prescription pattern of the physicians involved in the management of diabetes mellitus with co-morbidities and recognized the need for improvement in their prescription pattern for treating the diabetes mellitus. PDB120 Factors Associated With Discontinuation of Sulfonylurea Therapy In Type 2 Diabetes Patients Who Initiate Insulin PDB123 The Relationship Between Macular Edema and Health Outcomes Among Patients With Diabetes in Western Europe 1Merck Laires P 1, Fu A Z 2, Iglay K 3, Qiu Y 3 Sharp & Dohme, Oeiras, Portugal, 2Georgetown University, Washington, DC, DC, USA, 3Merck Sharp & Dohme Corp., Whitehouse Station, NJ, USA Pignot M 1, DiBonaventura M 2 1Kantar Health, München, Germany, 2Kantar Health, New York, NY, USA Objectives: Sulfonylureas (SU) represent a common treatment for type 2 diabetes (T2DM), but they are associated with hypoglycemia, weight gain, and possibly cardiovascular events. The purpose of this study is to evaluate factors associated with SU discontinuation after insulin initiation. Methods: Patients ≥ 21 years old with a T2DM diagnosis between 2005 and 2012 were identified using the GE electronic medical records database. Index date was defined as the first insulin prescription (Rx) between 2006 and 2011. Patients were required to be on SU at the index date. Patients were excluded if they did not have medical records available ≥ 12 months before and after index date, were receiving insulin in the 12 months prior to index date, or had other forms of diabetes. Treatment with other diabetes medications in addition to SU and insulin were permitted in this study. SU discontinuation occurred when the gap between the end date of current SU Rx supply and the start date of subsequent SU supply was ≥ 90 days apart. Multivariate logistic regression was performed to identify factors associated with SU discontinuation. Results: A total of 8,185 patients were selected, with mean age 64 years and 49% were male. 60.4% discontinued their SU Rxs within 1 year, with a median time from insulin initiation to SU discontinuation of 88 days. In the logistic regression, baseline diagnosed hypoglycemia (OR= 2.29 [95% CI 1.03 - 5.10]; p= 0.04) and baseline HbA1c (OR= 1.04 [1.01 - 1.06]; p= 0.004) were identified as factors associated with SU discontinuation. Additional factors included BMI (< 25 kg/m2 vs. ≥ 30 kg/m2; OR= 1.23 [1.03 - 1.46]; Objectives: Diabetes is associated with a number of microvascular and macrovascular complications. Diabetic macular edema (DME) is one of these complications and is among the leading causes of vision impairment. However, little data exists as to the patient-related burden of DME in Europe and the aim of the current study was to address this gap. Methods: Data from the 2013 5EU (France, Germany, Italy, Spain, and UK) National Health and Wellness Survey (NHWS) were used (N= 62,000). The NHWS is a patient-reported survey administered to a demographically representative sample of adults (with respect to age, sex, and region). Patients who reported experiencing DME were compared with a propensity-scored matched control group of patients with diabetes but without DME. Matching variables included demographics, comorbidities (Charlson comorbidity index [CCI]), and diabetes history. Post-match, patients with DME and matched controls were compared on health outcomes (SF-36v2, Work Productivity and Activity Impairment, and selfreported health care resource utilization) using general linear models. Results: 4,088 patients reported diabetes (6.6%). Of these, 296 (7.2%) reported having DME. Patients with DME were more likely to have type 1 diabetes (26.4% vs. 8.5%), had been diagnosed with diabetes for longer (54.4% vs. 20.7% were diagnosed 16+ years), were more likely to use insulin (68.6% vs. 27.3%), and had a greater comorbidity burden (CCI = 2.2 vs. 1.6) (all p< .05), among other differences. After matching on these variables, patients with DME (n= 286) reported significantly worse physical health status (PCS: 41.1 vs. 43.2), greater overall work impairment (36.2% vs. 25.64%), and . . . . . . . A354 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 more health care provider visits in the past six months (11.7 vs. 8.9) compared with matched controls (n= 286) (all p< .05). Conclusions: DME was more commonly reported by patients with diabetes and its presence was associated with a significant humanistic and economic burden in the 5EU. PDB124 EQ-5D Scores In Patients Receiving Tolvaptan for the Treatment Of Hyponatraemia Secondary to the Syndrome of Inappropriate Antidiuretic Hormone Secretion Trueman D 1, Hancock E 1, Robinson P 2, Dale P 2, O’Reilly K 2, Gisby M 2 1Decision Resources Group, London, UK, 2Otsuka Pharmaceutical Europe Ltd, Wexham, UK . . . . . . Objectives: Hyponatraemia (HN) is estimated to occur in 15% of all hospitalised patients with syndrome of inappropriate antidiuretic hormone secretion (SIADH) being one of the most common aetiologies. Patients treated with tolvaptan have demonstrated improvements in health related quality of life (HRQL) in the SALT I & II randomised controlled trials. The objective of this study was to map the SF-12 responses from SALT I & II to EQ-5D using a publically available algorithm and predict the change in EQ-5D associated with tolvaptan. Methods: SF-12 scores from the pooled SALT I & II studies were converted to EQ-5D scores using a mapping algorithm developed by Gray, et. al 2006. Simulated EQ-5D scores were then used to estimate changes in EQ-5D from baseline at day 30 using ordinary least squares regression (OLS) as a function of baseline characteristics, treatment arm and achievement of sodium correction (NA+) at day 4 (> 135 mmol/l). A preferred model was selected based on the highest adjusted R-squared. Secondary analyses looked at change from baseline in EQ-5D at day 7, day 14 and at 7 day follow-up following treatment discontinuation. Results: The preferred model included baseline age, gender, sodium level, EQ-5D and a tolvaptan indicator variable. Based on this model, tolvaptan was associated with a positive increase in simulated EQ-5D of 0.10 (n=164, p=0.03) at day 30 vs. placebo. After 7 days follow-up, tolvaptan was associated with a positive, but non- statistically significant effect of 0.04 (n=74, p= 0.54). No effect was observed at day 7 (n= 76, p= 0.90) or at day 14 (n= 85, p= 0.84). Sodium correction did not appear to be a statistically significant predictor of HRQL. Conclusions: The preferred model indicated a statistically significant improvement in HRQL associated with tolvaptan use at day 30. Further research is required to establish whether sodium correction has an effect of HRQL. PDB125 Patient Experience With The Single-Use Pen For Injection of Once Weekly Dulaglutide in Injection-Naive Patients With Type 2 Diabetes Van Brunt K 1, Ignaut D A 2, Zimmermann A G 2, Threlkeld R J 2, Matfin G 3 Lilly and Company, UK, Windlesham, Surrey, UK, 2Eli Lilly and Company, Indianapolis, IN, USA, 3International Diabetes Center, Minneapolis, MN, USA . . . . . re-use of eDiaries previously used by patients doing fingerstick tests poses a low and acceptable risk; flexibility for handling measurements should be incorporated in the eDiary design including batch and individual reporting, reporting hypoglycaemic events as part of a meal event or as a standalone event and edit checks should be included to identify where a number of low measurements relate to the same event; there should be clear guidance to patientson how to transfer measurements to the eDiary; a method for managing control test measurementsshould be incorporated; and integrating glucometer measurements decreases patient burden and increases patient engagement. . . . PDB127 The Development of an Integrated Ecoa Solution To Improve The Quality of Data Capture In Diabetes Clinical Trials McEvoy K CRF Health, London, UK . Objectives: The 347 million people who live with diabetes face an array of daily disease management tasks, e. g. measuring blood glucose, keeping track of the nutritional value of meals, and monitoring insulin usage. Diabetes trials vary in methodology, level of patient burden, volume of data captured and patient compliance. Our aim was to develop an electronic data capture system for use in clinical trials which would decrease patient burden while increasing compliance and improving data quality. Methods: We developed a diabetes specific, electronic, event driven diary for capturing data relevant for clinical trials. Iterations of the diary were tested in diabetes patients via focus groups and one-on-one usability evaluations. Feedback was analysed to understand the typical day-to-day experience of living with diabetes, and to examine the impact and acceptability of a tailored electronic solution. Feedback was incorporated into a refined solution for use in clinical trials. Results: The nature of diabetes means that patients are typically very actively engaged in managing their disease, although a lot of variation was seen in patients and their management routines. The requirement for multiple devices and high volume of data was also reported as burdensome. The focus groups and usability studies highlighted the benefit of providing a flexible solution, as well as redesigning the diary from unstructured blood glucose measurement reporting to event driven reporting. This redesign met the needs of patients, as well as the requirements of clinical trial protocols. Patients agreed that they would prefer this integrated, intuitive solution over traditional paper and electronic solutions. Conclusions: When faced with the task of recording patient data in diabetes, a well-designed and thoroughly tested electronic solution can reduce burden and increase patient satisfaction. This in turn improves compliance, data quality and overall study efficiency, while meeting the needs of all stakeholders. 1Eli Objectives: This 4-wk, Phase 3b, multicentre, open-label, single-arm, outpatient study evaluated the safe and effective use of the Single-Use Pen (SUP) in patients with type 2 diabetes (T2D) who were naïve to self-injection or injecting others. The SUP contains a pre-filled syringe and automates needle insertion, retraction, and drug delivery; specifically designed for once-weekly glucagon-like peptide-1 receptor agonist, dulaglutide. Patient-reported outcomes (PROs) related to self-injection and to the SUP were important secondary outcomes. Methods: Patients (N= 211) were trained on correct injection technique with the SUP containing 0.5 mL placebo prior to initial self-injection. PRO measures were completed by patients after final injection or at early termination to evaluate device ease of use, experience (including satisfaction/confidence), and key device features. Site trainers rated how easy/difficult it was to train the patient to use the SUP. Results: 99.0% of patients found the device easy to use. Patients found it easy to hold the SUP when injecting and to push the button to inject (97.6% and 98.1%, respectively). 97.1% of patients were satisfied with the overall injection experience, 96.7% of patients were confident they could identify when the full dose was delivered, and 99.0% were confident in their overall ability to use the SUP. Patients liked not having to attach (99.0%), touch (98.6%), or see (95.7%) the needle. 95.7% of patients also liked hearing the click indicating the dose was complete and 94.8% liked the automatic needle insertion. Finally, 92.4% of site trainers found it easy to train patients on how to use the SUP. Conclusions: This study demonstrated the SUP could be used safely and effectively by injection-naïve patients with T2D; PRO results indicated patient satisfaction with the SUP injection experience. A positive injection experience may be an important factor for some patients and providers when initiating injectable therapy. PDB126 Best Practices in Integrating Home Glucometer Measurements With Electronic Patient Reported Outcomes (Epro) In Clinical Trials Garner K CRF Health, London, UK . Objectives: Monitoring glycaemic control is important in diabetes clinical trials and is also relevant for some oncology trials. It is possible to seamlessly integrate blood glucose measurements with clinical trial data via Bluetooth. This integration reduces transcription errors and improves accuracy and completeness of data, while reducing burden on patients. Our aim was to explore best practices in integrating measurements with ePRO. Methods: A review of best practice and experience in blood glucose measurement data integration was undertaken and the findings were incorporated into a solution for use in clinical trials. A systematic literature review was also undertaken to assess the contamination risk to individuals handling and re-using electronic diaries (eDiary) previously used by patients doing fingerstick tests for blood glucose measurement. We defined search terms and undertook a PubMed and internet search. All documents that were identified as possibly relevant were reviewed and summarised. Results: The review revealed key features and activities related to planning, design, testing and delivery that are important for successful integration of blood glucose measurements with ePRO data. The literature review confirmed that the risk in handling and re-using eDiaries previously used by patients doing fingerstick tests was very low. Conclusions: Handling and PDB128 German Patients’ Preferences for Attributes Of Type 2 Diabetes Medications Gelhorn H 1, Stringer S 2, Reinders S 3, Schreeb K 4 1Evidera, Bethesda, MD, USA, 2Evidera, Inc., Bethesda, MD, USA, 3Evidera, London, UK, 4Boehringer Ingelheim Pharma GmbH & Co. KG, Ingelheim am Rhein, Germany . . . . Objectives: Treatments for Type 2 Diabetes Mellitus (T2DM) are associated with varying effectiveness and safety profiles. Patients’ preferences for each of the medication characteristics that yield these varying profiles can be assessed through discrete choice experiments (DCEs). This study expands on the knowledge from two previous T2DM DCE studies conducted in the United Kingdom (UK) and the United States (US), and was aimed at evaluating the relative importance of medication attributes that influence medication choice among participants with T2DM from Germany. Methods: A web-based DCE was conducted among patients with self-reported T2DM from Germany. The DCE was designed to examine 7 attributes of T2DM medications (efficacy, urinary tract infection/genital infection side effects, nausea/other gastrointestinal side effects, weight change, and hypoglycemic events, treatment in case of low blood sugar, and blood pressure). Part-worth utilities were estimated using multi-logit models, and relative importance [RI] values were calculated for each attribute. Results: N= 600 Participants with T2DM completed the study (50% male; mean age= 58.2 years SD= 10.0; BMI= 32.4, SD= 6.8). The RI values for the attributes in order of importance were: treatment in case of low blood sugar (22.5%), hypoglycemic events (18.1%), weight change (17.5%), efficacy (15.0%), nausea/other gastrointestinal side effects (12.9%), UTI/genital infection side effects (7.9%), and blood pressure (6.0%). Conclusions: The results of this study suggest that hypoglycemic events and the interventions required in the case of such events are of great importance to patients; these two attributes represent over 40% of the variance in patients’ medication decisions. Change in body weight as a consequence of treatment was also an important attribute to patients. The results may help treatment providers and payers to understand the preferences of patients with T2DM. Understanding these preferences may be useful in devising strategies for successfully engaging and maintaining patients on T2DM treatments. PDB129 Self-Reported Frequency and Impact of Non-Severe Hypoglycaemia in Insulin-Treated Adults in the UK Chubb B 1, Jensen M M 2, Frier B M 3 Nordisk Limited, Gatwick, UK, 2Novo Nordisk Scandinavia AB, Copenhagen, Denmark, 3The Queen’s Medical Research Institute, Edinburgh, UK . . . . . 1Novo Objectives: Hypoglycaemia is the main side-effect of insulin therapy and can prevent optimal diabetes management. Real-world data on the frequency and impact of non-severe (self-treated) hypoglycaemic events are scarce. Self-reported frequency of non-severe hypoglycaemic events (NSHEs), their impact on personal well-being, work productivity and health care resource use, and patient–physician communication following non-severe events, were examined in people taking insulin. Methods: Adults in the UK aged >15 years with Type 1 or insulin-treated Type 2 diabetes mellitus (T1DM or T2DM) completed ≤ 4 questionnaires at weekly intervals (7-day recall). Severe hypoglycaemic events (requiring external assistance) are not A355 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 reported here. Results: Overall, 1,038 respondents (T1DM= 466, T2DM= 572) completed 3,528 questionnaires. Mean insulin treatment duration was 11.7 years and mean HbA1c was 8.2% (66.2 mmol/mol). Mean NSHE/week was 2.4 (T1DM) and 0.8 (T2DM); 23% (T1DM) and 26% (T2DM) occurred at night. Fatigue and reduced alertness were the commonest sequelae of NSHEs (78% and 51% of respondents, respectively). Effects of nocturnal NSHEs lasted significantly longer than effects of daytime events: T1DM= 10.6 vs. 4.9 hours (p= 0.0002); T2DM= 15.3 vs. 5.1 hours (p< 0.0001). In the week following a NSHE, blood glucose testing increased 12% (T1DM) and 21% (T2DM). In employed respondents (47% of total), 20% of NSHEs caused loss of work-time, which was longer following nocturnal NSHEs: T1DM= 2.7 vs. 1.1 hours (p= 0.0184); T2DM= 2.5 vs. 1.6 hours (p= 0.1340). Over a third of employed respondents experienced difficulty concentrating at work following NSHEs (T1DM= 39%; T2DM= 44%). Respondents contacted a health care professional (HCP) after 3% (T1DM) and 7% (T2DM) of NSHEs. Overall, respondents rarely or never informed HCPs about NSHEs (T1DM= 82%; T2DM= 69%). Conclusions: NSHEs are common in adults with insulin-treated diabetes in the UK, and have a negative impact on personal well-being, work productivity, and health care resource use. As they are seldom reported to HCPs, the burden of hypoglycaemia may be underestimated. PDB130 Patients’ Preferences in Oral Diabetes Treatment: A Discrete Choice Experiment in Type2 Diabetes Mellitus Mühlbacher A C , Bethge S Hochschule Neubrandenburg, Neubrandenburg, Germany . . . Objectives: The aim of the empirical study is to evaluate patients’ preferences for different characteristics of treatment in type2 diabetes mellitus (T2DM). As T2DM treatment asks for strict adherence, patient’s needs and preferences should be taken into consideration. Methods: Based on a qualitative and quantitative analysis a Discrete Choice Experiment (DCE) was applied to identify patient preferences. Apart from six identical attributes (adjustment of glycated hemoglobin [HbA1c], prevention of hypoglycemia, risk of genital infection, risk of gastrointestinal problems, risk of urinary tract infection and weight change) one continuous variable of either healthy life years equivalents (HYE) or money equivalents (ME) was included. The DCE was conducted using a fractional factorial design and the statistical data analysis used random effect logit models. Results: In total N= 626 (N= 318 HYE + N=308 ME) T2DM patients participated in the survey. The estimation revealed a clear dominance for prevention of hypoglycemia (coefficient: 0,937) and adjustment of HbA1c (coefficient: 0,541). The attributes, additional healthy life years (coefficient: 0,458) or additional cost (coefficient: 0,420) were in the middle rank and both of significant impact. The side effects, risk of genital infection (coefficient: 0,301), risk of gastrointestinal problems (coefficient: 0,296) and risk of urinary tract infection (coefficient: 0,241) followed in this order. Possible weight change (coefficient: 0,047) was of less importance (last rank) to the patients in this evaluation. Conclusions: These survey results demonstrate how much a treatment characteristic of a (hypothetical) oral diabetes treatment affects the treatment decision Understanding how patients perceive and value different aspects of oral T2DM treatment is vital to the optimal design and evaluation of treatment options. (IB 209403011/14). PDB131 Hye and me as Identical Currencies in Preference Studies? A Discrete-Choice Experiment in Type2 Diabetes Mellitus Mühlbacher A C , Bethge S Hochschule Neubrandenburg, Neubrandenburg, Germany . . . Objectives: Only a few studies have explored deriving clinically relevant timeequivalents in comparison to money-equivalents from Discrete-Choice Experiment (DCE) data. By separating the decision model and including a) healthy life-years equivalents (HYEsurvey) or b) money equivalents (MEsurvey) it could be derived if both lead to similar preference patterns and allow answering the question: Can HYE or ME serve as identical “currency” for patients with type2 diabetes mellitus (T2DM) patients? Methods: A DCE, separated into two versions, was applied to identify patient preferences in oral diabetes treatment. Six identical attributes and one continuous variable of either HYE or ME were included. To be able to test the rescaling effect a scope test has been included, by using different level ranges of the time/money attribute. DCE used a fractional factorial design and random effectlogit-models (Stata, xtlogit and gllamm). Results: The estimation of the N= 626 (N= 318HYE+N= 308ME) datasets of T2DM patients revealed similar preference patterns for both survey version were prevention of hypoglycemia (coefficient. HYE: 0,937; coefficient. ME: 0,847) and adjustment of hemoglobin A1c (HbA1c) (coefficient. HYE: 0,541; coefficient. ME: 0,649) occurred on first rank. Additional healthy life years (coefficient. HYE: 0,458) or additional cost (coefficient. ME: 0,420) ranked in middle positions. Side effects of risk of genital infection (coefficient. HYE: 0,301; coefficient. ME: 0,416), risk of gastrointestinal problems (coefficient. HYE: 0,296; coefficient. ME: 0,408) and risk of urinary tract infection (coefficient. HYE: 0,241; coefficient. ME: 0,355) followed accordingly. Possible weight change (coefficient. HYE: 0,047; coefficient. ME: 0,067) showed no significant effect in this evaluation. Conclusions: For the first time the methods of HYE and ME were used in one study to be able to compare patients’ preferences regarding those two continuous currencies (HYE/ ME) of treatment in T2DM, as well as the influence of those criteria on the patient decision patterns and patient benefit. Therefore, as HYE and ME led to comparable preference patterns, both can be discussed as summary measures of health outcome, in interchangeable ways, but further research is needed. (IB 209203011/14) PDB132 Sagit©: A Novel Clinician-Reported Outcome for Managing Acromegaly in Clinical Practice Giustina A 1, Bevan J 2, Bronstein M 3, Casanueva F 4, Chanson P 5, Petersenn S 6, Truong Thanh X M 7, Massien C 7, Dias-Barbosa C 8, Guillemin I 8, Arnould B 8, Melmed S 9 1University of Brescia, Brescia, Italy, 2Aberdeen Royal Infirmary, Aberdeen, UK, 3Hospital das Clinicas, Sao Paulo, Brazil, 4Universidad de Santiago de Compostela, Santiago de Compostela, . . . . . . . . . . . . . Spain, 5Hôpital Bicêtre, Kremlin-Bicêtre, France, 6ENDOC Center for Endocrine Tumors, Hamburg, Germany, 7Ipsen Pharma, Boulogne Billancourt Cedex, France, 8Mapi, Lyon, France, 9Cedars Sinai Medical Center, Los Angeles, CA, USA Objectives: SAGIT is a clinician-reported outcome tool designed to help endocrinologists describe acromegaly patients and disease activity in their everyday practice. SAGIT records five elements of the disease: Signs and symptoms (S), Associated comorbidities (A), GH concentration (G - either assessed as GH nadir after oral glucose tolerance test or GH random or series), IGF-1 concentration (I) and Tumor (T). The objectives of the present work were to assess acceptability, understanding and possible uses of SAGIT with endocrinologists in real conditions. Methods: Endocrinologists from France (n= 2), Germany (n= 1), Italy (n= 2), Spain (n= 2) and United States (n= 2) completed SAGIT for patients with active acromegaly (n= 9), controlled/stable acromegaly (n=10) and treatment-naïve patients (n=7). After completion of SAGIT, endocrinologist’s reported their perception and opinion of the tool using the PRAgmatic Content and face validity Test. Results: Endocrinologists had no difficulties completing the S, A, I and T elements of SAGIT and reporting the respective scores based on their patients’ data. Both GH nadir after oral glucose load and GH random or series were informed by most endocrinologists while information for one of the two elements only is required. The majority of endocrinologists deemed the information retrieved from SAGIT useful, either for research purpose (n= 7), decision-making (n= 6), or response therapy assessment (n= 5). They found it concise, easy to understand and unbiased (n= 8), and simple, quick to complete and informative (n= 7). Scores and decision rules and interpretation were among the aspects that required improvement. Scores of each elements of SAGIT vary according to patient’s acromegaly status. Conclusions: Face and content validity of SAGIT are demonstrated, as is its applicability in clinical practice and research. It seems a promising tool for staging and classifying acromegaly patients. The planned validation study will allow the definition of scoring rules, interpretation and recommendations for managing patients in clinical practice. PDB133 Patient-Reported Outcomes of Dipeptidyl Peptidase-4 Inhibitors: A Systematic Review Franch J 1, Pérez A 2, Lizán L 3, Aceituno S 3, Lopez F 4, Fuster E 4, Granell M 4 1EAP Raval Sud- Institut Català de la Salut - USR Barcelona ciutat - IDIAP Jordi Gol, Barcelona, Spain, 2Hospital de la Santa Creu I Sant Pau, Barcelona, Spain, 3Outcomes’10, Castellon, Spain, 4Novartis Farmaceutica, Barcelona, Spain . . . . . . . Objectives: To synthesize the available information on the therapeutic value of dipeptidyl peptidase-4 inhibitors (DPP-4) for the treatment of type 2 diabetes mellitus (T2DM) from the point of view of the patient-reported outcomes (PROs). Methods: A systematic review was performed on International (Pub Med, WOK, Scopus, Cochrane Library) and Spanish (IBECS, MEDES) databases. Observational studies and narrative or systematic reviews regarding T2DM patients and use of DPP-4 inhibitors until June 2013 were selected. Results: We identified 1,713 publications; 317 were excluded after duplicate review, 1,383 by title/abstract review and 9 after applying inclusion criteria (n= 9). A total of 4 studies conducted in Italy (n= 1), Germany (n= 1), USA (n=1), and one international (including Spain) were selected. Three publications had a retrospective design and 1 was prospective. Two studies reported information about adherence/persistence, one about satisfaction and one about preferences. No information about HRQoL was identified. Patients receiving DPP-4 inhibitors were more likely to be adherent than those treated with Glucagon-like peptide-1 (GLP-1) agonists [Odds Ratio= 0.40; 95% CI= 0.37-0.42], sulfonylurea [OR= 0.49; 95% CI= 0.460.52] or thiazolidinediones [OR= 0.54; 95% CI= 0.51-0.57]; moreover, DPP-4 inhibitors were associated with a lower risk of treatment discontinuation [Hazard Ratio= 0.74; 95% CI= 0.71-0.76], explained by a greater tolerability and a lower risk of hypoglycemia. Combination of DPP-4 inhibitors and metformin increased patient’s satisfaction by a 30%, associated with higher control of glucose levels. Most of the patients preferred DPP-4 inhibitors to GLP-1 agonists (85% vs. 15%; p< 0.001) as first option. In the Spanish population, the proportion of patients preferring DPP-4 inhibitors was even higher (90.4% vs. 9.6%; p< 0.001). Conclusions: PROs in DPP-4 inhibitors are poorly described in the literature. Nevertheless DPP-4 inhibitors are preferred as first option and are associated with higher persistence and satisfaction, mainly due to higher perception of glycemic control of glucose level and lower hypoglycemia risk. PDB134 Assessment of Effect of Continuous Subcutaneous Insulin Infusion Treatment, Insulin Analog And Human Insulin of Children With Diabetes Petkova E 1, Petkova V 1, Petrova G 2, Dimitrova M 1, Konstantinova M 3 1Medical University of Sofia, Faculty of Pharmacy, Sofia, Bulgaria, 2Medical University Sofia, Faculty of Pharmacy, Sofia, Bulgaria, 3Medical University of Sofia, Pediatric Endocrinology Clinic, Sofia, Bulgaria . . . . . Objectives: The objective of the study is to assess the cost of continuous subcutaneous insulin infusion (CSII) application for children with diabetes type 1 in Sofia, Bulgaria and to compare the changes in BMI and HbA (1c), of three groups of children- with diabetes – applying insulin pens with human insulin, with analog insulin and children with insulin pumps. Methods: The study was performed from the point of view of the health insurance fund and patients. The data collected from the patients’ dossier including demographics about their age, sex, weight, duration of disease and HbA (1c) and type of treatment (CSII or analogue insulin treatment with a pen device). Cost of CSII, blood glucose monitoring system and strips was calculated. The primary outcome observed was the variation in HbA (1c) and the secondary was the BMI change. 51 children were observed. The data collected from the patients’ dossier including demographics about their age, sex, weight, duration of disease and HbA (1c) and type of treatment (CSII or analogue insulin treatment with a pen device). Results: The total yearly cost weighed with the duration of the disease is 1850 Euro (30% reimbursed). The average improvement of HbA (1c) after the CSII introduction is 1.72 and the average BMI was 37.03. Conclusions: Improvements in glycemic control associated with CSII led to reduced HbA (1c) A356 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 that can guarantee good diabetes management, but its control over BMI in growing children is still unclear. The insulin pumps have too many advantages: reduction in insulin requirements, reduced variability of insulin absorption, decreased hypoglycemia incidents, avoided pain, improved quality of life. PDB135 Patient Preferences in Treatment of Diabetes Mellitus: A Systematic Review of Stated Preference Surveys Mühlbacher A C , Kaczynski A Hochschule Neubrandenburg, Neubrandenburg, Germany . . . Objectives: Diabetes Mellitus is one of the most expensive common diseases. Because of the great socio-economic importance of this indication, it seems necessary to consider the expectations and needs of all patients with regard to treatment characteristics. To evaluate patient preferences stated-preference methods are increasingly used. The aim was to analyze and compare the available evidence from patient preference studies regarding the treatment characteristics and to display which target criteria are most important from the patients view. Methods: A literature review in PubMed was conducted to identify stated preference studies in the indication of diabetes and show which properties of a treatment are relevant to the patient’s benefit. By means of level difference calculation and a subsequent normalization uniform weights for the attributes of the studies were determined. Based on a final ranking the key criteria according to their value proposition from the perspective of affected populations are demonstrated. Results: N= 13 studies could be included in the analysis. By calculating the level difference and performing a normalization it was possible to obtain a uniform representation of all attributes of each study. The analysis includes the presentation and analysis of DCE-studies which have illustrated the determined coefficients. The studies show that blood sugar control, side effects (such as hypoglycemia, nausea/stomach discomfort, weight changes), long-term complications (especially cardiovascular disease) and mode of administration represent the most patient-relevant outcomes in diabetes therapy. Blood glucose control achieved the highest ranking within the relative importance score. Conclusions: Based on the findings of this systematic review the evidence of patient preferences in diabetes treatment is synthesized. The results show the relative importance based on a derived ranking score of the used patientrelevant endpoints. The determined ranking allows the comparison of the results of the identified preference studies and might solve the problem of scale heterogeneity. PDB136 Psychometric Evaluation of the Hypoglycaemia Perspectives Questionnaire in Patients With Type 2 Diabetes Mellitus Ong S H 1, Kawata A K 2, Kulich K 1, Wilson H 3, Coyne K S 2, Evripidou P 4, Koutsides P 5, Kyriakidou-Himonas M 6, Loizou T 7, Olympios G 8, Pastellas C 4, Picolos M 9, Stylianou A 8, Toufexis C 10, Therapontos C 11 1Novartis Pharma AG, Basel, Switzerland, 2Evidera, Bethesda, MD, USA, 3Evidera, Seattle, WA, USA, 4Private Practice, Limassol, Cyprus, 5Limassol General Hospital, Limassol, Cyprus, 6Private Practice, Larnaca, Cyprus, 7Iatrodiagnosis Medical Centre, Nicosia, Cyprus, 8Larnaca General Hospital, Larnaca, Cyprus, 9Alithias Endocrinology Center, Nicosia, Cyprus, 10Hippocrateon Private Hospital, Nicosia, Cyprus, 11Novartis Pharma Services Inc., Nicosia, Cyprus . . . . . . . . . . . . . . . . . . Objectives: The Hypoglycaemia Perspectives Questionnaire (HPQ) is a patientreported outcomes (PRO) instrument assessing diabetic patients’ experience and perceptions of hypoglycaemia. The aim of this study is to evaluate the factor structure and psychometric characteristics of the HPQ in type 2 diabetic patients (T2DM). Methods: HPQ was administered to adults with T2DM in a clinical sample from Cyprus and a community sample in the US from 2011 US National Health and Wellness Survey. Demographic and clinical data were collected. Participants completed the Audit of Diabetes Dependent Quality of Life (ADDQoL-19), treatment satisfaction items, and EuroQol-5 Dimensions (EQ-5D) (Cyprus only). HPQ items assess hypoglycaemia attitudes and behaviours on an 11-point numeric rating scale (NRS). Item performance and factor structure were examined and measurement properties (reliability, construct validity, known-groups validity) evaluated. Results: Cyprus (n=500) and US (n= 1,257) T2DM samples were of similar age (Cyprus 61.0±10 years, US 59.9±11 years). Cyprus had more males (67.4% vs. 54.2%) and fewer obese subjects (BMI ≥30) 45.6% vs. 67.8%) than the US. More US subjects reported hypoglycaemia events in the past seven days (27.7% vs. 16.6% with ≥ 1 event). Prescription oral diabetes medications were used by 90.3% of Cyprus participants and 83.6% of US participants, and insulin by 32.9% and 25.3%, respectively. Analyses supported three HPQ domains: Symptom Concern (six items), Compensatory Behavior (five items), and Worry (five items). Internal consistency was high for all three domains (all ≥0.75), supporting reliability. Convergent validity was supported by moderate correlations between HPQ domain scores and ADDQoL-19 total score. Patients with recent hypoglycaemia events had significantly higher HPQ scores supporting known-group validity. Conclusions: HPQ is a valid and reliable measure capturing the experience and impact of hypoglycaemia and can be useful in clinical trial and community-based settings. PDB137 The Prevalence of Hypoglycemia and Its Impact on The Quality of Life of Type 2 Diabetes Mellitus Patients in Greece (The Hypo Study) Yfantopoulos J 1, Hatzikou M 2, Rombopoulos G 2, Panitti E 2, Latsou D 3 and Kapodistrian University of Athens, Athens, Greece, 2Novartis Hellas, Metamorfosis, Greece, 3University of Peloponnese, Corinth, Greece . . . . . 1National Hypoglycemia (Hypo) is a common adverse effect of type 2 diabetes (T2D) therapy. Hypo has a negative impact on health care resources and quality of life (QoL) and can affect compliance and T2D control. Objectives: To estimate the impact of Hypo on the QoL as well as its prevalence on T2D patients in Greece. Methods: A cross-sectional epidemiological study was conducted in 6631 patients with T2DM. Hypo events with different treatment regimens, T2D control rates and QoL were assessed. Hypo episodes were defined as laboratory-confirmed (< 70 mg/dl) symptomatic events. QoL was measured using the patient-administered ADDQoL-19 questionnaire. Diabetes control was defined as Hb1Ac ≤ 7%. Results: The majority of the sample were male (55%) and overweight (59% had BMI 25-29.9). 20.4% of T2DM patients had a history of laboratory-confirmed hypoglycemia. In total, 59% had HbA1c > 7%. The mean age was 60 and the mean T2D duration was 10 years. The mean QoL score of the total sample was -3.1± 1.9, the mean score of non-hypoglycemic patients was -3.05 ± 2.0 while the respective score of hypoglycemic patients was -3.26 ±1.8, (p≤ 0.005). Similar results were observed in the group of controlled patients and uncontrolled patients, who scored -2.73±1.7 and -3.33±1.9, respectively (p≤ 0.005). Comparable results were identified in the majority of the ADDQoL-19 instruments’ dimensions. According to the logistic regression analysis the majority of the ADDQoL-19 dimensions were significantly affected by hypoglycemia, as well as by high levels of HbA1c. Conclusions: The study confirmed that the QoL of Greek T2D patients is negatively affected by hypoglycemic events and the level of the disease control. In T2D, treatment should attain good glycemic control without debilitating hypoglycemic episodes, which compromise patients’ QoL. PDB138 The Impact Of Daytime And Nocturnal Non-Severe Hypoglycaemic Events On People With Diabetes In Turkey Emral R Ankara University Medical Faculty, Ankara, Turkey . Objectives: Two surveys were conducted to assess the impact of nocturnal and daytime hypoglycaemia on the individual and on the Turkish economy. Methods: People with diabetes who had experienced a non-severe hypoglycaemic event in the 4 weeks prior to the survey were eligible for inclusion. The surveys were conducted face to face; all information, including hypoglycaemic events, was selfreported. Results: In the Turkish cohort, 95 people responded (50 for the nocturnal survey [N]; 55 for the daytime survey [D]). Mean age was 49 years/48 years (N/D), mean weight was 76.9 kg/75.2 kg (N/D), 40%/31% (N/D) of respondents were male, and 64%/80% (N/D) had type 2 diabetes. Among respondents with type 2 diabetes, 63%/48% (N/D) received insulin ± OADs and 63%/66% (N/D) received OADs ± insulin. Non-severe hypoglycaemic episodes were experienced at least once/week by 20%/35% (N/D) and at least once/month by 58%/73% (N/D) of respondents. After the non-severe nocturnal (N) /daytime (D) hypoglycaemic event, 16% and 9% (N/D) of respondents decreased their insulin dose and 16% and 20%, respectively, contacted a health care professional. On average, respondents used 1.5 or 1.6 (N/D) extra blood glucose tests the following week, and 34%/26% (N/D) of those surveyed reported a high level of fear of a hypoglycaemic event. Of respondents who worked for pay (n= 22/17), 59%/77% (N/D) reported a high/medium impact on work productivity following a hypoglycaemic event. Mean time to return to sleep following a nocturnal event was 60 minutes; 28% reported that the event had a high impact on sleep, and 54% had difficulty focusing at work the next day. Conclusions: In Turkey, nocturnal and daytime non-severe hypoglycaemic events impact quality of life, utilisation of health care resources and productivity in people with diabetes. PDB139 Health Related Quality of Life Outcomes (Hrqol) and Resource Use Associated With Type 2 Diabetes Patients Taking Steps To Lose Weight In 5 European Countries Nuhoho S 1, Vietri J 2, Isherwood G 3, Worbes-Cerezo M 4 Health, Milan, Italy, 3Kantar Health, Epsom, UK, 4Janssen-Cilag UK, High Wycombe, UK . . . . 1Janssen-Cilag A/S, Birkerød, Denmark, 2Kantar Objectives: Weight control is a cornerstone of type 2 diabetes (T2DM) management, but the impact of losing small to moderate amounts of weight relative to gaining weight on patient-reported outcomes and resource use among T2DM patients is not widely reported. This study was conducted to describe the association of weight change in T2DM patients taking steps to lose weight (TSLW) with HRQoL, work productivity and resource use in European T2DM patients. Methods: Data came from the 2013 5EU National Health and Wellness Survey, representative of adults in France, Germany, Italy, Spain and the United Kingdom. Respondents TSLW who lost ≤5kg were compared to respondents who gained weight. Outcome measures included SF-36v2, Work Productivity & Activity Impairment Questionnaire, and 6-month self-reported health care use. Comparisons used pairwise t-tests and chi-square tests for continuous and categorical variables, respectively. Results: Of 1,985 respondents taking steps to lose weight, 61% were male, 37% were employed with mean BMI 32.3 kg/ m2. Relative to respondents who gained weight (n= 354), respondents who lost ≤5kg (n= 560) had higher scores on mental component summary by 3.9 points (p<0.05), higher physical component summary by 2.3 points (p<0.05), and SF-6D health utility by 0.05 points (p< 0.05). This group reported lower work and activity impairment as well as lower health care resource utilisation relative to those who gained weight, with lower mean presenteeism (22% vs. 28%, p< 0.05), activity impairment (34% vs. 44%, p<0.05), health care provider visits (7.8 vs. 10.2 p<0.05), emergency room visits (0.2 vs. 0.6; p<0.05) and hospitalizations (0.2 vs. 0.4; p< 0.05). Conclusions: Treatments which facilitate small to moderate weight loss may offer T2DM patients better HRQoL and potentially offer savings to the health care system when compared to treatments which promote weight gain. Further research is required to explain this observation. PDB140 Racial Disparities in Type 2 Diabetes Health Care Utilization in Medicaid Adults With Developmental Disabilities Patel I 1, Erickson S 2, Bagozzi R P 1, Chang J 3, Caldwell C 4, Woolford S 4, Balkrishnan R 1 1University of Michigan, Ann Arbor, MI, USA, 2The University of Michigan, Ann Arbor, MI, USA, 3Samford University, Birmingham, AL, USA, 4University of Michigan, ANN ARBOR, MI, USA . . . . . . . . Objectives: Adults with developmental disabilities have higher prevalence of chronic disease conditions such as diabetes, obesity, high blood pressure, arthritis, CVD and chronic pain. They also have poor health care utilization and screened less for chronic disease conditions. The objective of this study was to examine the association of race, and medication adherence, and their interaction with health care utilization and health care costs in Medicaid enrollees with developmental disabilities (DD) and type A357 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 2 diabetes. Methods: This was a retrospective cohort study that identified the DD adults with type 2 diabetes from the MarketScan® Multi-State Medicaid Database. Enrollees aged 18-64 years who received new anti-diabetic medications from January 1, 2004 and December 31, 2006 were included. An index diagnosis date was assigned to each patient and adults with a continuous enrollment for at least 12 months were included. Multivariate negative binomial regression was used to measure the rate of change in type 2 diabetes related health care utilization (inpatient, outpatient and emergency department visits) in DD patients. Multivariate linear regression with logtransformation was used to determine type 2 diabetes related health care costs in DD Medicaid enrollees. Results: The study population comprised of 1529 patients. After controlling for all the covariates, compared to DD Caucasians, DD African Americans were more likely to have type diabetes related inpatient (OR= 1.71; 95% CI, 1.02-2.85) and emergency department visits (OR, 1.67; 95% CI, 1.02-2.73). After holding all the other covariates constant, among DD patients with type 2 diabetes, compared to Caucasians, African American patients had 23% significantly higher medication costs, 26% significantly higher medical costs and 21% significantly higher overall costs respectively. Conclusions: Racial disparities exist in health care utilization in DD Medicaid patients with type 2 diabetes. Access to culturally competent health care providers, providers who accept Medicaid patients and continuous care can reduce inpatient visits and emergency room visits in racial minorities. PDB141 Association of Hypoglycemia With Different Oral Antidiabetic Treatments and Its Impact on Quality of Life and Disease Control In Patients With Type 2 Diabetes Mellitus (The Hypo Study) Rombopoulos G 1, Hatzikou M 1, Panitti E 1, Latsou D 2, Yfantopoulos J 3 1Novartis Hellas, Metamorfosis, Greece, 2University of Peloponnese, Corinth, Greece, 3National and Kapodistrian University of Athens, Athens, Greece . . . . Objectives: The current prevalence of Diabetes Mellitus (DM) in Slovakia ranges in about 340000 cases. Insulin therapy is realised in the group about 44000 patients, including about 21000 men and 23000 women in the patients older than 19 years. Till now in the Czech Republic and in the Slovak Republic there was not realised the study oriented on the quality of life (QolL) in relation to young people with DM type I, treated by insulin pens or insulin pumps. Methods: 102 patients with type 1 diabetes mellitus aged from 18 to 35 attending diabetes centres and clinics in Brno and its surroundings were studied. The method of quantitative research was used to collect the data obtained by standardized questionnaire ADDQoL and DTSQs with the confirmed copyright. Results: The general value of QoL was 5.65 on the numeric scale from 0 to 10 (0 for the worst results, 10 for the best results). The positive statistical significance with “p” lesser than 0.05 was find out comparing insulin pump vs insulin pen: hobby and free time physical activities, family life, recommendation treatment by pump, and satisfaction with the form of treatment and continuing of the treatment. The evaluation of these categories has no statistical significance: too high level of glucose in the blood, too small level of glucose in the blood, the adaptation to the treatment, and the understanding to the illness. Conclusions: Statistical evaluation of the hypotheses confirmed a better quality of life in patients with insulin pump. On the contrary, the difference in perception of the quality of life between the sexes was not proved. Our research confirmed that the insulin pump is a modern way of compensation and has a positive impact on the quality of life in young patients with diabetes. DIABETES/ENDOCRINE DISORDERS – Health Care Use & Policy Studies . Hypoglycemia (Hypo) is a common adverse effect of type 2 diabetes (T2D) therapy. Oral antidiabetic (OAD) agents differ in their propensity to cause Hypo. Hypo has a negative impact on health care resources and quality of life (QoL) and can affect compliance and T2D control. Objectives: To estimate the impact of different OADs on Hypo prevalence and on the QoL of T2D patients in Greece. Methods: A cross-sectional epidemiological study was conducted in 6631 patients with T2D. Hypo events with different treatment regimens, T2D control rates and QoL were assessed. Hypo episodes were defined as laboratory-confirmed (< 70 mg/dl) symptomatic events. QoL was measured using the patient-administered ADDQoL-19 questionnaire. Diabetes control was defined as Hb1Ac ≤ 7%. Results: Mean age was 60 and mean T2D duration 10 years. 59% of patients had HbA1c > 7%; 20.4% had Hypo. Most patients (82%) were on combination therapies, while 14.5% were on oral monotherapy and the remaining were on insulin or GLP-1 analog treatment. The overall QoL in the three OAD monotherapy groups [biguanides, DPP4is and sulfonylureas (SUs)] were not statistically significant between groups, regardless of Hypo status. However, patients treated with biguanide or DPP4i monotherapy had lower rates of Hypo vs. patients on SU monotherapy (p≤ 0.001) and QoL in Hypo patients treated with biguanides or DPP-4is was better compared to SUs (p< 0.001). Similarly, more patients on biguanides and DPP4i monotherapies achieved HbA1c ≤ 7% than on SU (p≤ 0.001). Patients receiving combinations of biguanides with DPP4is vs. biguanides with SUs had significantly less Hypo (p≤ 0.001), better diabetes control (respectively; p≤ 0.001) and better QoL (p≤ 0.05). Conclusions: Biguanides, DPP4is and their combinations are associated with lower rates of Hypo, better QoL and T2D control, compared with SUs alone or in combination with biguanides. In T2D, treatment should attain good glycemic control without debilitating hypoglycemic episodes, which compromise patients’ QoL. PDB144 Is the rule of halves applicable in Diabetes Type 2? Evidence from Greece Tsiantou V , Karampli E , Zavras D , Athanasakis K , Pavi E , Kyriopoulos J National School of Public Health, Athens, Greece . . . . . . Objectives: The “rule of halves” (ROH) applies in most common chronic disorders and suggests that half of the patients are undiagnosed, half of those diagnosed are not treated, and half of those treated are not controlled. The aim of the present study was to examine whether the ROH is applicable in the case of Diabetes type 2 in Greece. Methods: Physicians who monitor patients with type 2 diabetes were invited to participate and complete the online version of a purpose-made questionnaire during a specified time period. An invitation was sent to the national associations of general practitioners (GPs), diabetologists and endocrinologists. The study took place between April and May 2014. Fully completed questionnaires were included in the analysis. Data analysis was conducted using STATA 9. Results: 176 physicians completed the questionnaire of whom 68.7% were men. Mean age was 47.8 years. The majority were internists and GPs (38.3% and 32.7%, respectively). Physicians estimated that in Greece 62% of those diagnosed with diabetes type 2 receive some kind of treatment (lifestyle or pharmacotherapy). 59.3% of those diagnosed will present at least one complication because of the disease. Less than half (43.3%) of those treated are controlled and achieve treatment targets (HbA1c≤7%). Of those not controlled, 73.3% will present at least one complication because of diabetes type 2. Conclusions: According to our findings, in Greece more than half of the diagnosed patients with diabetes type 2 receive treatment in a regular basis. However, less than half achieve treatment targets and manage to stay controlled. Emphasis should be given in selfmanagement, patient education and policies raising awareness for the early detection and management of diabetes type 2 in order to increase controlled patients and decrease the incidence of complications due to the disease. PDB142 Evaluation of The Impact of Pharmaceutical Cognitive Service on Quality of Life in Diabetic Patients PDB145 Physicians Estimation Regarding The Impact Of Recession on Patient Adherence To Treatment In Diabetes Type 2 In Greece Yordanova S 1, Petrova G 1, Naseva E 2, Manova M 1 1Medical University Sofia, Faculty of Pharmacy, Sofia, Bulgaria, 2Medical University Sofia, Faculty of Public Health, Sofia, Bulgaria Tsiantou V , Zavras D , Karampli E , Athanasakis K , Pavi E , Kyriopoulos J National School of Public Health, Athens, Greece . . . . Objectives: Comparison of the changes in QoL of diabetic patients after pharmaceutical cognitive service with SF-6D. Methods: Tree step methodology was created and applied towards the patients with type 1 and type 2 diabetes. On first the QoL of 146 diabetic patients was assessed with the Bulgarian version of SF-6D. Then educational program for the diabetic patients was developed and applied during the day of their visits in community pharmacy from June to September 2013. At the end the changes in the QoL after the pharmaceutical cognitive service was assessed in the group of 24 patients. Results: The mean age of patients included in the first step was 57.39 years (SD 17.087; 95%CI 54.60-60.19). 76% of them were with type 2 diabetes and 73% suffers from diabetes complications. The mean SF-6D score was 0.629 (SD 0.289; 95%CI 0.58-0.68). After the pharmacy program the type 2 diabetic patients showed better quality of life index 0.7133 that was 10% higher than the initial score 0.6089. In the patients with type 1 diabetes the improvement of QoL score is with 3 %. The patients were with mean SF-6D scores: 0.667 on the first visit in the pharmacy and 0.690 after the educational service. Conclusions: The study suggests the SF-6D is a valuable tool for QoL assessment of diabetes patients in Bulgaria. It also shows that SF-6D could be used for pharmacy program evaluation in diabetes patients. Regarding the pharmaceutical cognitive service in diabetic patients is evident that they have a potential to improve the QoL in patients with diabetes mellitus and their impact could be measured with SF-6D. PDB143 Quality Of Life in Teenagers and Young Adolescent Patients With Diabetes Mellitus Type I With Insulin Pens Or Insulin Pumps In The Czech Republic Mastiliakova D 1, Cikova Z 1, Gerlichova K 1, Matisakova I 2, Bielik J 1 1Trencin University, Trencin, Slovak Republic, 2Trencin University of Alexander Dubcek, Trencin, Slovak Republic . . . . . . . . . . . Objectives: Adherence to treatment plan and lifestyle advice is crucial for good health outcomes in diabetic patients. Given that Greece in the last years faces an unprecedented economic crisis which affects all sectors of economic and social life, the aim of the present study was to examine whether recession, austerity measures and implemented health policies had an impact on patient adherence to treatment in Diabetes type 2 (dt2) in Greece. Methods: A questionnaire was developed for the purpose of the study. Only Physicians who monitor patients with dt2 (general practitioners (GPs), diabetologists and endocrinologists) were invited to participate in the study and complete the online version of a purpose-made questionnaire during a specified time period. Physicians were recruited through their national associations. The study took place between April and May 2014. Results: Of the 176 physicians completing the questionnaire, 68.7% were men. Mean age was 47.8 years. The majority were internists (38.3%) and GPs (32,7%). Physicians estimated that during the previous year 22.9% of their patients had to quit or modify their treatment plan due to economic reasons. A somewhat higher percentage (26.9%) was found regarding the modification of the dietary habits during the previous year due to economic reasons which led to negative health outcomes. According to the participating physicians the main reasons for which patients modified or quitted their treatment were the higher copayments for pharmaceuticals, loss of insurance coverage and barriers of access to a physician in order to prescribe their medicines. Conclusions: Recession and recent health policies had a negative impact on patients’ adherence to treatment leading to poor adherence almost a quarter of patients with dt2, based on the experience of physicians. Equally critical is the fact that patients were forced to modify their dietary habits. PDB146 Effect of A Pharmacy-Based Diabetes Management Program on Glycemic Control in an Inpatient General Medicine Population Efird L 1, Shermock K M 2 . . . A358 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1The Johns Hopkins Hospital, Baltimore, MD, USA, 2The Johns Hopkins Medical Institutions, Baltimore, MD, USA Objectives: A pharmacy-based inpatient diabetes management program was evaluated to determine if improved glycemic control could be achieved in a general medicine patient population. Methods: A retrospective chart review of 151 patients with blood glucose (BG) values outside the range 70-180 mg/dL was conducted. Observations for the baseline group (n=84) were derived from July 2010 and for the intervention group (n= 67) in October 2010. The odds of poor glycemic control for patients in the intervention versus baseline groups were assessed by multivariate generalized estimating equations. These methods were also used to assess patient characteristics associated with poor glycemic control. Results: Across all patients, no evidence was observed indicating the pharmacy program decreased the proportion of days spent out of the targeted blood glucose range [70-180 mg/dL: OR 0.91 (95% CI: 0.83 – 1.02); 70-250 mg/dL: OR 1.03 (95% CI: 0.88 – 1.24)]. However, the subgroup of patients whose admission blood glucose was less than 200 mg/dL (55% of intervention group) experienced a significant reduction in days out of range for both ranges [70-180 mg/dL (OR: 0.72, 95% CI: 0.61– 0.88) and 70-250 mg/dL (OR: 0.5, 95% CI: 0.33 – 0.71)]. No improvement in glycemic control was observed in patients with an admission BG 200 mg/dL or greater. These patients had more disease- and social-related factors associated with poor glycemic control. Conclusions: A subpopulation, patients whose admission glucose was less than 200 mg/dL, experienced improvement in glycemic control in the pharmacy-based program. The remaining patients were generally more complicated from a disease-state and social perspective and experienced no improvement. These patients may require a more intense, multi-disciplinary approach that is better matched to the constellation of factors responsible for their condition. PDB147 Compliance to Hemoglobin A1c Testing Recommendations Following Initial Diabetes Diagnosis Palmer L A 1, Hansen L G 2 1Truven Health Analytics, Bethesda, MD, USA, 2Truven Health Analytics, Northwood, NH, USA . . . . Objectives: The hemoglobin A1c test (HbA1c) is the favored measure of glycemic control for patients with diabetes. Compliance to recommended testing continues to be a challenge. The current analysis evaluates how compliance to HbA1c testing varies based on initial HbA1c results. Methods: Newly diagnosed patients with Type II diabetes were identified using the Truven Health MarketScan Lab Database (1/1/2010-10/31/2013). Continuous eligibility for the 6 months prior and 12 months following diagnosis were required for study inclusion. Patient cohorts were created based on first HbA1c test value (< 7.0% [controlled], ≥ 7.0% [uncontrolled]). Presence of a subsequent HbA1c test, including test result and time to test, were evaluated for these cohorts. Results: A total of 133,011 patients met the study inclusion criteria; approximately 40% had evidence of an initial HbA1c test (n= 33,616 with HbA1c < 7.0%; n= 19,033 with HbA1c ≥ 7.0%). Approximately 64% (n= 21,497) of controlled patients had a subsequent HbA1c test (86% with HbA1c < 7.0%, 14% with HbA1c ≥ 7.0%); 91% (n= 17,344) of uncontrolled patients had a subsequent HbA1c test (36% with HbA1c < 7.0%, 64% with HbA1c ≥ 7.0%). Mean times (in days [d]) to subsequent HbA1c test (and result) were as follows: Among initially controlled patients: 231d (controlled) and 370d (uncontrolled); among initially uncontrolled patients: 238d (controlled) and 212d (uncontrolled). Mean time to HbA1c test was significantly longer for controlled patients with HbA1c ≥ 7.0% for subsequent test relative to controlled patients with HbA1c < 7.0% and to uncontrolled patients with HbA1c ≥ 7.0% for subsequent tests. Conclusions: Compliance to recommended timing for HbA1c testing is suboptimal in the majority of patients regardless of initial glycemic control. Importance of regular HbA1c evaluation should continue to be part of patient education – particular for patients who may initial appear to have favorable glycemic control. PDB148 Cost-Effectiveness of Theintroduction of A National Adherence Program for Type 2 Diabetes In Hungary Zsólyom A 1, Nagy B 1, Nagyjánosi L 2, Dessewffy Z 3, Steiner T 4, Kaló Z 5, Vokó Z 1 1Eötvös Loránd University, Budapest, Hungary, 2University of Debrecen, Debrecen, Hungary, 3Novartis Hungary, Budapest, Hungary, 4St. John’s Hospital and North-Buda United Institutions, Budapest, Hungary, 5Eötvös Loránd University (ELTE), Budapest, Hungary . . . . . . PDB149 Which Newly-Diagnosed Diabetics Should Receive Dietary Counseling Services? Estimating Individualized Treatment Allocations that Optimize cost-Effectiveness in Real-World Data Signorovitch J 1, Betts K A 1, Meng X 1, Zhuo Y 2, Wu E Q 1, Shi L 3 1Analysis Group, Inc., Boston, MA, USA, 2MIT, Cambridge, MA, USA, 3Tulane University, New Orleans, LA, USA . . . . . . . . Objectives: All people with type 2 diabetes should receive dietary advice. Some patients may benefit from additional dietary counseling. This study used recently developed statistical methods to estimate the efficiency frontier for individualized allocation of dietary counseling services. I. e., for each level of total health care expenditure, we estimate the individualized allocation of services that maximizes clinical benefit in the population. Methods: People newly diagnosed with type 2 diabetes were identified retrospectively from electronic health records and classified as receiving vs. not receiving dietary counseling. An individualized effectiveness score for achievement of HbA1c < 7% with vs. without counseling was estimated using multivariable logistic regression. Demographics, vitals, comorbidities and lab values served as candidate predictors. Prediction models were validated in a held-out sample. The efficiency frontier was estimated as the convex hull of the set of HbA1c control rates and total costs achievable by allocation of dietary counseling based on the effectiveness score. Results: Among 11,819 patients newly diagnosed with type 2 diabetes, 22% received dietary counseling and 74% achieved HbA1c control. Greater HbA1c, body mass index and age at the time of diagnosis were associated with greater effectiveness of dietary counseling. Allocation of all newly diagnosed diabetics to dietary counseling was estimated to increase the rate of HbA1c control to 80% at a cost of 56 USD per patient vs. current practice. An efficient allocation rule (counseling only the 55% of patients predicted to benefit most) achieved the same 80% control rate with an incremental cost of 6 USD per patient. Conclusions: Retrospective analysis of real-world data identified opportunities to improve diabetes outcomes vs. current practice with minimal expense through individualized allocation of dietary counseling. This result warrants validation in separate data. The analytical methods warrant broader use to investigate efficient allocation of treatments. PDB150 Chronic Care Management Gajdos O , Jurickova I Czech Technical University in Prague, Kladno, Czech Republic . . Objectives: Chronic illness is a prolonging health defect, which is causing a big use of financial means for its medical care. These means are not being coordinated and used effectively. The aim of this study was to suggest a management care system of the chronically ill in the Czech Republic, which should increase the quality of medical care and decrease the costs of it. Methods: There was chosen an appropriate diagnosis and concept for the management care, which were based on analysis of the current state. The type 2 diabetes mellitus was chosen as an appropriate chronic illness and for the management care concept was chosen Patient-Centered Medical Home.. The randomized selection of 100 patients was made in ordinary diabetes ambulance. The cost of illness was counted from the direct costs from the perspective of the society, of the payer and of the patient. The cost effectiveness analysis, which was comparing a standard treatment and chosen concept, was based on randomized selection, studies of Patient-Centered Medical Home and recommended standards of professional society. There were also used methods of value engineering especially Saaty matrix and multi-criteria decision making, mainly TOPSIS method for setting the scales of criteria and effect. Results: The average costs of one patient are from the perspective of the society 29 531 CZK, the payer 20 976 CZK and of the patient 9 196 CZK. The Patient–Centered Medical Home has taken the first place in comparison with a standard treatment, which was based on the cost of effectiveness analysis. The payer will obtain a 25.7x10-5 of the effect for Patient-Centered Medical Home according a spent monetary unit. Conclusions: The costs of the chosen concept can be more effective. The concept would provide greater prevention, quality and coordinated care and can be used for other chronic diseases. . Objectives: The Syreon health economic model was developed to predict long term effects of screening, treatment and control of type 2 diabetes, by taking into account baseline patient characteristics, history of complications, changes in physiological parameters, diabetes treatment and management strategies and screening programs. The aim of this analysis was to assess the cost-effectiveness of introducing a national public health program in Hungary to improve diabetes patient’s adherence in comparison to not introducing the program. Methods: According to the guideline of the Hungarian National Diabetes Association, the target HbA1c level is below 7%, except for special cases, where it is 8% or less. Without an organized patient education program 45% of the patients with known diabetes have higher than the target HbA1c level. In the studied scenario, the education program improves the patient adherence by 30% and increases the proportion of diabetic patients achieving the target HbA1c level to 72%. Patients reaching the target HbA1c level fully enjoy the benefits of efficient treatment. Non-adherent patients have higher HbA1c levels and face higher risk for diabetes-related complications, e. g. stroke, neuropathy or retinopathy. Results: The Syreon diabetes model is capable of analyzing the consequences of introducing the adherence program for patients with diagnosed diabetes. The results of the cost-effectiveness analysis were sensitive to the starting age of the target population and the effectiveness of the training program. Conclusions: Organized patient education program was predicted to be cost-effective compared with no program in Hungary. The education program contributes to better patient adherence resulting in better health and less disease related complications. PDB151 A Comparative Analysis on The Reimbursement Status Of Sensor Augmented Pump Therapy in Turkey and Other Selected Countries Seyhun O , Erdol S , Can H , Erdogan E Medtronic, Inc., Istanbul, Turkey . . . . Objectives: Sensor augmented therapy (SAP) with automated insulin suspension is the most advanced technology for the treatment of severe and moderate hypoglisemia in patients with type 1 diabetes mellitus. In order to sustain a better uptake of sensor augmented therapy for patients, it is crucial for this technology to be included in the reimbursement scheme of Turkey and in the other countries. Thus, we aimed to study and analyze reimbursement status of this technology in Turkey and across other selected countries of Western Asia, North America and Western Pacific. Methods: Mainly official web resources such as health authority web pages, direct contact with authority responsibles and published articles on SAP are utilized. Results: Most of the countries examined have either reimbursement or limited status of SAP. European countries such as Ireland, The Netherlands, Sweden, Estonia, Czech Republic; Israel in Western Asia, Japan in Western Pacific and USA are the ones where this technology is reimbursed mainly for patients with Type 1 Diabetes. Within these selected countries, Turkey has a position of having reasonably well defined reimbursement status for SAP despite insufficient number of sensors reimbursed - 2 sensors instead of 5 sensors a month; which potentially causes an incomplete therapy for the patients. Conclusions: Despite the operational hurdles and insufficient number of sensors reimbursed per patient in Turkey, the country still stands as a successful example of reimbursement practice, by making this technology available for the indicated patients. A359 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PDB152 Prescription of Rosiglitazone and Pioglitazone Folowing Safety Warnings: A Comparative Analysis of Trends in Dispensing Patterns in Denmark and Germany From 2007 To 2013 Hostenkamp G 1, Fischer K E 2 of Health Economics Research (COHERE), Odense M, Denmark, 2University of Hamburg, Hamburg, Germany . . . 1Center Objectives: Investigate how official drug safety warnings for thiazolidinediones (TZDs, rosiglitazone and pioglitazone) affected prescription trends for TZDs in Denmark and Germany and to compare the prescription patterns to data published for Australia, UK and the US. Methods: Identified the timing and content of all drug warnings for TZDs using the FDA’s and EMA’s databases. Using pharmaceutical claims data on all diabetes patients in Denmark and a large public health insurance fund in Germany, plotted the number of defined daily dosages (DDDs) per month for rosiglitazone and pioglitazone normalized per 1000 insurees for both countries from January 2007 -December 2013. Results: In Denmark rosiglitazone monotherapy prescriptions remained almost constant after publication of the safety warnings in May 2007 until drug suspension in September 2010, but declined fairly fast in Germany. Prescription of rosiglitazone combined with metformin declined by 1/3 compared to their May 2007 levels in both countries. Overall market penetration of rosiglitazone prior to the drug warnings was ~1 DDD per day per 1000 insurees in Germany and 0.36 in Denmark, and therefore much lower than in Australia where rosiglitazone peaked at about 2 DDDs per 1000 insurees. The FDA’s safety warning for increased risk of heart failure for all TZDs in August 2007 did not affect prescription growth rates for pioglitazone in Germany or Denmark. However, pioglitazone prescriptions declined after the FDA’s and EMA’s safety warning for possibly increased risk of bladder cancer for pioglitazone patients in June and July 2011 in both countries. Conclusions: TZD prescriptions in Germany and Denmark did not decline as radically after the publication of safety warnings as in other countries. Differences in reaction speed may reflect different levels of market penetration prior to the drug safety warnings and differences in the intensity of the safety warning issued by the drug administration authority. PDB154 Assessing The Medication Use Among Diabetic Patients With Comorbid Diseases Azmi S 1, Mustapha F I 2, Abdul Aziz S H 1 Burhani Consulting, Petaling Jaya, Malaysia, 2Disease Control Division, Putrajaya, Malaysia . . . . . 1Azmi Objectives: Type 2 Diabetes is on the rise in many Asian countries with growing economic prosperity. Treatment patterns of diabetic patients in South East Asia have rarely been studied using administrative data. Our objective was to understand medication use patterns among Malaysian patients with diabetes with comorbidities in the primary health care setting. This may be useful for understanding treatment strategies in subgroups of patients with diabetes who tend to pose more complex challenges in clinical decision-making. Methods: Data was obtained from diabetic patients who were audited in the NDR in 2012. Descriptive analysis was used to describe medication usage for patients with various comorbidities. Medications of interest were antiplatelet medications, lipid-lowering agents, anti-hypertensive and various classes of anti-diabetic medications including insulin. Results: A total of 130,270 patients were identified with comorbidities. Mean age of patients was 59.8 years old and 60.2% were female. A majority of patients were hypertensive (70.3%), more than half were dyslipidemic (55.1%). Other comorbidities such as retinopathy, ischemic heart disease (IHD), cerebrovascular disease, nephropathy, diabetic foot ulcer, amputation, hypertension and dyslipidemia were recorded among less than 10% of patients. Among non-diabetes medications, statins was the most commonly used ranging from 64.6% to 86.3% for various comorbidities. This was followed by ACE-inhibitors and aspirin. Among anti-diabetic medications, metformin was the most commonly prescribed followed by sulfonylureas and insulins. Conclusions: The registry data was able to shed light on patterns of medication usage among diabetic patients with comorbidities. Perhaps in the future, this data can be used to evaluate long-term outcomes among patients on various treatment plans and finally to inform more evidence-based clinical management strategies relating to medication use. PDB155 Six Years Of Incretin-Based Agents in Diabetes 2nd Type Treatment in the Czech Republic: The Utilization And Expenditure Fuksa L 1, Vytrisalova M 2 1General Health Insurance Company of the Czech Republic, Praha, Czech Republic, 2Charles University, Faculty of Pharmacy, Hradec Kralove, Czech Republic . . Objectives: The objective was to analyze evolving consumption, public costs as well as prescription practice pattern of the newer anti-diabetic drugs interfering with incretin system, i. e. DPP-IV inhibitors (“gliptins”) and GLP-1 mimetics, and compare it to other drugs in the anti-diabetic portfolio. Methods: The pharmacy claims-based database of the General Health Insurance Company of the Czech Republic (VZP CR) was used as the data source. An insured person with a recorded prescription for any of the drugs of interest (ATC3 classification A10B: blood glucose-lowering drugs excl. insulins) between 2008 and 2013 was defined as a patient. Information on pricing and reimbursement were obtained from the files of administrative proceedings at the State Institute for Drug Control. Results: Since the introduction of sitagliptin in autumn 2008 its utilization rose to 2.1 DDD/ TID in 2013, which makes it the most prescribed incretin-based drug. Overall consumption of all incretin-based therapies was 5.4 DDD/TID in 2013 and the patient number represents 9 % of all patients taking any anti-diabetic drug of the ATC A10B group (except for insulins). However, in terms of public costs incretin-based agents represent already 55% of all the blood glucose-lowering therapy of the evaluated A10B group (insulines excluded). In comparison, the utilization of the 1st line drug metformin between 2008 and 2013 has been annually rising by 5 % to 10 %, while in other groups, namely sulfonylureas and thiazolidinediones, it has been declining each year on average by 2 % and 15 %, respectively. Conclusions: The uptake of incretin-based drugs since 2008 has been rapid and their utilization keeps rising, apparently at the expense of other drugs, namely older sulfonylurea derivatives and thiazolidinediones. Consequently the overall public expenditure on anti-diabetic therapy has been increasing. PDB156 Non-Adherence And Non-Persistence Related To Glp-1 Therapy In Patients With Diabetes Mellitus Type 2 (T2dm): Analysis of A Large German Claims-Based Dataset And Comparison To Oral AntiDiabetics Wilke T 1, Groth A 1, Berg B 1, Sikirica M 2, Martin A A 3, Fuchs A 4, Maywald U 5 1IPAM - Institute for Pharmacoeconomics and Medication Logistics, Wismar, Germany, 2GlaxoSmithKline, King of Prussia, PA, USA, 3GlaxoSmithKline, Uxbridge, UK, 4AOK Plus, Dresden, Germany, 5AOK Plus - Die Gesundheitskasse für Sachsen und Thüringen, Dresden, Germany . . . . . . . . Objectives: This study describes the extent of non-adherence (NA) and non-persistence (NP) with Glucagon-like peptide 1 agonists (GLP-1) and oral anti-diabetic (OAD) therapy in T2DM patients in clinical practice in Germany. Methods: Claims data from a German sickness fund (AOK Plus) from 2010-2012 were used to identify patients with ≥ 1 diagnosis for T2DM. NA/NP were measured for patients initiating new T2DM medication (no prescription of respective medication in prior 6 months) over a minimum of 12 months. A 100% adherence to therapy was assumed equivalent to the drug-specific DDD (defined daily dosage). NP was defined as a medication gap ≥ 90 days. Adherence was calculated via medication possession ratios (MPR), NA was defined as MPR< 80%. Descriptive analyses with percentage of patients affected by NP and/or NA were conducted. Results: The NP analysis, included 2,490/42,891 T2DM patients initiating GLP-1s/OADs. Among GLP-1 patients, 408/1,727/135 received twice-daily (BID; exenatide)/once-daily (OD; liraglutide)/onceweekly (OW; exenatide) therapies. Average age for GLP-1/OAD users was 56 (SD 10.8) /67 (SD 12.6) years. After 12 months, the percentage of patients with NP was 36.83% (all GLP-1s), 42.65% (BID), 37.64% (OD), 27.41% (OW), and 55.87% (OAD). The NA analysis included 2,154/34,128 T2DM patients initiating GLP-1/OAD therapy with ≥ 1 follow-up prescription. Average age was 56 (SD 10.6) /66 (SD 12.2) years. Average MPR was 88.2% (all GLP-1s), 87.8% (BID), 88.2% (OD), 95.0% (OW), and 63.2% (OAD). Percentage of patients affected by NA was 21.96% (all GLP-1s), 23.98% (BID), 22.00% (OD), 5.74% (OW), and 65.82% (OAD). Conclusions: In this German dataset, overall patients with T2DM had low rates of adherence and persistence despite the chronic nature of the disease and clinical sequelae. Higher adherence and persistence rates were observed with GLP-1s than with OAD medications and also with less frequently dosed GLP-1s. PDB157 Type 2 Diabetes Treatment Patterns Across Europe Heintjes E 1, Overbeek J A 1, Blin P 2, Hall G C 3, Lapi F 4, Prieto Alhambra D 5, Bezemer I D 1 Institute for Drug Outcomes Research, Utrecht, The Netherlands, 2INSERM CIC Bordeaux CIC1401, Univ. Bordeaux, Bordeaux, France, 3Grimsdyke House, London, UK, 4Genomedics SRL, Firenze, Italy, 5IDIAP Jordi Gol, Barcelona, Spain . . . . . . . . . . 1PHARMO Objectives: To describe the sequential treatment classes of type 2 diabetes (T2DM) patients initiating antidiabetic drug therapy in the Netherlands, UK (UK), Spain, Italy and France. To describe treatment scaling (intensification or de-intensification) in the overall T2DM population. Methods: Antidiabetic drug use during a 5-year study period (2007-2011/2008-2012) (n= 639,088) was obtained from electronic health care databases. A standardized analytic tool performed treatment pattern analyses in each database for the overall population and those initiating treatment. Oral monotherapy was defined as first line, oral dual therapy as second line, multiple oral treatments or oral in combination with an injectable as third line and injectables only as fourth line therapy. Results: Newly treated patients represented 33-42% of the overall T2DM population. Metformin monotherapy was the most common initial therapy (65%-87%). Around 35% (Netherlands, Italy, UK) to 45% (Spain, France) switched treatment within the study period. The first switch was most often to metformin plus a sulfonylurea (SU) in the Netherlands (47%), UK (45%), Spain (22%), Italy (17%), but to DPP4 inhibitors in France (15%). DPP4 inhibitor use increased during the study period (France 0% to 27%, UK and Spain < 1% to 9%) but remained limited in the Netherlands (4%) and Italy (2%). In the total cohort, first line treatment was most prevalent over all the years in all countries (around 50%). Intensification was the most common switch. Switching patients mostly stepped up or down one line at a time, but larger steps were also observed. Fourth line therapy was uncommon in France (1-2%) but accounted for about 10% in the other countries. Conclusions: SU remained the most common add-on treatment to metformin in most European countries while DPP4 inhibitor use was common in France and increased in other countries. Most T2DM patients are treated with oral monotherapy. PDB158 Exploring The Determinants of Endocrinologist Visits In France Andrade L F 1, Rapp T 2, Sevilla-Dedieu C 1 Foundation for Public Health, Paris, France, 2University of Paris Descartes, Paris, France . . . . 1MGEN Objectives: Recent studies reported that seeing an endocrinologist is associated with better compliance with guidelines on diabetes management. This study aims at investigating the socioeconomic determinants of endocrinologist visits among a French population of diabetics. Methods: We used claim data collected for 66,508 affiliates of a French social security provider (MGEN). Patients were aged over 18 and treated for diabetes (types I and II). We ran logistic regressions to explore which factors were associated with the probability of consulting an endocrinologist. We controlled for patients’ socioeconomic characteristics (sex, age, income, marital status, activity status), type of treatment, medical care and health status. We also controlled for macro-level variables, such as the cost for a visit, the distance to the office and the medical density. Results: Women, married patients, and those working are more likely to consult an endocrinologist (p< 0.001). Age has a non-linear A360 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 association with the probability of seeing an endocrinologist: rates of specialist visits are the largest under the age of 25 (p< 0.05). In addition, people treated with insulin and those with another endocrine disorder are more likely to see a specialist (p< 0.001). In contrast, the newly diagnosed and patients followed by a general practitioner (GP) present lower probabilities of specialist visits (p< 0.001). Moreover, the probability of specialist visit rises when the density of endocrinologists is large in the patients’ neighbourhood, while it decreases with the distance (in kilometres) from the patients’ house to the endocrinologist’s office. Finally, the chances of seeing a specialist increase with income and decrease as the cost for a visit rises. Conclusions: Our results are consistent with previous literature showing evidence of the existence of a substitution effect between GPs and specialists in diabetes care. We show that financial barriers exist even in a population of patients receiving national health insurance coverage. PDB159 Payer’s Perceptions of Glucagon Kits Aimed at Reducing Administration Complexity During Severe Hypoglycemic Events Leinwand B , Hughes K E , Bartelme A Avalere Health, Washington, DC, DC, USA . . . . Objectives: Severe hypoglycemic events (SHE) contribute to health care utilization for insulin-dependent diabetics, and may drive expenditures by health care payers. Rescue treatment for SHE includes glucagon kits, which are administered by caregivers and difficult to use. Consequently, there are high rates of unsuccessful administration, leading to increased health care service use. A new kit under development aims to ease the burden of administration by reducing the complexity. The new device aims to increase successful use and lower rates of health care service utilization. This study aimed to understand payers’ receptivity to this value proposition. Methods: Seven medical directors from US payers were interviewed, representing commercial, Medicare, Medicaid and other covered lives. The interview was designed to understand plans’ focus on diabetes and hypoglycemia, coverage/reimbursement of current kits, and impressions of characteristics of the new kit. Results: The prevention of SHE is not a concern for insurers; the risk of SHE is largely a type 1 problem, representing a small subset of their overall diabetic population. While payers are confident in current kits’ effectiveness, they are not focused on managing access to such a low-volume treatment. Current kits are covered by insurers without restrictions, and contracting is not prevalent due to low volume. Payers believe the new kit will be an improvement over current kits, but do not believe it will reduce the cost of SHE. Conclusions: Payers feel that the new kit will be an improvement over currently available kits, but are skeptical that the ease of use will translate into lower health care service use. Additionally, SHE is not a major cost driver among their diabetic population; therefore, SHE is not a highpriority condition. Consequently, for optimal market access and reimbursement, the new kit must demonstrate significant reductions in high-cost health care utilization (e.g., ED, hospitalizations) to justify a premium price without coverage restrictions. PDB160 An Observational Cohort Study of Diabetes-Associated Secondary Health Care Utilisation in Patients With Type 2 Diabetes Prescribed Dual Combination Therapy With Oral Anti-Hyperglycaemic Agents in the UK Das R 1, Langerman H 1, D’Oca K 2, Strongman H 3 1Merck Sharp & Dohme, Hoddesdon, UK, 2Merck Sharp and Dohme Ltd, Hoddesdon, UK, 3CPRD, London, UK . . . . Objectives: The primary objective was to assess the impact of using ‘metformin plus sulphonylureas’ (Met+SU) in comparison with ‘metformin plus other oral anti-hyperglycaemic agents’ (Met+OHA) in patients with Type 2 Diabetes (T2D) on diabetes-associated secondary health care utilisation in the UK. The secondary objectives included investigating individual components of the primary objective. Methods: This retrospective cohort study used data from the Clinical Practice Research Datalink (CPRD) linked to Hospital Episodes Statistics. Adults (≥ 40 years) with T2D initiated on dual therapy with Met+SU or Met+OHA following metformin monotherapy were identified during the period April 2003–March 2012 and comprised the two study cohorts. Propensity scores were estimated and Met+SU patients caliper matched to Met+OHA patients to balance the covariates (including HbA1C and duration of diabetes at baseline). Diabetes-associated secondary health care utilisation (inpatient admissions and outpatient visits) were measured from > 6 months post-initiation of dual therapy until treatment change or end of follow-up. Outcomes were calculated as rate ratios (RR), adjusted for over dispersion using negative binomial regression and propensity score for covariates. Results: 14,416 patients in total were identified and in the directly matched analysis, 1,704 patients were included in each cohort. For the primary objective, the Met+SU cohort had a numerically higher rate of diabetes-related secondary health care utilisation than Met+OHAs (adjusted RR 1.12, 95% confidence interval [95%CI]: 0.97-1.29). For the secondary endpoints examining individual components, the adjusted RR for Met+SU cohort for inpatient admissions was 1.34 (95%CI 0.92–1.96) and 1.10 (95%CI 0.95-1.28) for outpatient visits. Macrovascular complications, accounting for 68.7% of inpatient admissions, occurred at a significantly higher rate in the Met+SU cohort than Met+OHA (adjusted RR 1.77, 95%CI: 1.15-2.71). Conclusions: Glucose-lowering combination therapy with metformin plus sulphonylurea is associated with a directionally higher rate of secondary health care utilisation than metformin plus other oral anti-hyperglycaemic agents. PDB161 Factors Associated With Weight Gain and Hypoglycaemia and The Impact Upon Hospitalisation in Type 2 Diabetes Patients Managed With Metformin Plus Sulphonylurea McEwan P 1, Gordon J 1, Evans M 2, Puelles J 3 Economics and Outcomes Research Ltd, Cardiff, UK, 2University Hospital Llandough, Cardiff, UK, 3Takeda UK Ltd, London, UK . 1Health . . . Objectives: The relationship between resource utilisation and patient phenotype in type 2 diabetes (T2D) is not well characterised. This study aimed to assess factors associated with weight gain (WG) and the occurrence of hypoglycaemia in T2D patients managed with metformin plus sulfonylurea (M+S) and any associated impact upon hospital resource utilisation. Methods: The study was a retrospective cohort study using the UK Clinical Practice Research Datalink (CPRD) and the Hospital Episode Statistics (HES) database. The association between phenotypic factors at baseline (therapy escalation from metformin to M+S) with WG (> 2kg weight change over 12 months) and presence of any recorded hypoglycaemia over 12 months following therapy escalation was assessed using logistic regression. Hospitalisation associated with increasing body mass index (BMI) and hypoglycaemia was also assessed. Results: A total of 11,071 patients met the study inclusion/ exclusion criteria. WG was observed in 28.35% of patients and was significantly associated with baseline age (OR= 0.99), female gender (OR= 0.87), baseline weight (OR= 1.003) and HbA1c (OR= 1.06). Hypoglycaemia occurred in 1.3% of patients and was significantly associated with duration of diabetes (OR= 1.04), baseline HbA1c (OR=0.86) and prior complications status (OR=1.92). Hospitalisation occurred in 10% of patients and was significantly associated with BMI (OR= 1.02) but not hypoglycaemia. The mean number of hospital admissions over the follow-up period was 1.7, 1.8, 1.9 and 3.1 in those with BMIs at the time of admission in the normal, overweight, obese and morbidly obese categories respectively. Conclusions: This real-world observational analysis suggests there are identifiable phenotypic characteristics predictive of WG and hypoglycaemia. This study also shows a general relationship between increasing BMI and hospitalisation that may not be adequately captured in widely used vascular risk equations such as UKPDS in which BMI has minimal influence on risk. Consequently, the value of diabetes management strategies that minimise WG may be underestimated. PDB162 The Use of Real World Data in the Decision-Making Process: An Example Using Blood Glucose Test Strip Use In Patients With Type 2 Diabetes In Ireland O’Leary A 1, Usher C 2, Barlow J 1, Gallagher P 1 College of Surgeons in Ireland, Dublin, Ireland, 2National Centre for Pharmacoeconomics, Dublin, Ireland . . . . 1Royal Objectives: Frequency of self-monitoring of blood glucose (SMBG) varies in patients with type 2 diabetes mellitus (T2DM) and there is debate as to the true benefit it plays in the management of the condition. The aim of this study was to compare the frequency of SMBG using patient medication record (PMR) dispensed data, with self-reported data, and to assess the relationship between type of glucose-lowering therapy to SMBG frequency among patients with T2DM dispensed glucose test strips. Methods: Patients with T2DM were identified in communitypharmacies (n= 116) using PMRs in 2012. Patients who consented to participation were surveyed in a face-to-face interview where self-reported (SR) frequency of SMBG was determined. Dispensed data were classified by the mechanism of action of the blood glucose lowering regimen, (secretagogues vs. sensitisers). Test strip use over a 12 month period was calculated and compared to SR use from the patient survey. Analysis was performed in MS Excel® and SAS®. Results: Data from 484 patients were analysed (mean age 65yrs). Sensitising agents alone accounted for 45% of the cohort, secretagogues alone - 8%, a combination of both - 42%, and diet 5%. Approximately 52% of patients reported frequency of SMBG as once or twice daily vs 56% in the dispensed data. 13% of patients reported a frequency of at least 3 times daily vs 21% for PMR data. There was no statistically significant difference in SR frequency among those on the various glucose-lowering regimens. No correlations were found between frequency of use and age, gender, place of residence or number of prescribed medicines. Conclusions: Results from this study highlight (i) differences in frequency of blood glucose monitoring between dispensed data and SR data and (ii) the difficulty in accurately capturing this type of data. A reliance on either type of data alone may be insufficient to accurately inform decision making. PDB163 Cost of Severe Hypoglycaemia in Hospitalized Patients In Poland- Is It Financed and Reported in a Right Way? Czech M 1, Jaron N 1, Pietrzyk A 1, Franek E 2 University of Technology, Warsaw, Poland, 2Central Clinical Hospital, Ministry of Interior in Warsaw, Warsaw, Poland . . . . 1Warsaw Objectives: Hypoglycaemia is common in diabetic patients, often as a side effect of a treatment. Its occurrence can influence patients’ professional life and generate extra direct costs to the health care system. Severe (requiring another person’s assistance) hypoglycaemic events (SHEs), especially leading to hospitalisation, are associated with the highest burden and for that reason their real frequency and costs should be calculated properly. The aim of the study was to assess the accuracy of reporting of hypoglycaemia to the National Health Fund in Poland. Methods: A combination of prospective and retrospective designs was applied. A detailed retrospective analysis of 117 hospitalised patients’ case histories was performed from the period of February-March 2013. In addition, prospectively, in March and April 2013, 45 patients were interviewed and their clinical records were analysed. Information was collected in special case report forms. Real hospitalisation costs were also calculated. Results: In the retrospective arm hypoglycaemia affected 42 patients (55% of type 1 and 25% of type 2) and 85 SHEs were recorded. Only 2 patients were reported in the hypoglycaemia DRG group compared to a broader diabetes/ hyperglycaemia DRG group, where the financing level is higher. A financial difference equal to 7,082 EUR was generated due to more favourable classification of these cases. In the prospective arm 20 patients experienced hypoglycaemia, whereas only one was coded as hypoglycaemic. It has generated 2,247 EUR difference. Even with this financially more favourable, (still clinically justified but less precise) coding hypoglycaemic patients generated losses equal to 23,628 EUR in retrospective and 5,053 EUR in prospective group. Conclusions: Our study provided a direct proof A361 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 of underreporting of hypoglycaemia in clinical hospital settings in Poland caused by DRG- based financing standards. There is a need of reflecting real treatment cost in the level of reimbursement in this group of patients. PDB164 Patient Specific Landscape of Information Technology Solutions for Diabetes Self-Management Xenakis J 1, Garfield S 2 . . 1GfK, Wayland, MA, USA, 2GfK Market Access, Wayland, MA, USA Objectives: Diabetes Mellitus affects approximately 60 million Europeans, with prevalence increasing. Diabetes is associated with a number of severe comorbidities, including visual impairment, lower extremity conditions/amputation, neuropathy, renal disease and cardiovascular disease. Managing underlying disease can significantly improve disease outcomes, though adherence remains a challenge. With the major expansion of consumer technology-based health care solutions and European initiatives like the REACTION project, diabetic patients are more involved in care than ever before. Our study looks at patient centric drivers of adherence to technology based diabetes self-management in Europe. Methods: An in-depth analysis of current European initiatives implementing digital health management solutions was performed, as well as overall demographic and statistical research of the European diabetes population. Additionally, using dual data collection methods, involving telephone-based interviews and internet-based questionnaires, a total of 2,307 diabetes patients in the US were sampled. Questions focused on demographic, lifestyle, treatment, access to information, and socioeconomic status. Data from US diabetic patient preferences related to digital health solutions and patient engagement strategies were reviewed in the context of diabetes selfmanagement solutions being deployed in Europe. Results: Socioeconomic disadvantage amongst European citizens has been shown to significantly contribute to development of diabetes and related complications. Among our study population, diabetics with a lower socioeconomic status and the elderly showed drastically lower rates of access to internet at home or at work (58.1% and 66.3%, respectively), as compared to those who were employed with an average income >$60,000 (92.2%). This same group preferred to get information from their health care providers versus the internet. Conclusions: While programs are being developed to impact diabetic treatment adherence via IT based diabetes self-management solutions, patient characteristics and preferences need to be understood to optimize impact. Additional research is required to understand the best methods to reach these at-risk populations. PDB165 Using Modern Information Technology For Medical and Pharmaceutical Care Of Patients With Diabetes Mellitus In Ukraine Boyko A , Parnovskiy B Danylo Halytsky National Medical University, Lviv, Ukraine . . Objectives: In Ukraine, 1.2 million people have diabetes mellitus (DM). 80 insulins and 169 oral hypoglycemic preparations are registered. However, incomplete information on DM patients results in problems with providing their medicines. To develop a computer information retrieval system (CIS) for the registration of patients with DM and their pharmacotherapy (including financing sources) and to optimize the requirements for medicines; the CIS can be used to analyse the pharmaceutical care of patients with DM. Methods: This study includes designin a CIS, entering pharmacotherapy information of DM patients, and use of the CIS to analyze the quality of pharmaceutical care of DM patients. Results: The study was conducted at the Ternopil Region, Ukraine. The study was based in 2 hospitals, 2 dispensaries, and 12 pharmacies and examined the public and private sectors. The CIS was developed for use at national level and contains information on individual pharmacotherapy (27 insulins, 46 oral hypoglycemic preparations) of 2,186 patients (282 DM type I). The CIS consists of 4 blocks: “Patient” (935 men, 1261 women; from them 12 child) “Pharmacotherapy” (371 - insulin therapy, 1727 oral hypoglycemic agents) and “Financing sources” blocks contain information on the centralized and local budgets, funds on maintenance of people that suffered from the Chernobyl accident, humanitarian help, and self-financing. The 4th block contains information on interactions of medicines. The analyses of individuals’ pharmacotherapies has shown that the CIS is an effective tool and can support endocrinologists in monitoring consumption of medicines by each patient. The system also provides expert analyses of changes of schemes, doses of medicines, and financing sources for individual patients. Conclusions: The CIS provides the doctor with the ability to precisely define what medicines are necessary for the patient, define availability of medicines, and raise rationality and safety of use of antidiabetic medicines. PDB166 Quality of Diabetes Care Among Obese and Overweight Patients at Nhg Polyclinics Thaung Y M , Toh MPHS, Lee L J National Healthcare Group, Singapore . . . . Objectives: The study aimed to examine the disparity of risk factor screening and clinical outcomes depending on bodyweight of patients with diabetes mellitus at 9 National Healthcare Group (NHG) Polyclinics in Singapore. Methods: Data were extracted from NHG Diabetes Registry. Diabetes care bundle measures of 8 process indicators (biannual blood pressure (BP), bodyweight and HbA1c test measurement, annual serum cholesterol level, smoking, eye, foot and nephropathy assessment) and 3 intermediate outcome indicators (HbA1c ≤ 7%, BP < 130/80 mmHg and LDL-C < 2.6mmol/L) were measured. Rates of these indicators were compared by BMI according to WHO classification. Multivariate logistic regression for achieving the individual indicators was referenced to the normal BMI group. Results: In 2012, 87,552 patients were enrolled for diabetes care at 9 NHG Polyclinics. Mean age was 64.1± 11.6. 51.8% were female. 70.6% were Chinese. 40.4% were overweight and 20.8% were obese. After adjusting for age, gender and ethnicity, overweight and obese patients were more likely to have annual cholesterol test, smoking status assessed, eye and renal function screening than normal weight patients. However, the overweight and obese class I/II patients were less likely to achieve optimal HbA1c (7% or less) and BP (< 130/80 mmHg) than normal BMI patients. The AdjOR for achieving optimal HbA1c decreased from 0.88 (95%CI 0.86-0.91) for overweight to 0.77 (0.720.83) for obese class II; and BP decreased from 0.74 (0.99-1.05) for overweight to 0.48 (0.45-0.52) for obesity class II. More obese class II and III patients had optimal LDLcholesterol control compared to normal BMI patients, AdjOR 1.11 (1.03-1.19) and 1.16 (1.04-1.29) respectively. Conclusions: Overweight and obese patients with diabetes were less likely to achieve the optimal HbA1c and BP control than normal BMI patients. Clinicians can identify the patients with higher BMI and poor risk factor control for more intense intervention to improve diabetes care and outcomes. PDB167 Data Sources of Disease Epidemiology in German Drug Reimbursement Dossiers: Case Study Of Diabetes Mellitus Type II Zhou M , Sandmann F G , ten Thoren C , Voelskow V , Mostardt S , Gerber-Grote A U Institute for Quality and Efficiency in Health Care (IQWiG), Cologne, Germany . . . . . . . . Objectives: Investigating and comparing the sources of prevalence data used in German drug reimbursement dossiers within the AMNOG-framework in order to explore causes for deviating prevalence estimates. Methods: All publicly available dossiers on diabetes mellitus type II were retrieved from the Federal Joint Committee’s homepage (http://www.g-ba.de). We investigated the epidemiological sources in the dossiers on their representativity (i. e., population-level epidemiological data), actuality (year of data collection), method of data collection, data quality, and open access. Results: As of early June 2014, 13 dossiers were published on diabetes mellitus type II. Overall, 23 different sources for quantifying the prevalence were used, which can be grouped into 5 categories: clinical trials, registries, health insurances’ claims data, commercial data-providers, and a governmental data-provider. Data was mostly collected 5 or more years ago (except for the commercial and governmental data-providers). Data was retrieved by means of physicians’ statements; patient surveys with voluntary disclosure; diagnostic tests (such as the oral glucose tolerance test); national registries; secondary data from health insurances, and (raw) secondary data of pharmacy prescriptions with physician diagnoses. Besides differences in diagnostic criteria, reasons for deviation among sources include: the sample size and making adjustments to improve representativity for Germany; the chosen approach to analyze secondary data; differentiation of type II from type I diabetes mellitus, and (not) considering undiagnosed or untreated cases. While accessing the data of publications and of the commercial providers requires a payment, health insurance data is not freely available in Germany (yet). The governmental data-provider is accessible without restrictions and free of charge. Conclusions: The need for accurate population-level data and access to it is highlighted by the present case-study. Sources for prevalence data in diabetes mellitus type II were characterized by a variety of differences in the methods applied to derive (the) data, leading to deviating estimates. PDB169 Examining the Economic Benefits of Lifestyle Intervention in Segments of A Prediabetic Population Storm M 1, Dall T 1, Semilla A P 1, Su W 1, Blanchard T 2 Global Inc., Washington, DC, USA . . . . . . 1IHS, Washington, DC, USA, 2IHS Objectives: Published studies have shown that lifestyle intervention in adults with prediabetes can improve health outcomes and reduce or delay onset of type 2 diabetes. Improved health translates to lower medical costs, reduced mortality, and higher levels of employment and productivity. This study examined patient characteristics that help predict future medical expenditures avoided and economic benefits to society associated with lifestyle intervention. Methods: A Markovbased microsimulation model utilizing Monte Carlo simulation was constructed to simulate patient health and economic outcomes over ten years with and without a lifestyle intervention. The population examined was a nationally representative sample (n= 2,887) of US adults with prediabetes from the 2003-2010 National Health and Nutrition Examination Survey who meet the American Diabetes Association’s screening criteria for diabetes. Prediction equations for diabetes and sequelae, medical expenditures, economic outcomes, mortality, and quality of life came from the published trials and studies, such as the UK Prospective Diabetes Study. The analysis simulated body weight and hemoglobin A1c benefits reported in the 10-year follow-up results from the lifestyle arm of the Diabetes Prevention Program and Outcomes Study (DPPOS). OLS regression was subsequently used to analyze the initial patient characteristics and their relation to subsequent outcomes. Results: Simulated outcomes suggest that age, sex, race, obese status, and HbA1c at time of intervention are strong predictors of successful economic outcomes (medical expenditures and societal benefit). Intervention impact on medical savings generally increased with age through age 65, while total economic benefits of the intervention increase through age 60. Women had $4,200 in higher medical savings then men. Obese participants had $3,000 higher medical savings than non-obese participants. Conclusions: The expected economic benefit of lifestyle intervention varies systematically by patient characteristics. Patients age 55-64 experienced the most medical savings, while patients age 45-54 experienced the largest societal economic impact. PDB170 Efficacy of The Automated Target Glucose Control: A Systematic Review Yoo K J NECA, Seoul, South Korea . . Objectives: Tight glucose control in critically ill patients is difficult and labor intensive, resulting in poor efficacy of glycemic control and increased hypoglycemia rate. The purpose of this study was to evaluate the safety and effectiveness of the automated target glucose control using computer-based insulin algo- A362 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 rithms compared to the conventinal glucose control using paper-based protocols. Methods: A systematic literature review and meta-analysis was used to evaluate the safety and effectiveness for the automated target glucose control. The 14 randomized control trials of 2537 articles were selected and two reviewers evaluated independently the quality of this selected articles using the Scottish Intercollegiate Guidelines Network (SIGN) tool. Results: The computer-based insulin protocols resulted in a shorter time to reach within target range (MD -1.03, 95%CI -1.70~-0.35, p= .003, I2= 77%) and a higher percentage of glucose readings within target range (MD 11.98, 95%CI 8.83~15.14, p< .001, I2= 89%) than the paperbased protocols. The incidence rates of above or below target range in computerbased insulin protocols were acceptable levels and the incidence rates of severe hypoglycemia below 40mg/dL in computer-based insulin protocol was significantly lower than the paper-based insulin protocols (MD -23.41, 95%CI -32.35~-14.46, p< . 001, I2= 0%). Conclusions: The automated target glucose control using computer-based algorithms resulted in tighter glycemic control without an increased risk of hypoglycemic events compared to the paper-based protocols. But taking into account the high heterogeneity between studies, these results must be interpreted with caution. PDB171 Comparison in Methods Between National Institute for Health And Care Excellence (Nice) Submissions For Sodium-Glucose Co-Transporter 2 Inhibitors (Sglt-2i) Kandaswamy P , Daacke I M Boehringer Ingelheim UK, Bracknell, UK . . . Objectives: A comparisons of economic methods across all NICE submissions for SGLT-2is (dapagliflozin, canagliflozin and empagliflozin) and exploring potential opportunities for optimising the appraisals of treatment classes. Methods: The NICE website was searched for: manufacturer submissions, clarification letters, evidence review group (ERG) reports and NICE guidance relating to the SGLT-2is. Data extracted included the sources of inputs and key assumptions to the economic analysis in particular the evidence synthesis, modelling approach and key cost-effectiveness results. Comparisons were then made across appraisals in reference to the ERG and NICE comments to identify any areas of uncertainty. Results: The results of the review indicated the following similarities between submissions: The lines of therapies explored; the use of United Kingdom Prospective Diabetes Study equations to link surrogate to clinically relevant outcomes; use of Bayesian network meta-analyses to derive clinical inputs which were all based on the same studies, outcomes and time period; the sources of cost and utility inputs were mainly derived from UKPDS studies; and the relative differences in costs and QALYS. The main differences identified included: the economic models used (pre-existing versus de Novo); treatments included in decision space; the modelling of weight gain; definitions of adverse effects included; and the total costs and QALYs between appraisals. Conclusions: The NICE appraisals for SGLT-2is were broadly similar in their overall design. This is apparent from the similar results of the appraisals in terms of relative outcomes. Given this, it is arguable whether the resources required to develop these separate submissions was essential. It would be of value for HTA organisations to develop reference cases including standard inputs for the modelling of significant conditions such as diabetes especially where there is such consistency. This would improve the comparability of analyses and reduce the costs to manufacturers and ERGs in developing and reviewing submissions respectively. PDB172 Adherence to Nice Guidance For Initiating Glp-1 Mimetics Among Patients With Type 2 Diabetes In Primary Care In England And Wales - An Evaluation Using The Clinical Practice Research Datalink (Cprd) Jameson K 1, D’Oca K 1, Murray-Thomas T 2, O’Regan C 3, Leigh P 1 1Merck Sharp and Dohme Ltd, Hoddesdon, UK, 2Medicines and Healthcare Products Regulatory Agency, London, UK, 3Merck Sharp & Dohme Limited, Hertfordshire, UK . . . . . Objectives: To assess adherence to UK NICE guidance/guidelines for initiating therapy with Glucagon-Like Peptide-1 [GLP-1] mimetics - exenatide, prolongedrelease exenatide [Exen-PR] and liraglutide – in patients with Type 2 diabetes (T2D). Methods: A retrospective cohort study of T2D patients in primary care identified in the CPRD. Patients were ≥ 40 years at GLP-1 initiation, had ≥ 12 months of CPRD history, with first GLP-1 prescription recorded on/after publication of the respective NICE guidance/guideline. The primary outcome was the proportion of patients initiating/not initiating GLP-1 as part of a NICE-recommended regimen (recommended dual therapy: liraglutide or Exen-PR added to metformin or sulphonylureas; recommended triple therapy: any GLP-1 added to metformin+sulphonylureas or metformin+thiazolidinedione). For patients initiating GLP-1 as triple therapy, the proportion meeting NICE initiation criteria (HbA1c ≥ 7.5% and body mass index [BMI] ≥ 35kg/m2) was assessed. Results: Mean age of the study cohort initiating GLP-1 (7,133 patients) was 58.2 years (SD 9.4), BMI 38.4 kg/m2 (SD 6.8) and HbA1c 9.2% (SD 3.2%). Overall, only 25% (1,772/7,133) of patients were initiated on GLP-1 as part of a NICE-recommended regimen. Approximately 13% of patients initiated liraglutide (443/3,438) and 13% initiated Exen-PR (45/346) as part of NICErecommended dual therapy regimen, with a higher proportion initiating a GLP-1 as part of NICE-recommended triple therapy (exenatide: 19% [637/3,349], liraglutide: 17% [588/3,438], Exen-PR: 17% [59/346]). Of those patients initiated on a recommended triple therapy regimen, 50% fulfilled both NICE HbA1c and BMI initiation criteria. The majority of the study cohort (75%) initiated GLP-1 in combination regimens not recommended by NICE: 12% non-recommended dual therapy; 21% nonrecommended triple therapy; 16% monotherapy, and 25% quadruple therapy/add-on to insulin. Conclusions: Many patients are initiated on GLP-1 in combination regimens which are not recommended by NICE. The rationale for these decisions may warrant further research to better understand the observed low adherence to NICE-recommended regimens. PDB173 Treatment Patterns Among Type 2 Diabetes Mellitus Patients in Saudi Arabia Levy A R 1, Szabo S M 1, Osenenko K M 1, Korol E E 1, Qatami L 2, Al Jaser S 3, Al Saggabi A 3, Al Suwaidan S 3, Maclean R 4, Donato B M K 5 1ICON Epidemiology, Vancouver, BC, Canada, 2Bristol-Myers Squibb Company, Dubai, United Arab Emirates, 3King Fahad National Guard Hospital, Riyadh, Saudi Arabia, 4Bristol-Myers Squibb Company, Plainsboro, NJ, USA, 5Bristol-Myers Squibb Company, Wallingford, CT, USA . . . . . . . . . . . . . . . . Objectives: Due to its dramatically increasing prevalence, type 2 diabetes mellitus (T2DM) is a serious public health concern in Saudi Arabia. Few data have been published that describe how persons with T2DM are managed in clinical practice; we aimed to characterize recent treatment patterns among individuals treated for T2DM in Riyadh, Saudi Arabia. Methods: Charts from 455 adults with T2DM who visited the King Fahad National Guard Hospital from October 2009 to March 2010 (enrolment period) were systematically sampled until the target (n= 250) was reached. Treatment data from enrolment to September 2011 were abstracted from subject charts. Treatment regimens, their frequency of use, and the number of switches (drug replacement/removal/addition) over the study period were calculated. Analyses were stratified by T2DM duration; data for those recentlydiagnosed (< 5 years; n= 29) and with longstanding disease (≥ 20 years; n= 67) are presented. Results: Forty-four percent of subjects were male; at enrolment, mean (SD) age was 61 (13) years, and mean T2DM duration was 11 (8) years. At enrolment, 42% of subjects had received prior insulin treatment (recently-diagnosed: 23%; longstanding disease: 52%). During the study period, the most common regimens were oral combination therapy (41%) and insulin+oral combination therapy (32%). Overall, 44% received any insulin therapy during the study period (recently-diagnosed: 25%; longstanding disease: 54%). By study end, 49% had received any prior insulin therapy (recently-diagnosed: 29%; longstanding disease: 58%). On average, T2DM subjects had 1.3 treatment switches over the period; little variation was seen by T2DM duration. Conclusions: Although the frequency of insulin treatment was lower than reported in other studies from the region, treatment switches are indicative of attempts to improve T2DM control. Novel therapies may improve clinical outcomes among T2DM patients and this study provides valuable baseline data with which to compare the effectiveness of new T2DM treatments in Saudi Arabia. PDB174 Usage of Self-Monitoring of Blood Glucose (Smbg) By Diabetes Therapy Type in India Mast O 1, Tan A 2, Punjabi K 3 1Roche Diagnostics, Diabetes Care, Mannheim, Germany, 2Cegedim Strategic Data, Singapore, 3Cegedim Strategic Data, Mumbai, India . . . Objectives: SMBG is a core component of diabetes therapy, supporting a safe and effective drug therapy and providing feedback on the impact of diet and lifestyle. In 2014 an Indian guideline on glycemic monitoring was published. This study aims to assess the level of SMBG usage in patients by diabetes therapies. Methods: Source data for this explorative analysis is the CSD-PDS Diabetes survey covering drug treated patients with diabetes. All 2,250 cases were documented in the 2nd half of 2013. PDS Diabetes is a syndicated research with a fixed representative panel of doctors from larger Indian cities using a standardized documentation of cases. SMBG usage is analyzed by therapeutic subgroups: oral diabetes therapy (OAD 75%), basal supported oral therapy (BOT 9%), conventional insulin therapy (CT 8%), intensive insulin therapy (IIT 5%) and others (3%, excluded). Results: Diabetes type 2: 93%. 1,180 patients (52.4%) have a meter for home-testing. Shares of testers by therapies (OAD/ BOT/CT/IIT) are 50.7%, 63.4%, 53.1% and 67.3%. An HbA1c-goal of 8% is achieved by 62/49/45/37%, an HbA1c-goal of 7% by 29/20/19/18%. Average weekly test frequencies are 2.4, 2.8, 2.9 and 2.2 respectively. In the OAD/BOT/CT groups 13/14/16% test at least once daily (recommended), less than 2% in IIT test 3 times daily (recommended). HbA1c levels are by 0.3%, 0.2%, 1.1% and 0.0% lower in SMBG users. Conclusions: Many patients in this sample do not achieve HbA1c targets and many do not test their blood glucose. Below 1 in 5 patients follows the recommended test frequency. In IIT SMBG is needed to support insulin dose adjustments: below 1 in 50 test enough. In CT patients the differences in HbA1c between testers and non-testers are largest. Lacking benefit in IIT treated patients might indicate the insufficiency of current testing. The new guideline might help to close these gaps. Gastrointestinal Disorders – Clinical Outcomes Studies PGI1 The Comparison Of Clinical Impact Between Entecavir And Tenofovir In Chronic Hbv Infection Shen W C , Tsai P Y , Kuo L N Taipei Medical University and Wan Fang Hospital, Taipei, Taiwan . . . . . . Objectives: To compare the efficacy and safety of Entecavir and Tenofovir in nucleos(t) ide-naïve CHB patients with HBV DNA in a medical center. Methods: We included patients in WFH had used Entecavir or Tenofovir from 2011/07 to 2012/12, and excluded patients who had (1) co-infection with the HCV, HIV, TB, (2) the history of malignancy within 3 years before the first time use of ETV or TDF, (3) the history of alcohol abuse, and (4) used other anti-HBV drugs. In Statistics, we used Kaplan–Meier survival analysis to present the HBV DNA suppression and the normalization of ALT. Results: We identified 124 consencutive patients treated with ETV and 24 patients treated with TDF. After 18 months, the probability of complete viral suppression was 67.7% in the ETV group and 45.8% in the TDF group (P= 0.49). The probability of complete ALT normalization was 87.1% in the ETV group and 83.3% in the TDF group (P= 0.159). In the adverse side effects of neurologic, malaise was observed in the patients treated with ETV more than TDF (11.3% vs. 0%), but in gastrointestinal disorders, diarrhea was observed in the patients treated with TDF more than ETV (25.0% vs. 3.2%). Conclusions: Entecavir and Tenofovir have no A363 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 significance difference in the efficacy and safety in nucleos(t) ide-naïve CHB patients with HBV DNA greater than 6 log10 in the medical center. PGI2 Effectiveness Of Probiotics In Irritable Bowel Syndrome: A Systematic Review With Meta-Analysis Nikfar S, Mozafari S, Didari T, Abdollahi M Tehran University of Medical Sciences, Tehran, Iran Objectives: To investigate the efficacy of probiotics in irritable bowel syndrome (IBS) patients, this meta-analysis was performed. Methods: The collected data contained twenty-four clinical trials which fifteen were eligible for meta-analysis and nine were reviewed systematically. All studies were randomized placebo-controlled trials in patients with IBS that investigated the efficacy of probiotics in IBS improvement. Trials were evaluated with Relative Risk (RR) with 95% confidence interval (95% CI). Cochran Q test was used to test heterogeneity with P value 0.05 (P< 0.05). Egger and Begg-Mazumdar tests in funnel plot were calculated as publication bias indicators. Results: Probiotics reduced pain severity, symptom severity score and induced adequate general symptom improvement. Distension, bloating, and flatulence were not improved after probiotics treatment when compared to placebo. Conclusions: Collectively, the results demonstrated the beneficial efficacy of probiotics compared with placebo in IBS patients. PGI3 Establishment Of A Hepatitis C Virus (Hcv) Cohort In A Large Israeli Hmo Shalev V 1, Weil C 2, Nwankwo C 3, Friedman M 1, Kenet G 2, Chodick G 4 1Medical Division, Maccabi Healthcare Services, Tel Aviv, Israel, 2Maccabi Healthcare Services, Tel Aviv, Israel, 3Merck, Whitehouse Station, NJ, USA, 4Maccabi Healthcare Services and Tel Aviv University, Tel Aviv, Israel . . . . . . Objectives: Hepatitis C virus (HCV) affects an estimated 130 million people worldwide and is a major cause of chronic liver disease. Real-world data is needed to better understand the epidemiology of HCV and its complications, as well as treatment patterns and outcomes. The objectve of this study is to establish a cohort of HCV carriers in a large health maintenance organization, with comprehensive data on treatment and disease outcomes. Methods: The HCV cohort will use data available since 1998 from the computerized databases of Maccabi Healthcare Services, the second largest HMO in Israel, with approximately 2 million members. HCV cases are included based on diagnostic codes (ICD9-CM), laboratory data (e. g. detection of HCV antibodies and RNA) and dispensed prescriptions for HCV treatment. The cohort includes demographic data (age, sex, immigration and socioeconomic status), clinical data (e. g. BMI; comorbidities), treatment patterns, virological outcomes, and HCV complications – including cirrhosis, hepatocellular carcinoma, liver transplants, and mortality. Results: At the end of 2012, HCV infection was identified among 10,648 patients, corresponding to an age-adjusted prevalence rate of 5 per thousand. The highest prevalence was found among males and in the age group 35-55 yrs. Two thirds of HCV patients were immigrants from Eastern Europe. HCV genotype 1 was predominant (67%). Over a third of patients had at least one recorded purchase of interferon, and treated patients were less likely to have chronic diseases such as diabetes or cardiovascular disease, compared to untreated patients. Conclusions: The establishment of a HCV cohort in MHS can serve as a basis for retrospective database studies and can be periodically updated to follow up existing patients and identify incident HCV cases. For example, future studies can examine the adherence and efficacy of treatments, and associations between HCV and chronic diseases such as chronic kidney disease. PGI4 The Incidence Of Upper And Lower Gastrointestinal Complications: A Retrospective Study Using A Japanese Health Care Database 1Ben Gurion University of the Negev, Beer-Sheva, Israel, 2Soroka University Medical Center, Beer-Sheva, Israel Objectives: To study possible associations and assess the impact of socio-demographic, quality-of-life and coping factors on the severity of Crohn’s disease patients in Israel. Methods: Consecutive Crohn’s disease patients undergoing clinical followup at the Inflammatory Bowel Diseases Clinic completed the following questionnaires: SF-36 quality-of-life, Ways of Coping, and socio-demographic details. Crohn’s disease severity was measured by Harvey-Bradshaw Index (HBI) expanded to include pharmaceutical and surgical therapies as well as medical and surgical hospitalization information. Based on these data, a multifactorial model was built using multivariate linear regression. Results: There were altogether 192 patients: 81 (42.2%) men (age 36.8 ± 14.5 years, disease duration 11.4 ± 8.6 years, education 13.7 ± 2.7 years, HBI 7.6 ± 4.9) and 111 (57.8%) women (age 41.4 ± 15.6*, duration 14.0 ± 9.3*, education 14.2 ± 2.8, HBI 8.4 ± 4.8; *p < 0.05 vs. men). The multifactorial model showed the following significant predictors of disease severity: age (beta = −0.250, p = 0.006), number of children (beta = 0.245, p = 0.008), SF-36 General Health score (beta = −0.378, p < 0.01), coping/sense of humor (beta = 0.209, p = 0.011), and coping/acceptance (beta = −0.183, p = 0.024). The model accounted for 29% of explained variance, with Adjusted R2 = 0.26. Gender, education and socio-economic status were not predictors of disease severity. Conclusions: All of the indicated factors (age, family size, coping skills) had a significant effect on the severity of Crohn’s disease. In addition to prescribing medications, physicians should pay special attention to these factors as part of an overall management plan for their Crohn’s disease patients. PGI6 Development And External Validation Of A Risk Calculator For Predicting Anemia In Patients Treated With Triple Therapy (Tt) Containing Boceprevir (Boc), Pegylated Interferon And Ribavirin (Pr) Nwankwo C 1, Kattan M 2, Luan A 3, Wahl J 1 Station, NJ, USA, 2Cleveland Clinic, Columbia, SC, USA, 3Agile 1, Torrance, CA, USA . . . . 1Merck, Whitehouse Objectives: To develop and validate a risk calculator for the prediction of anemia occurring after 8 weeks of TT using BOC+PR among Hepatitis C Virus (HCV) genotype 1 patients based on risk factors. Methods: Data for all randomized HCV genotype 1 patients starting therapy with BOC+PR and who received at least one dose of BOC in three phase 3 clinical trials (PN05101, PN5216 and PN5514) were included. The outcome of interest was the prediction of anemia, defined as hemoglobin < 10 g/L after 12 weeks of TT with BOC+PR (this included 4 weeks of lead-in treatment using PR). Logistic regression was used to develop the risk calculator model by analyzing the association of each variable with the likelihood of developing anemia while on treatment. Baseline variables were included as covariates in the model. Linearity assumptions were relaxed with the use of restricted cubic splines. Bootstrapping, with 1000 resamples, were used in conjunction with estimation of discrimination and calibration. Results: Following a stepdown procedure that eliminated predictors that did not contribute to the overall model concordance index, nine variables remained in the final model: age, hemoglobin, gender, cirrhosis, hematologic counts, Alkaline phosphatases levels, and creatinine. This model had a boostrap corrected concordance index of 0.775. The model was cross-validated by sequentially omitting each of the three randomized trials from the model development and using the omitted trial as a test set; the concordance indices following this procedure ranged from 0.75 to 0.85. Calibration of the model, assessed graphically, indicated reasonably close agreement between predicted and observed proportions. Calibration held following trial cross validation as well. Conclusions: The model calibrated well and demonstrated good predictive ability. This tool may be useful for identifying and subsequently managing HCV patients at relatively high risk for developing anemia if treated with BOC+PR. Kawaguchi I 1, Nakazuru Y 1, Yamamoto Y 2, Nagayasu S 2, Fujimoto Y 1 1Pfizer Japan Inc, Tokyo, Japan, 2MinaCare co. ltd, Tokyo, Japan Gastrointestinal Disorders – Cost Studies Objectives: The objective of this study was to investigate the incidence of complications (bleeding and perforation) with hospitalization from both upper and lower GI tract in Japanese population using a health care claims database (HDB). Methods: All of the claims data of the patients who have a history of hospitalization due to complications from GI tracts between January 2011 and December 2012 were extracted from the HDB which holds about 1.8 million peoples’ claims data under employmentbased health insurance. In order to identify upper and lower GI events precisely, we confirmed them when a diagnosis of a GI event in the claims (ICD-10 code) was accompanied by a record of examination and/or endoscopic or surgical treatment relevant to upper and lower GI complications. Results: The total number of person-years at risk was 1.2 million person-years in 2012. The incidence rates in upper and lower GI events were 48 and 41 per 100,000 person-years, respectively. Twenty-one percent of the lower events originated in bleeding from hemorrhoid or related treatments (eg. hemorrhoidectomy). Age-group analyses in the upper vs. lower events, except those from hemorrhoid, were 27 vs. 17, 57 vs. 46 and 184 vs. 104 per 100,000 person-years in 20-39, 40-59, and 60+ years groups, respectively. Data from 2011 were consistent with these observations on the ratio of upper to lower GI events and the age-based incidence, indicating the robustness of the results. Conclusions: This was the first study to investigate the incidence from both upper and lower GI complications with hospitalization in a real clinical setting in Japan using a single large data source. We confirmed that a number of GI events occurred in both upper and lower GI tract and the incidence rates of both events were increased with age in a real world setting. PGI7 Budget Impact Analysis Of Sofosbuvir For The Treatment Of Hepatitis C In The Veneto Region, Italy . . . . . PGI5 Associations Between Crohn’s Disease Severity And Specific SocioDemographic, Quality-Of-Life And Coping Factors Friger M 1, Sarid O 1, Slonim-Nevo V 1, Vardi H 1, Greenberg D 1, Ben Yaakov G 2, Gaspar N 2, Dizengof V 2, Moshkelo A 2, Munteau D 2, Rozental A 2, Abu Freha N 2, Schwartz D 2, Krugliak P 2, Eidelman L 2, Fich A 2, Odes S 2 . . . . . . . . . . . . . . . . . Adami S 1, Aiello A 2, Magrì M R 2, Nocera S 2, Piovesan S 3, Alberti A 3, Scroccaro G 1 Farmaceutico - Regione Veneto, Venezia, Italy, 2CRUF Centro Regionale Unico sul Farmaco Regione Veneto, Verona, Italy, 3Dipartimento Medicina Molecolare - Università degli studi di Padova, Padova, Italy . . . . . . . . 1Settore Objectives: Hepatitis C virus (HCV) infection is one of the main causes of chronic liver disease worldwide. Sofosbuvir, a nucleotide analogue inhibitor of HCV RNAdependent RNA polymerase, has been approved in Europe for the treatment of chronic hepatitis C genotypes 1-6. This study aims to estimate the budget impact of sofosbuvir in patients who live in the Veneto Region. Methods: Population data were obtained from a regional survey. The survey was conducted by the Department of Molecular Medicine of Padua University in order to identify the number of patients with advanced stage disease or more rapid disease progression. At the time of this study, the Italian price of sofosbuvir has not been defined so it was estimated that a full 12 week course of sofosbuvir would cost as Swedish price: 42.653 € (85.306€ for a 24 week treatment). Total costs include costs for other drugs which might be used in combination. Results: In Veneto Region, there are 835 patients with severe Hepatitis C genotype 1 (n= 493); genotype 2 (n= 94), genotype 3 (n= 178), genotype 4 (n= 70) and about 40% of them need a 24 week treatment. Moreover, there are 100 patients awaiting liver transplantation and 70 patients with posttransplant recurrence of HCV infection. Therefore, the total estimated expenditure would be about € 66.000.000 for all the 1005 patients and about € 34.000.000 for the 455 patients (45%) with very high priority to treatment. Conclusions: High prices for new drugs are a growing concern to payers, given the large number of innovative drugs in development and the limited health care resources. Due to high costs, economic analyses are needed to estimate the budgetary impact for the Regional Health System (RHS) with sofosbuvir. A364 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PGI8 Budget Impact Analysis Of Biosimilar Infliximab For The Treatment Of Crohn’s Disease In Six Central Eastern European Countries Brodszky V , Gulacsi L , Balogh O , Baji P, Rencz F , Péntek M Corvinus University of Budapest, Budapest, Hungary . . . . . Objectives: Biosimilar infliximab (CT-P13) is the first licensed biosimilar in EU for Crohn’s disease (CD). It is expected that spread of biosimilars in the treatment of CD will lead to cost savings and it might improve access to biological therapies. The main aim of this study was to analyse the budget impact of introducing biosimilar infliximab for the treatment of CD in six Central Eastern European (CEE) countries’ – Bulgaria, Czech Republic, Hungary, Poland, Romania and Slovakia – health care systems. Methods: This budget impact model estimates potential impact of biosimilar infliximab on health care budget over three-year time frame from thirdparty payer perspective. Spreadsheet-based country specific population model was developed functioning in quarter year time units. The model tracked movement of CD population between main states: 1) immune therapy, 2) infliximab, 3) biosimilar infliximab, 4) adalimumab. Switching between biologics and biosimilar infliximab was taken into consideration as well. In scenario analyses different rates of interchanging and switching were analysed. A -25% price difference was assumed for biosimilar infliximab compared to originator. Budget impact was calculated as difference in total cost of scenarios with and without biosimilar infliximab. Results: In 2013,4, 625 CD patients were treated with biologicals in the CEE. Over the 3-year period with gradually interchanging 80% of infliximab to biosimilar infliximab is expected to lead to a net benefit of 16,635,000 euros compared to the scenario in which biosimilar infliximab would not be available. In scenario in which interchangeabilty was disallowed the budget savings amounted to 7,842,000 euros. If budget savings were spent on reimbursement of additional biosimilar infliximab treatments, approximately further 889 or 420 patients could be treated in the six countries, respectively. Conclusions: Introduction of biosimilar infliximab treatment for CD in CEE is predicted to bring substantial cost savings or increase of the number of patients with access to biological therapy. PGI9 Proton Pump Inhibitors In South Africa: Role Of Branded Generics Truter I Nelson Mandela Metropolitan University, Port Elizabeth, South Africa . Objectives: Originator products dominate the market, and as soon as patents expire prescribing patterns change as branded generics become the most often prescribed due to mandatory generic substitution in South Africa. This study investigated the prescribing patterns of proton pump inhibitors (PPIs). Methods: A retrospective, cross-sectional drug utilisation study was conducted on prescription data of a medical insurance scheme administrator in South Africa for 2011. The database contained 2 298 312 records for medicine, medical devices and procedures. Results: A total of 22 198 PPIs were prescribed to 7 504 patients (average age: 45.09 (SD= 15.82) years) at a cost of R3 931 241.43. The average cost per PPI prescription was R177.10 (SD= R140.62). The average cost for a generic PPI prescription was R147.01 compared to R276.17 for a prescription for an originator product. Five different PPIs were prescribed. Omeprazole, lansoprazole and pantoprazole had branded generics available on the market, whereas esomeprazole and rabeprazole only had originator products on the market. Omeprazole was the most frequently prescribed (47.05% of all PPI prescriptions). Nearly all (98.90%) omeprazole prescriptions were for one of its eight branded generics. Esomeprazole had the highest average cost per prescription of R289.23. The 40mg tablet formulation of esomeprazole was the most commonly prescribed. Less than 2% of PPI prescriptions were for rabeprazole. Most prescriptions were for Schedule 4 products (prescription-only), with only 3.91% Schedule 2 prescriptions (over-the-counter, prescribed in lower dosages for acute symptoms over a short period of time). Overall in this study, 76.70% of PPI prescriptions were for branded generics. Conclusions: Omeprazole dominated PPI prescribing, whilst esomeprazole was the most expensive PPI. It will be important to further investigate the impact of patent expiry and the introduction of new branded generics on PPI prescribing patterns. The study confirmed the significant price differences for PPIs between branded generics and originator products. PGI10 Direct Treatment Costs Of Cirrhosis In The Brazilian Public Health Care System: A 2008-2012 Retrospecteve Analysis Magno L A 1, Morais A D 2 1Janssen, Sao Paulo, Brazil, 2Janssen Cilag Farmaceutica, São Paulo, Brazil . . . . Objectives: estimate direct health care costs associated with cirrhosis from a brazilian public payer perspective (SUS). Methods: A retrospective study of a public claims database (DATASUS) was used to assess direct health care costs associated with cirrhosis between 2008 and 2012. Patients with cirrhosis were identified based on ICD-10 codes (K70.3, K71.7, K74.3-K74.6). The retrospective analysis considered inpatient costs and outpatient treatment costs of patients during the 4 year time frame. Medication costs were defined as a weighted average during the period of analysis considering official government price lists available at www. comprasnet. gov. br. A sub analysis was carried out for cirrhotic patients with hepatitis C (HCV) treatment, defined as patients treated with peg-interferon with ribavirin (PR). Results: Between 2008-2012, 120,082 patients were hospitalized due to cirrhosis (49% of patients accounted by ICD K746) with a total treatment cost of R$ 404 million, 68% accounting for transplant (R$ 276 million), 30% for outpatient procedures (R$ 121 million), 2% for surgical procedures (R$ 6 million) and the remainder for exams and diagnostics. Patients were hospitalized for an average of 10,1 days per year, with an average cost of R$ 2,806 (R$2.149- R$ 3,757) per hospitalization event and R$ 3.378 (R$ 2.523-4.579) per patient. The average cost for transplant was R$ 52.912 (R$43.304-70.274). Between 2008 and 2012 2.933 patients were hospitalized for cirrhosis having received HCV treatment with a total cost of R$ 19 million. The average inpatient cost for HCV-cirrhosis patients was R$ 4,999 for 11,2 days of hospitalization. Organ transplants-associated costs accounted for R$ 15 million, 77% of total costs, with an average cost of R$ 67,319 per transplant. Conclusions: Organ transplant accounts for the majority of hospital health care costs in cirrhosis patients in the Brazilian public health care system. Considering patients with HCV treatment and cirrhosis, the average cost per patient was around 78% higher than the general cirrhosis patient. PGI11 Burden Of Hospitalizations Related To Chronic Hepatitis C In France: Evolution Between 2009 And 2012 Rotily M 1, Abergel A 2, Branchoux S 3, De Léotoing L 1, Vainchtock A 1, Akremi R 3, Gaudin A F 3 1HEVA, Lyon, France, 2centre hospitalier universitaire CHU Estaing, Clermont-Ferrand, France, 3Bristol-Myers Squibb, Rueil-Malmaison, France . . . . . . . . Objectives: To assess the burden of hospitalizations related to Chronic Hepatitis C (CHC) in France in 2012 and to benchmark it to the 2009 estimate. Methods: All hospital stays with chronic viral hepatitis C (ICD-10 code B18.2) as principal, related or significantly associated diagnosis were extracted from the PMSI 2012 (French Medical Information System Program) database. Through an algorithm and a medical review, hospitalizations not related to CHC were excluded. Patients were assigned to a liver disease stage: F0-F3; compensated cirrhosis; decompensated cirrhosis; hepatocellular carcinoma; liver transplant. Same database and method (without making the distinction between compensated/decompensated cirrhosis) were used in 2009 allowing descriptive benchmark. Valuation was performed on French official tariffs for 2009 and 2012 and expressed in 2013 value (Euros). Results: 22,056 hospitalizations were extracted in 2012 vs 27,258 in 2009 corresponding to 12,040 and 15,482 patients, respectively. Sex-ratio M/F increased between 2009 and 2012 (1.66 vs 1.78), as well as proportion of patients aged of 50 years old and over (60% vs 70%). In 2012, 5,834 patients were in F0-F3 fibrosis stage, 3,547 had cirrhosis and 2,105 had hepatocellular carcinoma. Crude estimates for the evolution for these stages between 2009 and 2012 was -32%, -36% and +17%, respectively. Between 2009 and 2012, number of liver biopsies was reduced by one third, and also less liver transplants were performed (293 vs 254). Overall, the economic burden of hospitalizations related to CHC was € 60 millions in 2012, 8% lower than in 2009. Conclusions: 3 major facts were highlighted: the decline in the number of liver biopsies, the decrease in the number of patients hospitalized for cirrhosis and the increase in patients hospitalized for hepatocellular carcinoma. This increase might be due to an improvement in screening techniques and aging of the population. PGI12 Mean Annual Cost Of Patients Hospitalized For Chronic Hepatitis C In France: The Hepc-Lone Study Abergel A 1, Rotily M 2, Gaudin A F 3, De Léotoing L 2, Vainchtock A 2, Akremi R 3, Branchoux S 3 1centre hospitalier universitaire CHU Estaing, Clermont-Ferrand, France, 2HEVA, Lyon, France, 3Bristol-Myers Squibb, Rueil-Malmaison, France . . . . . . . . Objectives: To assess the mean annual cost of patients hospitalized for Chronic Hepatitis C (CHC) in France stratified by liver disease stage. Methods: Patients hospitalized for chronic viral hepatitis C (ICD-10 code B18.2) were identified in the PMSI-MCO 2010-2012 database (French Medical Information System ProgramMedicine, Surgery, Obstetric) between January, 1st 2010 and December, 31st 2012 and tracked in the PMSI-SSR/HAD databases (post-acute care/hospitalizations at home: alternative management of care). An algorithm and a medical review excluded people under 18 years old, viral co-infected patients and non CHC related stays. During the study period, patients were followed from their first stay to censoring (i. e December 31st 2012 or death). For each patient, cumulative time in days in the different liver disease stages (F0-F3 fibrosis; compensated cirrhosis; decompensated cirrhosis; hepatocellular carcinoma; liver transplant) were calculated. Associated costs during these periods were added up and a mean annual cost per patient per liver disease stage estimated. As only inpatient deaths are captured into the PMSI, estimated outpatient deaths were imputed. Valuation was performed considering the national costs study (ENCC) expressed in 2013 Euro. Results: 58,405 stays were identified, extracted and considered as directly related to CHC corresponding to 26,621 patients. 743 additional hospitalizations were extracted from PMSI SSR/ HAD. Patients were 56±13.5 years old on average, 62% were male. Mean annual costs of patients hospitalized for CHC reason were: € 1,107 for F0-F3 fibrosis; € 2,104 for compensated cirrhosis; €10,400 for decompensated cirrhosis; €11,739 for hepatocellular carcinoma; € 62,098 for 1st year of liver transplant and € 7,449 for subsequent year. Conclusions: CHC patients are mainly managed in acute care. Costs associated to CHC complications are important. These estimated hospital-related costs will be useful to populate cost-effectiveness and disease cost models. PGI13 Direct Health Care Costs Associated With Opioid-Induced Constipation Lawson R 1, Haycock L 2, Laxman K 2, King F 3, Gardner K 2 . . . . . 1AstraZeneca, Cheshire, UK, 2Evidera, London, UK, 3AstraZeneca, Gaithersberg, MD, USA Objectives: Opioid-induced constipation (OIC) is the most common and distressing side effect of opioid treatment in patients with chronic pain. There are limited data to estimate the occurrence of OIC; however, estimates in non-cancer patients range from 40-50% and can be as high as 90% in cancer patients. Increased resource utilisation is associated with the diagnosis and management of OIC; however, the economic burden of OIC remains under-reported. Methods: This review considered the cost of illness associated with OIC. A targeted literature review was conducted for all publications since 2000 that evaluated the economic burden of OIC. Databases used for the literature search were PubMed, Embase, and HEED. All costs were converted to a 2014 USD cost base. Results: Eleven studies were identified which reported direct health care costs associated with OIC. All studies reported increased direct costs related to OIC; however, estimates per country varied significantly. Direct health care costs specifically related to OIC ranged A365 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 from $1,747 to $48,782 per patient year. Per patient episode, direct costs associated with OIC ranged from $54 to $11,705. This was further supported by a large international study which reported significantly more physician visits and alternative care provider visits among patients with OIC than those without OIC. Two studies reported on a subgroup of patients with OIC who failed to respond to laxatives; these patients reported higher direct costs than patients who had a response to laxatives. Conclusions: The management of OIC is associated with potentially significant health care resource utilisation and financial burden. Patients with OIC incur higher direct health care costs than those without OIC and costs are increased further if patients with OIC have failed to respond to laxatives. There remains a paucity of data on health care resource utilisation in OIC and further research into the economic burden of OIC is needed. PGI14 Economic And Quality-Of-Life Burden Of Moderate-To-Severe Irritable Bowel Syndrome With Constipation (Ibs-C) In Spain: The Ibis-C Study Mearin F 1, Cortes X 2, Mackinnon J 3, Bertsch J 3, Fortea J 2, Tack J 4 médico Teknon, Barcelona, Spain, 2Laboratorios Almirall S. A., Barcelona, Spain, 3TFS Develop S. L., Barcelona, Spain, 4University of Leuven, Leuven, Belgium . . . . . . 1Centro Objectives: This study is the first study to assess the burden of IBS-C in 6 European countries (France, Germany, Italy, Spain, Sweden, UK). Here we present the results for Spain. Methods: Observational, retrospective-prospective (6 months each) study in patients diagnosed with moderate-to-severe IBS-C in the last five years (Rome-III criteria). Moderate-to-severe IBS-C was defined as IBS-Symptom Severity Score (IBS-SSS) ≥ 175. Quality-of-life (QoL) was assessed with EuroQoL-5D (EQ-5D) and IBS-QoL. Results: 112 patients were included (58% severe, mean age (±SD) 46.8±13.7 years, 86% female). At baseline, symptom severity (IBS-SSS; severe > 300) was 315.4±82.9; presenteeism (WPA: IBS-C questionnaire; mean [±SD] % time in week prior to inclusion): 32.3±27.2; absenteeism: 6.1±15.8; work productivity loss: 29.2±27.5; and daily activity impairment 39.6±27.2. Mean IBS-QoL was 41.0±21.2, (scale: 0-100 [best-to-worst]), and the most affected domains were “food avoidance” (mean: 60.4) and “health worry” (54.6). Mean EQ-5D was 57±21 (scale: 0-100 [worstto-best]) and 86% and 63% of patients reported moderate-to-severe problems in pain/discomfort, anxiety/depression respectively. The most prevalent symptoms were: constipation (84%), abdominal pain (80%), abdominal distention (80%) and bloating (59%). Over the year, 88% of patients consulted a primary care physician, and 82% a gastroenterologist; mean (95%CI): 3.8 and 2.3 visits, respectively. 20% of patients required emergency department visits or hospitalization (mean stay: 1.5 (0.6-2.4) days). 58% of patients underwent a diagnostic test (mean: 2.4 (1.8-2.9)). 85% of patients took pharmacological medication (80% took some pharmacological medication for their IBS-C) 30% received complementary therapies. After 6 months of follow-up, only 1 in 5 patients had no or mild symptoms. The mean (95%CI) annual direct cost for the Spanish National Health System (NHS) was 1067€ (730-1447) and the mean cost for the patient was 568€ (333-841). 13% of patients took sick leave (mean: 6.3 times; mean duration: 52 days) and 28% had productivity losses (mean: 55 hours). Mean indirect costs were 1362€ (313-2866). Total costs amounted to 2997€ (1799-4515) /year. Conclusions: Moderate-to-severe IBS-C has a great impact on patient QoL, productivity, and health care resource utilization. PGI15 Hospitalization Costs Associated With Liver Cirrhosis Silva M 1, Laires P 2, Costa M 1, Leão R 1, Roque A 1, Calinas F 1 de Santo António dos Capuchos, Lisbon, Portugal, 2Merck Sharp & Dohme, Oeiras, Portugal . . . . . costs associated with the treatment of IBS in England. Methods: Hospital Episode Statistics (HES) data for 2012-13 for all clinical commissioning groups (CCGs) in England were analysed to calculate the tariff cost of IBS. IBS diagnosis codes were included in the analysis. Prescribing data and cost (PACT) data for 2012-13 were also analysed. Results: During 2012/13 there were 1,217,993 outpatient appointments in gastroenterology and colorectal surgery specialities, with a total tariff cost of £365,868,937. Despite this, only 1,982 patients were recorded with IBS-specific codes, with a total estimated tariff cost of £812,336. In addition, 28,849 patients were recorded with IBS-related symptom codes at a cost of £11,002,874. There were also 658,698 diagnostic lower GI endoscopies at a tariff cost of £169,676,704. Of these, 323,752 (49%) had no further follow-up in secondary care in the subsequent 12 months. PACT data indicated that £44,977,959 and £25,582,752 was spent on selected laxatives and antispasmodics, respectively, commonly used to treat IBS in primary care. Conclusions: Despite being poorly clinically coded, it is clear that IBS places a significant cost burden on the NHS. Notably, 49% of patients seen for lower GI endoscopies had no further activity provided by the Hospital Provider Trust as an inpatient or outpatient over the subsequent 12 months, implying functional symptoms. Better diagnosis and subsequent management of IBS within a primary care setting may provide direct savings in the cost of IBS management. This study was financially supported by Almirall. Hospital Episode Statistics Data were provided via Harvey Walsh Ltd under commercial reuse licence. PGI17 A Cost Of Care Model For Inflammatory Bowel Disease With A Uk Nhs Perspective Miles G 1, Leonard S A 1, Ghosh N 2, Premchand P 2 1Costello Medical Consulting, Cambridge, UK, 2Barking, Havering and Redbridge University Hospitals NHS Trust, Romford, UK . . . . . Objectives: There are an estimated 620,000 patients with inflammatory bowel disease (IBD) in the UK. The rising incidence of IBD combined with its incurability has significant cost implications, with the National IBD Audit estimating that cost to the National Health Service (NHS) exceeded £1 billion in 2010. The aim of this cost of care model was to calculate the annual cost per patient of treating ulcerative colitis (UC) and Crohn’s disease (CD) from an NHS perspective, and to enable areas of potential cost savings to be explored. Methods: The cost of IBD was calculated by summing the costs of treatment, treatment side effects and disease-related complications, accounting for the proportions of patients incurring these costs. The model included detailed costs for each treatment (eg. brand of mesalazine), major side effects for each treatment (eg. pancreatitis from thiopurines) and complications (eg. pyoderma gangrenosum). Default input values for costs, the percentage of patients receiving each treatment, and the percentage of patients experiencing side effects or complications were determined from national sources and published literature. However, the model permitted the user to input local or alternative data (eg. to reflect brand preferences for oral mesalazines) and conduct scenario analyses. Results: Using default input values, the annual cost of treating any UC patient was estimated to be £3,084. For a UC patient in remission, in relapse with mild-tomoderate UC or in relapse with severe UC, annual cost per patient was estimated to be £1,693, £2,903 and £10,760, respectively. The annual cost for any CD patient was estimated to be £6,156 (£1,800 for patients in remission; £10,513 for patients in relapse). Conclusions: IBD is a costly, chronic condition and this model facilitates calculation of annual costs per UC and CD patient. The models’ customisability will help hospitals to identify areas where savings could be made. . 1Hospital Objectives: The burden and cost of liver disease is known to be substantial, but accurate data on this subject is still scarce in Portugal. So far most estimates are obtained from diagnosis-related groups (DRGs) associated with hospital financing and from expert panels. In this analysis we aimed to calculate the real costs associated with hospital admissions due to liver cirrhosis (LC) in a large hospital. Methods: All hospitalizations in the gastroenterology department from a tertiary hospital in Portugal (Centro Hospitalar Lisboa Central) during 2012 were analyzed. Patient level data was used to retrieve relevant demographic and clinical information. Costs assumed to be specific for LC admissions, namely medication, imaging and other techniques, were estimated directly. Department-specific hospital accountancy was used to include the remaining hospital costs associated with these hospitalizations. Results: A total of 644 admissions were analyzed, from which 135 (21.0%) were due to LC, corresponding to 82 patients. 74.4% (n= 61) were males, median age was 55 years. The main cause of LC was alcohol (58.5%). In the admissions due to LC, average length of stay (LoS) was 15.3 days and total mean hospitalization cost was 3,979.5€ . For all analyzed admissions, average LoS and mean total cost were 8.1 days and 2,323.8€ , respectively. Average cost for medications in admissions for LC amounted to 492.7€ , compared to 228.7€ for all admissions. Within admissions for LC, costs were not significantly different for hospitalizations with or without hepatocellular carcinoma. Conclusions: The estimated overall cost of a hospital admission for LC in a gastroenterology department was superior to the average value of hospitalizations for all causes, and approximately twice the official DRG value. This study highlights the current economic burden of liver cirrhosis faced by a tertiary hospital and the possible need to revise some of the assumptions used for financing Portuguese hospitals. PGI16 The Cost Of Irritable Bowel Syndrome (Ibs) In England Soubieres A1, Wilson PDP1, Poullis A1, Wilkins J2, Rance M3 1St George’s Healthcare NHS Trust, London, UK, 2Harvey Walsh Ltd, Runcorn, UK, 3Almirall, Uxbridge, UK Objectives: The NHS is faced with increasing cost pressures that make the efficient use of resources paramount. Patients with IBS may consume considerable NHS resource through inpatient and outpatient visits, diagnostic tests and treatment. This study aimed to estimate admission costs and primary care prescribing PGI18 Health Care Cost Associated To Constipation Predominant Irritable Bowel Syndrome In Spain Raya A 1, Barrull C 2, Roset M 2, Cortes X 1, Fortea J 1 1Almirall S. A, Barcelona, Spain,, 2IMS Health, Barcelona, Spain . . . . . Objectives: To estimate health care resource use and costs associated with the management and treatment of constipation predominant irritable bowel syndrome (IBS-C) in Spanish clinical practice. Methods: The 2011 primary care (PC) IASIST database, which includes 3,678,522 clinical charts corresponding to PC sites from the National Health System (NHS), was used to estimate the number of patients with IBS-C (identified by a combination of ICD code and prescription of IBS-C drugs) and obtain data on the number of hospitalizations (service, frequency and duration), scheduled and emergency visits to PC, referrals to specialist care, and IBS-C treatments prescribed and dispensed in retail pharmacies and their associated costs. Based on data from IASIST database and unit costs retrieved from the e-Salud database, health care costs associated with IBS-C from the NHS perspective were calculated. Results: A total of 5,649 IBS-C patients were indentified in the database, corresponding to a prevalence of 0.15% of patients attending PC. Total costs associated with IBS-C in the sample were estimated to be €4,755,725; highest costs were associated with PC visits (€2,791,725, corresponding to 58.7% of total IBS-C costs). This high cost of PC visits is explained by the high number of visits by IBS-C patients (17.4 visits annually per patient). Costs of PC visits are followed by hospitalization costs (€ 610,859, 12.8% of total cost), medical visits to specialists (€556,164, 11.7%) and emergency room visits to PC (€491,142,10.3%). Costs of pharmacological treatment (€305,707) represented only 6.4% of total IBS-C costs. Conclusions: The low prevalence of IBS-C observed may be due to the underregistration or under-diagnosis of IBS-C. Costs associated with the management of IBS-C are driven mainly by the high number of PC visits, which may be associated with the current unmet medical needs in IBS-C. PGI19 Economic Evaluation Of Various Strategies For Antiviral Therapy For Previously Treated Patients With Chronic Hcv Genotipe 1 Infection Avxentyeva M , Pyadushkina E , Derkach E V The Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia . . . . A366 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To estimate health care costs of alternative therapeutic strategies: no treatment versus peginterferon and ribavirin (PR) versus protease inhibitor (PI) boceprevir added to PR for Russian treatment experienced chronic HCV patients, who had failure previous treatment, in short- and long-term time horizon. Methods: An Excel-based model was developed to evaluate costs and outcomes of chronic HCV treated with dual therapy with PR and triple therapy with boceprevir in combination with PR. Costs for two time periods were analysed: antiviral therapy costs (0-48 weeks) and disease-progression related costs (from 48 weeks to 25 years), depending on the chosen therapy. Antiviral drug costs were calculated on the base of registered prices from the list of vital and essential drugs. Incidence of compensated and decompensated cirrhosis, hepatocellular carcinoma, liver transplant and post-liver transplant in the outcome of chronic HCV in long-term period was derived from available published data. Treatment costs of liver disease progression events were estimated according to the tariffs of the Russian health care system in 2014. Results: Boceprevir +PR compared to no treatment strategy and PR therapy was associated with more avoided liver-disease progression events at lesser costs, resulting in boceprevir+PR as the dominant treatment option in patients with chronic HCV genotype 1 nonresponders to previous treatment in Russia. Additional costs per avoided event of liver-disease progression for boceprevir plus PR and dual PR therapy were € 12,654.26 and € 45,082.82 respectively. Conclusions: Antiviral therapy with boceprevir plus PR in comparison with only PR therapy and no treatment strategy is cost effective due to reduced frequency of disease progression events and associated costs. PGI20 Lubiprostone In Chronic Idiopathic Constipation: A CostEffectiveness Analysis Pennington B 1, Marriott E 1, Sophia H 2, Lichtlen P 3, Hatswell A J 1 1BresMed, Sheffield, UK, 2Sucampo Pharma Europe, Oxfordshire, UK, 3Sucampo, Zug, Switzerland . . . . . . Objectives: The clinical efficacy of lubiprostone in chronic idiopathic constipation has been demonstrated in three randomised clinical trials (RCTs). This analysis assesses the cost effectiveness of lubiprostone compared to prucalopride (the current standard of care), placebo and immediate referral to secondary care from the perspective of the UK National Health Service (NHS). Methods: A statetransition model was constructed to represent the treatment pathway for chronic idiopathic constipation from an NHS perspective with a 1-year time horizon. The model considered treatment continuation rules, at Week 2 for lubiprostone and Week 4 for prucalopride. Clinical data were taken from RCTs and an indirect comparison with prucalopride. Long-term duration of treatment was estimated by fitting curves to open-label study data. Costs included drug costs, medical resource use, and the cost of the pathway including primary and secondary care, obtained from published sources. Quality of life was modelled according to whether constipation was resolved or unresolved, with values taken from a large published study (n= 1,200). Results: Compared to placebo, lubiprostone delayed referral to secondary care and improved quality of life, but resulted in increased costs due to treatment, the incremental cost-effectiveness ratio was £2,924. Lubiprostone and prucalopride were found to have similar efficacy, with lubiprostone generating an additional 0.0014 QALYs in the base case. The cost per day for lubiprostone is lower than for prucalopride; leading to lower total costs (£1,596 v £1,655), and meaning lubiprostone dominated prucalopride (lower cost and higher QALYs). Probabilistic sensitivity analysis showed lubiprostone to have an 84% chance of being the most cost-effective treatment at a threshold of £20,000 per QALY. Conclusions: Treatment with lubiprostone provides substantial value to both patients and the NHS, being highly cost effective compared to placebo and dominant compared to the current standard of care. PGI21 Cost-Effectiveness Of Linaclotide: A Valuable Option In The Treatment Of Irritable Bowel Syndrome Almeida J , Vandewalle B , Félix J Exigo Consultores, Lisbon, Portugal . . . Objectives: Constipation-predominant irritable bowel syndrome (IBS-C) affects more than 2% of the population carrying a heavy burden in developed countries and leading to significant losses in patients’ quality of life. Treatment goals are to provide relief of abdominal pain, restore the bowel transit and alleviate associated symptoms. Linaclotide has been shown to significantly improve abdominal and bowel symptoms in two phase III trials being the only EMA approved therapy indicated for the treatment of IBS-C. Therefore, this study aimed to perform a cost-effectiveness analysis comparing linaclotide with no active treatment for the treatment of IBS-C from the Portuguese societal perspective. Methods: A Markov model was developed to simulate the natural course and treatment of the disease. Patient-level satisfaction data from linaclotide’s randomised clinical trials was used to define 4 health states: ‘Not Satisfied’, ‘Moderately Satisfied’, ‘Satisfied’ and ‘Dead’. These data were linked to utility scores based on patients’ responses to the EuroQol-5D questionnaire. Transitions between health states were assumed to occur every 4 weeks, with probabilities derived from observed efficacy data up to 20-weeks. Extrapolation beyond this period was based on last observation carried forward data. Effectiveness was measured in quality-adjusted life years (QALY). Only direct costs were incorporated. Resource utilization was estimated from a literature review. Unit costs came from official Portuguese databases and pricing lists. Time horizon was fixed at 10 years. Probabilistic sensitivity analysis was conducted with Monte Carlo simulations. Results: A mean gain of 0.22 QALY (95%CI: [0.1; 0.35]) was estimated for each patient treated with linaclotide versus no treatment. Additionally, linaclotide utilization led to an overall average cost reduction of 402€ (95%CI: [-1,735; 536]) thus representing a dominant option. Conclusions: When compared with no active treatment, linaclotide is a cost-saving and more effective therapeutic option for the treatment of IBS-C from the Portuguese societal perspective. PGI22 Cost-Effectiveness Analysis Of A Personalized Therapy For Genotype 1, Naive, Chronic Hepatitis C Patients In Italy Iannazzo S 1, Colombatto P 2, Bonino F 2, Brunetto M R 2 1SIHS Health Economics Consulting, Torino, Italy, 2University Hospital (AOUP), Cisanello (Pi), Italy . . . . . Objectives: Rapid Virologic Response (RVR) is currently used as the best predictor of Sustained Virologic Response (SVR) with dual therapy (DT) in genotype-1 chronic hepatitis C (G1-CHC), to optimize the adoption of the triple therapy (TT) with direct-acting antivirals (boceprevir: BOC and telaprevir: TVR). Bio-mathematical modelling of viral dynamics during DT has potentially higher accuracy than RVR in the identification of SVR patients. The objective of this study was to analyse the cost-effectiveness profile of a personalized anti-HCV therapy in F0-F2 G1-CHC based on a bio-mathematical model (model-guided strategy: MG) rather than on the RVR (guideline-guided strategy: GG). Methods: A deterministic bio-mathematical model of the infected cell dynamics was validated in a cohort of 135 G1-CHC patients treated with DT at the University Hospital in Pisa, Italy. A lifetime health economic (HE) model was then developed to compare MG and GG strategies in the perspective of the Italian National Health Service. The model was alimented with observed data in the validation cohort, clinical variables and economic data available in the literature. A 3.5% discount rate was applied to outcomes and costs. One-way and probabilistic sensitivity analyses were run. Results: the outcomes with MG- and GG-strategy were 19.1-19.4 and 18.9-19.3 qualityadjusted-life-years (QALY). Total per-patient lifetime costs were € 25,200-€ 26,000 with MG-strategy and € 28,800-€ 29,900 with GG-strategy. When comparing MGwith GG-strategy the former resulted more effective and less costly, being defined as dominant. Conclusions: The adoption of a SVR predictive criterion based on a bio-mathematical model, has the potential to improve the cost-effectiveness of a personalized anti-HCV therapy, allowing a more accurate identification of patients who can be effectively treated with DT and reserving high-cost BOC- and TVR-based TT for those who really need it. PGI23 Sofosbuvir For The Treatment Of Chronic Hepatitis C: A Comprehensive Cost-Effectiveness Analysis Across Hcv Genotypes, Pretreatment Conditions And Hiv Co-Infection Silva M 1, Félix J 1, Ferreira D 1, Vandewalle B 1, Guerra I 2, Cure S 2, Aldir I 3, Carvalho A 4, Macedo G 5, Marinho R T 6, Pedroto I 7, Ramalho F 6 Consultores, Alhos Vedros, Portugal, 2OptumInsight, Uxbridge, UK, 3Hospital Egas Moniz, Centro Hospitalar de Lisboa Ocidental, Lisboa, Portugal, 4Centro Hospitalar e Universitário de Coimbra, Coimbra, Portugal, 5Hospital de São João, Porto, Portugal, 6Centro Hospitalar Lisboa Norte. Hospital de Santa Maria, Lisboa, Portugal, 7Centro Hospitalar do Porto, Porto, Portugal . . . . . . . . . . . . . 1Exigo Objectives: Pegylated interferon-α plus ribavirin (PR) has been the standard of care for Chronic Hepatitis C (CHC). Therapy adding boceprevir (BOC) or telaprevir (TVR) to PR in HCV genotype-1 patients has failed to achieve generalized market acceptance, in part due to the approval of newer more efficacious and safer options like sofosbuvir (SOF), a new pan-genotypic RNA-polymerase inhibitor. Objective: To assess the cost-effectiveness of sofosbuvir based therapy for CHC across HCV genotypes, pretreatment conditions and HIV co-infection in Portugal. Methods: Costs and effectiveness were estimated based on discrete-time Markov-type model for CHC evolution accounting for different subpopulations in terms of HCV genotype, fibrosis progression, HIV co-infection status and previous treatment experience. The model incorporates 13 health states: 5 Metavir score, 2 SVR (with and without cirrhosis) and 3 advanced liver disease states (decompensated cirrhosis, hepatocellular carcinoma, liver transplant). Results are expressed in incremental costs per life year (LY) and quality-adjusted life year (QALY). Results: Overall sofosbuvircontaining regimens are expected to result in an increment of 3.49 LY (3.05 QALY) after weighting for the different subpopulations assuming Portuguese epidemiology. The corresponding weighted incremental cost-effectiveness ratios (ICER) was 14,649€ /LY (16,720€ /QALY). In the comparison against the BOC and TVR containing regimens (genotype-1), ICER of 10,675€ /LY (12,238€ /QALY) and 14,618€ /LY (16,495€ / QALY) were obtained, respectively. For HIV co-infected and PegIFN eligible/tolerant patients, estimated ICER varied between 6,463€ /LY (6,902€ /QALY) and 21,281€ /LY (28,245€/QALY), for G1 and G2, respectively, when comparing against treatment with PR. Additionally, in HCV/HIV coinfected patients ineligible/intolerant to PegIFN, the comparison against lack-of-therapy resulted in ICER of 15,656€ /LY (17,756€ /QALY) and 12,915€ /LY (19,077€ /QALY), for HCV G1 and G3, respectively. Conclusions: Sofosbuvir-containing regimens for the treatment of adult CHC patients, irrespective of HIV co-infection status, are expected to result in significant health gains at an incremental cost within the range of European Health Authorities acceptability. PGI24 Cost-Effectiveness Analysis Of Antiviral Pharmacotherapies For Treatment Of Chronic Hepatitis C Virus Infection In Russia Pyadushkina E 1, Avxentyeva M 1, Omelyanovsky V V 1, Treur M 2, Westerhout K Y 2 1The Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia, 2Pharmerit International, Rotterdam, The Netherlands . . . . . . . Objectives: To evaluate the clinical and economic expediency of adding simeprevir (SMV) to pegylated-interferon and ribavirin (PR) versus PR only, or triple therapies with PR plus boceprevir (BOC) or telaprevir (TVR) for treatment of genotype 1 chronic hepatitis C (CHC) virus infection in patients who have failed previous therapy with interferon and ribavirin in Russia. Methods: A Markov model developed earlier was adapted to the Russian settings. The analysis consists of 2 time periods: 1st the antiviral therapy (AVT) phase and 24 week follow-up (0-72 weeks) and 2ndthe disease progression of CHC (72 weeks - lifetime). Incidences of disease progression health-states (decompensated cirrhosis, hepatocellular carcinoma, liver transplant, post-liver transplant and death, associated with high health care costs, high mortality rates) based on available published data. Costs and outcomes (life years, LY) were analyzed for treatment experienced CHC patients from Russian health care system perspective. The incremental cost-effectiveness ratio (ICER) per A367 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 LY was calculated. Sensitivity analyses were performed. Results: Model results for treatment-experienced patients show that SMV is the dominant treatment compared to TVR+PR and BOC+PR therapies as more total LYs are saved and less costs accrued. ICER of SMV+PR vs PR was € 22,967 per LY. Results were robust in sensitivity analyses. Conclusions: SMV + PR is cost-effective compared to dual PR-therapy and appears the dominant strategy compared to other PI (telaprevir, boceprevir) for CHC treatment-experienced patients in Russia. PGI25 Cost-Effectiveness Analysis Of Triple Therapy With Peginterferon, Ribavirin, And Boceprevir For The Treatment Of Chronic Hepatitis C Virus Genotype 1 With Severe Fibrosis Under “Real-Life” Conditions Ruiz-Antorán B 1, Pascasio J M 2, Gea F 3, Barcena R 4, Larrubia J 5, Pérez Álvarez R 6, Sousa J 2, Romero Gómez M 7, Solá R 8, de la Revilla J 9, Crespo J 10, Navarro J 11, Arenas J 12, Delgado M 13, Fernández Rodríguez C 14, Planas R15, Buti M 16, Forns X 17, Calleja J L 9 1University Hospital Puerta de Hierro, Madrid, Spain, 2Unidad de Gestión Clínica de Enfermedades Digestivas, IBIS (Instituto de Biomedicina de Sevilla (IBIS), H. Virgen del Rocío, Sevilla, Spain, 3University Hospital La Paz, Madrid, Spain, 4University Hospital Ramón y Cajal, Madrid, Spain, 5Guadalajara University Hospital. University of Alcalá. Spain, Guadalajara, Spain, 6Asturias Central Hospital University, Oviedo, Spain, 7Valme University Hospital, University of Seville, Sevilla, Spain, 8Hospital del Mar, Universitat Autònoma de Barcelona,, Barcelona, Spain, 9University Hospital Puerta de Hierro, Madrid, Spain, 10Hospital Universitario Marqués de Valdecilla, Insituto de Investigación Valdecilla. IDIVAL, Santander, Spain, 11Hospital Costa del Sol. Marbella, Malaga, Spain, 12University Hospital Donostia, San Sebastian, Spain, 13University Hospital La Coruña, La Coruña, Spain, 14Hospital Universitario Fundación Alcorcon, Madrid, Spain, 15Hospital Germans Trias i Pujol, Barcelona, Spain, 16Vall d’Hebron University Hospital, Barcelona, Spain, 17Hospital Clinic, Barcelona, Spain . . . . . . . . . . . . . . . . . . . . . Objectives: Studies based on the data of clinical trials have proved that the triple therapy for hepatitis C is cost effective. This study we assessed the cost-effectiveness of triple therapy in treatment of Chronic Hepatitis C with Severe Fibrosis under “real-life” conditions. Methods: The analysis was conducted from the data included in the prospective, multicentre, Spanish registry that includes patients with HCVgenotype-1 infection, who had severe fibrosis and were treated with triple therapy (peginterferon alfa-2a or 2b, ribavirin, and boceprevir). The cost effectiveness analysis of antiviral treatment includes the costs of antiviral treatment, of concomitant treatments and costs in relation to health care resources (in relation to clinical practice and the adverse events). Results: 170 patients were included. 68.2% male, mean age of 53 (29-76) years. 80% had received prior treatment. 36.5% of patients reported at least one SAEs. The overall percentage of patients with SVRw12 was 46.5%. The cost of triple therapy represented a total of 4,916,652.84€ , the pharmacological cost (triple therapy+concomitant treatment) involved a total cost of 5,161,168.98€ . The consumption of health resources generated an additional cost of 240,000 € , which is about 1,500€ /patient. The total cost per patient cured was 70,262€ . This cost varies greatly based on different baseline characteristics of the patients, with significant differences in patients with albumin <3.5, 120,597€; prior null response 120,727€ and platelets <90,000,104,464€. Conclusions: The current scenario of the hepatitis C treatment is changing. Triple therapy is more costly for patients with severe fibrosis and predictors of poor response. However, keeping in mind that the timeframe for the release of IFN-free regimens remains uncertain and considered that the actual access to the new DAA in the real world setting could be delayed, boceprevir could remain as an option for patients with intact liver function and a high unmet medical need, regardless of the degree of liver fibrosis, in locations where a delay in the access to the newer therapies is foreseen and hepatic transplant would not be readily available. PGI26 The Cost Effectiveness Analysis Of The Oral Anti-Viral Treatments Alternatives For The Chronic Hepatitis B In Turkey Objectives: Prolonged exposure to CNI-based immunosuppressant therapy (IS) in liver transplant (LTx) recipients is associated with long-term complications. In the global registration trial H2304, patients receiving everolimus + reduced tacrolimus (EVR + reduced TAC) demonstrated non-inferior efficacy and superior renal function at Month 12 that was sustained at 36 months compared to tacrolimus alone (TAC). A peer-reviewed Markov model has been adapted to the Italian setting to explore the cost-effectiveness of EVR + reduced TAC compared to TAC, in de novo liver-recipients. Methods: The model estimates long-term outcomes associated with IS following LTx along two independent pathways: 1. liverrelated (acute rejection, hepatocellular carcinoma, hepatitis C [HCV] recurrence, graft loss); 2. kidney-related (chronic kidney disease, dialysis, renal transplantation) and death. All patients, stratified by liver diagnosis, entered the model at time of LTx and followed both pathways, allowing for multiple combinations of liver and kidney health states. The lifetime model used an annual cycle length except for the 1styear post LTx (quarterly). Efficacy and safety of IS strategies were assessed through the risk of acute rejection, change in renal function, HCV fibrosis progression and frequency of adverse events. Utilities and costs were assigned to each renal and liver state. Subgroup and sensitivity analyses were performed. Results: With a mean life expectancy of 18 years, the model predicts patients treated with EVR + reduced TAC gain on average 1.84 years of life and 1.55 QALYs vs. TAC. The risk of acute rejection was reduced by 20%. The incremental cost of EVR + TAC was € 38,884 per life year gained and € 46,103 per QALY gained vs. TAC. Conclusions: This model shows a strategy of EVR + reduced TAC postLTx improves survival and quality of life. Higher treatment costs are offset by slower progression of renal deterioration predicted in the first 10 years and fewer lifetime liver complications. PGI28 Cost-Utility Analysis Of Screening Strategies For Nonalcoholic Steatohepatitis Zhang E 1, Wartelle-Bladou C 1, Lepanto L 1, Lachaine J 2, Cloutier G 1, Tang A 1 Hospital, Montreal, QC, Canada, 2University of Montreal, Montreal, QC, Canada . . . . . . 1CHUM-St-Luc Objectives: Nonalcoholic fatty liver disease (NAFLD) is the most common liver condition in Western countries. To date, no studies have examined the costeffectiveness of screening for nonalcoholic steatohepatitis (NASH), its advanced form. Methods: We performed a cost-utility analysis of annual non-invasive screening strategies using a third-party payer perspective in a general population and compared it to screening in a high-risk obese or diabetic population. Screening algorithms involved well-studied non-invasive techniques including NAFLD fibrosis score, ultrasound transient elastography (TE), and ultrasound acoustic radiation force impulse (ARFI) imaging for detecting advanced fibrosis (≥ F3); and plasma cytokeratin-18 for NASH detection. Liver biopsy and magnetic resonance elastography (MRE) were compared as confirmation methods. Model uncertainties were tested using sensitivity analyses. Canadian dollar costs were adjusted for inflation and discounted at 5%. Incremental cost-effectiveness ratio (ICER) of $C50,000 per quality-adjusted life year (QALY) or less was considered cost-effective. Results: Compared with no screening, screening with NAFLD fibrosis score/ TE/CK-18 algorithm with MRE as confirmation for advanced fibrosis had an ICER of $C26,143 per QALY gained. Screening in high-risk obese or diabetic populations was more cost-effective, with an ICER of $C9,051 and $C7,991 per QALY gained respectively. Screening algorithms with liver biopsy confirmation were not found to be cost-effective. Sensitivity analyses revealed that the screening starting age, the annual transition probability from simple steatosis to NASH, and the cost of a TE exam had the most impact on the results. Conclusions: Our model suggests that annual NASH screening in high-risk obese or diabetic populations can be cost-effective. Kockaya G 1, Kose A 1, Dalgic C 1, Yenilmez F B 2, Ozdemir O 3, Kucuksayrac E 1, Tayfun K 1 1Gilead Science, Istanbul, Turkey, 2Hacettepe University, Ankara, Turkey, 3Yorum Consultancy, ISTANBUL, Turkey PGI29 The Cost-Effectiveness Of Sofosbuvir And Ribavirin Treatment In Hcv-Infected Patients Listed For Liver Transplantation Objectives: The aim of this study is to compare the cost effectiveness of oral antiviral treatment strategies in CHB for Turkey using lamuvidine, telbuvidine, entecavir, and tenofovir as medications. Methods: The analysis was conducted using Markov model. Inadequate response or resistance after receiving 12 months of the treatment with entecavir and telbivudine were compared to the results found from switching from entecavir to tenofovir or from switching from telbuvidine to tenofovir. In additional, inadequate response or resistance after receiving 6 months of the treatment for lamivudine was compared to the results found from switching from lamivudine to tenofovir. The model duration was constructed to evaluate a treatment strategy duration of 40 years. Years of Potential Life Lost (YPLL) was used as the health outcome. An incremental cost-effectiveness ratio (ICER) analysis of the results was conducted. Results: In a life time period, the lowest YPLL and the cost of treatment were calculated for the NS. Tenofovir treatment with 0.54 years and 37,213.75 TL. Depending on the results, the lowest YPLL and the cost of treatment were served by NS. Tenofovir treatment with 2.06 years and 276,468.45 TL. The highest YPLL and the cost of treatment were calculated for the NS. The ICER analysis found that all treatment strategies were dominated by NS. Tenofovir and S. Entecavir. Only these two treatment strategies were found to be cost-effective. Conclusions: The cost of providing 40 years of treatment for patients with CHB, if reimbursement agencies includes Tenofovir and Entecavir as part of the first line treatment strategy for CHB, it can be expected that this approach would result in a positive contribution to the health budget in Turkey. Cortesi P A 1, Mantovani L 2, Ciaccio A 1, Rota M 1, Cesana G 1, Strazzabosco M 1, Belli L S 3 1University of Milano - Bicocca, Monza, Italy, 2Federico II University of Naples, Naples, Italy, 3Niguarda Hospital, Milan, Italy . . . . . . . . PGI27 Cost-Effectiveness Of Everolimus Plus Reduced Tacrolimus In De Novo Liver-Recipients In The Italian Setting Bianic F 1, Campbell R 2, De Simone P 3, Roccia A 4, Gregson G 5 1OptumInsight, Paris, France, 2OptumInsight, Uxbridge, UK, 3University of Pisa Medical School Hospital, Pisa, Italy, 4Novartis Farma, Origgio, Italy, 5Novartis Pharma AG, Basel, Switzerland . . . . . . . . . . . . . . Objectives: Sofosbuvir in combination with ribavirin (SOF/RBV) is a novel treatment able to suppress HCV viremia when applied to HCV patients listed for transplant, preventing HCV recurrence. Aim of this study was to assess the costeffectiveness of this regimen in HCV patients listed for transplant for cirrhosis (HCVcirrhosis) or for hepatocellular carcinoma (HCV-HCC). Methods: a semi-Markov model was developed. The model simulates the progression of HCV-cirrhosis or HCV-HCC patients from the time of listing until death considering the risk of HCV recurrence post-transplant. The model compared 2 different strategies: 1) SOF/RBV up to a maximum of 24 weeks or until OLT if performed before the 24th week, 2) No antiviral treatment. The model estimated the costs related to the treatment with SOF/RBV, the costs associated to each health state, the life-years (LYSs), the quality-adjusted life-years (QALYs), and the incremental cost-effectiveness ratio (ICER) expressed as € per QALY gained. The analysis was performed from the Italian National Health System perspective with a lifetime time horizon and one-month Markov cycles. Future costs and clinical benefits, expressed as QALYs, were discounted at 3% per year. Results: in the base-case analysis the ICER for 24 weeks of SOF/RBVR was € 30,518 per QALY gained in HCV-cirrhosis patients and € 41,610 in HCV-HCC patients. The reliability of our results was confirmed by the one way sensitivity-analysis and by the cost-effectiveness acceptability curve. Further, SOF/ RBV cost-effectiveness was clearly sensitive to the duration of treatment; assuming 12 weeks SOF/RBV treatment duration, the ICER decreased to € 19,317 in HCVCirrhosis and € 29,540 in HCV-HCC. Conclusions: our study shows that treating patients with HCV-cirrhosis or HCV-HCC listed for transplant with SOF/RBV is costeffective and may become the new standard of care for these patients. However a well-defined prospective study is needed to confirm the value of the parameters assumed in the model and the results. A368 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PGI30 Cost-Utility Analysis Of Sofosbuvir For Treatment Of Genotype2 Chronic Hepatitis C In Japan Igarashi A 1, Tang W 1, Cure S 2, Guerra I 2, Lopresti M 3, Tsutani K 1 of Tokyo, Graduate School of Pharmaceutical Sciences, Tokyo, Japan, 2OptumInsight, Uxbridge, UK, 3Junicon Japan Inc., Minato Tokyo, Japan . . . . . . 1University Objectives: To conduct a cost-utility analysis of sofosbuvir for genotype 2 chronic hepatitis C virus (HCV) infection in Japan. Methods: The Markov-model, “Sofosbuvir cost-effectiveness model”, which was constructed originally for similar study in UK, was modified and used for this analysis, while imputed data were replaced with Japanese data, as far as possible. Various health states, such as noncirrhotic hepatitis, sustained virological response (SVR), compensated cirrhosis, decompensated cirrhosis and hepatocellular carcinoma were incorporated to the model. Analyses were conducted for 4 scenarios, classified by treatment history (naive/experienced) and eligibility for interferon. Peg-interferon alpha with ribavirin was set as a comparator for those who were eligible for interferon. No treatment was selected for those who were not eligible for interferon. Probability of SVR was derived from clinical trials conducted in Japan. Other transition probabilities and utility scores of each health state were obtained from published data in Japan. Cost data for interferon-alpha and ribavirin were derived from national drug tariff (2014). For sofosbuvir, average European price was adopted since it was not yet approved in Japan. Other cost data, such as costs related to health states, were mainly obtained from claim data, provided by JMDC (Japan Medical Data Center). Inc. Time-horizon was set to lifetime. Costs and outcomes were discounted with 2% per annum, according to Japanese guideline. Results: For interferon-unsuitable patients, sofosbuvir was dominant to no-treatment. Sofosbuvir would save overall costs for JPY990,000 (USD9,900, JPY100= USD1) and prolonged 6.20QALY for treatment naive patients. It would save JPY 837,000 and prolonged 6.08QALY for treatment experienced group, For interferon-suitable patients, sofosbuvir would increase overall costs for JPY3,270,000 and prolonged 2.23QALY for treatment-naives. It would increase JPY1,551,000 and prolonged 2.36QALY for treatment-experienced. ICER were JPY1,470,000 and JPY657,000 per QALY gained, respectively. Conclusions: Sofosbuvir was considered to be cost-effective for treatment of genotype-2 HCV patients in Japan. PGI31 Cost-Utility Analysis Of Fidaxomicin Compared To Vancomycin In The Management Of Severe Clostridium Difficile Infection In Poland Petryszyn P , Well A Wroclaw University of Medicine, Wroclaw, Poland . . Objectives: In recent years a number of infections caused by Clostridium difficile has been significantly increasing. In Poland oral metronidazole constitutes the therapy of choice of non-severe infection and first-recurrence, while oral vancomycin is recommended to be given in case of severe disease and subsequent recurrences. Fidaxomicin is a novel treatment for Clostridium difficile infections (CDI). The aim of this study was to perform a cost-utility analysis of fidaxomicin for the treatment of severe CDI compared to vancomycin. Methods: A meta-analysis of two randomized clinical trials phase III comparing oral fidaxomicin and oral vancomycin in CDI was conducted. A Markov model was used to determine the cost-utility of fidaxomicin in patients with severe CDI. The cycle length was 10 days and the time horizon was 1 year. The patient entered the model in the severe CDI health state and was given either fidaxomycin or vancomycin for 10 days. The analysis was performed from the third-party payer perspective – the Polish National Health Fund. Only direct health care costs (drug costs, hospitalization) were included. Given the lack of formal utility measures for CDI, the utilities for the alternative health states described in the literature were adapted. Results: In the base case, fidaxomicin was dominant compared to vancomycin, resulting in cost savings of PLN 905 and an incremental QALY gain of 0.015. Fidaxomicin was associated with higher cost savings (PLN 30,883) assuming that patients with severe CDI would be hospitalized at intensive care unit. One-way sensitivity analyses revealed that fidaxomicin remained dominant even if considering marginal values of both antibiotics’ acquisition cost. Conclusions: Fidaxomicin was dominant compared to vancomycin, generating additional QALYs with cost-savings in severe CDI patients in Poland. PGI32 Economic Evaluation Studies In Gastroenterology In Brazil: A Systematic Review Haddad L 1, Decimoni T 2, Turri A 1, Leandro R 2, Soarez P 2 Paulo University, Sao Paulo, Brazil, 2Sao Paulo University, São Paulo, Brazil . . . . . 1Sao Objectives: The aim of this study was to systematically review the economic assessment studies carried out in Brazil, published between January 1980 and December 2013, assessing the technologies studied, study types, the and temporal evolution and quality. Methods: We systematically searched in MEDLINE (PubMed), EMBASE, LILACS, SciELO, NHS EED, HTA Database (CRD), BVS ECOS, SCOPUS, Web of Science, and SISREBRATS. We selected partial and full economic evaluation studies in gastroenterology, where at least one of the authors was affiliated to a Brazilian institution. Two authors performed study selection and data extraction independently. Disagreements were resolved through discussion or through consultation with a third reviewer. The study characteristics were summarized in figures and summary tables. Results: Forty studies were included. The first studies were published in the 80s, but most occurred after 2000, with greater frequency in the last 4 years. Seventeen economic evaluations were incomplete (42.5%) and 23 complete (57.5%). In the 23 complete reviews, 11 (47.8%) studies were cost-utility analysis, 7 (30.4%) were cost-effectiveness analysis, 4 (17.4%) costconsequence analysis, and 1 (4.3%) cost-minimization analysis. The type of technology evaluated was mainly medications in 25 studies (62.5%), 7 (17.5%) medical and surgical procedures, 3 (7.5%) medical and hospital equipment, 1 (2.5%) vaccines and 4 (10%) evaluated more than one type of technology. When classified by disease, 22 (55%) were studies on viral hepatitis, and in its most published after the year 2010 (63.4%). Five studies were related to digestive cancers and other included peptic diseases, hernias and other. Conclusions: There was a considerable increase in publications of economic evaluations in Gastroenterology in Brazil, being mostly studies of drugs for treatment of viral hepatitis. The high cost of these treatments and increased of lawsuits seem to account for this increase. PGI33 Estimating The Cost Of Liver Transplantation In Patients Diagnosed With Chronic Hepatitis C And B In The Uk Singh J , Longworth L Brunel University, Uxbridge, UK . . Objectives: Liver transplantation is an effective treatment option for end-stage liver disease and acute liver failure, including patients with hepatitis C (HCV) and hepatitis B (HBV). Recent health technology assessments of treatments for HCV and HBV have relied on data from a large cohort study of transplanted patients to inform estimates of costs of liver tranplantations; however this was conducted in the 1990s. The overall aim of this study was to estimate the current cost of liver transplant for patients with HCV and HBV in the UK. Methods: Historical summary data from the original cohort study were updated to reflect current unit costs and key changes in clinical practice. Semi-structured interviews were conducted with experts and a computer-based user-interface was developed to elicit estimates of key resource use items. Uncertainty in the experts’ estimates was captured by eliciting probability distributions for each item from each expert. Updated unit costs were obtained from national sources. Data were analysed by phase of the transplant procedure. Results: The expert elicitation exercise included two hepatologists, three transplant surgeons and one liver transplant coordinator. Few patients with HBV are now being transplanted due to improvements in anti-viral treatments. Mean total costs for patients with HCV were £18,055 pre-transplantation, £64,452 during the transplant phase and £36,009 in two years post-transplant. The average cost per transplanted patient with HCV from assessment to two years post-transplant is £111,810. Conclusions: There have been some significant changes in clinical practice since the original study such as change in standard immunosuppressant therapy, more patients with co-morbidities being placed on the transplant waiting list, increased use of sub-optimal organs and reluctance to re-transplant patients with graft failure and recurrence of HCV. Gastrointestinal Disorders – Patient-Reported Outcomes & Patient Preference Studies PGI34 Adherence Rates For Peginterferon + Ribavirin Compared With Telaprevir + Peginterferon + Ribavirin In Medicaid And Commercial Patients Treated For Chronic Hepatitis C Samp J C , Walker D , Manthena S , Juday T AbbVie, North Chicago, IL, USA . . . . . Objectives: Prior to approval of telaprevir (TPV), the treatment for chronic hepatitis C virus (HCV) included peginterferon (P) weekly injections and ribavirin (R) orally twice daily. In 2011, TPV was approved for coadministration with P+R during the first 12 weeks. Though TPV improved viral clearance, it also increased the treatment complexity by 2 pills given 3 times a day. The impact of increased regimen complexity on adherence is not well understood. This study compared treatment adherence over 24 weeks in HCV patients treated with TPV+PR compared to those on PR. Methods: Large US commercial and Medicaid health insurance claims databases were used to identify HCV patients initiating treatment with PR (pre-TPV [2007 to 2009]) or TPV+PR (post-TPV [2011 to 2013]). The index date was the date of HCV treatment initiation. Adherence was measured by medication possession ratio for all patients at 4 week intervals thru 24 weeks. Regression analyses adjusted for age, sex, comorbidities, liver disease severity, and pill count prior to HCV treatment. Results: The study included 7,601 and 1,487 treated HCV patients in the commercial and Medicaid databases. Unadjusted and adjusted adherence was high for both cohorts throughout the study period (> 88% for Medicaid and > 82% for the commercial at 24 weeks). Adherence was not significantly different between the PR and T+PR cohorts at any time point in the Medicaid patients (88.9% [TPV+PR] and 90.5% [PR] at 24 weeks). Adherence was also similar between the cohorts in the commercial patients (82.7% [TPV+PR] and 83.2% [PR] at 24 weeks) but was statistically different at weeks 8 and 12, though not clinically meaningful. Age was the only factor consistently associated with adherence. Conclusions: Among HCV patients, adherence rates were high and were similar between the cohorts, despite the higher daily pill count for patients on TPV+PR. PGI35 Quality Of Life Of Diarrheal Children And Caregivers In Thailand Rochanathimoke O 1, Postma M 2, Thavorncharoensap M 1, Riewpaiboon A 1, Thinyounyong W 3 1Faculty of Pharmacy, Mahidol University, Bangkok, Thailand, 2Unit of PharmacoEpidemiology & PharmacoEconomics (PE2), Department of Pharmacy, University of Groningen, Groningen, The Netherlands, 3Phetchabun Provincial Public Health Office, Phetchbun, Thailand . . . . . Objectives: To estimate the utility scores for diarrheal children aged under 5 years and their caregivers and to identify the influencing factors which affected on these. Methods: Hospitalized diarrheal children aged between 2 months and 5 years and their caregivers at were recruited in this cross-sectional study at three hospitals in Phetchabun province. The EQ-5D instrument was used to collect the quality of life (QoL) data at the first date of admission. Quality of life of diarrheal children was measured as proxy report from caregiver while QoL of caregiver was measured as self-report. The raw data was converted to utility values using the Thai algorithm. The clinical severity of diarrheal children was rated using the Vesikari clinical severity scoring system. Stepwise multivariate linear regression was applied to explore the impact of the various factors on the utility value of children and A369 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 caregivers. Results: 468 children and caregivers were included in this study. Mean children’s age was 1.77 years. The caregivers rated the mean child’s utility and themselves as 0.604 (95%CI: 0.592, 0.615) and 0.618 (95%CI: 0.606, 0.629), respectively. Mainly domains of diarrheal children were affected as pain/discomfort and anxiety/ depression similarly to their caregivers. On multivariate regression analysis, factors which affected the children’s utility significantly were Vesikari score, child’s age and child’s gender. Next, reduced QoL of their caregivers related to caregiver’s gender and Vesikari score. Conclusions: Diarrhea had an impact on quality of life of children and their caregivers. It appeared that QoL of children with diarrhea was impacted worse in girls, those with high severity score and was associated with age. In addition, the diarrheal severity and female gender reduced the impact of diarrhea on QoL of caregivers. These results can be useful to evaluate the cost-effectiveness of vaccines against diarrhea such as rotavirus vaccines. PGI36 How Does Non-Malignant Opioid Induced Constipation (Oic) Impact Health State Utility? Lawson R 1, Marsh K 2, Altincatal A 3, King F 4 1AstraZeneca, Cheshire, UK, 2Evidera, London, UK, 3Evidera, Lexington, MA, USA, 4AstraZeneca, Gaithersberg, MD, USA . . . . Objectives: Little is known about the impact of OIC and treatments for OIC on health state utility. Studies often focus on collecting data on changes in OIC status. The objective of this paper is to examine if the utility impact of treatment is driven by change in OIC status, and what the magnitude of the change in utility associated with changes in OIC status is. Methods: 1352 patients with non-malignant OIC were allocated to one of two, phase III, 12 week randomised controlled trials to study naloxegol. These trials were pooled and prospective analyses on these data were undertaken. Both trials collected the three level EQ-5D at baseline, week 4 and week 12. EQ-5D scores were converted into estimates of utility using a tariff generated based on UK general population preferences. A repeated measure mixed model (RMMM) regression analysis was conducted to identify the impact of the following factors on utility: age, gender, race, BMI, duration of opioid use, treatment (naloxegol 12.5mg, 25mg or placebo), baseline utility and OIC status (OIC or non-OIC). Results: Baseline utility across all patients was 0.559. The regression demonstrated that baseline utility score (β = -0.532, SE= 0.023) and OIC status (β = 0.032, SE= 0.012) were the only significant predictors of change in utility score (p< 0.0001and p= 0.008 respectively). Further univariate analyses examined the effect of OIC status in patient subgroups that had different experiences of laxative treatment. OIC status had an increased and meaningful impact on patients who had previously responded inadequately to laxatives. Conclusions: OIC status is a significant factor on the impact of treatment on patient’s utility. Furthermore the impact of OIC status is increased in patients who had previously responded inadequately to laxatives. PGI37 Mapping May Cause Straining: The Inconsistent Relationship Between A Disease- Specific Questionnaire (Pac-Qol) And Eq-5d Mapping In Constipation Vegter S 1, Hatswell A J 2 1Vegter Health Economic Research, Amersfoort, The Netherlands, 2BresMed, Sheffield, UK . . . Objectives: A recent double-blind, placebo-controlled clinical study with lubiprostone in opioid-induced constipation (OIC), OBD-1033, included the EQ-5D generic quality-of-life instrument, and the PAC-QOL, a constipation-specific disease measure. This study calculated utility values for patients with OIC using the direct EQ-5D responses, and compared the resulting utilities to those calculated from a published mapping formula between the PAC-QOL and EQ-5D that was derived in chronic idiopathic constipation. Methods: EQ-5D responses from OBD-1033 were converted to utilities using the EQ-5D UK value set. These were compared with utilities generated with the published mapping algorithm. Following this step, an attempt was made to map the PAC-QOL to the EQ-5D in OIC. The root mean squared error (RMSE), adjusted R2, and predicted/observed plots were used to assess the quality of mappings. Results: Patients in OBD-1033 had low utility values at baseline: mean= 0.45 (Standard Deviation 0.33, n=439). Using the published algorithm, the predicted mean utility was much higher: 0.74. This led to a high RMSE (0.43), indicating a poor fit to the data. Replicating the mapping using OBD-1033 PAC-QOL and EQ-5D data showed the PAC-QOL, although correlated with the EQ-5D, had a poor predictive value (RMSE= 0.31; R2< 0.10). High utilities were underestimated and low utilities overestimated. Conclusions: Mapping algorithms are a vital tool for generating utility values when none are available. However, the relationship derived between instruments should be assessed cautiously. Mappings with the same instruments may not be reliable if crossing disease areas – even if the symptoms experienced by patients appear similar. Data show patients in OBD-1033 entered the study with poorer health status than those in the chronic constipation mapping (utility of 0.45 vs 0.81), likely due to comorbid conditions (the reason for opioid prescribing). This led to a different relationship between the PAC-QOL and EQ-5D, compared to the previous estimate. PGI38 A Comparison Between The Health-Related Quality Of Life Reported By The General Population And By Patients With Major Liver Diseases Cortesi P A 1, Rota M 1, Scalone L 1, Cozzolino P 2, Cesana G 1, Mantovani L 3, Okolicsanyi S 1, Ciaccio A 1, Gemma M 1, Fagiuoli S 4, Valsecchi M G 1, Belli L S 5, Strazzabosco M 1 1University of Milano - Bicocca, Monza, Italy, 2CHARTA Foundation, Milano, Italy, 3Federico II University of Naples, Naples, Italy, 4Papa Giovanni XXIII Hospital, Bergamo, Italy, 5Niguarda Hospital, Milan, Italy . . . . . . . . . . . . . . . . Objectives: the impact of liver diseases (LDs) on health-related quality of life (HRQoL) is an important aspect to understand the burden of these conditions and improve their management. A well characterized impact of the major LDs on HRQoL of the general population is still lacking. The aim of our study was to fill this GAP. Methods: a dataset with HRQoL data of a representative sample of the general population of most populated Italian region was matched with the dataset from a multicenter study conducted in the same region and time period to generate and validate a set of health care outcomes indicators for the major LDs (hepatitis B (HBV), hepatitis C (HCV), compensated cirrhosis (CC), decompensated cirrhosis (DC), hepatocellular carcinoma (HCC), autoimmune hepatitis (AIH), primary biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC), NAFLD/ NASH and patients listed for liver transplant (LTL)). Within both datasets, HRQoL data were collected using the EQ-5D-3L. Multivariate logistic and Tobit regressions were then performed adjusting for possible confounders (age, sex, education and working status). Results: a total of 6,800 “healthy subjects” and 3,105 subjects with LDs were included in the analyses. Multivariate logistic analyses showed that DC, HCC, and LTL had significantly (p< 0.05) higher risk to have problems in mobility, self-care, and usual activities compared to “healthy subjects”. AIH had significantly higher risk to have problems in self-care; while HCV, CC, DC, and NAFLD/NASH in Anxiety/depression. Similar results were obtained with the Tobit model performed using VAS and Utility-index. DC, HCC, AIH and LTL reported the highest decrease in VAS and Utility score. Conclusions: HRQoL decreased in advanced LDs (DC, HCC, LTL) and AIH. This study provides an actual true estimate of the impact of major LDs on the patients’ HRQoL compare to the general population, and therefore is a key tool for decision-making in care delivery for liver diseases. PGI39 Translation And Cultural Adaptation Difficulties Encountered During Linguistic Validation Of The Bristol Stool Scale Edwards A , Williams H , Anderson H ICON plc, Oxford, UK . . . Objectives: The aims of this study were: (1) to investigate translation difficulties encountered during linguistic validation of the Bristol Stool Scale; (2) to discover whether certain items consistently posed problems across different languages; and (3) to analyse how solutions were reached and what types of solutions were appropriate. Methods: The investigation was made up of the following stages: (1) collation of back translation reviews of the Bristol Stool Scale for 30 European and Asia-Pacific languages; (2) identification of problematic words and phrases, based on written discussion between lead translators and project managers; (3) investigation of patterns that became apparent across different languages; (4) review of methods used to overcome the translation difficulties. Results: For more than a third of the languages reviewed (2 European and 9 Asia-Pacific, including 7 Indian), difficulties were experienced when attempting to translate certain items word-for-word. The majority of these difficulties centred on the food-related similes used in the scale to describe the different stool types. In all of the Indian languages in this study, alternative food-related similes were deemed necessary (e. g. ‘banana’ replacing ‘sausage’) in order to ensure that the wording was culturally relevant. Of the other languages, Thai and Romanian preferred similes that did not relate to food (‘bullets’ and ‘beads/little round bits’ respectively, instead of ‘nuts’), while for Singapore Chinese and Polish it was decided to remove certain similes altogether (‘like nuts’/‘like a sausage’). Conclusions: Certain items in the Bristol Stool Scale, in particular the two food-related similes, cause translation difficulties for some languages. For such items, it may not be appropriate to produce a word-for-word translation. In these cases, alternative solutions must be sought, taking into account cultural considerations, in order to achieve conceptual equivalence. PGI40 Self-Reported Health Related Quality Of Life Of Hepatitis C Virus (Hcv) Genotype 1 Patients With And Without Comorbid Conditions Nwankwo C 1, Sung A H 2, Pike J 3 Station, NJ, USA, 2St. John’s University, Queens, NY, USA, 3Adelphi Real World, Macclesfield, UK . . . . 1Merck, Whitehouse Objectives: To describe and compare self-reported health-related quality of life (HRQL) in HCV Genotype-1 infected patients with and without comorbid conditions (e. g. depression, hypertension, obesity, arthritis, anemia and diabetes). Methods: A Cross-sectional survey of physicians and their consulting patients was conducted from October 2012 to January 2013 as part of the Adelphi Real World Hepatitis C Disease Specific Programme. Overall, 348 patients from USA and France (231 with comorbidities and 117 without comorbidities) completed an EQ5D-3L and/or a Multidimensional-Assessment-of-Fatigue (MAF) scale. HRQL was evaluated by estimating EQ5D health-state-index (EQ5D-HSI), Visual-AnalogueScale (EQ5D-VAS) score, Domain-dimension score and MAF Global-Fatigue-Index (MAF-GFI). Further analysis compared HRQL reported by patients with comorbidities who completed treatment for HCV and were cured versus those who were not cured. Results: HRQL reported by untreated patients were better for those without comorbidities than those with comorbidities. For three of the five EQ5D domains, more patients without comorbidities versus patients with comorbidities reported no problems performing usual activities (85% versus 49%, p< 0.001), no pain or discomfort (70% versus 33%, p< 0.001) and not being anxious/depressed (67% versus 37%, p= 0.004). The corresponding mean HQRL estimates were EQ5DHSI 0.89 vs 0.77, p= 0.001; EQ5D-VAS 79 vs. 69, p= 0.014 and MAF-GFI 11 vs. 19, p= 0.016 for patients without comorbidities versus those with comorbidities. Among treated patients with comorbidities, those cured had significantly higher HRQL than those not cured (EQ5D-HSI 0.84 versus 0.67, p= 0.008, EQ5D-VAS 74 vs. 57, p= 0.003 and MAF-GFI 18 versus 32, p< 0.001). Conclusions: The results from this study suggest that patients with comorbidities have a poorer HRQL than patients without comorbidities, and that the treatment and cure of HCV in these patients is associated with higher HRQL compared with treatment and no cure. This implies that treatment and subsequent cure of HCV genotype 1 patients with comorbidities may help improve their HRQL. A370 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PGI41 Should Cost-Effectiveness Analysis Include The Cost Of Consumption Activities? An Empirical Investigation Adarkwah C C 1, Sadoghi A 2, Gandjour A 2 University, Maastricht, The Netherlands, 2Frankfurt School of Finance & Management, Frankfurt, Germany . . . . 1Maastricht Objectives: There has been a debate on whether cost-effectiveness analysis should consider the cost of consumption and leisure time activities when using the quality-adjusted life year as a measure of health outcome under a societal perspective. The purpose of this study was to investigate whether the effects of ill health on consumptive activities are spontaneously considered in a health state valuation exercise and how much this matters. Methods: The survey enrolled patients with inflammatory bowel disease in Germany (n = 104). Patients were randomized to explicit and no explicit instruction for the consideration of consumption and leisure effects in a time trade-off (TTO) exercise. Results: Explicit instruction to consider non-health-related utility in TTO exercises did not influence TTO scores. However, spontaneous consideration of non-health-related utility in patients without explicit instruction (60% of respondents) led to significantly lower TTO scores. Conclusions: Results suggest an inclusion of consumption costs in the numerator of the cost-effectiveness ratio, at least for those respondents who spontaneously consider non-health-related utility from treatment. Given the importance of this question for the conduct of cost-effectiveness analysis in health care, confirmation in additional studies that are conducted outside Germany and consider other health-state valuation techniques and diseases is recommended. PGI42 Impact Of Irritable Bowel Syndrome With Constipation On Work Productivity And Daily Activity Among Commercially Insured Patients In The United States Stephenson J J 1, Buono J L 2, Spalding W M 3, Cai Q 1, Tan H 1, Carson R T 2, Doshi J A 4 1HealthCore Inc, Wilmington, DE, USA, 2Forest Research Institute, Jersey City, NJ, USA, 3Ironwood Pharmaceuticals, Cambridge, MA, USA, 4University of Pennsylvania, Philadelphia, PA, USA . . . . . . . . . . . . Objectives: To assess work productivity and daily activity impairment among irritable bowel syndrome with constipation (IBS-C) patients in the United States. Methods: IBS-C patients aged ≥ 18 years with continuous medical and pharmacy benefit eligibility (03/01/2011–02/28/2013) were identified from claims in the HealthCore Integrated Research Database and invited to complete a crosssectional patient survey. Survey questions based on modified Rome III criteria confirmed IBS-C claims-based diagnoses. Patients who met both claims-based and Rome criteria completed the full survey, including the Work Productivity and Activity Impairment Questionnaire: General Health (WPAI: GH), which assessed the impact of general health problems on patients’ ability to work and function. WPAI: GH scores for absenteeism (work hours missed), presenteeism (lost productivity at work), and overall work productivity loss (absenteeism + presenteeism) over the previous week were calculated for employed respondents; daily activity impairment (functional impairment) was computed for all respondents. Follow-up questions adapted from the WPAI: GH evaluated work and activity impairment due to IBS-C. Scores are expressed as percentages, with higher percentages indicating greater impairment and less productivity. Indirect costs were calculated based on overall work productivity loss due to IBS-C using the human capital method. Results: Of 53 respondents (mean age: 40±15 years; 84.9% female), 35 were employed. Absenteeism averaged 10.6%, presenteeism averaged 37.4%, overall work productivity loss averaged 39.3%, and daily activity impairment averaged 45.7% due to general health problems over the past week. Of this, 7.5%, 21.5%, 23.8%, and 14.2% was attributed to IBS-C, respectively. The economic cost due to lost productivity attributable to IBS-C was estimated at $155±$321 USD per employed patient/ week. Conclusions: IBS-C–related impairment at work and in daily activities represents a significant burden for patients and employers. Treatments that effectively manage IBS-C symptoms may represent cost savings in the form of avoided work productivity losses associated with IBS-C. Gastrointestinal Disorders – Health Care Use & Policy Studies PGI43 Utilization Of Total Parenteral Nutrition In A South Indian Tertiary Care Hospital Sreedharan N 1, Vinitha M 1, John P 1, Girish T 1, Vijayanarayana K 1, Gabriel R 2 1Manipal University, Manipal, India, 2New Medical Center Specality Hospital, Madinat Zayed, United Arab Emirates . . . . . . Objectives: Total Parenteral Nutrition (TPN) is an essential pharmaceutical preparation used in hospitalized patients to whom enteral feeding is not possible or for critical care patients with compromised gastrointestinal tract function. Use of TPN reduce the incidence of malnutrition, which is a leading complication associated with various medical and surgical conditions. Hence the purpose of the study is to assess TPN utilization in surgical in-patients and its outcomes. Methods: Retrospective analysis of surgical in-patients receiving TPN from Jan 2011 to Dec 2012 in a tertiary care hospital was carried out. Patients who were administered TPN were included in the study. Patient characteristics and treatment details were collected. Data were analyzed using SPSS ® version 20.0. Results: A total of 120 patients were enrolled in the study. The mean age of patients taking TPN was 48.9±17.7 years. Majority of patients (67.5%) were males. A large proportion (40.8%) of the patients receiving TPN were those w underwent surgical procedures and had intestinal obstruction. Major metabolic complication included hypernatremia (26.5%) followed by hyperglycemia. Higher (79.5%) recovery rate was observed in patients who received TPN peripherally compared to those who received it via a central line. Among the patients receiving TNP, mortality was higher in patients with infections (31.9%) than without infection. In 93.3% of the cases, TPN starts were considered to be appropriate indications and rest inappropriate. From an eco- nomical standpoint, the total avoidable cost with TPN mounted to 2,48,200 Indian Rupees. Conclusions: Proper use of TPN reduced mortality in post-surgical patients. Greater attention to nutritional assessment to determine calorific need and nutritional requirement for individual patients should further improve benefits, reduce mortality and save treatment costs in hospitalized patients. PGI44 A Health Technology-Related Cost Description Concerning Italian Ibd Centres Dealing With Crohn’s Disease: Results From Sole Study Lazzaro C 1, Cappello M 2, Cortelezzi C 3, Costantino G 4, Fiorino G 5, Mastronardi M 6, Giannotta M 7, Galletti B 8, Cicala M 9, Vadalà di Prampero S 10, Gualberti G 11, Caprioli F 12, Gasbarrini A 13, Fattore G 14 1Studio di Economia Sanitaria, Milan, Italy, 2Università di Palermo, Palermo, Italy, 3AOU di Circolo - Fondazione Macchi, Varese, Italy, 4Università di Messina, Messina, Italy, 5Istituto Clinico Humanitas, Rozzano (MI), Italy, 6IRCCS S. De Bellis, Castellana Grotte, Italy, 7Azienda Ospedaliero Universitaria di Careggi, Firenze, Italy, 8Ospedale S. Salvatore, L’Aquila, Italy, 9Università Campus Bio Medico, Roma, Italy, 10Azienda Ospedaliero-Universitaria S. Maria della Misericordia di Udine, Udine, Italy, 11AbbVie, Campoverde di Aprilia (LT), Italy, 12Università degli Studi di Milano, Milano, Italy, 13Università Cattolica del Sacro Cuore, Rome, Italy, 14Bocconi University, Milan, Italy . . . . . . . . . . . . . . Objectives: To investigate the health technology-related costs of Italian inflammatory bowel disease (IBD) centers dealing with Crohn’s disease (CD). Methods: Following the hospital standpoint, a questionnaire-supported cost description was performed on a convenience sample of 38 Italian IBD centers participating in the ongoing Survey on Quality Of Life in Crohn’s Patients (SOLE). Consistently with their average useful life, a 5-year straight-line depreciation approach was adopted for calculating the yearly cost for each health technology. Cost description was undertaken either considering all centers as an undifferentiated sample, or stratifying them according to their complexity (number of beds for inward and dayhospital; personnel dedicated to CD patients; number of cross-border CD patients; availability of dedicated rooms for biological drugs administration; feasibility of electronic patient forms). Costs (€ 2012) were reported as mean (standard deviation, SD). Results: Half of centers (19/38) were public teaching hospitals, whereas 39.5% were regional referral centers for CD (15/38). The study sites were located in Northern (12/38, 31.6%), Central (11/38, 28.9%) and Southern (15/38, 39.5%) Italy, and could be classified as high (32/38, 84.2%), moderate (1/38), mild (3/38), and low (2/38) complexity centers. Endoscopy, capsule endoscopy and ultrasonography were the most widespread health technologies available in 92.1%, 78.9% and 34.2% centers, respectively. Considering the undifferentiated sample, mean yearly cost for health technologies amounts to € 23,557.50 (€ 24,277.90). High complexity centers report the highest mean yearly cost of € 25,580.38 (€ 25,706.92), whereas the lowest mean yearly cost of € 5,113 (€ 0) refers to the unique moderate complexity center. Regardless of site complexity, the cost-driver was endoscopy, which accounts for a percentage of the mean yearly cost that ranges from 36.9% (high complexity) to 97.8% (moderate complexity). Conclusions: SOLE results show that when Italian IBD centers complexity is taken into account, remarkable differences exist about costs for health technologies for managing CD patients. PGI45 Diagnosis And Management Of Moderate-To-Severe Irritable Bowel Syndrome With Constipation (Ibs-C) In Spain: The Ibis-C Study Mearin F 1, Cortes X 2, Mackinnon J 3, Bertsch J 3, Fortea J 2, Tack J 4 médico Teknon, Barcelona, Spain, 2Laboratorios Almirall S.A., Barcelona, Spain, 3TFS Develop S.L., Barcelona, Spain, 4University of Leuven, Leuven, Belgium . . . . . . 1Centro Objectives: The IBIS-C study assessed the burden of IBS-C in 6 European countries (France, Germany, Italy, Spain, Sweden, and UK). Here we present the diagnosis and management results for the Spanish cohort. Methods: Observational, retrospective-prospective (6 months each) study in patients diagnosed with moderate-tosevere IBS-C in the last five years (Rome-III criteria). Moderate-to-severe IBS-C was defined as an IBS-Symptom Severity Score (IBS-SSS) ≥ 175. Results: 112 patients were included (58% severe, mean age [±SD] 46.8 ± 13.7 years, 86% female). Mean time since diagnosis: 2.3 ± 2.7 years; mean symptom duration: 9.6 ± 9.9 years. Diagnostic procedures were highly variable; the most common were blood tests (71%), colonoscopy (56%) and abdominal ultrasound (54%). At inclusion the most prevalent symptoms were constipation (84%), abdominal pain (80%), abdominal distention (80%) and bloating (59%). Main ongoing comorbidities were dyspepsia (41%), anxiety (38%), depression (21%), headache (25%), or insomnia (25%). 58% of patients had an average of 4.1 ± 2.5 diagnostic tests during follow-up. 85% of patients took pharmacological medication (80% took some pharmacological medication for their IBS-C). The most common prescription drugs were plantago ovata (35%), otilonium bromide (22%), macrogol plus electrolites (13%) and cinitapride tartrate (10%). Likewise, common drug combinations were laxative monotherapy (21%), laxatives and antispasmodics (14%), and antispasmodic monotherapy (5%). In addition, 30% of patients received complementary therapies. Overall, marginal improvement was noted in symptom severity (IBS-SSS total score) between baseline (315±83) and the 6-month visit (234±98). Conclusions: Moderate-to-severe IBS-C symptoms often remain undiagnosed for many years. With frequent visits to health care professionals IBS-C continues to be a burden despite the availability of therapeutic interventions. Finally, current health care resource utilization is high even though there is a high degree of prescription medication use. PGI46 Real World Studies Using Japanese Administrative Databases: Chronic Hepatitis C Treatment Pattern And Resource Use Crawford B 1, Tang A 2, Li H 3, Burns L 4, Wada K 1, McDonald J 5 KK, Tokyo, Japan, 3Bristol Myers Squibb, Wallingford, CT, USA, 4Bristol-Myers Squibb, Hopewell, NJ, USA, 5Adelphi Values, Boston, MA, USA . . . . . . 1Adelphi Values, Tokyo, Japan, 2Bristol-Myers Objectives: While discussions on health technology assessment (HTA) in Japan continue, platforms for real-world population-based studies are lacking. We attempt A371 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 to compare two large commercial databases focusing on CHC patient characteristics and treatment patterns. Methods: We analyzed the Japan Medical Data Center (JMDC) database containing employer-based health insurance claims and the Medical Data Vision (MDV) database containing administrative data from acute care hospitals. Patients aged 20 years or above and diagnosed with CHC were included in the study. Patients co-infected with hepatitis B and/or human immunodeficiency viruses were excluded. Patient characteristics and treatment patterns for CHC were compared. Results: 3,590 (JMDC) and 29,702 (MDV) confirmed CHC patients were included in the analysis. Patients in JMDC were younger than those in MDV (age 51.6±11.9 and 63.2±13.4) and had fewer comorbidities (Charlson Comorbidity Index 1.3±1.7 and 1.9±2.1). More patients in JMDC were found to have prior CHC treatment (37.2%) than in MDV (24.6%). With an index date of September 2011, about a third of the patients (n=1,271, 35.4%; JMDC and n=9,791, 33.0%; MDV) received treatments for CHC, comprised of 10.2% (n= 365; JMDC) and 5.5% (n= 1,620; MDV) on antiviral treatments (interferon or direct acting antiviral combinations) and 25.2% (n= 906; JMDC) and 27.5% (n= 8,171; MDV) on liver protection drugs only. Among patients treated with triple therapy (telapravir/peginterferon/ribavirin), the mean total treatment duration was 19.3 weeks (JMDC) versus 21.9 weeks (MDV). The mean duration of telaprevir therapy was 10.8 weeks (JMDC) and 10.1 weeks (MDV), followed by 11.7 weeks (JMDC) and 14.1 weeks (MDV) of peginterferon/ribavirin therapy. Conclusions: Both databases found low treatment rates for CHC. Although the findings are consistent, there are differences in database populations and treatment patterns that warrant further research. Using these administrative databases for real-world research may be useful depending on research objectives. PGI47 Cohort Of Hcv Patients In Italy: Sizing And Treatments In A Sample Of Italian Hepatology Centers Lanati E P 1, Lidonnici D 1, Gasbarrini A 2, Ruggeri M 3, Sacchini D 3, Caporaso N 4, Fagiuoli S 5 Provider, Milano, Italy, 2Policlinico Gemelli, Rome, Italy, 3Università Cattolica del Sacro Cuore, Rome, Italy, 4Universita’ di Napoli Federico II, Napoli, Italy, 5Papa Giovanni XXIII Hospital, Bergamo, Italy . . . . . . . . 1MA Objectives: The aim of this study is to give an overview of the treatment available for the patients with Hepatitis C Virus (HCV) through the segmentation of HCV patients in Italy. Methods: The study was carried out in 9 hepatology centers, treating HCV patients in 9 Italian Regions. The structures selected constitute a representative sample of the Italian scenario, being Centers of Excellence in HCV management in northern, southern, and center of Italy (representing about 24% of the 2.000 patients treated with first generation Triple Therapy in Italy according to AIFA Data). The patients’ distribution has been investigated in terms of fibrosis stage (F0 to F4), therapy type (Triple, TT, or Double therapy, DT) and treatment status (naive or experienced patients). Results: Data (collected from structures and Workshop of Pharmacoeconomics in Hepatology) show that HCV patients are more concentrated in the two fibrosis stage extremes: 43% in F0-F1 Range, 23% in F2, 32% in F3-F4 Range and 2% unclassified. Data about drug administration demonstrate that, at national level, patients are equally distributed between therapy type (56% TT and 44% DT) and treatment status (49% naive and 51% experienced). On the contrary, at Regional level many differences were found in all of the three parameters examined. In the structure investigated in Campania, for example, 72% of patients receive TT and 76% are experienced, while in Lazio 72% of patients receive DT and 72% are naive. Furthermore, considering the fibrosis stage, the 46% of patients treated in a center operating in Bari is in the range F3-F4, whilst in the center in Milan the 56% of the HCV patients treated is in range F0-F1. Conclusions: The study demonstrate that, concerning the treatment of HCV, there are significant differences among the hepatology centers, both in terms of patients’ health status and therapy pathways. PGI48 The Costs-Effectiveness Of Sofosbuvir VersUS Standard Of Care (Soc) In Chronic Hepatitis C From A Belgian Reimbursement Perspective De Groote K 1, Delwaide J 2, Michielsen P 3 1INNOSENS bvba, Brakel, Belgium, 2Centre Hospitalier Sart-Tilman, Liège, Belgium, 3UZ Antwerp, Edegem, Belgium . . . Objectives: Hepatitis C virus (HCV) infection resulting in chronic liver disease has an estimated prevalence in Belgium of 0,87% (Beutels, 1997) with 59% of patients having genotype 1 HCV (GT1), 6% GT2, 19% GT3 and 16% GT4-5-6. Sofosbuvir (SOF), a novel HCV treatment, has demonstrated high rates of sustained virological response (SVR) when given with ribavirin to subjects with chronic HCV infections (all GT’s). The objective of this analysis was to assess the cost-effectiveness of sofosbuvir vs SoC in the treatment of chronic hepatitis C (CHC) in the following indications: GT 1 (Treatment Naïve (TN), INF-eligible and IFN-ineligible patients); GT2 & 3 (TN & Treatment Experienced (TE), both INF-eligible and IFN-ineligible patients) and GT 4/5/6, treatment-naïve patients in Belgium. Patients ineligible to IFN have no treatment options today. Methods: Based on a Markov model, this cost-utility analysis models the cost-effectiveness of SOF versus SoC in Belgium from the perspective of the RIZIV/INAMI and taking into account the proposed reimbursement criteria for SOF in Belgium and the guidelines of the Knowledge Centre (KCE) (KCE report 78C, 2008). Results: Weighted ICER’s were calculated taking into account patient eligibility for treatment with IFN and treatment duration with SOF: GT 1, 3, 4, 5 or 6 patients who are IFN-eligible and are being treated with SOF (+ subcutaneous pegylated interferon-alpha (IFN-α ) plus daily oral ribavirin (RBV)) for 12 weeks (€ 19,954/QALY); GT 1, 3, 4, 5 or 6 patients who are IFN-ineligible due to intolerance and/or contra-indications and are being treated with SOF (+ RBV) for 24 weeks (€ 35,086/QALY) and GT 2 patients that are being treated with SOF (+ RBV) for 12 weeks (€ 28,121/QALY). Overall, the weighted PANgenotypic ICER was € 21,651. Conclusions: PAN-genotypic cost-effectiveness has been demonstrated for sofosbuvir in comparison to the current standard of care in HCV in Belgium. PGI49 Which Metric To Choose For Indirect Comparison Of Treatments When Multiple Comparisons Are Feasible: Lubiprostone VersUS Prucalopride In Chronic Constipation Hatswell A J 1, Griffiths A 2, Lichtlen P 3, Losch-Beridon T 4, Pennington B 1 Economics, Solihull, UK, 3Sucampo, Zug, Switzerland, 4Sucampo Pharma Americas, Bethesda, MD, USA . . . . . . 1BresMed, Sheffield, UK, 2Research Objectives: For a recent health technology appraisal in the treatment of chronic idiopathic constipation, direct evidence of the effectiveness of a new intervention (lubiprostone) against the standard of care (prucalopride) was not available. The aim of this study was to review the available data from clinical trials and perform indirect comparisons between the two treatments where possible. Methods: A literature search (in Medline and other databases) was conducted in December 2013 for trials of lubiprostone or prucalopride. Data for any comparable endpoints were extracted from the papers, and indirect comparisons performed using the Bucher method. Results: Four clinical trials for lubiprostone were identified (three company-sponsored, and a small clinician-led trial), as well as three companysponsored clinical trials for prucalopride. After data extraction, indirect comparisons were possible for seven different endpoints, including the primary efficacy parameter of the lubiprostone studies (Spontaneous Bowel Movements; the relative risk was 1.12 in favour of lubiprostone, 95% CI 0.77-1.64). Other endpoints included the primary efficacy parameter of the prucalopride studies (Spontaneous Complete Bowel Movements), and a range of symptom comparisons. In total, five of the seven indirect comparisons favoured lubiprostone, with statistical significance reached in favour of lubiprostone once and prucalopride once. Conclusions: The indirect comparisons showed that lubiprostone is likely to be at least as effective as prucalopride, with numerical superiority in five out of seven comparisons. However, the number of feasible indirect comparisons on a range of endpoints raises a wider question: which to use in cost-effectiveness modelling? Although analyses generally have a ‘base case’, each of the indirect comparisons adds different information about the relative efficacy of the two products. Given the range of endpoints with associated relative risks, to reduce these to a single comparison (as is current practice) may omit important and relevant information about relative efficacy. PGI50 High Therapeutic Efficiency With Sofosbuvir For The Treatment Of Chronic Hepatitis C Félix J 1, Silva M 1, Ferreira D 1, Vandewalle B 1, Guerra I 2, Cure S 2, Aldir I 3, Carvalho A 4, Macedo G 5, Marinho R T 6, Pedroto I 7, Ramalho F 6 1Exigo Consultores, Alhos Vedros, Portugal, 2OptumInsight, Uxbridge, UK, 3Hospital Egas Moniz, Centro Hospitalar de Lisboa Ocidental, Lisboa, Portugal, 4Centro Hospitalar e Universitário de Coimbra, Coimbra, Portugal, 5Hospital de São João, Porto, Portugal, 6Centro Hospitalar Lisboa Norte. Hospital de Santa Maria, Lisboa, Portugal, 7Centro Hospitalar do Porto, Porto, Portugal . . . . . . . . . . . . . Objectives: Chronic hepatitis C (CHC) is a major public health problem contributing to more than 86,000 premature deaths in Europe. Pegylated interferon-α plus ribavirin (PR) based therapy, including regimens with boceprevir (BOC) or telaprevir (TVR) in HCV genotype-1 patients, have failed to provide more extensive therapeutic benefit leaving space for substantial outcomes improvement. Sofosbuvir (SOF) – a pangenotypic RNA polymerase inhibitor – has shown unprecedented sustained virologic response rates and tolerability profiles. The objective of this study was to estimate SOF contribution to public health by exhausting CHC therapeutic efficiency. Methods: Therapeutic efficiency was defined as maximum capacity to benefit from treatment in terms of life years (LY) relative to the general population’s life expectancy. The natural history of CHC and treatment implication was modelled with a discrete-time Markov allowing for long term assessment in terms of HCV genotype, fibrosis progression, HIV co-infection status and previous treatment experience. Treatment options compared were dependent on interferon eligibility/tolerance and genotype: PR, SOF/PR, BOC/ PR and TVR/PR in elegible/tolerant patients (BOC/TRV regimens in genotype-1 only; SOF/ribavirin in genotype-2). For ineligible/intolerant patients, comparison of SOF/ ribavirin was performed against lack-of-therapy. Results: In mono-infected HCV genotype-1 patients SOF/PR treatment is estimated to result in 4.3 LY, 7.0 LY or 8.0 LY gained in comparison to TRV/PR, BOC/PR or PR, respectively. In genotype-1 and genotype-2 HIV-coinfected patients elegible for interferon treatment, the estimated LY gained with SOF treatment is 11.8yrs and 5.0yrs, respectively. In patients ineligible for interferon treatment, SOF is expected to almost double life expectancy irrespective of the genotype, with therapeutic efficiency ranging from 79% to 95%. In co-infected patients, therapeutic efficiency of SOF is expected to range between 84.3% and 92.4% of general population life expectancy. Conclusions: Sofosbuvir-containing regimens are expected to maximize years of life lived and maximize efficiency relative CHC patients residual life expectancy. PGI52 Comparison Of The Burden Of Ibs With Constipation On HealthRelated Quality Of Life (Hrqol), Work Productivity, And Health Care Utilization To Asthma, Migraine, And Rheumatoid Arthritis In The Us, Uk, And France Taylor D C A 1, Kosinski M 2, Reilly K 2, Lindner L 3 Pharmaceuticals, Inc., Cambridge, MA, USA, 2QualityMetric, OptumInsight Life Sciences, Lincoln, RI, USA, 3Almirall S.A., Barcelona, Spain . . . . . . 1Ironwood Objectives: Studies have shown that irritable bowel syndrome (IBS) and its subtype with constipation (IBS-C) are associated with poorer HRQoL, decreased work productivity and increased health care utilization. However, no studies have compared the burden of IBS-C to similar chronic conditions. Objective was to evaluate burden of IBS-C compared with no functional gastrointestinal disorders (non-FGID), asthma, migraine, and rheumatoid arthritis (RA), on HRQoL, work productivity and health care utilization. Methods: Data come from the 2011 National Health and Wellness Survey (nationally representative sample of adults including the US, UK and France), which includes health-related topics such as HRQoL (SF-12 Health Survey), work productivity (Work Productivity and Activity Impairment) and health care utiliza- A372 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 tion (6-month ER visits and hospitalizations). Comparisons in each of these healthrelated topics were made between respondents with IBS-C versus non-FGID, asthma, migraine, and RA. Comparisons controlled for age and gender. Results: Prevalence of IBS-C was 0.67%, 1.03%, and 0.91% in the US, UK, and France, respectively. Most respondents were female (> 65%) and mean age ranged from 47.7 to 49.8 years. In each country, mean SF-12 physical (PCS) and mental (MCS) summary scores were statistically (p< 0.001) and clinically meaningfully (> = 2.5) lower than the non-FGID group. With few exceptions, PCS and MCS scores of IBS-C were significantly (p< 0.05) lower than scores for asthma, migraine, and RA. In each country, IBS-C reported significantly (p< 0.001) greater absenteeism and worse presenteeism and greater overall work productivity loss and daily activity impairment than non-FGID, and with few exceptions, worse than asthma, migraine, and RA. Lastly, in each country the number of ER and hospital visits was significantly higher (p< 0.01) for IBS-C compared to non-FGID and, with few exceptions, higher than asthma, migraine, and RA. Conclusions: In the three countries respondents with IBS -C showed significant and clinically meaningful deficits in HRQoL and work productivity and elevated health care utilization compared with other chronic conditions. PGI53 Assesment Of Serial Transverse Enteroplasty: Systematic Review Yoo K J NECA, Seoul, South Korea . . Objectives: The purpose of this study was to evalute the safety and effectiveness of serial transverse enteroplasty (STEP). Methods: The clinical utility of STEP was first reviewed in three textbooks and secondly the safety and the effectiveness of the STEP were assessed based on a systematic revew. 575 articles were searched using database (KoreaMed, Ovid-MEDLINE, Ovid-EMBASE and Cochrane Library, etc.) and eliminated acoording to inclusion/exclusion criteria, duplicate search results. Finally, 9 articles (2 cohort studied and 7 case studies) were used. Two evaluators performed independently throuout each process. The article’s quality evaluated by the SIGN’s tool and the grade of recommendation were selected. Results: A review of the textbooks revealed that STEP is conducted on patients with refractory SBS. And the advantages of STEP are that it is technically straightforward, can form a uniform bowel channel regardless of variable underlying bowel dilation, and can be repeated if the bowel subsequently radilated. The safety of STEP was assessed based on procedure-related complications in 6 articles. The complication rate of STEP was overall 6.3~25%, bleeding 6.3~12.5%, obstruction 6.3~22%, anastomosis leak 2.3~16.7% and stricture 11~20%. The complication rate of the STEP versus the Bianchi LILT was no significantly difference. The effectiveness of STEP was assessed based on the enteral autonomy and survival rate in 9 articles. There was a trend toward a decreased rate of PN after STEP. As a result, the 33~56% patients who underwent STEP were weaned off within one year and 83~88% after one year. Survival was 83% with maximum followup of 68 months. Conclusions: The safety of STEP was at an acceptable level as complications reported in studies generally occur after an intestinal operations and such problems can be corrected surgically. The clinical utility of STEP was described in the textbooks as a non-transplantation procedure and all the selected articles was reported on wean-off parenteral nutrition. Therefore, there is evidence for the safety and effectiveness of STEP performed on patients with refractory short bowel syndrome for promoting eneteral nutrition (Recommendation grade C). PGI54 Prescribing Pattern Of Drugs For Ulcerative Colitis In Japan . PGI55 Patients Diagnostic Therapeutic Pathways For Hcv Patients In Italy: Impact Of Regionalization In Treatments And Guidelines Lanati E P 1, Lidonnici D 1, Gasbarrini A 2, Ruggeri M 3, Sacchini D 3, Caporaso N 4, Fagiuoli S 5 Provider, Milano, Italy, 2Policlinico Gemelli, Rome, Italy, 3Università Cattolica del Sacro Cuore, Rome, Italy, 4Universita’ di Napoli Federico II, Napoli, Italy, 5Papa Giovanni XXIII Hospital, Bergamo, Italy . Murata T , Hanada K , Shibahara H CRECON Research & Consulting Inc., Tokyo, Japan . . . Objectives: This retrospective study aimed to evaluate the actual chronic hepatitis C (CHC) treatment in Japan including interferon (IFN) treatment using Japanese claims data provided by Japan Medical Data Center Co., Ltd (JMDC). Methods: All claims data for 9,634 patients having a history of diagnosis related to CHC, liver cirrhosis (LC), or hepatocellular carcinoma (HCC) among Japanese claims data from January 2005 to October 2013 provided by JMDC were analyzed in this study. Treatment patterns and costs were evaluated in CHC patients newly diagnosed during the study period. Results: 3,546 newly diagnosed CHC patients were included in this analysis. The median follow-up period was 21.0 months. The CHC treatment including IFN was not implemented in most cases (76.8%) during the follow-up period. The rate of IFN treatment was 7.0%. The combination therapy of IFN and ribavirin (RBV) was the most frequently used in IFN treatment. Median period of first line treatment to the discontinuation was 7.0 months in patients with IFN + RBV. The mean annual total cost in CHC patients with IFN, CHC patients without IFN with other drugs (e. g. ursodeoxycholic acid), CLC patients were 3.5 million yen ($35,000), 2.0 million yen ($20,000), 3.1 million yen ($31,000), respectively. The mean annual number of outpatient visits in patients with IFN + RBV, patients without IFN with other drugs were 56.7 and 37.5 visits, respectively. Conclusions: The CHC treatment including IFN was not implemented in most cases despite having the CHC diagnosis. Furthermore, IFN treatment completion rate are low. One of the main causes of these facts would be adverse effects associated with IFN treatment. New drugs with fewer adverse effects are awaited for the future. . Objectives: Ulcerative colitis is one of 56 designated diseases for the Specified Disease Treatment Research Program that provides co-payment reduction or exemption according to disease severity and patients’ income levels. Under the fee-for-service payment system, such financial assistance program may influence drug utilisation. The objective of this study is to assess the overall drug prescribing pattern for ulcerative colitis. Methods: Claims data with indication for ulcerative colitis were extracted from the Social Health Insurance claims data processed from December 2009 to February 2010 and from February to April 2011. Extracted data were analysed and compared in terms of patients’ age, types of public subsidy, prescribed places (clinic/ hospital or pharmacy) and pharmacologic classes. Results: Total numbers of patients were 67,480 for the former duration and 84,450 for the latter duration. Among them, patients who were entitled to the Specified Disease Treatment Research Programme were 47,103 and 60,231 respectively. The most commonly used drug was mesalazine which accounted for 56% and 59%. Infliximab was prescribed mainly for those entitled to the Specified Disease Treatment Research Program. Of total infliximab dose, 23 % and 30% were with steroid and 33% and 38% were with mesalazine. Generic utilisation rate for patients entitled to the Specified Disease Treatment Research Programme or other financial assistance programs was low. Conclusions: The highest utilization of infliximab and low generic utilization were observed for patients entitled to the Specified Disease Treatment Research Programme. Prescribing pattern of drugs suggested that some cases did not follow clinical guidelines. Adherence to clinical guideline recommendations should be promoted. . PGI56 Claims Database Analysis Of Patients With Chronic Hepatitis C In Japan Muscular-Skeletal Disorders – Clinical Outcomes Studies Tomita N 1, Kanatani Y 1, Nakagawa Y 2 1National Institute of Public Health, Saitama, Japan, 2National Hospital Organization Shikoku Medical Center for Children and Adults, Zentuji city, Japan . the PDTAs of the most representative Regions. The analysis investigates the following Regional characteristics: establishment of a working group to define the PDTA, the gap of criteria for prescribing centers with the AISF guidelines, indication of therapeutic scheme for the different kind of patients, budget allocation, characteristics of the management model and existence of a follow up system. Results: The results of the study show that, besides some common characteristics observed in all Regions (e. g. the presence of working group and criteria for the eligibility of patients and identification of prescribing centers), there is a substantial heterogeneity in Regional guidelines, especially regarding budget allocation, management model and follow up system. Indeed, if in some Regions like Basilicata and Veneto there is a specific budget allocation (respectively €1,2 ml and € 12 ml in 2014), the management model is based on Hub and Spoke system and the follow up mechanism is clearly set up, in other Regions like Lombardy and Liguria the guidelines are more vague, detailing only few criteria. These Regional differences are confirmed by other two data: the timespan between the market authorization of the Triple Therapy for HCV patients and the integration in Regional guidelines (e. g. 17 days in Veneto and 147 days in Emilia Romagna) and the number of prescribing centers per million inhabitants (6,92 in Basilicata and 1,85 in Veneto). Conclusions: The study demonstrate that, with a view to the future novel drugs, a common disease management model, including common guidelines and organizational model, is essential to avoid regional disparities in HCV therapies access. . . . . . . PMS1 Impact of Apremilast on Physical Function over 52 weeks in Patients with Active Psoriatic Arthritis Mughal F 1, Tencer T 2, Clancy Z 2, Zhang F 2 Corporation, Uxbridge, UK, 2Celgene Corporation, Warren, NJ, USA . . . . 1Celgene Objectives: The PALACE studies compared the efficacy and safety of apremilast (APR) with placebo in patients with active psoriatic arthritis (PsA) despite prior conventional disease-modifying antirheumatic drugs and/or biologics. The objective was to assess the impact of APR on physical functioning in patients enrolled in the PALACE trials. Methods: The pooled analysis included data from PALACE 1-3, three 52-week, randomized, placebo-controlled studies evaluating APR in subjects with active PsA. Patients were randomized (1: 1: 1) to placebo, APR 20 mg BID (APR20), or APR 30 mg BID (APR30). Patients with < 20% reduction from baseline in swollen and tender joint counts at Week 16 were required to be re-randomized (1: 1) to APR20 or APR30 if initially randomized to placebo, or continued their initial APR dose. At Week 24, all remaining placebo patients were re-randomized to APR20 or APR30. The analysis reports data from the APR-exposure period (Weeks 0 to 52). Physical function, a pre-specified secondary end point, was measured using the Health Assessment Questionnaire-Disability Index (HAQ-DI) and the 36-item Short-Form Health Survey version 2 Physical Functioning (PF) domain and physical component summary (PCS) scores. Results: At Week 16, the observed physical function change from baseline was improved with APR20 and APR30 vs. placebo, as measured by the HAQ-DI (-0.17 [P< 0.001] and -0.23 [P< 0.001] vs. -0.07), PF (2.73 [P< 0.007] and 4.08 [P< 0.001] vs. 1.52), and PCS (3.44 [P< 0.002] and 4.46 [P< 0.001] vs. 2.03). At Week 52, among patients who were treated with APR continuously, the physical function change from baseline for APR20 and APR30 was improved, as measured by the HAQ-DI (-0.30 and -0.33), PF (5.55 and 5.53), and PCS (6.37 and 6.23). Conclusions: Patients treated with APR30 reported improvement in physical function compared with placebo, as measured by the HAQ-DI, PF, and PCS. This response was maintained over 52 weeks. 1MA Objectives: The aim of this study is to describe the differences in Patients Diagnostic Therapeutic Pathways (PDTA) among Italian Regions, as the Italian scenario shows substantial differences in terms of Regional guidelines, organization and allocated budget. Methods: The Regional policies were examined analyzing PMS2 Work Productivity Improvement Associated With Apremilast, An Oral Phosphodiesterase 4 Inhibitor, in Patients With Psoriatic Arthritis Results Of A Phase 3, Randomized, Controlled Trial Zhang F 1, Tencer T 1, Li S 1, Strand V 2 Corporation, Warren, NJ, USA, 2Biopharmaceutical Consultant, Portola Valley, CA, USA . 1Celgene . . . A373 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: The Work Limitations Questionnaire (WLQ) measures the degree to which employed individuals are experiencing health-related limitations at work, as well as health-related productivity loss. The PALACE 1 study compared the efficacy and safety of apremilast (APR) with placebo in patients with active psoriatic arthritis (PsA) despite prior or concurrent conventional disease-modifying antirheumatic drugs (DMARDs) and/or prior biologics. The objective of the current analysis was to assess the effect of APR on the work productivity of employed patients in PALACE 1. Methods: Patients were randomized (1: 1: 1) to receive placebo, APR 20 mg BID (APR20), or APR 30 mg BID (APR30). Treatment efficacy was assessed at Week 16 based on the intent-to-treat population. Employed patients completed the WLQ at baseline and Week 16. Work limitations were categorized into 4 domains, which were then used to calculate the WLQ index: physical demands (PDS), mental demands (MDS), time management demands (TMS), and output demands (ODS). Improvement in the WLQ index, and its 4 domains, is represented by a negative change from baseline. Improvement in work productivity is represented by a positive improvement in percentage of productivity loss. Results: 504 patients were randomized (mean age: 50.4 years; male: 49.4%). Of these, 261 who were employed and completed at least 1 component of the WLQ were analyzed. At Week 16, APR20 and APR30, vs. placebo, were associated with a greater mean change from baseline in PDS (-5.58 and -6.24 vs. -2.14), MDS (-2.22 and -5.18 vs. 1.15), TMS (-4.03 and -8.76 vs. -4.25), and ODS (-5.92 and -10.3 vs. -1.34), resulting in a greater mean improvement in the WLQ index (-0.01 and -0.03 vs. 0.00), which corresponds to a higher median percent improvement of productivity loss (18.9% and 24.7% vs. -3.7%). Conclusions: APR20 and APR30 increased work productivity among patients with active PsA. PMS3 Network Meta-Analysis With Baseline Risk Adjustment To Assess The Relative Efficacy Of Ustekinumab In Adult Patients With Active Psoriatic Arthritis Van Sanden S 1, Diels J 2, Van Laer J 2, Hemels M 3 1University Hospital Leuven, Leuven, Belgium, 2Janssen Pharmaceutica N.V., Beerse, Belgium, 3Janssen A/S, Birkerød, Denmark . . . . Objectives: To compare the relative efficacy of ustekinumab to alternative therapies in anti-TNF treatment-naïve adult patients with active PsA, while adjusting for the variability in placebo response rates across trials. Methods: A Bayesian network meta-analysis (NMA) was performed to compare ustekinumab with adalimumab, golimumab, etanercept, certolizumab pegol and infliximab. Four outcomes (PASI75, PASI90, PSARC, ACR20) were analysed after 12-16 and 24 weeks. The NMA was conducted using a meta-regression model, with the trial-specific estimated baseline risk included as a covariate (Dias et al., 2013). A sceptical prior was used for this covariate. Both fixed (FE) and random effects models were considered. Results: Nine placebo-controlled trials were identified for inclusion in the NMA based on systematic literature review, including 2 ustekinumab trials. The placebo response rates varied significantly across trials, with ustekinumab trials having generally higher values. The median value of the meta-regression coefficient ranged between -0.02 and -1.69 (P [coef<0] between 50% and 95%) over the different scenarios and models, suggesting an interaction effect between baseline risk and treatment effects. At week 24 using the FE, the probability for ustekinumab 45mg and 90mg to be more effective than the comparators based on the PASI75 ranged from 87% (etanercept) to 46% (golimumab 100mg) and from 88% (etanercept) to 50% (golimumab 100mg), respectively. The variability around the point estimates was however large. Conclusions: This analysis indicates baseline risk in PsA-trials to be a treatment effect modifier. Any NMA not correcting for baseline risk might generate biased results. After adjusting for differences in baseline risk between the trials, ustekinumab indicates comparable efficacy to alternative therapies, however with high uncertainty around the point estimates. PMS4 Comparison Of Disease Status And Outcomes Of Patients With Psoriatic Arthritis (Psa) Receiving Adalimumab Or Etanercept Monotherapy In Europe Narayanan S 1, Lu Y 2, Hutchings R 2, Baynton E 2 1Ipsos Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, London, UK . . . . Objectives: To compare the disease status and outcomes of patients with PsA receiving adalimumab and etanercept monotherapy in Europe. Methods: A multi-country, multi-center medical chart-review study of AS patients was conducted among rheumatologists in UK/France/Germany/Italy/Spain to collect deidentified data on patients who were recently treated with a biologic as part of usual care. Physicians were screened for duration of practice (3-30yrs) and patient volume (incl. > 5PsA biologic patients/month) and recruited from a large panel to be geographically representative in each country. Eligible patient charts (≥ 3) were randomly selected from a sample of prospective patients visiting each center/practice during the screening period. Physicians abstracted patient diagnosis, treatment patterns/dynamics and patient symptomatology/disease status/outcomes. Patients on adalimumab/etanercept monotherapy were analyzed. Results: 249 eligible PsA patient charts were abstracted; 120 on adalimumab (male: 58%, age: 46.2yrs, average months on adalimumab: 23.9, 95% on first biologic) and 81 on etanercept (male: 51%, age: 47.6yrs, average months on etanercept: 22.7, 94% on first biologic). Top-3 comorbidites (adalimumab vs. etanercept) were obesity: 12% (range: 4% (France) -20% (Germany)) vs. 9% (range: 0% (France) -25% (UK)), dyslipidemia: 8% (range: 0% (Italy) -13% (UK)) vs. 16% (range: 0% (UK) -24% (Italy)) and anxiety/depression: 12% (range: 7% (UK) -22% (France)) vs. 6% (range: 0% (UK) -12% (Spain)). Among patients with available data, latest lab measures documented were (adalimumab vs. etanercept): ESR: 19.2mm/h (range: 16.0 (France) -23.6 (Italy)) vs. 22.0mm/h (range: 13.0 (Germany) -30.1 (Italy)), CRP: 6.5mg/dl (range: 1.3 (Spain) -10.8 (UK)) vs. 13.7mg/dl (range: 6.7 (Germany) -24.0 (UK)), rheumatoid factorpositive: 10% (range: 0% (UK/Spain) -19% (Germany)) vs. 17% (range: 7% (Spain) -50% (UK)), and anti-CPP-positive: 5% (range: 0% (France/Italy/Spain) -50% (UK)) vs. 7% (range: 0% (Germany/Spain) -25% (UK)). Latest disease severity measures documented were (adalimumab vs. etanercept): Swollen Joint Counts: 1.6 (range: 0.4 (Spain) -2.9 (UK)) vs. 2.7 (1.4 (Germany) -8.0 (UK)), Tender Joint Counts: 2.4 (range: 0.9 (Spain) -3.5 (Italy)) vs. 4.1 (range: 1.9 (Germany) -9.8 (UK)), and HAQ rating: 1.2 (range: 0.4 (Spain) -2.7 (Germany)) vs. 1.5 (range: 1.3 (Germany) -2.0 (France)). Conclusions: Among PsA patients receiving adalimumab or etanercept monotherapy, disease severity differed within the EU5, with patients on etanercept, and patients in Italy and UK, having relatively higher burden and poorer outcomes. Factors influencing the observed patterns of geographic variation and the impact of specific biologic treatments on observed patterns warrant further scrutiny to optimize therapeutic interventions and improve outcomes. PMS5 Relative Efficacy And Safety of Ustekinumab Compared to Anti-TNF-Alfa Therapies in Patients With Active Psoriatic Arthritis Huszti Z National Institute for Quality- and Organizational Development in Healthcare and Medicines, Budapest, Hungary . Objectives: Assess the relative efficacy and safety of ustekinumab compared to reimbursed anti-TNF-alfa therapies in patients with active psoriatic arthritis (PsA). Methods: Randomized, placebo-controlled, at least 24-week-long, pivotal studies were searched in PubMed and EMBASE databases. Patients, participating in these trials, had active psoriatic arthritis. They were treated with DMARD and/ or NSAID therapies and had not received any biological treatment before. Metaanalysis was based on ACR20, ACR50, ACR70, PASI75, HAQ-DI, AE and SAE endpoints. The quality assessment of evidences was based on EUnetHTA guidelines. For the applicability evaluation Aetkin’s method was used, while the meta-analysis used Frequentist approach. The mean and the 95% confidence interval of odd ratios (OR) were estimated with the Mantel-Haenszel test and with the fixed effect method. Statistical heterogeneity was tested using the Chi-squared-test, publication bias was checked with funnel plot. Results: During the literature search 8 studies were identified. One study was excluded from the evaluation due to inadequate treatment dose. Due to altering placebo arm treatment and a high risk of bias at study level, two studies were involved in sensitivity analyses. Anti-TNF-alfa treatments had significantly better results on primary and accentuated secondary endpoints (ACR20 week 24 - ustekinumab vs. placebo: OR= 2,56 (95%CI= 1,74; 3,74); anti-TNF-alfa vs. placebo: OR= 7,89 (95%CI= 5,65; 11,1); ACR50 week 24 – ustekinumab vs. placebo: OR= 3,46 (95%CI= 1,94; 6,17); anti-TNF-alfa vs. placebo: OR= 14,21 (95%CI= 8,01; 25,23)). On the other secondary endpoints the differences were not significant, however results have shown subservient tendency for anti-TNF-alfa treatment against ustekinumab. During the safety comparisons we did not find a substantial difference between these treatments. Conclusions: Based on our evaluation the anti-TNF-alfa treatment appears more effective than ustekinumab in the first line biologic treatment of psoriatic arthritis. PMS6 Drug Usage Analysis and Health Care Resources Consumption in Patients With Rheumatoid Arthritis Degli Esposti L 1, Sangiorgi D 1, Calandriello M 2, Buda S 1 1CliCon Srl, Ravenna, Italy, 2HE OR Unit - Bristol-Myers Squibb S.r.l., Rome, Italy . . . . Objectives: To prevent irreversible joint damage in rheumatoid arthritis (RA), biological drugs have been developed; when insufficient response is observed, the initial dosage could be increased (dose escalation). The aim of the study was to assess dose escalation among different therapeutic strategies and associated health care resources use. Methods: An observational retrospective cohort analysis based on 3 Local Health Units administrative databases was conducted. Patients who filled at least one prescription for biologic agents with a diagnosis of RA between 01/01/2009-31/12/2011 (enrolment period) were included. Patients were followedup for 12 months and characterized on the basis of the previous 12 months. Dose escalation and cost of illness were calculated on naïve patients (patients with no prior prescriptions of the index biologic); dose escalation was defined as having ≥2 consecutive claims with an average weekly dose 130% greater than the initial average weekly dose. Results: 594 patients were analyzed, female: male ratio=3: 1, age 54±14. Biologic at index date were Etanercept (39%), Adalimumab (25%), Infliximab (14%), Abatacept (10%), Tocilizumab (9%), Golimumab (3%). Naïve patients to index biologic were 293 (49%); among them, dose escalation was observed in 21.4% patients on Infliximab, 11.5% Adalimumab, 5.6% Abatacept, 4.0% Tocilizumab, 3.8% Etanercept. Hospitalized patients with RA as main discharge diagnosis were 9.6% Infliximab, 6.7% Golimumab, 6.6% Adalimumab, 2.9% Tocilizumab, 2.8% Etanercept, 2.4% Abatacept. Overall, 95% of cost of illness was driven by biologics, 2% traditional DMARDs, 3% non pharmacological costs; annual cost of illness was € 13,622 for Golimumab, € 12,803 Adalimumab, € 11,924 Etanercept, € 11,830 Tocilizumab, € 11,201 Infliximab, € 10,943 Abatacept. Dose escalation was associated with a higher expenditure for biologic (€ 12,248 VS € 11,232, p= 0.023) and no difference on other costs (€ 464 VS € 566, p= 0.349). Conclusions: Among patients with RA, Abatacept was associated to a lower cost of illness and hospitalization ratio, due to a low dose escalation rate. PMS7 Comparative Effectiveness of Biological Agents With Diseases Modifying Anti-Rheumatic Drugs For Methotrexate Failure Rheumatoid Arthritis Patients: A Baysian Mixed Treatment Comparison Choi M 1, Hyun M K 1, Choi S 1, Tchoe H J 1, Lee S Y 2, Son K M 2, Jung Y O 2, Kim H A 2 Evidence-based Healthcare Collaborating Agency, Seoul, South Korea, 2Hallym University, Chuncheon, South Korea . . . . . . . . . . . . . . 1National Objectives: Recently biological agents with a mechanism of action targeting the pathogenesis of rheumatoid arthritis (RA) have emerged as an important treatment. Especially, as anti-TNF (anti-tumor necrosis factor antagonist) drugs are effective A374 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 for patients who do not respond to traditional disease modifying anti-rheumatic drugs (DMARDs). This study aimed to compare the efficacy of biological agents with traditional DMARDs for methotrexate (MTX) treatment failure patients. Methods: Four DMARDs (hydroxychloroquine, sulfasalazine, methotrexate, leflunomide) and five anti-TNF drugs (adalimumab, etanercept, golimumab, infliximab, certolizumab) were selected according to expert consensus. A systematic search of the published systematic reviews was performed including MEDLINE, EMBASE and Cochrane Library. Among the identified 52 systematic reviews, 3 systematic reviews were finally selected and updated to July 2013. Data extraction and methodological quality assessment using Cochrane Risk of Bias was performed in pairs. Comparative efficacy was analyzed using Baysian mixed treatment comparison (MTC). Results: A total of 85 trials from 7,938 citations were included. Nineteen trials were grouped as MTX failure patients (mean age: 52.9 years, mean of rheumatoid factor positive rate: 76.6%). Nine studies were included in the analysis of Health Assessment Questionnaire (HAQ). The best treatment was certolizumab combined with MTX (MD -0.40,95% CrI -0.95 to 0.13). For comparative effects on DAS 28-ESR < 2.6 (remission), 4 trials were included in analysis. The best treatment was golimumab combined with MTX (OR 24.5, 95% CrI 3.51 to 99.52). For comparative effects on ACR 70, the best treatment was certolizumab combination with MTX (OR 10.46, 95% CrI 3.66 to 24.41) in 11 trials. Conclusions: In the MTX failure patients, certolizumab combination with MTX lowered HAQ score than MTX. The result of DAS 28- ESR< 2.6 (remission), golimumab combined with MTX was the most effective treatment. Certolizumab combined with MTX was best treatment for the ACR 70 response. PMS8 Quality Of Life Assessments In Korean Patients With Rheumatoid Arthritis (Ra): An Analysis From The Phase Iii Trial To Evaluate Equivalence Of The Etanercept Biosimilar Hd203 And Enbrel® In Combination With Methotrexate (Mtx) In Patients With Ra; The Hera Study Bae S C 1, Lee S R 2, Ahn Y 3 1Hanyang University Hospital for Rheumatic Diseases, Seoul, South Korea, 2Hanwha Chemical Biologics, Seoul, South Korea,, 3Hanwha Chemical, Daejeon, South Korea . . . . . Objectives: Quality of Life (QoL) is important for patients with RA and Enbrel® has demonstrated QoL benefits in this patient group. HD203 is an etanercept biosimilar, which reported pharmacokinetic bioequivalence to the reference product Enbrel® in a Phase I trial and, more recently, demonstrated equivalent efficacy and comparable safety in a phase III randomized trial in Korean patients with RA. Assessing QoL was a secondary objective of the HERA study, the results of which are reported here. Methods: Patients (male or female aged ≥ 20 years) with active RA were randomized (1: 1) to 25 mg HD203 or Enbrel®, administered subcutaneously twice weekly with MTX for 48 weeks. QoL assessments (Short Form 36, SF-36; Functional Assessment of Chronic Illness Therapy-Fatigue, FACIT-F; EuroQol-5 dimension, EQ5D) were undertaken at weeks 24 and 48. Results: In total, 294 patients were randomized (147 to HD203; 147 to Enbrel®). There was no significant difference between groups on QoL assessments at baseline. QoL assessments were similar for HD203 and Enbrel® at week 24 and 48 overall. However, SF-36 Role Emotional and Bodily Pain subscales showed significant increase in favour of HD203 at week 24 (p= 0.0252) and week 48 (p= 0.0243) respectively. No significant difference was observed between HD203 and Enbrel® for FACIT-F scores at week 24 or 48, except at week 48 for the Emotional Well-being domain, which was significantly improved with Enbrel® vs. HD203 (p= 0.0360). No significant differences between groups in EQ5D scores were observed at any time. Conclusions: Together with previous reports of equivalent pharmacokinetics and efficacy, and comparable safety, these data support the biosimilarity of HD203 vs. Enbrel®. PMS9 Comparison of Disease Status And Outcomes of Patients With Rheumatoid Arthritis (Ra) Receiving Adalimumab or Etanercept Monotherapy In Europe Narayanan S1, Lu Y 2, Hutchings R 2, Baynton E 2, Hautamaki E 1 Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, London, UK . . . . 1Ipsos Objectives: To compare the disease status and outcomes of patients with RA receiving adalimumab and etanercept monotherapy in Europe. Methods: A multi-country, multi-center medical chart-review study of RA patients was conducted among rheumatologists in UK/France/Germany/Italy/Spain to collect deidentified data on patients who were recently treated with a biologic as part of usual care. Physicians were screened for duration of practice (3-30 yrs) and patient volume (incl. > 5 RA biologic patients/month) and recruited from a large panel to be geographically representative in each country. Eligible patient charts (≥ 3) were randomly selected from a sample of prospective patients visiting each center/practice during the screening period. Physicians abstracted patient diagnosis, treatment patterns/dynamics and patient symptomatology/disease status/outcomes. Patients on adalimumab/etanercept monotherapy were analyzed. Results: 428 eligible RA patient charts were abstracted; 152 on adalimumab (female: 68%, age: 52.1yrs, average months on adalimumab: 22.8, 90% on first biologic) and 142 on etanercept (female: 70%, age: 50.5yrs, average months on etanercept: 24.6, 97% on first biologic). Among patients with available data, latest lab measures documented were (adalimumab vs. etanercept): ESR: 21.0mm/h (range: 16.1 (UK) -24.6 (Germany)) vs. 23.4mm/h (range: 13.5 (France) -29.8 (Italy)), CRP: 9.9mg/dl (range: 5.0 (Italy) -17.0 (Germany)) vs. 13.1mg/dl (range: 7.7 (France) -15.6 (Italy)), rheumatoid factor-positive: 81% (range: 68% (Germany) -86% (Spain)) vs. 85% (range: 62% (Germany) -95% (Spain)), and anti-CPP-positive: 69% (range: 47% (Germany) -83% (France)) vs. 74% (range: 61% (Spain) -83% (France)). Latest disease severity measures documented were (adalimumab vs. etanercept): Swollen Joint Counts: 2.7 (range: 2.2 (Spain) -3.5 (Germany)) vs. 2.6 (range: 0.7 (Spain) -6.3 (UK)), Tender Joint Counts were 3.9 (range: 3.2 (France) -5.1 (Germany)) vs. 3.7 (range: 1.5 (Spain) -7.1 (UK)), Health Assessment Questionnaire (HAQ) rating: 1.6 (range: 0.1 (Spain) -2.3 (Germany)) vs. 1.4 (0.3 (Spain) -2.5 (Germany), DAS28 score: 3.8 (range: 2.5 (UK) -4.8 (Spain)) vs. 3.2 (1.5 (Germany) -6.6 (Spain), VAS score: 3.8 (range: 3.0 (UK) -4.8 (Italy)) vs. 3.1 (range: 1.9 (Germany) -3.9 (Italy)). Conclusions: Among RA patients receiving adalimumab or etanercept monotherapy, disease severity was similar across the EU5, with patients in Spain and France having relatively lower, and patients in Italy, Germany, and UK having relatively higher burden and poorer treatment response. Factors influencing the observed patterns of geographic variation warrant further scrutiny to optimize therapeutic interventions and improve outcomes. PMS10 Efficacy Of Novel Dmards In Early Active Rheumatoid Arthritis: An Indirect Comparison Sawyer L 1, Chang S 1, Diamantopoulos A 2, Dejonckheere F 3 1Symmetron Limited, Herts, UK, 2Symmetron Limited, Borehamwood, UK, 3F. Hoffmann-La Roche, Basel, Switzerland . . . . Objectives: We evaluated the effectiveness of traditional disease-modifying antirheumatic drugs (tDMARDs) and novel DMARDs, alone and in combination, in methotrexate- and biologic-naive adults with moderate to severe early rheumatoid arthritis (ERA; < 3-year duration). Methods: Literature review identified randomized controlled trials (RCTs) of tDMARDs and novel DMARDs reporting efficacy outcomes (American College of Rheumatology [ACR] 20/50/70/90 response and Disease Activity Score at 28 joints [DAS28] remission). Data were pooled using Bayesian network meta-analysis techniques. For ACR response, data were analyzed using a fixed-effects ordered probit model, which makes efficient use of ordered categorical data and guarantees coherent prediction of multinomial response probabilities. For DAS28 remission, data were analyzed with a fixed-effects binomial logit model. Sensitivity analyses tested the effects of grouping treatments by class and broadening/narrowing inclusion criteria. Results: Results from a synthesis of 16 RCTs of tDMARDs (methotrexate, sulfasalazine, hydroxychloroquine) and novel DMARDs (biologics [abatacept, adalimumab, etanercept, infliximab, golimumab, tocilizumab] and tofacitinib) indicated that biologics+methotrexate, triple tDMARDs, and tocilizumab and tofacitinib monotherapy significantly increased response across all ACR categories versus methotrexate. ACR response probabilities for biologics+methotrexate were not significantly different between agents. ACR response probabilities to novel DMARD monotherapy varied, trending toward higher values for tofacitinib and tocilizumab than etanercept or adalimumab. In studies reporting DAS28 remission, treatment with tofacitinib or biologics±methotrexate, except adalimumab alone, improved remission likelihood versus methotrexate. Tocilizumab±methotrexate generated the highest probability of remission among biologics and was significantly more effective than other biologics±methotrexate and tofacitinib. Results across outcomes were robust to alternative grouping of interventions and change in inclusion criteria. Conclusions: Based on ACR response, the expected efficacy of biologics+methotrexate, tofacitinib and tocilizumab monotherapy, and triple tDMARD therapy appeared higher than MTX in ERA. Tocilizumab±methotrexate was expected to have the highest probability of generating DAS28 remission and was significantly more effective than other biologics±methotrexate and tofacitinib. PMS11 Using Health Assessment Questionnaire – Disability Index To Estimate Eq-5d Utility Values for Patients With Rheumatoid Arthritis in Taiwan Tang C H 1, Hsu J Y 1, Hsu P N 2 1Taipei Medical University, Taipei, Taiwan, 2National Taiwan University, Taipei, Taiwan . . . . . . Objectives: This study aims to provide estimates of EQ-5D as a function of Health Assessment Questionnaire – Disability Index (HAQ-DI) scores in patients with rheumatoid arthritis (RA) in Taiwan. Methods: Face-to-face patient interviews on a total of 140 patients aged between 30 and 70 years old were recruited at the rheumatology outpatient clinics of four hospitals located in northern, central and southern Taiwan during June 2013-May 2014. The severity distribution of patients was mild RA (Disease Activity Score [DAS 28] < 3.2) (N= 57), moderate RA (3.2□… DAS< 5.1) (N= 44), and severe RA (DAS≥ a5.1) (N= 39). Socio-demographic and clinical information were collected, and the HAQ-DI and the EQ-5D questionnaires were completed. Generalized linear regression models were used to predict EQ-5D utility values as functions of HAQ-DI scores, age, and gender. Results: Patient mean age was 50.8 years old (standard deviation [SD], 11.3 years); 81.4% of the patients were women and mean disease duration was 9.65 years (SD, 6.84 years). HAQ-DI <0.5, 58%; 0.5□…HAQ-DI<1.1, 16%; 1.1□…HAQ-DI<1.6, 9%, 1.6□… HAQ-DI<2.1, 12%; and HAQ-DI≥ a2.1, 4%. HAQ-DI and EQ-5D mean scores were 12.01 (SD, 7.8) and 0.67 (SD, 0.34), respectively. The models were able to predict actual EQ-5D across the range of the HAQ DI. Age and gender were found to be significant determinants in estimating the utility functions. Conclusions: Utility values have very often not been assessed in the data collection process in a clinical trial. This study showed that HAQ-DI scores can be used to derive EQ-5D utility values for patients with RA in Taiwan to facilitate conducting a cost-utility analysis. PMS12 Heal Rate In 4,190 Fresh Fractures Treated With Low-Intensity Pulsed Ultrasound (Lipus) Zura R 1, Mehta S 2, Della Rocca G 3, Jones J 4, Steen R G 4 1Duke University Medical Center, Durham, NC, USA, 2Hospital of the University of Pennsylvania, Philadelphia, PA, USA, 3University of Missouri, Columbia, MO, USA, 4Bioventus, LLC, Durham, NC, USA . . . . . . Objectives: Patient age is one of many potential risk factors for fracture nonunion. We evaluate impact of age on heal rate (HR) in patients with fresh fracture (≤ 90 days old). Our hypothesis is that age is not a risk factor for nonunion if fractures are treated with LIPUS. Methods: A LIPUS device was approved in 1994 to accelerate fresh fracture healing, though the FDA required a Post-Market Registry. Patient data were collected from October 1994 until October 1998 and were reviewed and A375 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 validated by a registered nurse. We required 4 data elements to report a patient: date when fracture occurred; date when treatment began; date when treatment ended; and outcome (healed vs. failed, by clinical and radiological criteria). Data were used to calculate: days to treatment (DTT); and days on treatment (DOT). All fresh fractures with DTT, DOT, and outcome are reported. Results: 5,765 patients in the registry had fresh fracture; 73% of patients (N= 4,190) are reported; 13% of patients were lost to follow-up; 11% withdrew or were noncompliant; and 3% died or are missing outcome. Among compliant patients, HR was 96.2%. Logistic estimates of the odds ratio for healing are equivalent for patients aged 30 to 79 years. Nevertheless, patients who failed treatment were 4.5 years older than patients who healed (p < 0.0009). DTT was significantly shorter for patients who healed (p < 0.0001). Data show that obesity, smoking, diabetes, vascular insufficiency, osteoporosis, cancer, rheumatoid arthritis, and chronic use of NSAIDs reduce HR. Conclusions: LIPUS mitigates the effect of age on fracture HR. Patients who used LIPUS had a 96% HR, whereas the expected HR averages 93%. Time to treatment was significantly shorter among patients who healed (p < 0.0001), suggesting it is beneficial to begin treatment early. Comorbid conditions in conjunction with aging can reduce fracture HR. PMS13 Pain Therapy for Osteoarthritis in Germany: Analysis of Sickness Fund Claims Data Tomeczkowski J 1, Wimmer A 2, Pappert K 2 1Janssen-Cilag GmbH, Neuss, Germany, 2Janssen-Cilag, Neuss, Germany . . . Objectives: Using sickness fund claims data, we sought to determine osteoarthritis rate, drug compound class, pain therapy prevalence and type of medical specialists providing treatment. Methods: A group of company health-sickness funds (approx. 2.1 million insured patients in 2010; 2.5 million insured patients in 2011) was used. Osteoarthritis was identified based on ICD-10 diagnoses (M16.0-9, M17.0-5, M17.9, M19.05, M19.25, M19.85, M19.95), then linked to prescriptions using the ATC codes: M01A (nonsteroidal anti-inflammatory drugs, NSAIDS), N02B (analgesics and antipyretics), and N02A (opioids). Furthermore, we determined which groups of medical specialists prescribed the drugs. Results: Osteoarthritis was diagnosed in 7.8% (in 2010) and in 7.1% (in 2011) of patients. In one year, 65.4% of patients received a prescription for at least one drug from the analysed ATC codes: 81.4% of patients received at least one NSAID, 36.4% an analgesic and antipyretic, and 27.4% an opioid. For M01A, diclofenac (54%) was most frequently prescribed; the proportion of coxibs was 6%. For N02B, 99% of prescriptions were for metamizol; 1% for paracetamol. For N02A, most prescriptions were for tramadol (29%) or tilidin (28%). General practitioners most frequently prescribed these drugs (42.2% [M01A]/46.2% [N02B]/45.9% [N02A]). Conclusions: In Germany in 2010-2011, OA prevalence was 7-8%, and associated with analgesic prescriptions for the majority of evaluated patients. Diclofenac (NSAIDs, metamizol (analgesics and antipyretics), and tramadol or tilidin (opioids) were most frequently prescribed in each group. General practitioners were the most frequent painkiller prescribers. Muscular-Skeletal Disorders – Cost Studies PMS14 A Budget Impact Analysis of Ustekinumab in the Management of Psoriatic Arthritis in Greece Kousoulakou H 1, Natsikas C 2, Geitona M 1 of Peloponnese, Corinth, Greece, 2Janssen Pharmaceutical Greece, Pefki, Greece . . . 1University Objectives: Ustekinumab was approved for the treatment of psoriatic arthritis (PsA) in September 2013. The objective of the study was to estimate the budget impact of ustekinumab in the management of PsA in Greece. Methods: A budget impact model was developed in Excel 2010 comparing the total PsA treatment costs in the current treatment pathway (including golimumab, adalimumab, etanercept and infliximab) with the respective costs of a treatment mix with the inclusion of ustekinumab. Market share data for the current treatment pathway were based on market research. Epidemiology data were taken from the published literature. Due to lack of published data on resource use, a 60-field questionnaire was developed in order to collect local data relating to the management of PsA in Greece. Two expert panels were convened, one with 8 KOL dermatologists and one with 8 KOL rheumatologists, with the Delphi technique. Unit costs were retrieved from publically available sources. The time horizon was five years and the analysis was conducted from the Social Insurance Fund perspective. Results: The total number of eligible patients (incident and prevalent cases) was estimated to increase from 6,448 in Year 1 to 7,754 in Year 5. The total cost in the current treatment pathway was estimated to range between €48.4 million in Year 1 and €20 million in Year 5. The costs in the treatment pathway including ustekinumab were €47.8 and €18.5 million, in the respective years. Therefore, the addition of ustekinumab in the treatment mix can lead to cumulative savings for the Social Insurance Funds of € 7.7 million, over the 5-year time horizon. This cost reduction is mainly attributed to the less frequent administration of ustekinumab. Conclusions: Inclusion of ustekinumab in the treatment mix appears to be a cost saving treatment option in the management of PsA in Greece. PMS15 Budget Impact Analysis of Certolizumab Pegol in the Management of Patients With Moderate-To-Severe Active Rheumatoid Arthritis in Greece Tzanetakos C 1, Maniadakis N 1, Kourlaba G 2, Tzioufas A 3, Goules A 3, Theodoratou T 4, Christou P 4 1National School of Public Health, Athens, Greece, 2Collaborative Center for Clinical Epidemiology and Outcomes Research (CLEO), Athens, Greece, 3Athens Medical School, Athens, Greece, 4UCB Pharma, Athens, Greece . . . . . . . Objectives: To investigate the budgetary impact of increasing the patient share of certolizumab pegol (CZP) versus the other recommended biologic disease modifying anti-rheumatic drugs (bDMARDs; etanercept, adalimumab, golimumab, infliximab, abatacept, tocilizumab) for the treatment of moderate-to-severe active rheumatoid arthritis (RA) in Greece. Methods: A budget impact model was adapted from a thirdparty payer perspective (National Organization for Healthcare Services Provision [EOPYY]) to evaluate economic aspects of RA treatment over 5 years (2014–2018). The model assumed Greek epidemiological data and local reimbursement requirements. Two main scenarios, following either a conservative or an increased market uptake of CZP in the Greek health care market, were estimated and individually compared to the current market trend scenario, which incorporates original biologics erosion from biosimilars entry in the coming year. Costs pertaining to drug acquisition, administration (only for intravenous drugs), and monitoring were included in the analysis and corresponded to 2014 costing year. Officially published sources were used to derive unit costs. The outcome measures were the annual cost of treatment with bDMARD presented as total cost and disaggregated by drug cost, administration cost and monitoring cost, as well as the incremental cost savings per year. Results: Comparing CZP current versus conservative market uptake scenarios, the total budget was slightly increased by € 0.05 million. In contrast, comparing CZP current versus increased market uptake scenarios, the total budgetary savings were €0.23 million. In the latter comparison setting, the cost savings were attributed to reduced drug and administration costs. More specifically, the greater replacement of an intravenously administered bDMARD (infliximab) conduced to the greater reduction of administration costs than in the former comparison setting (cost savings: € 0.17 vs. € 0.14 million). Conclusions: A potential increased use of CZP treatment was shown to be associated with cost savings over the next 5 years in Greece. PMS16 Pharmacoeconomic Evaluation of Biologic Therapies in Russian Patients With Rheumatoid Arthritis and Intolerance or Inadequate Response to Conventional Basic Therap Ryazhenov V V 1, Gorokhova S G 1, Maximkin S A 2 First Moscow State Medical University, Moscow, Russia, 2Center for strategic research in healthcare, Moscow, Russia . . . . . . 1I.M. Sechenov Objectives: About 2.79% of patientswith rheumatoid arthritis (RA) in Russia receive TNF-α inhibitors; this value may vary significantly and depends on limitations of regional budgets. In this respect, the aim of our study was to conduct a comparative pharmacoeconomic analysis of the most prevalent TNF-α inhibitors in Russian patients with rheumatoid arthritis and intolerance or inadequate response to conventional basic therapy. Methods: An pharmacoeconomic model was developed based on the data from indirect comparison of anti-TNF-α agents. The model considers the use of infliximab, etanercept, certulizumab pegol, adalimumab in patients with RA who lost response to conventional basic therapy. Cost-effectiveness and costs of TNF-α inhibitors for health care budget were estimated. The cost analysis included costs of pharmacotherapy. Infliximab and etanercept are included into the list of vital and essential medicines and were considered as accepted technologies in budget impact analysis: certulizumab pegol and adalimumab were novel tecnologies in our model. A 24-weeks horizon was adopted. Sensitivity analysis (SA) was performed by changing costs of medicines Results: the costs of therapy in certulizumab pegol and etanercept groups were significantly lower than in infliximab and adalimumab groups. The cost-effectiveness ratios (CERs) in terms of ACR20 in 24 weeks were 703 625.00, 587 776.09, and 4 119 260.82 for certulizumab pegol, etanercept and infliximab groups, respectively. The same was observed in case of ACR50 and ACR70: a strategy of drug use in certulizumab pegol and etanercept groups was preferable in comparison with infliximab and adalimumab groups. Budgetary costs for health care system were higher in case of infliximab and adalimumab. SA confirmed the robustness of the model Conclusions: The study demonstrated that certulizumab pegol and etanercept are an economically effective strategy for Russian patients with RA and lost response to conventional basic therapy. PMS17 Mast (Minimal Access Spinal Technologies) Versus Open Surgery: Cost Analysis From Hospital Perspective Lukyanov V 1, Melik-Gusseinov D 1, Borgman B 2, Demourenko D 3, Mlyavykh S 4 1Center of Social Economics, Moscow, Russia, 2Medtronic International, Tolochenaz, Switzerland, 3Medtronic LLC, Moscow, Russia, 4Scientific Research Institute of Traumatology and Orthopedics, Nizhniy Novgorod, Russia . . . . . Objectives: The objective of this study was to determine and compare the resource consumption associated with open vs. minimal invasive surgery in patients with degenerative spinal disorders in the Russian hospital setting. Methods: Conducted cost-analysis was based on Moscow hospital setting, where resource utilization associated with average one-level spinal operation was determined through interviews with KOLs in spinal surgery. Costs were retrieved from public sources and hospital data for the following categories 1) hospital stay; 2) blood transfusion 3) consumables (suture materials, hemostatic sponges, disposable instruments); 4) time in the operating room; 5) spinal implants/instrumentation; and 6) complications. Results: The results of the calculations have confirmed MAST economic advantages over open surgery (OS). MAST was associated with fewer costs, mainly due to shorter stay in intensive care unit (1 vs. 2 days) and general ward (9 vs. 15 days), no need for blood transfusion and less rate of complications. The difference in the duration of surgery, which depends mainly on the speed of approach and the installation of implantable structures, is approximately 20 minutes in favor of MAST. With the cost of one-hour long surgery at about 6,000 rubles (167$), excluding the cost of implantable structures, the use of MAST instead of OS translates into savings of 2,000 rubles (56$) per each surgical intervention. As for overall budget savings, the use of MAST translates into savings of between 14,783 ($410) and 35,000 ($970) rubles per whole hospital visit, depending on what materials and structures are used. Conclusions: The economic evaluation confirms economic domination of MAST over OS. Despite initial higher investments, MAST appears to be a cost saving alternative to OS, in terms of diminution of actual surgery time, reduction of blood transfusion costs, and prevention of post-surgery complications and shorter overall length of hospital stay. A376 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1HERESCON GmbH, Hannover, Germany, 2Medtronic International, Tolochenaz, Switzerland, GmbH, Meerbusch, Germany, 4Spine Center Osnabrueck, Osnabrueck, Germany, 5Klinik für Wirbelsäulenchirurgie, Dortmund, Germany PMS18 Cost Comparison of Surgical and Non-Surgical Treated Lumbar Spinal Stenosis Patients 3Medtronic Jacob C 1, Brasseur P 2, Haas J S 1, Braun S 1 1HERESCON GmbH, Hannover, Germany, 2Medtronic International Trading Sarl, Tolochenaz, Switzerland . . . . . Objectives: Lumbar spinal stenosis (LSS) occurs as a degeneration of the spine in aging populations. Treatment options comprise surgical and non-surgical intervention. The aim of this study was to compare annual costs between LSS patients treated with instrumental spinal surgery (ISS) and those non-surgically treated. Methods: A retrospective claims data analysis was conducted using anonymized claims data from the Health Risk Institute research database. The study period comprised 01 January 2009 to 31 December 2011. LSS patients receiving an ISS were compared to an age and gender matched non-operated control group with comparable disease state. Patients were identified by ICD-10-GM code M48.0* in the inpatient setting. Operation and procedure codes (OPS) were used to identify ISS. Comparable disease state was achieved by matching total costs in an individual period of 12 months before the first LSS caused hospitalization. Annual costs after surgical treatment were compared for LSS patients receiving ISS and those with no surgical treatment. Results: A total of 2,027 patients with LSS were identified in 2010 records. Surgical treatment with ISS was applied in 542 cases whereas 1,485 individuals received non-surgical treatment. After matching both groups for age, gender, and comparable disease state, 393 patients in each group were available for the cost comparison. Mean annual costs were €8,458 higher in the ISS-treated group; clearly due to average ISS-cost of € 9,644. In contrast, costs for outpatient care and pharmaceuticals decrease after the surgery. Conclusions: Surgical treatment for LSS patients exceeds the cost of non-surgical treatment in the first year after surgery. A cost offset is not achievable in this period due to the high cost of the surgical intervention. Nevertheless, cost savings were already observed in pharmaceutical therapy and outpatient care. Further research is needed to determine if overall cost savings could be achieved in an extended timeframe. . Tencer T , Clancy Z , Zhang F Celgene Corporation, Warren, NJ, USA . . Objectives: The purpose of this study was to estimate the annual costs and the cost per responder for psoriatic arthritis (PsA) patients treated with apremilast, etanercept, and adalimumab in adults with PsA in the United States. Methods: Comparative efficacy data were obtained from a Bayesian network meta-analysis of biologic and non-biologic disease-modifying antirheumatic drugs as of October 2013. The primary outcome was ACR20 response at Week 24. Response rate differences from the clinical trials were assumed to be maintained for 52 weeks. US wholesale acquisition cost as of June 2014 and approved labeled dosing were used to derive drug treatment costs. Results: At Week 24, the adjusted ACR20 response rate was 40.3% for apremilast, 53.4% for etanercept, and 57.8% for adalimumab. The cost per ACR20 responder at Week 24 was $23,562 for apremilast, $30,346 for etanercept, and $25,978 for adalimumab. By Week 52, the cost per ACR20 responder was $53,704 for apremilast, $65,750 for etanercept, and $56,273 for adalimumab. The annual cost to achieve 100 responders was $5,370,387 for apremilast, $6,574,981 for etanercept, and $5,627,336 for adalimumab. Conclusions: Apremilast had the lowest wholesale acquisition costs per ACR20 responder and the lowest annual cost to achieve 100 ACR20 responders, as compared with etanercept and adalimumab through 52 weeks in PsA patients. PMS20 Direct Non-Medical Costs Of Rheumatoid Arthritis By Disease Level In Portugal Mateus C , Moura A Escola Nacional de Saúde Pública, Universidade Nova de Lisboa, Lisboa, Portugal . . Objectives: Direct non-medical costs may be difficult to assess, so these are often disregarded in cost-of-illness studies. This analysis aims to estimate the annual direct non-medical costs of treating a patient with Rheumatoid Arthritis (AR) in Portugal, per level of disease activity. Methods: Patients with RA followed by 5 different rheumatologists across the country, were asked to participate in a survey, and informed consent was given by all patients. Besides socio-economic data, the survey included questions regarding the following cost components: number of medical appointments and treatments attended by the patient and its associated travel expenses; expenditures related to adaptations made to their homes and the purchase of medical devices; amount of domiciliary support received, both paid and unpaid. In case of paid domiciliary support patients were also asked about the associated amount. That figure was then used in order to assess the value of unpaid domiciliary support. Disease activity was measured by the DAS28 score. Results: Our final sample consists of 90 patients and descriptive statistics are in line with RA epidemiology (73 women, mean age 56.9). The percentage of patients with low disease activity was 27.7%, whereas, of 43.3% and 28.8% for those with moderate and high disease activity, respectively. Estimated direct non-medical costs ranged from €1,339 to €5,475 per year, for patients with low and high disease activity, respectively. More than 70% of these costs were related to paid and unpaid domiciliary support. Conclusions: Direct non-medical costs are positively associated with disease activity. Estimated costs have a non-negligible magnitude and are entirely borne by patients. Our results confirm the heavy economic burden that RA places on patients and their families. PMS21 Claims Data Analysis on the Annual Frequency and Incremental Cost of Reoperations in Instrumental Spinal Surgeries in Germany Jacob C 1, Annoni E 2, Haas J S 1, Witthohn A 3, Braun S 1, Grünert J 3, Winking M 4, Franke J 5 . . . . . . . PMS22 Current and Future Costs of Osteoporotic Fractures in the Netherlands Lötters F J B 1, de Vries F 2, van den Bergh J 2, Rutten-van Mölken M 1 1Erasmus University Rotterdam, Rotterdam, The Netherlands, 2Maastricht University Medical Center, Maastricht, The Netherlands PMS19 Cost Per Responder Of Apremilast Versus Etanercept And Adalimumab In Patients With Active Psoriatic Arthritis . Objectives: Complications in instrumental spinal surgeries (ISS) pose a considerable burden on patients. Necessary reoperations are associated with significant resource utilization and cost and from the perspective of the German Statutory Health Insurance (SHI). Dependable data on the frequency of reoperations and associated costs are lacking for Germany. The aim of this study was to estimate the incidence of ISS and consecutive reoperations, and to calculate the related costs. Methods: We conducted a retrospective claims data analysis using the Health Risk Institute research database, which contains anonymized claims data and covers approximately 5.42% of the German population. The study period comprised 01 January 2009 to 31 December 2011. An algorithm of operation and procedure codes (OPS) identified primary ISS and following reoperations. Reoperation rates were calculated for an individual period of 12 months after the primary ISS in 2010. Annual costs for reoperations were calculated based on group comparison of patients with reoperation and those without reoperation (control group). Existing differences in cost levels in the year before the primary ISS were adjusted by the difference in differences approach. Results: A total of 3,316 individuals had a primary ISS in 2010. The reoperation rate was 9.98% (95% CI = 8.98% to 11.02%). Mean cost per ISS was € 11,331 for all patients (€ 13,358 reoperation group, € 11,106 control group). The mean adjusted annual cost for a reoperation was € 11,370, with € 8,432 directly attributed to the reoperation procedure and € 2,938 to excess costs in the first year after the primary ISS. Conclusions: The direct cost of ISS has a significant impact on health insurance budgets. With 10% of primary ISS patients requiring a reoperation in Germany, their associated annual costs are relevant from the SHI perspective. As demonstrated elsewhere, these cost might be partly avoidable by using intra-operative 3-D imaging with navigation. . . . . . . . Objectives: This study aims to estimate the incidence and costs of osteoporotic fractures in The Netherlands in 2010 and project them to 2030. Methods: The incidence and health care costs of fractures were derived from claims data of all health care insurers in The Netherlands. We obtained 5-year age- and gender-specific costs of patients with and without fractures. Cost included hospital admission, physical therapy, occupational therapy, general practitioner and medication. In order to attribute fractures to osteoporosis we used a large dataset from a general hospital that included patients with a fracture screened with Dexa scan. Future projections were based on four different scenarios: 1. demographic scenario, 2. demographic+trend in incidence scenario, 3. demographic+trend in incidence+trend in cost scenario, and 4. increased treatment scenario. Results: Of all registered fractures 32% could be attributed to osteoporosis. In women this percentage was larger than in men (36 versus 21 %). This resulted in an incidence for all osteoporotic fractures of 964 per 100.000 in women and 245 per 100.000 in men for 2010. Over time (2010-2030) the overall increase in incidence of osteoporotic fractures was 40% (scenario 1). The increase in hip fractures ranged from 60% (scenario 1) to 79% (scenario2). In 2010 approximately 200 million Euros was spend on treatment of osteoporotic fractures. The costs for osteoporotic fractures increased with 50% from 2010 till 2030 (scenario 1). The increase in costs for hip fractures was highest, ranging from 60% (scenario1) to 148% (scenario 3&4 combined), resulting in cost estimates in 2030 of 161 and 249 million, respectively. Prevention of osteoporotic fractures in general can lead to cost-savings of 92 million in 2030 (scenario 4 & 1 combined). Conclusions: The expected high increase in incidence and costs of osteoporotic fractures calls for a wider use of prevention and treatment options. PMS23 Burden Of Disease Analysis Of Ankylosing Spondylitis In Hungary Hegyi R 1, Nagy B 1, Koncz A 2, Huybrechts I 3, Lavicky J 4, Ferenczik A 1 Consulting Ltd., Budapest, Hungary, 2UCB Pharma, Budapest, Hungary, 3UCB Pharma, Brussels, Belgium, 4UCB Pharma, Prague, Czech Republic . . . . . . 1Healthware Objectives: Ankylosing spondylitis (AS) entails an individual burden to patients and ties up resources. This study aimed to assess the total costs of AS, including the indirect burden of AS patients in Hungary and to obtain an overview of patients’ status, demographics, morbidity, working capacity and other characteristics. Methods: Between January–March 2014, a questionnaire survey was conducted among AS patients, which was filled out voluntarily and anonymously. Missing data was not imputed in the analysis; considered patient number is presented next to results if lower than total patient number. Results: 152 patients completed the questionnaire, of which 37% were women. Mean age was 51 years (Standard Deviation [SD]: 13 years) and average disease duration was 17 years (SD: 12 years). At primary diagnosis of AS, 80% of patients had a full-time job, 2% a part-time job, and only 8% received disability pension. At time of survey, only 36% of patients worked full-time, 1% part-time, and the proportion of disability pensioners increased to 42%. Cost calculation results* showed that the average annual total cost per AS patient was 5,155€ . Within this, average annual direct nonmedical cost was over 1,976€ and average annual indirect cost per patient was approximately 3,129€ (145 patients). Wage loss due to disability pension generated the highest average annual indirect cost per patient (3,290€ – 124 patients). In the working-age population (118 patients, 31–62 years), total average cost per patient was 5,996€ . Conclusions: Due to their disease, AS patients can become partially or completely disabled, which imposes a significant burden directly on their environment and indirectly to society. Average costs were associated with disease duration. Patients may already be driven out from the labour market in their active A377 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 ages because of the disease’s symptoms and variability, leading to reduced social engagement. *exchange rate: 303,59HUF/EUR PMS24 Early Retirement Indirect Costs Attributable to Rheumatic Diseases in Portugal Laires P A 1, Canhão H 2, Gouveia M 3 Portuguesa de Reumatologia, Lisbon, Portugal, 2Faculdade de Medicina, Universidade de Lisboa, Lisboa, Portugal, 3Católica Lisbon School of Business and Economics, Lisbon, Portugal . . . . 1Sociedade Objectives: Rheumatic diseases (RD) cause physical disability that may lead to early retirement, generating high indirect costs to society. We estimate these costs in the Portuguese population approaching the statutory retirement age. Methods: Health, sociodemographic and occupational data was retrieved from the 4th National Health Survey (2005/2006), for all people between 50 and 64 years of age (3,762 men and 4,241 women), whilst an official wage national database was used to estimate productivity values by gender, age-group and region, using the human capital approach. The effects of RD on the likelihood of early retirement and the attributable fractions estimates were obtained at the individual level by logistic regression. Results: At the time of the survey, 37.2% of the population with ages 50-64 years old self-reported at least one RD and among these 22.7% were officially retired, compared with 17.6% of those without RD (p< 0.001). Females had higher prevalence of RD (49.8% vs. 23.4% for males; p< 0.001). Presence of RD was associated with early retirement regardless of age, gender and region (OR=1.3; CI 95%: 1.0 - 1.6). The estimated annual indirect costs following premature retirement attributable to RD were € 367 million (€ 504 per RD patient). Age-groups closer to the statutory retirement age contribute more to the overall indirect costs and also have higher individual productivity values. For instance, 60-64 years old age-group contribute with 52% of overall costs and had an average indirect cost of 699€ per RD patient. Females were responsible for about 55% of these costs, however males contribute with higher individual productivity losses (cost per patient: 788€ vs. 390€ , for males and females respectively). Conclusions: Annual early retirement indirect costs attributable to RD are considerable. These results highlight the need to target patients with rheumatic conditions in order to obtain health and productivity gains and reduce early retirement in society. PMS25 Burden of Disease Analysis of Psoriatic Arthritis In Hungary Hegyi R 1, Nagy B 1, Koncz A 2, Huybrechts I 3, Lavicky J 4, Ferenczik A 1 Consulting Ltd., Budapest, Hungary, 2UCB Pharma, Budapest, Hungary, 3UCB Pharma, Brussels, Belgium, 4UCB Pharma, Prague, Czech Republic . . . . . Objectives: Psoriatic arthritis (PsA) entails an individual burden to patients and ties up resources. This study aimed to assess the total costs of PsA, including the indirect burden of PsA patients in Hungary and to obtain an overview of patients’ status, demographics, morbidity, and working capacity. Methods: Between January–March 2014, a questionnaire survey was conducted among PsA patients and filled out voluntarily and anonymously. Missing data was not imputed in the analysis; considered patient number is presented next to results if lower than total patient number. Results: 145 patients completed the questionnaire, of which 57% were women. Mean age was 54 years (Standard Deviation [SD]: 14 years) and average disease duration was 17 years (SD: 11 years). At primary diagnosis of PsA, 79% of patients had a full-time job, 4% a part-time job and only 3% received disability pension. At time of survey, only 32% of patients worked full-time, 4% part-time, and the proportion of disability pensioners increased to 28%. Cost calculation results* showed that the average annual total cost per PsA patient was 3,842€ . Within this, average annual direct non-medical cost was approximately 1,318€ (141 patients) and average annual indirect cost per patient was approximately 2,635€ (142 patients). Wage loss due to disability pension generated the highest average annual indirect cost per patient (2,734€ – 119 patients). In the working-age population (95 patients, 31–62 years) total average cost per patient was 5,334€ . Conclusions: PsA can cause patients to become partially or completely disabled, which imposes a significant burden directly on their environment and indirectly to society. Average direct/ indirect costs are higher with longer disease duration and greater skin manifestations. Patients may already be driven out from the labour market in their active ages because of the disease’s symptoms and variability, leading to reduced social engagement. *exchange rate: 303,59HUF/EUR PMS26 Economic Modeling of The Use of Botulinum Toxin A in A Homogenous Patient Population Based on Real-Life Clinical Practice: Ulis-Ii (The Upper Limb International Spasticity Study) Dinet J 1, Lambrelli D 2, Balcaitiene J 1 1IPSEN Pharma, Boulogne-Billancourt, France, 2Evidera, London, UK . PMS27 Hospitalization Burden Among Dialysis Patients In Brazil: an Analysis of The Public Health System Database Coutinho M B 1, Custodio M R 2, Pecoits-Filho R 3, Borges L 4, Guersoni A C 1 Brazil, Sao Paulo, Brazil, 2Federal University of Uberlandia, Sao Paulo, Brazil, 3Pontifícia Universidade Católica do Paraná, Curitiba, Brazil, 4Evidências, Campinas, Brazil . . . . . . . . 1Amgen Objectives: The objective is to compare the hospitalization length of stay and costs associated with comorbidities between dialysis patients and non-dialysis patients. Methods: This was a cross-sectional observational study (from January to December 2010) of a national hospitalization database from the Brazilian Unified Public Healthcare System (Sistema Único de Saúde – SUS). Patients included in the study were hospitalized due to 3 predefined comorbidities categories: heart disease, vascular disease, and osteometabolic diseases, identified using ICD-10 codes. Results: A total of 491,644 admissions were observed in patients not on dialysis (control group), while 2,627 admissions were identified in dialysis patients. The comparative analysis of causes of hospitalization showed that a larger proportion of admissions due to heart disease was observed in patients on dialysis (61%) compared to the control group (47.9%), the opposite was observed for vascular and osteometabolic hospitalizations. For all 3 categories, cost of hospitalization for dialysis patients is at least two-fold higher and the length of stay almost 3 times longer (21 days for dialysis patients versus 8 days for non-dialysis) than non-dialysis patients. The greatest differences between these two patient populations are in average costs among dialysis patients hospitalized for osteometabolic disorders versus control group and the highest burden regarding length of hospital stay, due to heart disease. Conclusions: There is a high hospitalization burden among CKD-5D patients in the Brazil health care system. Since inpatient costs were the key cost drivers for CKD, strategies that reduce the risk of hospitalization and increase prevention of comorbidities may substantially decrease the overall health care economic burden. . 1Healthware . 235 U (108)/200 U (100-440) was injected. Based on a BoNT-A injection interval of 12 weeks, the annual cost per patient in the UK would be 1,068 GBP for abobotulinumtoxinA, 1,198 GBP for onabotulinumtoxinA and 1,399 GBP for incobotulinumtoxinA. Conclusions: Considering the real life practice of BoNT-A injections and the comparison of treatment groups treated for the same limb segment, this analysis suggests that the use of abobotulinumtoxinA would result potentially in a reduction in the health care cost for the treatment of spasticity and that more patients could be treated with abobotulinumtoxinA with a given budget. . Objectives: To evaluate the real life practice of the use of botulinum toxin A (BoNT-A) in post-stroke upper limb spasticity and the economic consequences of fair comparisons of the dosing between either abobotulinumtoxinA (Dysport®) or onabotulinumtoxinA (Botox®) or incobotulinumtoxinA (Xeomin®). Methods: ULIS-II is an observational, prospective study, conducted in 84 centers in 22 countries. Of 456 adults with post-stroke upper limb spasticity presenting for treatment with BoNT-A, 193 patients with the same injected limb segments “upper arm and lower arm” were analyzed for the dose injected for one cycle of BoNT-A. Treatment and concomitant interventions were in accordance with routine local clinical practice. Sample size, mean (SD)/median dose (min-max) in Unit for each BoNT-A and annual cost per patient were calculated using the mean dose administered and considering no vial sharing. An injection interval of 12 weeks was simulated for all BoNT-A treatments and the UK listed public price was used to ascertain annual cost. Results: For the abobotulinumtoxinA group (N= 141) a mean (SD)/median (min-max) dose of 665 U (280)/500 U (150-1500) was injected, for the onabotulinumtoxinA group (N= 37) a mean (SD)/median dose of 183 U (99)/200 U (50-500) was injected and for incobotulinumtoxinA (N= 15), a mean (SD)/median dose of PMS28 The Costs of Diagnosis and Treatment of Ankle Sprains and Fractures, 1980-2013: A Systematic Review Bielska I A , Wang X , Johnson A P Queen’s University, Kingston, ON, Canada . . . . . Objectives: Ankle sprains and fractures are common injuries affecting many individuals, often requiring substantial and costly medical interventions. Ankle injuries can have significant physical and economic consequences. Therefore, the objectives of this study were to systematically review and describe the literature on the direct and indirect costs related to the diagnosis and treatment of ankle sprains and fractures. Methods: A systematic literature review of Ovid MEDLINE, EMBASE, Cochrane DSR, ACP Journal Club, AMED, Ovid Healthstar, and CINAHL was conducted for English-language studies on ankle sprains and fractures published from January 1980 to December 2013. Two reviewers assessed the articles for study quality using available giudelines and abstracted the data. Results: Overall, 1,415 studies were identified of which 16 were selected for analysis. A majority of the studies were published in the last decade. The costs of ankle sprain diagnosis and management ranged from $495 to $4,667 per patient (2014 USD). The costs of stable ankle fracture diagnosis and management ranged from $89 to $602 per patient. However, the management costs were higher for unstable ankle fractures ($2,680 to $15,095) and open fractures and varied depending on the severity of injury. The economic evaluations were conducted from the societal or health care system perspective. Conclusions: The costs of diagnosing and treating ankle sprains and fractures varied among the studies mostly due to differences in injury severity and the study characteristics. Future studies undertaking economic evaluations should follow the available guidelines and ensure that their methods are transparent and understandable especially the study perspective and the valuation of the costs and outcomes. PMS29 The Pension Costs of Musculoskeletal Diseases. Estimation of the Economic Burden Borne by the Italian Social Security System Russo S 1, Mariani T T 2, Migliorini R 2, Marcellusi A 1, Mennini F S 3 1University of Rome “Tor Vergata”, Rome, Italy, 2Istituto Nazionale della Previdenza Sociale, Rome, Italy, 3University of Rome “Tor Vergata”, Italy, Rome, Italy . . . . . . . Objectives: The aim of the study is to estimate the pension costs (social security system in Italy is financed by public expenditure) induced by patients with musculoskeletal disorders (MD) and specifically for rheumatoid arthritis (RA), ankylosing spondylitis (AS) and psoriatic arthritis (PsA) in Italy, between 2009 and 2012. Methods: We analysed the database of National Institute of Social Security (INPS) to estimate for MD, RA, AS and PsA, the total costs of for three types of social security benefits: disability benefits (for people with reduced work ability), disability pensions (for people who are not eligible to be considered as workers) and incapacity pensions (for people without work ability). Also was estimated the productivity loss for RA in the 2013 with data from the National Institute of Statistics and from national literature review. Results: From 2009 to 2012 were paid about 320 thousand benefits at a cost of approximately € 1.7 billion, for an average of just over 80,000 performances a year with a cost of € 432 million per year. Specifically the total pension burden for RA was about € 99 million, for AS was € 26 million and for PsA was € 12 million. The loss of productivity for AR in 2013 amounted to € 707,425,191 due to 162,360 workers with RA that determine 9,174,221 working days A378 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 lost. Conclusions: The most important indirect costs in Italy from 2009 to 2012 was represented by disability benefits (69% of the total indirect cost), followed by disability pensions (28% of cost) and incapacity pensions (3% of cost). A better prescription appropriateness and rapid access to innovative treatments (Italy, among the EU Countries, is the one with the greatest delay in access) would reduce the costs incurred by the social security system accompanied by an improvement on the effectiveness of interventions. PMS30 Systematic Review of Societal Cost of Illness Evidence in Patients With Psoriatic Arthritis King D 1, Mitchell S A 1, Keyzor I 2, Kiri S 2 International, Bicester, UK, 2Novartis Pharmaceuticals UK Limited, Camberley, UK . . . . . 1Abacus Objectives: Societal cost evidence relating to patients with psoriatic arthritis (PsA) is required to understand the overall economic impact of the disease. The objective of this research was to systematically review the published data relating to the societal burden of PsA. Methods: Embase, Medline, and Cochrane databases (accessed November 2013) were interrogated. Studies reporting on non-health sector costs or work disability associated with PsA were eligible for inclusion with no geographical restriction. Results: In total, 19 studies met eligibility criteria of the review; nine studies reported indirect costs associated with PsA and 11 studies reported PsArelated work disability. Studies were conducted across Europe, USA, Canada, and Asia. Methodology used to calculate indirect cost estimates differed among studies; seven studies defined indirect costs as costs associated with productivity loss only, and two studies adopted a wider definition whereby indirect costs included productivity loss and out-of-pocket costs expenses (e. g. transportation, informal care, and home aids). Annual indirect cost associated with PsA, calculated using the human capital or friction capital approach, ranged from € 1,261 (Spain) to € 7,919 (Germany), with the majority of studies (71%) reporting annual indirect costs of less than €4,000 per person. Considerable heterogeneity was reported, with indirect costs of € 74,009 over a two year period estimated in Norway, compared with 6-month costs of € 576 in Italy. Indirect costs accounted for between 38% and 72% of total costs. Work disability estimates ranged from 5% to 39%. Disease severity, age, and gender were reported as main drivers of indirect costs and work disability. Conclusions: Cost associated with productivity loss represented the majority of indirect costs associated with PsA. Comparison of studies is confounded by heterogeneity between studies. Further research into the societal costs associated with patients with PsA in the UK is required as no studies reporting UK specific costs were identified. PMS31 Systematic Review of Societal Cost of Illness Evidence in Patients With Ankylosing Spondylitis King D 1, Mitchell S A 1, Keyzor I 2, Kiri S 2 International, Bicester, UK, 2Novartis Pharmaceuticals UK Limited, Camberley, UK . . . . . 1Abacus Objectives: Ankylosing spondylitis (AS), one of the most common spondyloarthritides, is a chronic condition with a significant impact on patient quality of life (QoL), particularly physical functioning. The objective of this research was to systematically review published data relating to the societal burden of AS. Methods: Embase, Medline, and Cochrane databases (accessed November 2013) were interrogated. Studies reporting on non-health sector costs or work disability associated with AS were eligible for inclusion. Results: In total, 42 studies met eligibility criteria; 19 studies reported indirect costs associated with AS and 23 studies reported AS-related work disability. Studies were conducted across Europe, Africa, North America, and Asia. All studies defined indirect costs as costs associated with productivity loss. Annual indirect costs, calculated using the human capital approach, ranged from €595 (Belgium) to €9,837 (Germany). Friction costs calculated ranged from € 428 (France) to €6,843 (Spain). Considerable heterogeneity was reported, with 2-year indirect costs of €66,907 estimated in Norway. Despite heterogeneity between studies, 75% of publications reported indirect costs of less than € 4,500 per patient/year, representing between 34% and 74% of total costs. Non-medical costs such as informal care, and transportation ranged from $65 (US) to $3,663 (US), indicating that loss of productivity was the main driver of indirect costs. Estimates of work disability varied greatly across studies primarily due to heterogeneous populations and the lack of standardised questionnaires. The work disability of AS was multifactorial with regression analyses identifying several factors associated with both indirect costs and disability including demographic factors, disease and job characteristics. Conclusions: Although reported indirect costs and work disability estimates varied between studies, findings from this systematic review indicate that AS represents a substantial burden both on society and individuals with AS. A lack of UK-based data illustrates the challenge of incorporating societal costs into future UK reimbursement submissions. PMS32 Determination Of The Annual Health Insurance Cost Of Outpatient Care Physiotherapy Services For Low Back Pain 3.83% prevalence. The 20 most frequent treatments accounts for 82.12 % (1016057) of total services. The following top-10 medical procedure were responsible for 58,05% (718.276) of total activities: 1) Ultrasound therapy 118806 (9.60%), 2) Iontophoresis 100788 (8.15%), 3) Passive movement 80424 (6.50%), 4) Mid-frequency treatment 77434 (6.26%), 5) Muscle strengthening exercises 62049 (5.02%), 6) Hand Massage 62026 (5.01%), 7) Ergo therapy 60462 (4.89%), 8) Spinal Mobilization 57609 (4.66%), 9) Diadynamic interference management 54092 (4.37%), 10) Training for circulation improvement 44586 (3.60%). Physiotherapy out-patient care for Low Back Pain patients were 1.23 million cases with 294 million Hungarian Forint (0.96 million Euro) health insurance subsidy. Conclusions: The list of the 20 most frequent types of services reflects to the demand for the combination of active and passive exercises in Hungary. In case of capacity planning for diagnostic and therapeutic procedures, these findings should be considered. PMS33 Cost of Treating Hip Fractures With Cephalomedullary Nails: A Retrospective Claims Database Review Menzie A M 1, Daskiran M 2, Abbott T A III2 and Johnson, Boston, MA, USA, 2Johnson & Johnson, New Brunswick, NJ, USA . . . . . 1Johnson Objectives: The number of hip fracture hospitalizations in the U. S. was approximately 352,000 in 2011 and expected to rise. The consequences of hip fractures are significant regarding the associated negative impacts on functioning and quality of life. Intertrochanteric hip fractures are frequently stabilized surgically using cephalomedullary nails (CM). Complications may include post-operative femoral-shaft fracture, mechanical complications, and delayed/non-union, requiring reoperations that increase risk to the patient and cost to the health care system. Methods: A retrospective analysis of the Premier research database was conducted. Patients diagnosed with open and/or closed intertrochanteric fractures (ICD-9-CM codes 820.2x or 820.3x) and treated for Hip Fracture (MS-DRG 480-482) from 1-JAN-2008 through 30-JUN-2012 were included in the study. Among this population, patients treated with IM were identified using standard charge descriptions within orthopedic-implant hardware charges. To identify patients who had a reoperation, patientlevel admission records starting from index admission up to 12-months post-index procedure were investigated within the same MS-DRG classifications. Results: Overall, 73,745 patients qualified for the study (71% female; 74% ≥ 75-years-old). Of these, 32,175 (44%) were treated using CM. The average hospital cost of the index procedure was $15,798USD per patient (inclusive of fixed and variable costs), of which the average cost of the orthopedic nails was $1,704USD. Among these patients, a total of 419 (1.3%) patients were readmitted at 427 occasions within the 12-month study period. The mean hospital cost of treating these readmissions was $16,352USD, and the primary diagnoses codes for these readmissions included femoral fractures, nonunions, and device/mechanical complications. Conclusions: Although the incidence of readmission following a hip fracture is low, the economic burden placed on the health care system and the humanistic burden are both significant. The incidence rate and long-term sequelae for hip fractures are well documented in the literature; however, the mean costs to a hospital have not yet been published. PMS34 Cost Related to the Waiting List of Patients With Vertebral Malformation Gresz M 1, Varga S 2, Boncföldi K 3, Boncz I 4 Institute for Quality- and Organizational Development in Healthcare and Medicines, Budapest, Hungary, 2University of Pécs, Pécs, Hungary, 3Semmelweis University, Budapest, 4 Hungary, Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . 1National Objectives: The Hungarian health budget does not allow all candidates to be promptly operated on with vertebral malformation. The budget is enough for approximately 4000 stabilizations by implant. The rest have to be put on the waiting list. The aim of our study is to analyse the costs related to the patients being on waiting list for surgery of vertebral malformation. Methods: Data derived from the nationwide financial database of the Hungarian National Health Insurance Fund Administration. 71 patients on waiting list of an institute in 2005-2006 were studied. We analysed the average waiting time and the costs related to patients either operated on or without operation. Results: The average age of all candidates was 55 year and the average waiting time was 32 months. The treatment cost during the waiting time was US$ 103 per month per patient. The average age of candidates operated on was 57 years and their average waiting time was 15 months with a costs of US$ 155 per month per patients. The average age of candidates not operated on was 54 years and their average waiting time was 15 months with a cost of US$ 92 per month per patient. Conclusions: Half of candidates have not been operated on, because at the appointed date they postponed the operation. But they seem not to really want to be operated on. The reason could be they were not sure enough to be able to get job after the operation at their age. Járomi M 1, Hanzel A 1, Endrei D 1, Zemplényi A 1, Csákvári T 2, Danku N 1, Boncz I 3, Molics B 1 of Pécs, Pécs, Hungary, 2University of Pécs, Zalaegerszeg, Hungary, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary PMS35 Treating to the Target of Das28 < 2.6 in Rheumatoid Arthritis: the Impact of Efficacy on Cost Effectiveness Objectives: The aim of our study is to assess amount and frequency of the physiotherapy services in Low Back Pain disease and determine the total health insurance reimbursement. Methods: Data were derived from the countrywide database of Hungarian Health Insurance Administration (HHIA), based on official reports of outpatient care institutes in 2009. The total numbers of different physiotherapy services were determined by selecting the reported specific diagnoses codes and counting the number treatments provided for that specific diagnosis code. The different types of treatment codes are listed in the chapter of the Guidelines of HHIA for `Physiotherapists, massage-therapists, conductors and other physiotherapy practices`. The Low Back Pain was reported according to WHO ICD diagnosis code M5450. Results: Low Back Pain account for 1237263 cases in the annual number of the physiotherapy-related activities (32318413 cases) showing an approximately Alemao E 1, Johal S 2, Al M J 3, Rutten-van Mölken M 3 Squibb, Princeton, NJ, USA, 2Parexel International, London, UK, 3Erasmus University, Rotterdam, The Netherlands . . . . . . . . 1University . . . . . 1Bristol-Myers Objectives: Target measures such as DAS-28 < 2.6 are the goal of treatment in rheumatoid arthritis (RA). In general, DAS-28< 2.6 is associated with improved functionality and reduced resource use but with higher treatment costs. This study evaluated whether treatment strategies that were more effective but more expensive in achieving DAS-28< 2.6 in moderate to severe RA patients were cost effective (CE). Methods: A micro-simulation model of 10,000 RA patients was developed that estimated lifetime Health Assessment Questionnaire (HAQ) progression, quality adjusted life years (QALYs) and direct costs from a UK NHS perspective. The model incorporated an initial response criterion of achieving a DAS-28 < 2.6 at 6 months A379 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 after initiating treatment that triggered treatment switching to subsequent therapies upon failure. Scenario 1 compared a treatment sequence of a high efficacy biologic therapy followed by conventional DMARD therapy versus a less efficacious biologic therapy followed by a cDMARD. Scenario 2 compared two multiple biologic treatment sequences with a difference in efficacy between the first biologic in each sequence. A sensitivity analysis and a CE efficiency frontier analysis examined the effect of varying the difference in DAS-28 < 2.6 rates (Δ Rem) between first-line therapies. Results: In scenario 1, a Δ Rem of 10% between the bDMARDs yielded a mean per-patient gain of 0.117 QALYs and a mean per-patient increase in total cost £2,590, giving an incremental cost-effectiveness ratio (ICER) of £22,112 per QALY. In scenario 2, the higher efficacy bDMARD strategy had an ICER of £6,720 when Δ Rem = 10%. Conclusions: Over a lifetime, treating to the target of DAS28< 2.6 with more efficacious therapies in RA patients was cost-effective at various levels of efficacy differences. The higher costs of longer term treatment were offset by the increased quality of life for patients with DAS-28 < 2.6 as well as reduced hospitalisation costs. PMS36 Cost-Effectiveness Of Routine Testing For Hla-B*5801 In Caucasian Patients Newly Diagnosed With Gout In Portuguese Nhs Hospitals Araújo M 1, Pinto C G 2 1Faculty of Pharmacy - University of Coimbra, Coimbra, Portugal, 2Research Centre on the Portuguese Economy – CISEP, ULisboa, Lisboa, Portugal . . . Objectives: Routine testing for HLA-B*5801 in European patients has been proposed before allopurinol treatment aiming to reduce the incidence of StevensJohnson syndrome/toxic epidermal necrolysis (SJS/TEN). Having some discretion in the allocation of financial resources, Portuguese NHS hospitals may be particularly interested in this issue as extra costs with the treatment of SJS/TEN episodes may be avoided. This study aims to evaluate the cost-effectiveness of HLA-B*5801 genotyping versus no genotyping (usual care) in Portuguese adult patients newly diagnosed with gout before allopurinol treatment. Methods: A decision tree model was constructed to structure the sequences of clinical events that may occur after the diagnosis of the disease, over a period of three months. Effectiveness data were estimated for Portuguese patients based on published literature and Portuguese population databases. All costs were obtained from official prices published for the Portuguese NHS. Admitting that NHS hospitals objective is to lower costs in the treatment of gout, the incremental cost per SJS/TEN reaction avoided was the incremental cost-effectiveness ratio (ICER) considered. One way sensitivity analysis was performed for each relevant parameter. Results: The use of HLA-B*5801 genotyping versus no genotyping was associated with a decrease of 1.24 SJS/TEN reactions and an incremental total cost of 42, 401.03€ , resulting in an ICER of 34, 194.38€ per SJS/TEN reaction avoided. Sensitivity analysis results showed that base-case results are very sensitive to variations in the baseline odds ratio of gene-disease association, epidemiological data used to estimate the incidence of SJS/TEN and genotyping cost. Conclusions: Widespread use of HLA-B*5801 genotyping in Portuguese hospitals is not cost-effective at the current unit price of 45.30€ as it leads to a substantial increase in expenditure compared to usual care. However, it may be an advantageous option if its maximum price would equal 6,37€ . PMS37 Cost-Effectiveness Analysis of Tocilizumab VersUS Infliximab for the Patients With Rheumatoid Arthritis in Kazakhstan Bektur C 1, Kayir F 2, Nurgozhin T 1 University, The Center for Life Sciences, Astana, Kazakhstan, 2Nazarbayev University, Astana, Kazakhstan . . . 1Nazarbayev Objectives: The purpose of present study was to evaluate the cost-effectiveness of tocilizumab (TCZ) vs. infliximab (IFX) for the treatment of patients with rheumatoid arthritis (RA) from the perspective of Ministry of Health (MoH) over a patient lifetime horizon in Kazakhstan. Methods: A decision analytic model was constructed in MS-Excel to estimate effects and directs costs of methotrexate (MTX) combination with TCZ or IFX in hypothetical cohorts of RA patients. Treatment efficacy and transition probabilities were synthesized through a meta-analysis of data from relevant clinical trials and literature. The characteristics of patient cohort and treatment costs (drug, monitoring, RA-related hospitalisation, adverse effects management) in year 2014 Kazakhstani tenge (KZT) were estimated from republican official sources. Primary efficacy outcome was defined as an improvement of 70% in ACR response (ACR70), which can be regarded as a close measure of remission. Annual 3% discounting rate and 6-month cycles were utilized for the model. Robustness of the model parameters was explored by one-way and probabilistic sensitivity analysis. Results: For an average 70-kg patient, discounted direct costs per year were estimated at 4 456 000 KZT for TCZ+MTX and 3 190 000 KZT for IFX+MTX. Adjusted ACR70 rates were higher in TCZ+MTX group (23%) than in IFX+MTX group (17%). As a result of a lifetime stimulation (average 12 years) of the model, TCZ incurred 1.89 times higher costs, and 1.67 times more achievements of ACR70 than IFX. The ICER was estimated to be 3 218 900 KZT per achieved ACR70, which is within the costeffectiveness threshold values recommended by WHO. Overall, model was sensitive to the changes in drug costs and patient weight. Conclusions: TCZ+MTX incurred significantly higher costs compared to IFX+MTX, although ICER per achieved ACR70 lays within the recommendations of WHO. Further research is necessary for other ACR rates. These findings may better inform decision makers regarding formulary inclusion and reimbursement in the treatment of RA patients in Kazakhstan. PMS38 Cost-Effectiveness of Certolizumab Pegol in the Treatment of Active Rheumatoid Arthritis, Axial Spondyloarthritis, and Psoriatic Arthritis In Romania Codreanu C 1, Mogosanu C 1, Joita M 2, Purcaru O 3 1Rheumatology, Dr. Ion Stoia Rheumatic Disease Center, Bucharest, Romania, 2UCB Pharma, Bucharest, Romania, 3UCB Pharma, Brussels, Belgium . . . . Objectives: To evaluate the cost-effectiveness of certolizumab pegol (CZP) versus other subcutaneous (SC) anti-tumour necrosis factors (anti-TNFs) currently licensed and reimbursed in Romania (etanercept [ETA] and adalimumab [ADA]) in the treatment of adult patients with active rheumatoid arthritis (RA), axial spondyloarthritis (axSpA; comprising ankylosing spondylitis [AS] and non-radiographic axSpA [nr-axSpA]) and psoriatic arthritis (PsA). Methods: Cost-utility models (Markov structure) were developed for each condition, following the EULAR, ASAS and NICE guidelines. Clinical efficacy data, disease history and resource use data came from published literature. Utilities were derived from EQ-5D data assessed in CZP pivotal clinical trials. Published 2013 unit costs were taken from local sources/expert opinion. A willingness-to-pay threshold of 3xGDP/capita (87600RON) was considered. Base-case analysis was conducted from the payer perspective, with lifetime horizon, costs and outcomes discounting rates of 3.5%. Sensitivity analyses were conducted to test the robustness of the results. Results: In AS, CZP dominates ADA and ETA (quality-adjusted life years [QALYs] gain of 0,098 and 0,021; total costs lower by 48499.73RON and 8350.28RON, vs ADA and ETA respectively). In nr-axSpA, CZP dominates ADA (QALYs gain of 0,257; total costs lower by 13955.14RON). In RA, CZP dominates ADA (QALY gain of 0,226; total costs lower by 108694RON) and is marginally as effective, but slightly more costly vs ETA (QALYs gain of 0,056, total costs higher by 6356RON). In PsA, CZP dominates ADA (QALY gain of 0.05; total costs lower by 9397RON) and is less costly, but marginally less effective versus ETA (QALYs difference of -0.08, total costs lower by 21484RON). Probabilistic sensitivity analysis indicated that CZP has a similar or higher probability of being cost-effective when compared to comparators. Conclusions: CZP is a cost-effective treatment compared to currently available SC anti-TNFs for the treatment of RA, axSpA and PsA in Romania. PMS39 Estimating the Cost-Effective Intervention Thresholds for Osteoporotic Fractures Based on Frax® in the Greek Setting Athanasakis K 1, Makras P 2, Boubouchairopoulou N 1, Rizou S 3, Kyriopoulos J 1, Lyritis G 3 1National School of Public Health, Athens, Greece, 2251 Hellenic Air Force General Hospital, Athens, Greece, 3Hellenic Osteoporosis Foundation, Athens, Greece . . . . . . Objectives: Osteoporosis is a condition consisting in reduced bone mass and density, thus resulting in an increased risk of fragility fractures. Considering the aforementioned, and in conjunction with the complications that osteoporosis engenders, a considerable economic burden is imposed on industrialized countries. The purpose of this study was to determine the thresholds at which treatment initiation becomes cost-effective in Greece, using the fracture risk assessment algorithm (FRAX®) developed by WHO and recently adjusted for Greece. Methods: A previously developed state transition Markov model was populated employing epidemiological and economic data from Greece. The analysis calculated the costeffective intervention thresholds using the Greek FRAX® model, which estimates the 10-year probability of a major osteoporotic fracture in subjects with a fracture risk equivalent to that of a woman with a prior fragility fracture and no other risk fracture. The analysis was undertaken from a third-party payer perspective, assuming a willingness to pay of 30,000€ per QALY gained. Results: In Greece, the drug intervention aiming at reducing the fracture risk, was found to be cost-effective with a 10-year probability for a major osteoporotic fracture at/or above 12.3% (range: 7.8%-20.4%) for women and 18.2% (range: 9.6%-34.2%) for men aged from 50 to 59 years old. When considering women and men aged between 60 and 74 years, the threshold was estimated ≥ 8.9% (range: 8.5%-9.2%) and ≥ 9.4% respectively (range: 8.9%-10.0%), whereas for women and men over 75 years old, the cost-effective thresholds were calculated to be ≥ 15.0% (range: 13.0% to 16.0%) and ≥ 11.0% (range: 10.6%-11.2%) accordingly. Significant age-variations were generally not noticed, some exceptions apart, especially for the age-range of 50-55. Conclusions: As expected, the cost-effective thresholds obtained via FRAX® were age-dependent in both sexes. Therefore, considering these thresholds should lead to cost-effective access to therapy of patients with high fracture probability, and ipso facto significantly reduce the economic burden of osteoporotic fractures in Greece. PMS40 Cost-Effectiveness of Celecoxib and Non Selective Non Steroidal Anti-Inflammatory Drug (Nsaid) Therapy for the Treatment of Osteoarthritis in Spain: A Decision-Tree Model De Lossada Juste A 1, Rejas Gutiérrez J 2, Oteo Álvaro Á 3 1Pfizer, S.L.U, Alcobendas/Madrid, Spain, 2Pfizer S.L.U., Alcobendas/Madrid, Spain, 3Hospital General Universitario Gregorio Marañón, Madrid, Spain . . . Objectives: Treat Osteoarthritis (OA) with traditional non-steroidal anti-inflammatory drugs (t-NSAIDs) may be associated with significant gastrointestinal (GI) and cardiovascular (CV) events. The objective was to assess the cost-effectiveness (CE) of Celecoxib and non selective non steroidal anti-inflammatory drugs therapy for the treatment of OA in clinical practice in Spain. Methods: The CE was assessed in patients aged 55+ years with OA and was based on a retrospective decision-tree model using distribution, dose and duration of therapies and incidence of GI and CV events observed in the pragmatic PROBE-designed 6-month “GI Reasons” trial. Effectiveness was expressed in terms of event averted and quality-adjusted lifeyears (QALYs) gained. Utility loss in case of adverse events both in values and time duration were derived from the published literature. Perspective of the analysis was that of the National Health System (NHS) in Spain; cost calculations included actual prices of drugs plus cost of adverse events occurred during trial only. The analysis was expressed as an incremental CE ratio (ICER) per QALY gained and per event averted. Probabilistic and univariate sensitivity analyses, including celecoxibprice-threshold analysis, were carried out. Results: Data from 7,939 patients were included in the analysis. Compared with traditional NSAIDs, celecoxib treatment had higher drug costs than traditional NSAIDs (€ 119 vs. € 34), and the overall treatment cost was estimated at € 201 and € 157, respectively. Moreover, celecoxib was associated with slightly increase in QALY gain and significant lower incidence of gastrointestinal events (p< 0.001) with mean ICERs of € 13,286 per QALY gained and A380 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 € 4,471 per event averted. Probabilistic and univariate sensitivity analyses were robust and confirmed results of the base case scenario. Conclusions: This economic evaluation modeling suggests that celecoxib may be considered as a costeffective alternative vs. t-NSAIDs in the treatment of osteoarthritis in daily practice in the Spanish NHS. PMS41 The Cost-Effectiveness of Biologic Dmards in Patients With Severe or Mild-To-Severe Rheumatoid Arthritis After Conventional Dmards Wailoo A J 1, Stevenson M 2, Tosh J 2, Hernández M 2, Stevens J W 2, Archer R 2, Simpson E 2, everson Hock E 2, Scott D 3, Young A 4, Paisley S 5, Williams K 5 1NICE Decision Support Uni, Sheffield, UK, 2University of Sheffield, Sheffield, UK, 3KCL, London, UK, 4adam. young@nhs. net, Heertford, UK, 5The University of Sheffield, Sheffield, UK . . . . . . . . . . . . . . Objectives: To estimate the cost-effectiveness of biologic disease modifying antirheumatic drugs (bDMARDs) following failure of conventional disease modifying anti-rheumatic drugs (cDMARDs) in patients with severe or mild-to-severe active rheumatoid arthritis from a UK, NHS perspective, as part of an ongoing National Institute of Health and Care Excellence (NICE) appraisal. Methods: Systematic review of clinical effectiveness of seven bDMARDs: abatacept; adalimumab; certolizumab pegol; etanercept; golimumab; infliximab; and tocilizumab. Network meta-analyses (NMA) of randomised controlled trials (RCTs) reporting European League Against Rheumatism (EULAR) response and for RCTs reporting American College of Rheumatology (ACR) outcome data. An individual patient model was constructed to estimate costs and outcomes in terms of quality adjusted life years (QALY). Large observational databases, published literature and the results of the NMAs were used to provide data for the model. Following failure of two cDMARDs two broad strategies were evaluated i) a bDMARD, followed by, if necessary, rituximab then tocilizumab then cDMARDs and ii) remaining on cDMARDs. Results: The estimated incremental costs per QALY of bDMARD strategies compared with a cDMARD alone strategy were typically over £50,000 regardless of the severity of rheumatoid arthritis or whether the EULAR or the ACR RCTs were used. The cost per QALY is greater for those who receive bDMARD monotherapy. One key parameter affecting the results was the estimated trajectory of HAQ progression whilst a patient received cDMARDs; using rates previously assumed in NICE appraisals reduced estimates to approximately £30,000. Conclusions: bDMARDs are unlikely to be as cost-effective as has been estimated previously. The costs per QALYs generated within our base case analyses are greater than commonly reported cost-effective thresholds in England and Wales. PMS42 A Cost-Effectiveness Analysis for Total Knee Arthroplasty Telerehabilitation: Proof of Concept of A Decision Model Fusco F 1, Turchetti G 2 1Scuola Superiore Sant’Anna, Pisa, Italy, 2Scuola Superiore Sant’Anna, Pisa, Italy . (BMD) testing with dual energy x-ray absorptiometry (DXA) and received five years alendronate therapy if the percent young adult mean (%YAM) for BMD was less than 70%. Lifelong drug therapy for secondary fracture prevention was assumed for patients who had a osteoporotic fracture. For the base-case analysis, we ran the model with different age groups (65-69, 70-74, and 75-79 years). Results: In the women aged 65-69 years, screening strategy incurred an additional lifetime cost of $1,486 per person and conferred an additional 0.029 QALY, resulting in an incremental cost-effectiveness ratio of $51,195per QALY gained. For those aged 70-74 and 75-79 years, ICER was estimated to be $23,375 and $17,742per QALY, respectively. Probabilistic sensitivity analysis showed that in women aged 65-69,70-74 and 75-79 years, screening strategy was cost-effective in 48.9%, 58.9%, and 59.7% of the simulations, respectively, if society is willing to pay $50,000 per QALY. Conclusions: Osteoporosis screening and treatment strategy would be cost-effective in the Japanese women aged ≥70 years. PMS44 Cost-Effectiveness of Use of Barricaid® In Lumbar Discectomy Surgery in Turkey Tatar M 1, Senturk A 2, Dalgali Y 3 University, Ankara, Turkey, 2Polar Polar Health Economics & Policy, Ankara, Turkey, 3Intrinsic Therapeutics, Inc., Wouburn, MA, USA . . . 1Hacettepe Objectives: While widely perceived as a successful procedure, discectomy surgery has a high failure rate over time. The overall risk of recurrent disc herniation varies between 2-18% in reported literature. The Barricaid® anular closure device was designed as an adjunct to lumbar limited microdiscectomy to block large anular defects while maintaining as much native nucleus within the disc space. Patients that are considered for anular closure have a minimum posterior disc height of 5mm, and an intra-operatively measured anular defect between 5mm and 12mm wide. The aim of this study was determined as to assess the cost effectiveness of the use of Barricaid® in this group of patients in Turkey. Methods: A simple decision analysis model was used to assess the cost effectiveness of the use of Barricaid®. The primary clinical endpoint was determined as the number of prevented reherniations. According to the literature, the use of Barricaid® reduced the number of reherniations by 18%. Resource utilization data were obtained via expert clinical opinion and included pre-op, post-op and follow-up costs, etc. Unit costs were taken from the Social Security Institution’s official price list. Results were presented as incremental cost/number of prevented reherniations. The comparison was made between using and not using the Barricaid®. Results: According to the results of the cost effectiveness analysis, the incremental number of prevented reherniation was 4.398 with Barricaid® and incremental cost was 119.343.000 TL. The ICER was within the limits of the threshold recommended by the World Health Organization with 27.136 TL. Conclusions: Use of Barriciad® in lumbar discectomy surgery is a cost-effective treatment option in Turkey. . PMS45 Pharmacoeconomic Evaluation of Treatment With Tocilizumab in Russian Children With Systemic Juvenile Idiopathic Arthritis Objectives: The study aims to assess the cost-effectiveness of telerehabilitation for Total Knee Arthroplasty (TKA) patients in Italy. TKA was performed 64,936 times in Italy in 2012, reasonably leading to the same number of rehabilitation processes. The most recent cost analysis showed rehabilitation to account for Euro 158 million per year. Therefore, new strategies aiming at optimizing resources and preserve patients’ wellbeing are claimed. Methods: A four-state Markov model (successful TKA; revision; successful revision; death) forecasted costs and clinical outcome over 10 years (cycle length: 1 year) for 1,000 individuals undergoing usual care rehabilitation (UC) or a mixed UC-telerehabilitation (UC-T) program. Published literature provided transition probabilities and clinical outcome (active knee flexion Range Of Motion-ROM); while UC and telerehabilitation costs were estimated through Italian tariffs and panel of experts. Each surgery was assumed to lead to rehabilitation or telerehabilitation, resulting in direct medical and indirect costs (human capital approach). Results were adjusted applying half-cycle correction method and discount rate of 3%. A Probabilistic Sensitivity Analysis described parameters uncertainty and results were reported using Incremental Cost-Effectiveness Ratios (ICER) from societal and Italian-NHS perspectives. Results: Expected mean health care costs for UC were 1,253.2€ / patient over ten years, and UC-T costs were on average 33.7€ /patient higher (95%CI € 10.8). ROM-degrees for UC and UC-T were respectively 24.5 and 26.8 (mean difference= 2.3, 95%CI 0.002). The resulting ICER was 14.5€ /ROM-degree (ItalianNHS perspective). Adopting a societal perspective, UC-T was more effective yet appeared cheaper than UC (respectively 1,429 and 1,457€ /patient, mean difference -28/patient (95% CI € 10.8)). Conclusions: Although the preliminary results have shown that UC-T could be a cost-saving procedure if societal perspective is adopted; these findings are uncertain due to the model assumptions. Therefore, further investigations with patient-level data and generic outcome measures (e. g. QALY) are required to draw definitive conclusions about cost-effectiveness in telerehabilitation. Objectives: To evaluate the use of tocilizumab in Russian patients with systemic juvenile idiopathic arthritis (SJIA) in terms of cost-effectiveness and impact on social and economic burden of the disease. Methods: The model was based on TENDER clinical study (De Benedetti F et al., 2012). First, a pharmacoeconomical cost-efficiency of tocilizumab and a standard basic therapy of SJIA were compared. The analysis included direct medical costs in two comparable groups (1st with routine administration of methotrexate and prednisolone, and 2nd with tocilizumab prescribed in case of refractoriness to NSAID and glucocorticosteroids). The efficacy of therapies was evaluated according to ACR criteria. After that, the influence of tocilizumab on cost and burden of illness was assessed. The analysis included direct medical costs and government expenditures on hospitalization, work incapacity insurance, monthly social pensions, benefits for care for a disabled children below 18 y. o. and GDP losses. 12-week time horizon was adopted in the CEA model, and a 1-year horizon for the assessment of burden of illness. Results: The cost-effectiveness in terms of ACR 90 and 70 was 4 428 262.96 and 2 952 175.31 RUB for the group of standard treatment, and 1 166 111.66 and 615 218.74 RUB for tocilizumab group. Tocilizumab demonstrated the same benefits in terms of ACR 50, 30. Analysis of burden of illness revealed a different structure of financial expenditures in considered strategies: pharmacotherapy constituted > 50% costs in tocilizumab group, but hospitalization costs were 12 times less than in standard therapy. Annual budget losses due to social burden of this disease were 426 144.63 RUB per patient in the group of standard treatment and 226 729.10 RUB in tocilizumab group. Conclusions: The use of tocilizumab in SJIA is justified by better cost efficiency and reduced of social and economic losses of state budget connected with the burden of the disease. PMS43 Health Economic Evaluation of Osteoporosis Screening and Treatment Strategy in the Elderly Japanese Women PMS46 Cost-Effectiveness Analysis of Etanercept in the Treatment of Rheumatoid Arthritis In Portugal Yoshimura M 1, Moriwaki K 2, Noto S 3, Takiguchi T 3 1Graduate School of Health and Welfare, Niigata Universitiy of Health and Welfare, Niigata, Japan, 2Kobe Pharmaceutical University, Kobe, Japan, 3Niigata University of Health and Welfare, Niigata, Japan Mateus C 1, Moura A 2 1Escola Nacional de Saúde Pública, Universidade Nova de Lisboa, Lisbon, Portugal, 2Escola Nacional de Saúde Pública, Universidade Nova de Lisboa, Lisboa, Portugal . . . . Objectives: The objective of this study was to estimate the cost-effectiveness of osteoporosis screening and treatment with alendronate in the Japanese women aged ≥65 years without a fragility fracture history. Methods: A Markov model with ten health states (no event, seven types of post-fracture, bedridden, and death) was developed to predict lifetime costs and quality-adjusted life years (QALY) of screening and treatment strategy, comparing with no screening. In the screening arm, 1,000 hypothetical cohort experienced a bone mineral density Ryazhenov V V , Gorokhova S G , Emchenko IV I.M. Sechenov First Moscow State Medical University, Moscow, Russia . . . . . . Objectives: The present study aims to estimate the cost-effectiveness of etanercept compared to golimumab in the treatment of patients with Rheumatoid Arthritis (RA) in Portugal. Methods: A model was adapted to assess the costeffectiveness of etanercept in the treatment of RA. We performed a comparison of the combination of etanercept + methotrexate and golimumab + methotrexate. Dosage of etanercept was 50mg on a weekly basis, whereas for golimumab it was 50 mg once a month. The model is an individual simulation model and takes a lifetime perspective. Outcomes are expressed in QALYs, using the HAQ score A381 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 to measure quality of life. The clinical parameters used in the model take into account the results from a MTC of clinical trials. In order to adapt the model to the Portuguese context, several adjustments were made to its original version. These mainly relate to mortality rates by gender and to the unit costs of medical resources, such as drugs, medical visits, admissions, ancillary tests and so on, that were obtained from public official sources. We applied a 5% discount rate and conducted an analysis for a hypothetical cohort of 1,000 patients. Results: The treatment of RA with etanercept is more expensive than that with the comparator. Nevertheless, QALYs gained do compensate for the additional costs. Overall, ICER is € 12,853, which is below the usual willingness to pay threshold used in Portugal. A sensitivity analysis was carried out, which confirmed the robustness of these results. Conclusions: According to our analysis, etanercept is a cost-effective alternative versus golimumab for the treatment of rheumatoid arthritis. Its main advantage over the selected comparator relates to the associated improvements in health related quality of life. PMS47 The Impact of Disease Modification on the Cost-Effectiveness of Pegloticase for the Treatment of Severe Debilitating Chronic Tophaceous Gout In Adult Patients Wallerstein K 1, Tolley K 2, Vegter S 3 1Access Partnership, Horsham, PA, USA, 2Tolley Health Economics Ltd., Buxton, Derbyshire, UK, 3Vegter Health Economic Research, Amersfoort, The Netherlands . . . Objectives: To determine the cost-effectiveness of pegloticase (Krystexxa®) for patients with severe debilitating chronic tophaceous gout (SDCTG), from a UK health care perspective. Methods: Severe debilitating chronic tophaceous gout (SDCTG) is a debilitating disease, with high unmet medical need. Existing treatments mainly provide symptom relief and do not modify the disease course. Pegloticase has the potential to be a disease modifying agent. A decision analytical model was built to compare the use of pegloticase in patients with SDCTG with best supportive care (BSC) with a Markov model used to extrapolate outcomes to a 20 year time-horizon. In the basecase, the disease modifying properties of pegloticase were modelled. In scenario analyses only symptomatic relief of pegloticase on acute attacks and tophi was included. Results: In the basecase, the cost-effectiveness of pegloticase compared to BSC was £31,027 per QALY gained. In this basecase, pegloticase dramatically reduced the uric acid burden in over 60% of patients who completed a six month course and were assumed could then be controlled on xanthine oxidase inhibitor maintenance treatment. In a pessimistic scenario whereby pegloticase was assumed to only provide symptomatic relief, the ICER was £48,672/QALY gained. In a further scenario whereby utility benefits were limited to only short-term reduction in acute flares and presence of tophi, the ICER was £54,345/QALY. Apart from these two drivers, the cost-effectiveness estimates were relatively stable across a range of sensitivity analyses. Conclusions: In the context of a highly severe and debilitating form of gout, with small patient numbers and a lack of alternative effective treatment options, pegloticase can be considered good value-for-money. Further clinical evidence is required to demonstrate the disease modifying properties of pegloticase. However, such data collection and hence the ability to perform robust economic evaluations for HTA purposes is difficult especially when the sponsoring company is small with limited funds. PMS48 Comparison of Diagnostic Strategies to Detect Prevalent Vertebral Fracture for Adults Over Age 50: Use of Vertebral Fracture Assessment or Spine Radiography Oh S H 1, Lee Y E 1, Kim D Y 2, Lee J H 3, Kim D 2, Hwang J S 4, Bae S C 2, Ahn J H 1, Sung Y K 2 1National Evidence-based Healthcare Collaborating Agency, Seoul, South Korea, 2Hanyang University Hospital for Rheumatic Diseases, Seoul, South Korea, 3Inje University Ilsan paik Hospital, Seoul, South Korea, 4National Evidence-based Healthcare Collaborating Agency (NECA), Seoul, South Korea . . . . . . . . . . . . . . . . Objectives: To estimate the incremental cost-effectiveness ratio (ICER) of dualmobility cups (MBH) when used instead of conventional cups (FBH) to help reduce dislocation rates following total hip replacement (THR) in France. Methods: A Markov model simulated two cohorts of patients: one with MBH and one with FBH. Three different states of health were considered: “stable”, “dislocation/revision” and “death”. The model adopted a collective perspective and the time horizon of the model was lifetime. Dislocation/revision rates were estimated using two different sources: literature review and expert opinions (analysis 1) and analysis of the PMSI (French hospital database) of Patients having one THR in 2009 followed through the end of 2012 (analysis 2). Costs considered were hospitalization costs (reduction/ revision for dislocation performed in acute care unit and care in rehabilitation units), valued using the National Scale Costs with Common methodology (ENCC). Outpatients’ costs were considered for patients going back home after hospitalization and valued by expert opinions and literature data. The model estimated the number of dislocations/revisions for each cohort and differences between the two groups in terms of QALY gained, costs and cost-effectiveness ratio. Deterministic and probabilistic sensitivity analyses (PSA) were conducted. Results: In analysis 1 for a 100,000 THR cohort, 4,626 dislocations and 1,243 revisions were avoided. Total number of QALY gained was 894 and total economic gain € 44.7 million. In analysis 2, 3, 176 dislocations and 854 revisions were avoided for an economic gain of € 30.7 million and a gain of 611 QALY. Reported to 100,000 THR, the economic gain per MBH cup was € 447 or € 307. PSA estimated the mean ICER to € -55,693 per QALY gained. Conclusions: Because of the current rules of tariffs used in France and given the absence of additional costs associated with the use of DM prosthesis, this strategy can be considered dominant in THR. PMS50 Cost-Effectiveness of Multiple Anti-Osteoporotic Therapies for Secondary Fracture Prevention in Japan Moriwaki K 1, Yoshimura M 2, Izumi R 3, Noto S 3 Pharmaceutical University, Kobe, Japan, 2Graduate School of Health and Welfare, Niigata Universitiy of Health and Welfare, Niigata, Japan, 3Niigata University of Health and Welfare, Niigata, Japan . . . . 1Kobe Objectives: The purpose of this study was to estimate the cost-effectiveness of multiple anti-osteoporotic drug therapies for secondary prevention of fractures in elderly women with osteoporosis in Japan. Methods: A state transition model with nine health states (seven types of post-fracture, bedridden, and death) was developed to predict lifetime costs and quality-adjusted life years (QALY) of no antiosteoporotic therapy and eight drug therapies in patients with a previous vertebral fracture. Incidence of hip, vertebral, and other fracture associated with age and bone mineral density (BMD) was estimated by using a regression approach based on epidemiologic studies in Japan. Comparative effectiveness of anti-osteoporotic drug therapies was derived from a published network meta-analysis. For the base-case analysis, we ran the model with T-score of -2.5 and different age (65, 70, and 75 years). Probabilistic sensitivity analysis was performed to assess parameter uncertainty. Results: Alendronate therapy dominated all other strategies, resulting in 11.746 QALY and lifetime costs of $34,568. Compared to no preventive therapy, alendronate conferred an additional 0.458 QALY and saved lifetime costs of $13,753. Risedronate was equally cost-effective, resulting in 11.731 QALY and costs of $34,932. Applying a willingness to pay threshold of $50,000 per QALY, the probability of being cost-effective were estimated to 76.2 % and 23.8% for alendronate and risedronate, respectively. These results did not change in women aged 70 and 75 years. Conclusions: Bisphosphonate therapies for secondary prevention of fractures in elderly women would be cost-effective in terms of Japan health care system. . PMS51 Pharmacoeconomic Analysis Tofacitinib Use in Rheumatoid Arthritis Treatment Scheme Kulikov A , Komarov I First Moscow State Medical University named after I.M. Sechenov, Moscow, Russia . . Objectives: The prevalent vertebral fracture (VF) is a risk factor for future VF, which can be decreased with drug therapy. However, most VFs are not recognized clinically. Vertebral fracture assessment (VFA) by dual-energy x-ray absorptiometry (DXA) and spine x-ray can be performed to detect these prevalent VFs. This study aimed to estimate the costs, effectiveness, and radiation exposure of VF diagnostic strategies. Methods: Markov model over a 10-year period was used to calculate the medical costs for diagnostic tests and VF treatment, the reduction of incident VFs of patients who have experienced a VF, and the radiation doses in target population aged over 50. We compared three strategies: ‘VFA followed by confirmatory radiography (VFA screening)’, ‘only VFA’ and ‘only x-ray’ every 2 years, to ‘no screening before recognition’. We assumed that all patients tested positive for VF received drug therapy. A discount rate of 5% was applied in cost. Results: The results showed the incremental costs for women over age 50 who had VFA screening, only VFA, and only x-ray were $1,112, $1,546, and $1,270 per person, respectively. Future VF incidence was reduced by 29% in both VFA screening and only VFA and 35% in only x-ray as compared with no screening for 10 years. Radiation exposure was highest in the only x-ray strategy. Also, the effectiveness and medical costs were more increased in female and old age people than in male and over age 50. The sensitivity analyses showed that these results are robust to variety assumptions including cycle length, medical costs, and diagnostic accuracy. Conclusions: This study suggests that VFA screening strategy can be relevant option for new VF prevention as considering lower cost and less radiation. This study is expected to provide useful information as establishing the VF diagnostic strategy in clinical practice. Objectives: Evaluate most rationale medical technology in the rheumatoid arthritis therapy (RA) (comparison of alternatives – GEBDs Tofacitinib and biologics: Infliximab, Abatacept, Certolizumab pegol, Golimumab, Adalimumab and Tocilizumab) from pharmacoeconomic analysis point of view. Methods: Analysis based on the assessment for one statistically average patient suffering from RA, and weighing 70 kilograms, over a one year course of treatment (52 weeks). The analysis done of direct costs included: cost of DMARDs and biologics therapy use; costs of drug introduction; physician visits cost. Cost-minimization and missed opportunities analysis were used. Results: During the effectiveness analysis of RA treatment, based on the meta-analyses of randomized placebo-controlled trials data (including meta-analyses P. Kawalec, 2013; E. Salgado, 2013), Russian and international RA treatment recommendations, it was concluded that there was no statistically significant difference in efficacy and toxicity of the Tofacitinib and the biologics used in the RA treatment. One year treatment course with Tofacitinib, Infliximab, Abatacept, Golimumab, Certolizumab pegol, Adalimumab and Tocilizumab cost, include subcutaneous route of administration, will amount to 12.818 EUR, 20.932 EUR, 14.855 EUR, 18.104 EUR, 19.642 EUR, 20.120 EUR and 21.664 EUR, respectively. Conclusions: During pharmacoeconomic analysis was defined that therapy with Tofacitinib in comparison with biologics use will reduce the cost of a one year course of treatment for each RA patient from 2.037 EUR to 8.846 EUR. Transition of 100 RA patients onto a treatment regimen, includes Tofacitinib use, will make it possible to treat from 15 to 69 patients more suffering from this disease. PMS49 Cost-Effectiveness Model of Dual-Mobility Cups for Total Hip Replacement in France PMS52 Rehabilitation In Resurfacing Hip Arthroplasty Patients: Preliminary Cost-Effectiveness Results From A Clinical Trial Epinette J A 1, Robert J 2, Rodriguez J 3, Lafuma A 2 1Clinique Medico-Chirurgicale, Bruay-Labuissière, France, 2Cemka-Eval, Bourg la Reine, France, 3Stryker, PUSIGNAN, France Fusco F 1, Campbell H 2, Newman M 3, Barker K 2 1Scuola Superiore Sant’Anna, Pisa, Italy, 2University of Oxford, Oxford, UK, 3University of Oxford, Oxford, UK., UK . . . . . . . . . A382 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: A tailored accelerated physiotherapy (AP) program following Resurfacing Hip Arthroplasty (RHA) has been shown to be effective in improving hip function and range of motion in young male patients; however no evidence has been provided on its cost-effectiveness. The aim of this UK trial-based economic evaluation was to assess the cost-effectiveness of AP versus a standard rehabilitation protocol (SRP). Methods: Trial participants were randomized post-RHA to AP or SRP. The experimental arm followed an 8-week programme with no hip precautions, full-weight bearing from day one, tailored exercises and an additional physiotherapy session. The control group received the standard 8-week course of rehabilitation. At 6, 16, and 52 weeks, patients reported primary and secondary health care contacts, use of equipment, and private health care contacts. These data were valued using 2012/2013 national average unit costs. The 3-level EuroQol EQ-5D questionnaire was completed by patients at baseline, 6, 16 and 52 weeks and used to calculate Quality Adjusted Life Years (QALYs) to 12 months. Results: 80 young males (median age: 55.8 years) were randomized to AP (n= 40) or SRP (n= 40). Preliminary results showed mean (SE) health care costs to 52 weeks were £375 (£76) in the AP arm and £612 (£150) in the SRP arm (mean (95% CI) difference -£237 (-£582 to £108)). There were more visits to secondary care and primary care practitioners in the SRP arm. Mean (SE) QALYs were 0.84 (0.02) with AP and 0.72 (0.03) with SRP (mean (95% CI) difference 0.12 (0.04 to 0.21)). The probability that AP is cost-effective at a maximum willingness to pay of £20,000 per QALY is 99%. Conclusions: From the perspective of the health care provider, a tailored accelerated physiotherapy programme for younger male patients undergoing RHA appears cost-effective when compared to a standard rehabilitation programme. PMS53 Rituximab as First Choice for Patients With Refractory Rheumatoid Arthritis: Cost-Effectiveness Analysis in Iran Based on A Systematic Review and Meta-Analysis Ahmadiani S , Nikfar S , Karimi S , Jamshidi A R , Akbari-Sari A , Kebriaeezadeh A Tehran University of Medical Sciences, Tehran, Iran . . . . . . . Objectives: The aim of this study is evaluation of the effectiveness and costeffectiveness of using rituximab as first line of treatment for patients with refractory rheumatoid arthritis in comparison to continuing conventional DMARDs, from a perspective of health service governors. Methods: A systematic review was implemented through searching PubMed, Scopus and Cochrane Library. Quality assessment was performed by JADAD questionnaire. After meta-analysis of ACR (American College of Rheumatology) index results, QALY (Quality Adjusted Life Years) gained were calculated through mapping ACR index to HAQ (Health Assessment Questionnaire) and utility index. To measure the direct and indirect medical costs, a set of interviews with patients were applied. Thirty two patients were selected from three referral rheumatology clinics in Tehran with definite diagnosis of refractory rheumatoid arthritis one year before, and treatment regimen of either rituximab or DMARDs within last year. Incremental cost-effectiveness ratio were calculated for a period of six months for base case and generic rituximab scenario. Three fold of GDP (Gross Domestic Product) per capita was considered as threshold of cost-effectiveness. Results: Four studies were eligible to be considered in this systematic review. Total risk difference of 0.3 for ACR20 criteria, 0.21 for ACR50 and 0.1 for ACR70 were resulted from meta-analysis. Also mean of total medical costs of patients for 24 weeks were $3985 in rituximab group and $932 for DMARDs group in the base case analysis. Thus, the incremental cost per QALY ratio will be $45900 to $70223 in the base case, and $32386 to $49550 for generic scenario, while the threshold of cost-effectiveness was $21684. Conclusions: Rituximab cannot be considered as cost-effective for the treatment of patients with refractory rheumatoid arthritis in Iran. PMS54 Cost-Utility Analysis of Certolizumab Pegol Plus Methotrexate for the Treatment of Moderate-To-Severe Active Rheumatoid Arthritis In Greece Tzanetakos C 1, Maniadakis N 1, Kourlaba G 2, Tzioufas A 3, Goules A 3, Theodoratou T 4, Christou P 4 1National School of Public Health, Athens, Greece, 2Collaborative Center for Clinical Epidemiology and Outcomes Research (CLEO), Athens, Greece, 3Athens Medical School, Athens, Greece, 4UCB Pharma, Athens, Greece . . . . . . . Objectives: To evaluate the cost-effectiveness of certolizumab pegol (CZP) as an add-on therapy to methotrexate (MTX) versus etanercept, adalimumab or golimumab in patients with moderate-to-severe active rheumatoid arthritis (RA) who did not respond adequately to conventional synthetic disease-modifying anti-rheumatic drugs (csDMARDs) including MTX in Greece. Methods: A Markov model with a cycle length of 6 months was used to assess cost and health outcomes of CZP versus other TNF-α inhibitors recommended in Greece over a patient’s lifetime. On the discontinuation of first-line treatment with CZP or comparator, patients were switched to a second anti-TNF agent, and after failing that, they moved on to a third biologic agent with another mode of action. A sequential use of csDMARDs was assigned after the last biologic therapy. Clinical data and utility values were extracted from published literature. The analysis was conducted from a third-party payer perspective in Greece. Costs related to drug acquisition, administration, monitoring and patient management were considered (2014). All results were presented as incremental cost-effectiveness ratios (ICERs) per quality-adjusted life year (QALY). Probabilistic sensitivity analysis (PSA) was performed to ascertain the robustness of the base-case findings. Results: The base-case analysis indicated that compared with etanercept+MTX, CZP+MTX was cost-effective (ICER: €3,177/QALY), and versus adalimumab or golimumab, CZP was the dominant strategy (less costly and more effective). For all comparisons, CZP treatment resulted in greater improvements in life expectancy and QALYs. PSA indicated that at the willingness-to-pay threshold of € 34,000 per QALY gained, CZP+MTX was associated with a 71.6%, 97.9% or 99.2% probability of being cost-effective versus etanercept, golimumab or adalimumab as combination therapies with MTX, respectively. Conclusions: This analysis shows that CZP+MTX seems to be a cost-effective alternative when compared to approved subcutaneous anti-TNFs for the management of RA in Greece. PMS55 Cost-Effectiveness Analysis of Strontium Ranelate VersUS Alendronate for Management of Osteoporosis Among PostMenopausal Women in Malaysia Using A Markov Modelling Approach Wu D B C 1, Hussain S 2, Mak V 1, Lee K K C 1 University Malaysia, Selangor, Malaysia, 2University of Malaya, Kuala Lumpur, Malaysia . . . . . . . . 1Monash Objectives: Osteoporotic fractures are common in older adults and are often associated with high morbidity and mortality. As the incidence increases with age, it is natural that osteoporotic fractures have become a major health problem worldwide. Increasing number of patients with osteoporotic fracture will have a serious economic impact on the patient themselves and the society. The objective of this study is to study the cost-effectiveness of strontium ranelate compared to alendronate for patients with post-menopausal osteoporotic fractures in Malaysia. Methods: A Markov model was developed to project clinical and economic benefits of strontium in a hypothetical cohort of patients (N= 1,000) over a 5-year time horizon. This study was conducted from a payer perspective. Model parameters including transition probabilities and costs of treating fracture at various sites were Malaysia-specific. Drug costs were obtained from a public teaching hospital in Kuala Lumpur. Utilities were derived from previous literatures and efficacy data were derived from two pivotal trials, i. e. SOTI and TROPOS trials. Outcomes were presented as cost per quality-adjusted life year (QALY) gained. A discount rate of 3% was applied. Both 1-way and multivariate probabilistic sensitivity analyses were undertaken to evaluate robustness of results. Results: Compared to alendronate, strontium could prevent 328 wrist, 192 hip, 7 vertebra and 115 multiple fractures respectively over 5 years, which was translated into 27.9 QALYs gained. Using strontium can lead to cost reduction of MYR1,416,595 (USD442,685), MYR478,257 (USD149,455), MYR22,784 (USD7,120) and MYR61,883 (USD113,088) due to reduced episodes of fractures at wrist/hip/vertebra/multiple sites respectively. The total reduction of direct medical costs of MYR2,279,519 (USD712,349) was larger than the extra drug cost, hence making strontium a costsaving therapy. Conclusions: It was shown that strontium appeared to be more cost-effective compared to alendronate and hence should be recommended in the public sector in Malaysia. PMS56 Mabthera® (Rituximab) for the Treatment of Severe Granulomatosis With Polyangiitis (Gpa) and Microscopic Polyangiitis (Mpa) – A CostUtility Model for the United Kingdom Harland D 1, Naisbett-Groet B 1, Chang S 2, Sungher D K 3, Sawyer L 2, Diamantopoulos A 4 Products Limited, Welwyn Garden City, UK, 2Symmetron Limited, Herts, UK, 3Symmetron Limited, Elstree, UK, 4Symmetron Limited, Borehamwood, UK . . . . . . . 1Roche Objectives: To evaluate the cost-effectiveness of MabThera in patients with severe GPA and MPA in the United Kingdom (UK). Background: In March 2014 NICE issued positive guidance for the use of MabThera in patients with severe GPA and MPA [TA 308]. Methods: An economic model was developed to reflect the health care system and the current treatment pathway in the UK. The cost-effectiveness analysis employs a Markov model with four health states: complete remission, non- remission, uncontrolled disease and death. Patients were assumed to start in the non-remission health state, transitioning based on their response to treatment. Relapsing patients who have exhausted all available treatment options they are assumed to enter the uncontrolled disease health state where they remain until death. The efficacy data for the intervention and comparator arm were taken from the RAVE study (Stone et al 2010) which demonstrated that MabThera was noninferior to cyclophosphamide (CYC). In a subgroup of patients who had received prior therapy, MabThera was superior to CYC. Benefits were expressed as QALYs. Costs were calculated from a National Health Service and Personal Social Services perspective. The analysis calculated incremental costs and benefits associated with the addition of MabThera to the treatment paradigm which was assumed to consist of CYC and azathioprine. For patients intolerant to CYC, MabThera was assumed to substitute for CYC. The RAVE trial reports health related quality of life using SF-36. The SF-36 scores were converted to EQ-5D in a post-hoc analysis using a published model [Ara and Brazier 2008]. Results: Base case results estimated incremental costs of approximately £3,700 and incremental QALYs of 0.306. The incremental cost-effectiveness ratio (ICER) was £12,100 per QALY gained. Conclusions: The results of this analysis suggest that MabThera is a cost-effective treatment for severe GPA and MPA. PMS57 Work Productivity Loss Due To Rheumatoid Arthtiris (Ra), Crohn’s Disease (Cd) And Psoriasis (Ps) In Poland Wladysiuk M , Bebrysz M , Fedyna M , Haldas M , Rutkowski J HTA Consulting, Krakow, Poland . . . . . Objectives: To assess the indirect costs of RA, CD and Ps in an employed population in Poland. Methods: Data on presenteeism and absenteeism related with analyzed diagnoses were collected in a cross-sectional study from patients of ambulatory specialist care around Poland (30 rheumatology, 30 dermatology and 29 gastroenterology centers). Lost productivity was measured with Work Productivity and Activity Impairment (WPAI) questionnaire and patients’ disease activity was assessed on standardized, disease specific scales (DAS28, PASI, and CDAI). 328 (RA), 460 (Ps), 256 (CD) working patients were included in the analyses conducted separately for each diagnosis. Unit cost of lost productivity was estimated using 2012 GDP per worker per hour corrected for diminishing marginal productivity and added up to PLN 33.18. Results: Mean age of M2W respondents was 36 for CD, 42 for Ps and 46 for RA patients (only patients in productive age – 18-60/65 were included in the study). A383 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Ps patients had dominantly (54%) low disease activity, in RA group mostly (51%) the moderate activity was observed, while the most frequent disease activity category for CD patients was remission (39%). Mean annual cost of overall work impairment (presenteeism and absenteeism cost together) amounted to PLN 29 727 for RA, PLN 24 434 for Ps and PLN 23 682 for CD. Cost of loss of productivity due to RA ranged from PLN 15 069 for patients in remission to PLN 41 296 for highly active disease. For Ps it was respectively PLN 13 846 and PLN 44 009 and for CD PLN 15 543 and PLN 63 771. Conclusions: Productivity loss among workers with CD, Ps and RA generates significant costs for society which rises with disease activity. PMS58 Long-Term Work Productivity Costs Among Subjects With Early Rheumatoid Arthritis - A Nationwide Analysis Based on 7,831 Subjects’ Sickness Absence Days and Income Martikainen J A 1, Krol M 2, Rantalaiho V 3, Kautiainen H 4, Puolakka K 5 of Eastern Finland, Kuopio, Finland, 2Erasmus University, Rotterdam, The Netherlands, 3Tampere University Hospital, Tampere, Finland, 4Helsinki University Central Hospital, Helsinki, Finland, 5Lappeenranta Central Hospital, Lappeenranta, Finland . . . . . . 1University Objectives: To estimate the long-term productivity costs (PC) and their determinants associated with absenteeism and permanent work disability (WD) in Finnish patients with early rheumatoid arthritis (RA) available to workforce at baseline. Methods: A cohort of subjects with early RA was created by identifying the new cases of RA from the national register of the Social Insurance Institution (SII) in Finland, who were granted a special reimbursement of anti-rheumatic medications for rheumatoid factor positive (ICD-10 code: M05) or negative RA (M06) in 2000 – 2007. The obtained dataset was enriched by cross-linking with the national databases about the subjects’ annual incomes, WD days, and permanent disability pensions. The human capital approach was applied to estimate PC based on subjects’ the annual number of absenteeism days and incomes. The PC were estimated and expressed as per patientobservation year. Hurdle regression analysis was applied to study the determinants of PC. Sensitivity analyses were conducted to test the robustness of the obtained results. Results: The study cohort comprised 7,831 subjects with early RA in paid jobs. The mean age (SD) of subjects was 46 years (11) and 71 % were women. Mean (bootstrapped 95%CI) annual PC per patient-observation year was 4,574€ (95% CI 4,469€ to 4,680€). The PC increased progressively over the years. The use of methotrexate-based combination therapies during the first three months after RA diagnosis reduced significantly the cumulative PC during the follow-up. Conclusions: So far the majority of the productivity cost studies have been based on cross-sectional data. However, the results of the present study provide unique evidence about the longitudinal economic burden of RA over the course of disease. The results highlight the need for treatment strategies with predefined targets and tight control of disease activity in the early course of disease to reduce the long-term burden of RA. PMS59 Characterizing Work Productivity Loss In Incident Rheumatoid Arthritis In Sweden Banefelt J 1, Gustavsson A 1, Borgström F 1, Alemao E 2 Research, Stockholm, Sweden, 2Bristol-Myers Squibb, Princeton, NJ, USA . . . . 1Quantify Objectives: To study the trends in work productivity loss pre- and post-diagnosis of incident rheumatoid arthritis (RA) and compare it to patients with osteoarthritis (OA) undergoing knee or hip replacement surgery. Methods: Retrospective register study conducted using Swedish national registers. Patients of working age with an incident diagnosis of RA between 2003 and 2009 were identified in the National Patient Register (n= 14900). OA patients undergoing surgery were identified during the same period and considered as a reference group (n= 34240). Monthly productivity loss twelve months pre-diagnosis to twelve months post-diagnosis was evaluated in the base case. Productivity loss was defined as the sum of net sick leave days and net disability pension days as recorded in the Social Insurance Register. Results: Monthly productivity loss gradually increased during the months leading up to RA diagnosis, peaking the month after diagnosis (mean 14.7 days/month). The same pattern was observed in the OA group in relation to time of surgery, although the increase in productivity loss the months post-surgery was considerably larger than the corresponding increase in the RA patients, peaking the second month following surgery (28.5 days/month). Twelve months post-surgery, the OA patients returned to levels of productivity loss similar to those seen six months pre-surgery (12.3 days/month vs. 12.7 days/month, respectively). This reversal was not mirrored by the RA patients, who stabilized at an elevated level of productivity loss post- compared to pre-diagnosis levels (12.1 days/month vs. 9.6 days/month). Conclusions: This study illustrates the unmet needs in RA. The partial reversal in work ability post-diagnosis, in contrast to the almost complete reversal seen in OA patients post-surgery, highlights the need for improved treatment options in RA; while the gradual loss of work ability leading up to diagnosis highlights the need for intervention earlier in the disease process. PMS60 A Comparison of the Impact of Rheumatic Diseases and Other Chronic Diseases on Early Retirement in Portugal Laires P A 1, Canhão H 2, Gouveia M 3 Portuguesa de Reumatologia, Lisbon, Portugal, 2Faculdade de Medicina, Universidade de Lisboa, Lisboa, Portugal, 3Católica Lisbon School of Business and Economics, Lisbon, Portugal . . . individual level by logistic regression. Results: At the time of the survey, 19.5% of the Portuguese population with ages between 50 and 64 years old were officially retired. A larger average number of major chronic diseases per capita was found among those with early retirement when compared to active workers in the same range of age (2.0 vs. 1.4 p<0.001). RD were particularly prevalent among early retirees when compared with employees (43.3% vs. 32.1%, p< 0.001). The following OR and PAF (adjusted by age, sex and region and unadjusted) were obtained: RD (OR:1.31.4; PAF:7.3-9.7), chronic pain (OR:1.4-1.4; PAF:6.6-7.2), hypertension [OR:1 (NS) -1.3; PAF:1.1-7.1], diabetes [OR:1 (NS) -1.2 (NS); PAF:0-2.4], renal impairment (OR:3.3 -3.4; PAF:2.7-3.2), respiratory diseases (OR:1.5-1.7; PAF:1.5-2.1), stroke (OR:3.2-3.6; PAF:2.6-3.2), myocardial infarction [OR:1.2 (NS) -1.9; PAF:0.3-1.1], cancer (OR:2.11.9; PAF:2.1-2.1), depression (OR:1.5-1.2 (NS); PAF:4.8-2.8) and anxiety (OR:1.6-1.5; PAF:3.7-3.5) with NS standing for not statically significant. Conclusions: From a public health angle, PAF is a good measure of the importance of a risk factor, taking into account both the strength of the association with the outcome and its prevalence in the population. Among all major chronic diseases, RD had the highest PAF estimates for early retirement in Portugal. PMS61 Cost of Pharmacotherapy in Polish Patients With Rheumatoid Arthritis Szafraniec-Burylo S 1, Orlewska E 2 Institute of Public Health - National Institute of Hygiene, Warsaw, Poland, 2Jan Kochanowski University, Kielce, Poland . . 1National Objectives: To assess cost of pharmacotherapy in Polish patients with rheumatoid arthritis (RA) in relation to disease activity (DAS28-CRP) and disability (HAQ-DI). Methods: Data on drug consumption was collected during a prospective one center cohort observational study of non-selected RA patients discharged from tertiary academic hospital. At enrollment patients were divided according to DAS28-CRP and HAQ-DI. Observational period was 6 months. Spearman rang correlation coefficient and test of its significance were used to investigate the relationship of cost of pharmacotherapy with disease activity and disability. Mean total cost of drugs and mean cost of drugs covered by public payer were calculated in PLN at 2014 prices (1 EURO= 4.2 PLN in 2014). Results: DAS28-CRP was ≤ 5.1 in 124 patients (group A) and > 5.1 in 83 patients (group B). HAQ-DI was > 2 in 51 patients (group I), > 1 and ≤ 2 in 88 patients (group II), ≤ 1 in 66 patients (group III). Mean cost of pharmacotherapy per patient was 1010 PLN in group A and 858 PLN in group B, 1078 PLN in group 1, 981 PLN in group II and 745 in group III. The differences between groups were not statistically significant. Public payer covers only 36-40% of the drugs cost. Conclusions: These results represent the current use of drugs in the population of Polish RA patients under real-life conditions and indicate that there is no impact of disease activity and disability on the cost of pharmacotherapy due to RA. PMS62 Organizational and Management Impact Analysis of Using the New Subcutaneous Formulation of Tocilizumab in Selected Italian Rheumatology Centers Ravera S 1, Tomic R 1, Adami S 2, Viapiana O 2, Paolini D 1, Bianchino L 1, Canciani M 3, Farina M 3, Ciancio G 4, Govoni M 4 1Roche S.p.A., Monza, Italy, 2University of Verona, Verona, Italy, 3Emmeffe S.r.l., Milano, Italy, 4Ferrara University, Ferrara, Italy . . . . . . . . . . Objectives: In the light of future arrival of subcutaneous (SC) formulation of tocilizumab for the treatment of moderate to severe active rheumatoid arthritis, a multidimensional analysis has been performed in order to evaluate potential impact of introducing SC formulation vs. intra-venous (IV) one from the perspective of both patients and hospitals in Italy. Methods: The analysis was conducted in three Italian Rheumatology centers (AO-S. Anna-Ferrara, AO-Verona, UOS-Valeggiosul-Mincio) through a questionnaire administered to clinicians and nurses. A 60 minutes day-hospital administration for IV and 20 seconds-1 minute administration for SC selfinjector or pre-filled syringe administered at home (excluding the first administration) were assumed. Monitoring visits were considered as 1 per month for IV and 1 every three months for SC. Four impact areas were evaluated: patient’s drug-administration time and costs (including transportation time and cost and loss of productivity), drug-administration related hospital-personnel time, drug wastage and patient risk profile evaluated through Failure Models and Effect Analysis. A one year time horizon was considered. Results: The analysis showed that the new SC formulation, compared to IV, could have a significant impact in terms of: patient time saving (- 91% of the time for the administration of therapy), patient costs (-86%), clinicians and nurses time saving measured as Full Time Equivalent (- 59% for clinicians and -94 % for nurses), drug wastage (-100%), and patient risk profile (-93%). Conclusions: The subcutaneous formulation of tocilizumab could have several organizational and management impacts. From the hospital perspective it could lead to reduction of medical resources consumption with the possibility to re-allocate them in other medical activities. From the patient perspective the new SC formulation could lead to time savings and costs reduction with a potential improvement of patient quality of life. . 1Sociedade Objectives: Chronic diseases and in particular rheumatic diseases (RD) may lead to early retirement, generating substantial indirect costs to society. We compare RD to other major chronic diseases regarding their impact on the likelihood of early retirement in the Portuguese population. Methods: The study population consisted of all people between 50 and 64 years of age (3,762 men and 4,241 women) who participated in the Portuguese National Health Survey, conducted in 2005/2006. Self-reported data were collected on health, sociodemographic and occupational factors. The effects of RD and other chronic diseases on the likelihood of early retirement and the population attributable fractions (PAF) estimates were obtained at the Muscular-Skeletal Disorders – Patient-Reported Outcomes & Patient Preference Studies PMS63 Systematic Review and Meta-Analysis of Persistence With Denosumab in Patients With Osteoporosis Jonsson E 1, Cheng L I 2, Ström O 1, Intorcia M 3, Karlsson L 1 Research, Stockholm, Sweden, 2Amgen Inc., Thousand Oaks, CA, USA, 3Amgen (Europe) GmbH, Zug, Switzerland . . . . . . 1Quantify Objectives: Conduct a systematic review and meta-analysis of published literature on persistence with denosumab in patients with osteoporosis. Methods: A384 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 A systematic review identified retrospective and prospective observational studies reporting 12-, 18-, or 24-month persistence with denosumab, an osteoporosis therapy approved in 2010. Searches covered the period January 2011─May 2014 and were conducted in the PubMed and EMBASE databases, and conference abstract supplements from ACR, AMCP, ASBMR, WCO-IOF-ESCEO, and ISPOR. To be eligible, studies needed to report at least one estimate of persistence with denosumab in patients with osteoporosis, have a clear definition of persistence, and be in English language. Using a random effects model, pooled estimates of denosumab persistence were calculated for retrospective and prospective studies separately. A subgroup analysis was conducted based on geographical regions. Results: The search identified 338 unique citations in PubMed and EMBASE, and 200 conference abstracts. After applying the eligibility criteria, 11 studies were included in the final review and meta-analysis (7 retrospective; 4 prospective). All studies reported at least one estimate of 12-month persistence; 9 studies included females only. For retrospective studies, the pooled persistence estimates were 74.6% (95% CI: 65.4-82.9) at 12 months, 67.6% (65.2-70.0) at 18 months, and 57.2% (51.8-62.5) at 24 months. For prospective studies, the pooled estimate of 12-month persistence was 89.0% (95% CI: 84.6-92.7). In the subgroup analysis, European studies had higher pooled 12-month persistence estimates compared with North American studies (retrospective: 78.6% vs. 68.9%; prospective: 91.5% vs. 85.3%). Conclusions: To our knowledge, this is the first systematic review and meta-analysis of persistence with denosumab in patients with osteoporosis. We identified a growing body of evidence, both prospective and retrospective, suggesting persistence to denosumab is higher than previously reported for other OP therapies1.1. Kothawala P, et al. Mayo Clinic. 2007; 82 (12): 1493-501. PMS64 Persistence Rate With Subcutaneous Biologic Therapies in Patients With Rheumatoid Arthritis (Ra) Lyu R 1, Ding Q 2, Govoni M 3, Fan T 1 & Co., Inc, Whitehouse Station, NJ, USA, 2Temple University, Philadelphia, PA, USA, 3Merck Sharp & Dohme Limited, Rome, Italy . . . . 1Merck Objectives: This study examined persistence over 12 months for RA patients who were newly treated with subcutaneous biologics, and assessed if there are differences between patients with and without prior DMARDs experience. Methods: In this retrospective cohort study using Electronic Medical Record database of IMS Disease Analyzer-Germany, adult (≥ 18 years old) RA patients with exposure to a subcutaneous biologic between January 1, 2009 and June 30, 2012 were identified. The first prescription date for the subcutaneous biologic agent was defined as their index date. Patients were excluded from the study if they were prescribed a biologic agent during the pre-index period, and/or diagnosed with ankylosing spondilitis, psoriatic arthritis, or other conditions treated with subcutaneous biologics either pre- or post-index. A chi-square test was used to assess significant differences in the percentage of persistent patients between those with and without DMARD use and a logistic regression model was used to control for differences in baseline demographic and clinical characteristics. Results: A total of 576 RA patients without prior biologic experience met the study selection criteria; 471 were DMARD experienced and 105 were DMARD naive. The mean (SD) age of the patients was 57 (13), with 75% being female. The majority of patients indexed on etanercept (46%) or adalimumab (40%). Forty eight percent of the patients persisted on their index biologic over the 12 months post-index period, with the rate significantly higher among those with pre-index DMARD use (51% vs. 31%; p= 0.0012). After controlling for pre-index characteristics, patients with pre-index DMARD had 2.23 times the odds of being persistent compared to those without pre-index DMARDs (OR: 2.23, 95% CI: 1.37-3.62). Conclusions: Approximately half of biologic naïve RA patients were persistent over a 12-month period with their index subcutaneous biologic, with rates significantly higher among patients with pre-index DMARD. PMS65 Persistence Rate With Subcutaneous Biologic Therapies in Patients With Ankylosing Spondylitis (As) Govoni M 1, Lyu R 2, Ding Q 3, Fan T 2 Sharp & Dohme Limited, Rome, Italy, 2Merck & Co., Inc, Whitehouse Station, NJ, USA, 3Temple University, Philadelphia, PA, USA . . . . 1Merck Objectives: This study aimed to describe persistence with subcutaneous biologics among biologic naïve AS patients over 12 months, and to identify differences among patients with and without NSAID drug experience. Methods: This retrospective study used IMS Disease Analyzer-Germany, an electronic medical records database. Data for adult (≥ 18 years of age) AS patients with a prescription for subcutaneous biologic between January 1, 2009 and June 30, 2012 were used for this analysis. The index date was the date of the first subcutaneous biologic prescription. Prescription for any biologic during the pre-index period or diagnosis for rheumatoid arthritis, psoriatic arthritis or other conditions treated with subcutaneous biologics either in pre- or post-index qualified patients for exclusion from the study. Differences between pre-index NSAID naïve and experienced patients were measured using a chi-square test. A logistic regression model was used to further assess the impact of NSAID use on persistence, controlling for baseline characteristics. Results: The study cohort included a total of 108 biologic naïve AS patients, 72 with use of NSAIDs in the pre-index period. The mean age of the AS cohort was 42 years, with 70% of patients being male. Adalimumab, etanercept, and golimumab were initiated by 61%, 28%, and 11% of patients, respectively. Persistence for at least 12 months with the index subcutaneous biologic was observed in 46.3% of the overall cohort, with similar results among those with and without NSAID use (47% vs 44%, respectively). Multivariate analysis confirmed similar persistence between those with and without pre-index NSAID (OR: 0.97; 95% CI: 0.39-2.44). Conclusions: Findings from this German study showed that less than half of AS patients are persistent with the index subcutaneous biologic over a 12 month period. Results were similar irrespective of prior use of NSAIDs. PMS66 Determinants of Non-Persistence to Antiosteoporotic Drugs by Using Administrative Database Orlando V 1, Guerriero F 1, Monetti V M 1, Putignano D 1, Moretti A 2, Iolascon G 2, Menditto E 1 Center of Pharmacoeconomics, Naples, Italy, 2Second University of Naples, Naples, Italy . . . . . . . . 1CIRFF- Objectives: Osteoporosis treatment involves several therapeutic tools, including long-term drug therapy. Subjects with chronic disorders are more likely to be non-adherent and/or non-persistent to treatment than those with other diseases. Adherence is the extent to which patients take medication as prescribed by their physicians, whereas persistence is the time from treatment initiation to discontinuation. Lack of persistence is common among subjects using oral anti-osteoporotic drugs, and leads to increased risk of fragility fracture. The aim of our study is to analyze the rates and reasons for discontinuation of anti-osteoporotic drugs in the Campania Region. Methods: The study was a retrospective cohort study. Patients, aged ≥ 40 years, were enrolled if at least one prescription for any antiosteoporotic drugs had been filled from July 1, 2009 through June 30, 2010. Data were retrieved from an administrative database of medications prescription in Campania region. Patients were followed from the index date until the antiosteoporotic therapy discontinuation or end of the observation period (June, 30, 2011). Results: A total of 30,048 were incident users of anti-osteoporotic drugs: 1, 731 (5.8%) males and 28,317 (94.2%) females. The mean age [SD] of the cohort was 69.0 [10.0] years. Weekly bisphosphonate (51.1%), was the most commonly prescribed drug. In the overall cohort study, persistence rates were 34, 8% after 6 months, 13, 4% at one year. A multivariate Cox proportional hazard analysis showed that daily regimen (HR 1,9) treatments remained at a higher risk of early discontinuation compared to weekly regimen therapies. Patients who started treatment with a co-prescription with calcium and vitamin D had a lower risk of early discontinuation (HR 0.7). Conclusions: Our data showed that the persistence to osteoporosis therapy is significantly worse than reported in literature. A better osteoporosis management should include drugs with less frequent dosing, to obtain both an increase in rate of persistence and a reduction in side-effect. PMS67 Use of Medication Reminders in Patients With Rheumatoid Arthritis Bruera S 1, Lopez-Olivo M A 2, Barbo A 2, Suarez-Almazor M E 2 College of Medicine, Houston, TX, USA, 2The University of Texas, MD Anderson Cancer Center, Houston, TX, USA . . . . . . 1Baylor Objectives: In this study we determined the characteristics of patients with RA who used these aids, and the association of reminder use with adherence. Methods: 201 patients with RA were included in this prospective cohort study examining treatment adherence. At baseline patients were asked if they used any special reminders such as pill containers, calendars, or diaries. Patients completed two selfreported adherence questionnaires: the Compliance Questionnaire Rheumatology and the Adult AIDS Clinical Trials Group adherence questionnaire. Disease activity measures included number of swollen joints, number of tender joints, disease activity score (DAS28), and patient global assessment. Functional status was evaluated with the modified Health Assessment Questionnaire (MHAQ). Results: Mean age of the patients was 51 years, 75% were female, 53% were Hispanic, 25% white, and 21% African American. Sixty-eight (34%) patients reported using a reminder: 53 (26%) used special pill containers, 12 (6%) used calendars, and 3 (1%) diaries. Factors associated with the use of reminders were older age (p= 0.004), being white vs. Hispanic or African American (p= 0.003), being male vs. female (p= 0.005). Working patients were less likely to report using reminders (p= 0.006). No association was observed between education levels and use of aids. Use of reminders was associated with domains of self-reported adherence: adherence while away from home (r= 0.16, p= 0.03), when busy (r= 0.16, p= 0.03), and when running out of pills (r= 0.15, p= 0.04). Conclusions: Older patients, males, and whites were more likely to use these aids, more often pill containers. Our study shows that reminders can assist patients with RA in taking their medications, particularly in situations when they are most prone to forget including being away from home or busy. Use of reminders should be encouraged by providers as a low cost aid to enhance adherence. PMS68 Treatment Persistence With Subcutaneous Biologic Therapies in Patients With Psoriatic Arthritis (Psa) Lyu R 1, Ding Q 2, Govoni M 3, Fan T 1 & Co., Inc, Whitehouse Station, NJ, USA, 2Temple University, Philadelphia, PA, USA, 3Merck Sharp & Dohme Limited, Rome, Italy . . . . 1Merck Objectives: The objective of this study was to describe persistence with subcutaneous biologic over 12 months for newly treated PsA patients and evaluate the impact of prior DMARD use. Methods: This was a retrospective analysis using IMS Disease Analyzer-Germany, an electronic medical records database. Adult (≥18 years of age) PsA patients who initiated therapy with subcutaneous biologics between January 1, 2009 and June 30, 2012 were included in the analysis. The first subcutaneous biologic prescription date served as their index date. Continuous observation of at least 12 months pre- and post-index date was required. Patients who were prescribed any biologic during the pre-index period or diagnosed with rheumatoid arthritis, ankylosing spondylitis, crohn’s disease, or ulcerative colitis during the study period were excluded from the study population. A chi-square test was used to measure differences between patients with and without use of pre-index DMARD. A multivariate logistic regression was created to assess the impact of DMARD use on persistence, controlling for baseline characteristics. Results: A total of 197 biologic-naïve PsA patients were selected. Of these, 89 were free of PsO. The mean (SD) age of the patients was 49 (11) years, with 50% being female. The majority of patients (61%) indexed on adalimumab, while the remainder indexed on etanercept (35%) and golimumab (4%). In the overall PsA population, the persistence rate with the index subcutaneous biologic was 54.3%, with similar results among those with and without DMARD use (53% vs 56%, respectively). Multivariate analysis did not identify any significant predictors for persistence, including DMARD use (OR: 1.05; A385 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 95% CI: 0.53-2.07) and psoriasis (OR: 1.07; 95% CI: 0.56-2.03). Conclusions: More than half of the PsA patients were persistent with the index subcutaneous biologic over a 12-month period with similar persistence rates observed among those with and without psoriasis and DMARD use. PMS69 Impact of Medication Adherence by Using Indian Version Compliance Questionnaire Rheumatology (Cqr) and Medication Adherence Report Scale (Mars) Tools on Quality of Life of Patients With Rheumatoid Arthritis Shetty R 1, Reddy K 2, Inam S 2, Khera K 1 1Manipal College of Pharmaceutical Sciences, Manipal, India, 2Manipal Univiersity, Manipal, India . . . . Objectives: To assess medication adherence to DMARD in patients with Rheumatoid Arthritis using CQR and MARS tools, identification of factors affecting adherence and its effect on quality of life. Methods: A randomly selected sample of 110 adult patients with RA on DMARDs admitted to hospital were asked about their medication adherence, through self-report questionnaire [CQR and MARS] and quality of life was assessed by HAQ (Health Assessment Questionnaire). Additionally, various factors affecting adherence were identified. Results: According to the tools used, 86.4% (CQR), 74.29% (MARS -mean cut point) and 95.45% (MARS -prior study cut point) of patients showed adherence towards DMARD. Better adherence was seen in patients with primary education (COR- 94%) or secondary education (MARS -83%). Patients who suffered from RA for more than 2yrs showed better adherence (CQR- 93%) compared to those with recent disease (< 2yrs) (CQR- 89%). Non adherence was seen in patients having co-morbidities compared to patients with only RA (CQR- 91% vs 94%; MARS- 62% vs 82%). Mean HAQ of adherent patients was better (2.83±1.05) than non-adherent patients (3.23 ± 0.74). Adherent patients showed moderately active disease state (Mean DAS – 5.96 ± 1.67) whereas, non-adherent patients showed highly active disease state (Mean DAS – 6.70 ± 0.84). Conclusions: Patient reported questionnaires showed disease duration of less of 2yrs, and patients with co-morbidities lead to Non-adherence which worsened disease activity which lead to decreased quality of life. PMS70 Quality of Life in Psoriatic Arthritis: Consistent and Stable Across Datasets Hatswell A J 1, Almond C 1, Nassens D 2, Ganguly R 3, Ito T 4 1BresMed, Sheffield, UK, 2Janssen BVBA, Beerse, Belgium, 3Janssen R&D, LLC, Spring House, PA, USA, 4Janssen-Cilag Ltd, High Wycombe, UK . . . . . . Objectives: Psoriatic arthritis (PsA) is a multi-factorial disease that affects the skin, joints and soft tissues. Two of the commonly used measures for PsA are the Psoriasis Area and Severity Index (PASI, 0-100 scale) and the Health Assessment Questionnaire (HAQ, 0-3 scale) for skin and joints symptoms, respectively. Previous work in the area has estimated a relationship between these patient-reported instruments and utility (SF-36 mapped to the EQ-5D). The objectives of this study were to calculate patient-reported utility and investigate the consistency of the relationship between PASI, HAQ and utility with previously published estimates, based on the PSUMMIT trials of ustekinumab versus placebo. Methods: Patient level data from PSUMMIT1 (anti-TNFα naïve) and PSUMMIT2 (both anti-TNFα naïve and experienced) were analysed in Stata 11. SF-36 data were converted to EQ-5D using the mapping by Rowen et al., with regression analysis used to estimate the relationship between PASI, HAQ and the resulting utility (including multiplicative terms). Goodness of fit was determined by the adjusted R2 and Root Mean Squared Error (RMSE). Results: Anti-TNFα naïve and experienced patients had a baseline utility of 0.50 and 0.48, respectively. Utility improved over the 24-week blinded period by 0.04/0.06 in the placebo arms for anti-TNFα naïve and experienced, and 0.11/0.13 in the treatment arms. In regression analysis utility was predicted as 0.897 – 0.004xPASI - 0.298xHAQ (adjusted R2 0.60, RMSE 0.12), similar to previously published estimates. Adding a multiplicative term for PASI and HAQ did not improve goodness of fit statistics, although baseline methotrexate use was linked to a lower utility. Conclusions: Patients with PsA have a low level of health-related quality of life that improves with treatment. The determinants of utility in the PSUMMIT trials were the skin and joint symptoms faced by patients, in keeping with previous estimates. PMS71 Patient Preferences in the Choice of Disease Modifying Anti-Rheumatic Drugs Schiffner-Rohe J 1, Alten R 2, Krüger K 3, Behmer O S 4, Schiffhorst G 5, Rellecke J 5, Nolting H D 5 1Pfizer Deutschland GmbH, Berlin, Germany, 2Schlosspark Klinik, Berlin, Germany, 3n.a., Munich, Germany, 4Pfizer Pharma GmbH, Berlin, Germany, 5IGES Institut GmbH, Berlin, Germany . . . . . . . . . Objectives: There is a variety of biologic and non-biologic disease modifying antirheumatic drugs (DMARDs) available for the treatment of rheumatoid arthritis (RA). These DMARDs are associated with different characteristics in key attributes such as mode of administration, side effects, etc. The current study assessed the importance of treatment characteristics for RA patients’ preferences. Methods: In a discrete choice experiment (DCE), 1570 RA patients are asked to choose the most and the least preferred DMARD (best-worst-scaling) among hypothetical multi-attribute treatment alternatives with varying levels of key attributes, as defined in focus groups: mode of administration, frequency of administration, time till onset of drug effect, necessity of combination therapy with methotrexate, and side effects. The multi-profile case design simulates a real choice situation between different hypothetical treatment alternatives. Interim analysis was conducted after half the sample size had been reached. Results: Interim analysis included 836 patients from 33 office based rheumatologists across Germany. Majority of patients were female (74%), 50 to 64 years of age (46%), with < 10 years of disease duration (54%), and reported experience with injectable DMARDs (63%). Mode of administration appeared the most important attribute guiding patients’ preferences, with ‘oral application’ being most desired (selected as best option in 51% of the cases) and infusion being least preferred (worst option in 45% of the cases). The second most relevant attribute was “necessity of combination therapy with methotrexate”, with DMARDs not requiring such combination being most preferred (in 43% of the cases). Conclusions: Our data indicate that, of the included attributes, the most important ones are route of administration (oral being the number one choice by majority) and combination therapy with methotrexate (with DMARDs not requiring such combination being the most preferred) for RA patients’ choice. This research was funded by Pfizer GmbH. PMS72 Are Patients’ Preferences Transferable Between Countries? A CrossEuropean Discrete-Choice Experiment to Elicit Patients’ Preferences for Osteoporosis Drug Treatment Hiligsmann M 1, Dellaert B 2, Dirksen C 1, Van der Weijden T 3, Watson V 4, Goemaere S 5, Reginster J Y 6, Bours S 1, Roux C 7, McGowan B 8, Silke C 8, Whelan B 8, Diez Perez A 9, Papadakis G 10, Torres E 9, Rizzoli R 10, Cooper C 11, Pearson G 11, Boonen A1 1Maastricht University, Maastricht, The Netherlands, 2Erasmus University Rotterdam, Rotterdam, The Netherlands, 3CAPHRI, Maastricht, The Netherlands, 4Health Economics Research Unit, University of Aberdeen, Aberdeen, UK, 5Ghent University Hospital, Ghent, Belgium, 6University of Liège, Liège, Belgium, 7Paris Descartes University, Paris, France, 8Our Lady’s Hospital, Manorhamilton, Ireland, 9Hospital del Mar-IMIM and RETICEF, Barcelona, Spain, 10Geneva University Hospitals, Geneva, Switzerland, 11University of Southampton, Southampton, UK . . . . . . . . . . . . . . . . . . . Objectives: Discrete-choice experiments are increasingly used to assess preferences in health care. To date, very little is known about the transferability of patients’ preferences between jurisdictions. In this study, we aim to evaluate the preferences of patients with, or at risk of, osteoporosis for medication attributes in six European countries, and to assess whether preferences are transferable across these countries Methods: A discrete-choice experiment was conducted using a questionnaire in Belgium, France, Ireland, Spain, Switzerland and United Kingdom. Patients were asked to choose between two hypothetical unlabelled drug treatments (and an opt-out option) that vary in several attributes: efficacy in reducing the risk of fracture, type of potential common side-effects, mode and frequency of administration and out-of-pocket costs (only in countries with patients´ contribution on the cost of treatment). An efficient design was used to construct the treatment option choice sets and a mixed logit model was used to estimate patients’ preferences. Results: A total of 1,124 patients completed the experiment, with at least 100 patients per country. As expected, in all countries, patients preferred treatment with higher effectiveness and lower cost was preferred in the three countries in which a cost-attribute was part of the experiment. In all countries, patients preferred 6-monthly subcutaneous injection over weekly oral tablets. In most countries, patients also preferred monthly oral tablet and yearly intravenous injections over weekly oral tablets. Patients disliked being at risk of gastro-intestinal disorders more than being at risk of skin reactions and flu-like symptoms, except in Spain. There were significant differences between countries for some levels of attributes. Conclusions: This study suggests that the preferences of patients for osteoporotic drug therapy did not substantially differ between six European countries. However, for levels of some attributes, significant differences were observed. PMS73 Long-Term Maintenance of Improvements in Patient-Reported Outcomes With Certolizumab Pegol in Patients With Axial Spondyloarthritis, Including Ankylosing Spondylitis and NonRadiographic Axial Spondyloarthritis: 96-Week Results of the Rapid-Axspa Study Sieper J 1, Kivitz A 2, van Tubergen A 3, Deodhar A 4, Szegvari B 5, Nurminen T 6, Landewé R 7 Hospital Charité, Berlin, Germany, 2Altoona Center for Clinical Research, Duncansville, PA, USA, 3Maastricht University Medical Center, Maastricht, The Netherlands, 4Oregon Health and Science University, Portland, OR, USA, 5UCB Pharma, Brussels, Belgium, 6UCB Pharma, Monheim, Germany, 7Amsterdam and Atrium Medical Center, Heerlen, The Netherlands . . . . . . . 1University Objectives: To report the effect of certolizumab pegol (CZP), a PEGylated Fc-free anti-TNF, on patient-reported outcomes (PROs) in axial spondyloarthritis (axSpA), including ankylosing spondylitis (AS) and non-radiographic axSpA (nr-axSpA), over 96 weeks (wks) of the RAPID-axSpA trial. Methods: The RAPID-axSpA trial (NCT01087762) is double-blind and placebo-controlled to Wk24, dose-blind to Wk48, and open-label to Wk204. Patients fulfilled ASAS criteria and had active axSpA. Patients originally randomized to CZP (200mg Q2W or 400mg Q4W, following 400mg loading dose at Wks 0, 2, 4) continued on their assigned dose in the dose-blind phase and OLE. Here we report PRO data for the CZP-treated randomized set, including mean change from baseline and the proportion of patients achieving a Minimal Clinically Important Difference (MCID). Missing data were imputed by LOCF. Correlations between clinical and patient-reported outcomes were also investigated. Results: Of 218 patients randomized to CZP, 203 (93%) completed Wk24, 191 (88%) Wk48, and 174 (80%) Wk96. Rapid improvements from baseline to Wk24 were maintained to Wk96 in all patient subpopulations (overall axSpA, AS, nr-axSpA) in pain (Wk24: -3.2, -3.2, -3.3; Wk96: -3.6, -3.6, -3.7); fatigue (Wk24: -2.7, -2.5, -2.9; Wk96: -2.9, -2.8, -3.1); BASFI (Wk24: -2.4, -2.3, -2.4; Wk96: -2.6, -2.6, -2.6); ASQoL (Wk24: -5.1, -4.8, -5.5; Wk96: -5.7, -5.5, -6.1) and sleep (Wk24: -12.8, -10.5, -15.7; Wk96: -13.9, -11.6, -16.7). CZP-treated patients also maintained improvements in SF-36 components and domains. Sustained improvements in the proportion of patients (overall axSpA, AS, nr-axSpA) achieving MCID (%) were observed in fatigue (Wk24: 78.4, 76.0, 81.4; Wk96: 67.0, 70.2, 62.9); BASFI (Wk24: 67.4, 68.6, 66.0; Wk96: 64.2, 68.6, 58.8) and ASQoL (Wk24: 69.3, 71.1, 67.0; Wk96: 65.6, 66.9, 63.9). Similar outcomes were seen with both dosing regimens. Correlations were observed between improvements in PROs (pain/fatigue/SF-36) and clinical outcomes (ASDAS) (data not shown). Conclusions: Improvements in PROs (including pain, fatigue and ASQoL) were maintained over 96 wks in both the AS and nr-axSpA subpopulations. Sustained improvements in the proportion of patients achieving MCID were also reported. A386 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PMS74 Inadequate Pain Relief Among Patients With Primary Knee Osteoarthritis - Analysis From The Portuguese Sample Of The Survey Of Osteoarthritis Real World Therapies (Sort) Laires P 1, Laíns J 2, Miranda L 3, Cernadas R 4, Pereira da Silva J 5, Gomes J M 6, Peloso P M 7, Taylor S D 7, Silva J C 8 1Merck Sharp & Dohme, Oeiras, Portugal, 2Centro de Medicina e Reabilitação da Região Centro, Coimbra, Portugal, 3Instituto Português de Reumatologia, Lisbon, Portugal, 4ARS Norte, Oporto, Portugal, 5Centro Hospitalar e Universitário de Coimbra, Coimbra, Portugal, 6Clínica Reumatológica Dr. Melo Gomes, Lisbon, Portugal, 7Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Whitehouse Station, NJ, USA, 8Hospital Garcia de Orta, Almada, Portugal . . . . . . . . . . . . . Objectives: Despite widespread treatments for Osteoarthritis (OA), data on treatment patterns, adequacy of pain relief, and quality of life are limited. The prospective multinational Survey of Osteoarthritis Real World Therapies (SORT) was designed to investigate these aspects. This analysis aims to describe the clinical characteristics and the patient reported outcomes of the Portuguese dataset of SORT baseline. Methods: The statistical analysis included, from January to December of 2011, 192 participants ≥ 50 years or older with primary knee OA from 7 health care centers in Portugal who were receiving oral or topic analgesics. Inadequate Pain Relief (IPR) was defined as a score > 4/10 in item 5 of the Brief Pain Inventory (BPI), indicating moderate to severe pain. Results: Overall, the median age was 67.0 ± 8.7 years and 77.6% were female. Mean duration of knee OA was 6.3 ± 6.3 years. IPR was reported by 52.0% of the patients. The most prescribed analgesics were NSAIDs (88.1%), alternative therapies, including glucosamine, chondroitin and hyaluronate (44.3%) and paracetamol (28.6%). Patients with IPR scored higher than non-IPR patients in WOMAC – Stiffness (61.0 vs 39.7, p< 0.001) and WOMAC – physical function (59.2 vs 39.4, p< 0.001), meaning worse condition. Patients with IPR had worse quality of life related to knee osteoarthritis as measured by the SF-12 questionnaire (fair/poor: 86.9% vs. 72.0%, p< 0.001). 62.0% of patients with IPR were dissatisfied or very dissatisfied with the effects of analgesics versus 34.0% of patients with non-IPR (p< 0.05). Conclusions: Despite the use of analgesics, over half of the Portuguese patients in SORT reported moderate to severe knee pain. Worse outcomes were also observed in this group regarding other symptoms of knee OA and general quality of life. These findings suggest that if an improvement of pain management in knee OA can be achieved, it may have high impact on patients’ lives. PMS75 Qualitative Equivalence Between A Paper and Electronic Tablet Version of the Womac®Nrs3.1 and Patient Global Assessment Eremenco S 1, Fleming S 2, Riordan D 3, Stringer S 1, Gleeson S 1, Sanga P 4, Kelly K 5 1Evidera, Inc., Bethesda, MD, USA, 2Janssen Global Services, Titusville, NJ, USA, 3Janssen Research and Development, Raritan, NJ, USA, 4Janssen Research and Development, Titusville, NJ, USA, 5Janssen Research and Development L.L.C., Titusville, NJ, USA . . . . . . . Objectives: Prior equivalence work with the WOMAC® scale was published for the VAS scale and older touchscreen computer technology. Additional equivalence evaluation of the WOMAC®NRS3.1 and the Patient Global Assessment (PGA) in a newer tablet with stylus was needed to document suitability of this mode of data collection for these instruments in upcoming clinical trials. Methods: A cross-sectional qualitative study was conducted involving cognitive and usability interviews with patients diagnosed with osteoarthritis of the hip or knee who were taking pain medication for their condition. Interviews were conducted in two waves of 10 participants each, with revisions to the PGA made in between the rounds, which allowed for changes to the electronic version to be evaluated. Results: Mean age of the sample (N= 20) was 66 years, (range 43-78), 90% over 60 years old; 60% were female; 95% were white; 75% were retired; 70% had completed secondary school or some college, while 30% had completed college or a post-graduate degree. In wave 1, minor issues were found with completing the WOMAC®, mainly with using the stylus to select responses and glare on the screen. There were no issues identified in interpreting the response scale. For the PGA, 50% (5/10) used the wrong recall period (48 hours or longer). The PGA recall period was revised from “at this time” to “over the past 24 hours” and bolded for emphasis. In wave 2, similar issues with glare and stylus response were found, while 80% used the correct recall period on the PGA, with 20% using 48 hours. Conclusions: The study showed excellent qualitative equivalence between the paper and electronic WOMAC® with only minor usability issues. The two wave study design provided the opportunity to detect and make changes to the PGA recall period and formatting that showed improvement in the second wave. PMS76 Long-Term Maintenance of Improvements in Multiple Facets of Psoriatic Arthritis With Certolizumab Pegol: 96-Week PatientReported Outcome Results Of The Rapid-Psa Study Gladman D 1, Fleischmann R 2, Szegvari B 3, Peterson L 4, Mease P J 5 Research Institute, Toronto, Ontario, ON, Canada, 2Metroplex Clinical Research Center, Dallas, TX, USA, 3UCB Pharma, Brussels, Belgium, 4UCB Pharma, Raleigh, NC, USA, 5Swedish Medical Center and University of Washington, Seattle, WA, USA . . . . . . 1Toronto Western Objectives: To report the effect of certolizumab pegol (CZP), a PEGylated Fc-free anti-TNF, on patient-reported outcomes (PROs) in psoriatic arthritis (PsA) over 96 weeks (wks) of the RAPID-PsA trial. Methods: The RAPID-PsA trial (NCT01087788) is double-blind and placebo-controlled to Wk24, dose-blind to Wk48 and open-label to Wk216. Patients had active PsA and had failed ≥ 1 DMARD. Patients originally randomized to CZP (200mg Q2W or 400mg Q4W, following 400mg loading dose at Wk0, Wk2, Wk4) continued on their assigned dose in the dose-blind phase and OLE. Here we present PRO data for the CZP-treated randomized set, including mean change from baseline (CFB) and Minimal Clinically Important Differences (MCIDs). Data were also analysed for CZP-randomized patients with (19.8%) or without (80.2%) prior anti-TNF exposure. Missing data were imputed by LOCF. Correlations between clinical outcomes and PROs were also investigated. Results: Of 273 patients randomized to CZP at Wk0, 91% completed Wk24, 87% Wk48, and 80% Wk96. Rapid improvements observed to Wk24 were maintained to Wk96 for pain (Wk24 and Wk96; CFB: -28.5 and -31.3; MCID: 69.2% and 66.3%), fatigue (Wk24 and Wk96; CFB: -2.0 and -2.4; MCID: 64.1% and 60.4%), HAQ-DI (Wk24 and Wk96; CFB: -0.48 and -0.52; MCID: 48.7% and 48.0%), SF-36 PCS (Wk24 and Wk96; CFB: 8.01 and 9.01; MCID: 67.4% and 60.1%), SF-36 MCS (Wk24 and Wk96; CFB: 4.50 and 3.92; MCID: 50.9% and 43.6%), PsAQoL (Wk24 and Wk96; CFB: -3.87 and -4.50), and DLQI (Wk24 and Wk96; CFB: -5.8 and -6.0; MCID: 40.7% and 41.0%). Similar improvements were observed with both dosing regimens and in patients with or without prior anti-TNF exposure. Correlations were observed between improvements in PROs and DAS28 (data not shown). Conclusions: Improvements observed to Wk24 in generic and diseasespecific PROs were sustained to Wk96 of the RAPID-PsA trial for both CZP dosing regimens. PMS77 Usability Testing of A Novel Pain Medication Diary Administered Electronically Eremenco S 1, Fleming S 2, Riordan D 3, Stringer S 1, Gleeson S 1, Sanga P 4, Kelly K5 1Evidera, Inc., Bethesda, MD, USA, 2Janssen Global Services, Titusville, NJ, USA, 3Janssen Research and Development, Raritan, NJ, USA, 4Janssen Research and Development, Titusville, NJ, USA, 5Janssen Research and Development L.L.C., Titusville, NJ, USA . . . . . . Objectives: Pain medication diaries have traditionally been collected via paper due to challenges of patients entering unlimited medications, units, dosages, and administration schedules. This study developed an electronic diary that permits site staff to enter medications that patients are taking, enables the patient to update medication taken and to enter new medications within the reporting period, and reduces the possibility of cheating behaviors during the study. Usability of this electronic diary was evaluated to ensure that patients in a clinical trial setting could successfully update their diaries in real-time to accurately track pain medication intake. Methods: A cross-sectional qualitative study was conducted involving usability interviews with patients diagnosed with osteoarthritis of the hip or knee who were taking pain medication. Interviews were conducted in two waves of 10 participants each, allowing for evaluation of findings and revisions to the eDiary between waves. Results: Mean age of the sample (N= 20) was 66 years (range 43-78), 90% over 60 years old; 60% were female; 95% were white; 70% completed secondary school or some college. In wave 1, issues were noted with training, selecting responses, exiting to send data, and some wording. For wave 2, the training module was revised to more closely match the diary, wording was revised, and a screen added to facilitate exiting the diary. No issues were noted with training, 4 had trouble selecting responses, and 3 suggested additional instructions on the new screen. No additional changes were made following wave 2. Conclusions: The study showed it is possible to develop an electronic pain medication diary that allows patients to update their medications during a study. Extensive training was critical to the usability of the electronic version. The two wave study design provided the opportunity to detect and make changes to the eDiary with marked improvement in wave 2. PMS78 Quality of Life in Patients With Axial Spondyloarthritis in Clinical Practice In Sweden: Baseline Results From A Longitudinal Study Jacobsson L T 1, Husmark T 2, Theander E 3, Henriksson K 4, Johansson M 5, Büsch K 5 1Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden, 2Falu Hospital, Falun, Sweden, 3Lund University, Malmo, Sweden, 4Rheumatology city clinic, Stockholm, Sweden, 5AbbVie AB, Solna, Sweden . . . . . . . Objectives: Spondyloarthritis (SpA) is a group of diseases that share common clinical, radiographic and genetic features. Axial SpA is one major subgroup including patients with radiographic (rad-axSpA) and non-radiographic axSpA (nr-axSpA). There has been limited research on axSpA patients in clinical practice and the impact of the disease on patient’s health-related quality of life (HrQoL). The aim of this study was to characterize patients with axSpA in clinical practice and to investigate similarities/differences between rad-axSpA and nr-axSpA with respect to their HrQoL. Methods: This is a longitudinal, multi-center cohort study where patients were consecutively recruited from Swedish clinical practice and followed for 3 months. At baseline, the rheumatologists registered information on disease history, extra articular manifestations and treatments. The patients answered online questionnaires capturing patient demographics, disease activity, function and HrQoL. HrQoL was measured using the EQ-5D and the Ankylosing Spondylitis Quality of Life Questionnaire (ASQoL). While higher scores in the EQ-5D indicate better HrQoL, the opposite is true for the ASQoL. Results: 251 patients were included of whom 197 (78%) were classified as axial SpA. Of those, 125 (63%) were classified as rad-axSpA and 72 (37%) as nr-axSpA according to the ASAS axSpA criteria. There were more women in the nr-axSpA group (50%) compared with the rad-axSpA group (38%). The nr-axSpA patients had a shorter time between symptom onset and diagnosis than the rad-axSpA patients (6.7 vs. 9.0 years) and a significantly higher disease activity (BASDAI= 4.1 vs 2.7, p< 0.001). Mean EQ-5D score at baseline was 0.66 for rad-axSpA and 0.61 for nr-axSpA, lower than the Swedish general population (0.84). ASQoL scores was significantly higher in the nr-axSpA group (8.8 vs 6.4, p= 0.016). Conclusions: HRQoL is poorer in axial SpA patients compared to the general population and patients with nr-axSpA reported a higher impact on HRQoL than patients with rad-axSpA. PMS79 Functional Status, Quality of Life And Work Disability for Patients With Rheumatic Diseases in Greece Athanasiadi E 1, Fragoulakis V 2, Vozikis A 2 1Medical School of Athens, Athens, Greece, 2University of Piraeus, Pireaus, Greece . . . Objectives: Rheumatic diseases (RD) have been associated with functional and work-related disability due to the deliberating and progressive nature of these diseases and have many deleterious consequences on patients’ life. The aim of the A387 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 present study was to measure the functional status and quality of life in RD patients receiving the biologic agent golimumab in Greece Methods: A descriptive study was conducted estimating the annual Quality-of-Life (QoL) improvement for 148 patients diagnosed with rheumatoid arthritis (RA), psoriatic arthritis (PS) and ankylosing spondylitis (AS). QoL was estimated with standardized questionnaires such as EQ-5D and HAQ questionnaire. In addition, twenty specific parameters including “general health”, “severity of pain”, “productivity level” etcetera wereassessed on 10-point Likert scale. Data was collected by doctors across the country at 3 month intervals (4 waves of questionnaires during the year). Results: Patients with RA reported improved QoL by 14%, followed by 0.67 decrease of average HAQ score. They also reported a 7.5 hours gain of working hours and an overall 14% gain in productivity per week. Patients with AS were estimated to present 30% improvement in general health condition according to their doctors with similar self-reported estimations. Additionally, their QoL was improved by 17%. Similarly, patients with PS estimated their QoL improved by 17% with a 6hr/week gain in working hours otherwise missed, while the average HAQ score fell by 0.71 Findings are in accordance with similar published studies all contributing to the general assumption that patients receiving biological agents experience greater QoL improvement compared to conservative treatment options. Conclusions: Statistical analysis revealed significant improvement of functional status, quality of life, productivity gains and decrease of disease activity for those receiving Golimumab in Greece for all three disease groups. Amongst limitations, both number of the study group and follow-up period should be taken into consideration. PMS80 Quality of Life in Patients With Chronic Lumbosciatic Syndrome in the Slovak Republic Matisakova I 1, Kamenska E 1, Mastiliakova D 2, Melus V 1, Bielik J 2, Novak I 3, Andrasova E 3 University of Alexander Dubcek, Trencin, Slovak Republic, 2Trencin University, Trencin, Slovak Republic, 3Novartis Slovakia, Bratislava, Slovak Republic . . . . . . . 1Trencin Objectives: The current prevalence of Chronic Lumbosciatic Syndrome (CLSS) in Slovakia ranges in about 170000 cases. The CLSS has a great impact on quality of life (QoL) and the ability to work too. Till now in Slovakia was not realised the study like this. Methods: 86 patients with CLSS were studied. The “hospital” and the “out-patients clinic” group had 43 vs 43 patients. The average age was 50.36 vs 51.79 y., weight – 61.04 vs 77.95 kg, duration of illness – 5.3 vs 5.0 years, symptoms of illness before diagnosis – 2.27 vs 2.16 years. QoL and the ability to work was evaluated on the numeric scale from 0 to 10 (0 for the worst, 10 for the best) by patients themselves. Results: The “hospital” and the “out-patients clinic” group had these results: the average of hospitalisation – 1.51 vs 1.44 times, incapacity to work – 3.06 vs 2.90 months. QoL in the time of good health was 8.2 vs 8.04, in the time of diagnosis – 5.86 vs 6.74, and in the current time – 4.58 vs 3.95. The work ability (WA) had these results: WA in time of good health was 8.97 vs 9.04 in the time of diagnosis – 8.16 vs 8.13, and in the current time – 3.83 vs 3.83. The impact of the treatment on the QoL was 5.37 vs 6.06 and on the patients families QoL it was 6.16 vs 6.13. The willingness to pay for the perfect cure was 477.90 € vs 524.41 € per month (the average salary in Slovakia in 2013 was 824 € ). Conclusions: CLSS has a great impact on QoL and on the WA too. There was not statistical difference between both “hospital and out-patient clinic” patients in the QoL and WA. Early diagnosis is important to effective treatment. PMS81 Reasons for Treatment Discontinuation of Biologics – Development and Validation of A Questionnaire Meyer-Moock S 1, Schiffner-Rohe J 2, Kohlmann T 3 Greifswald, Greifswald, Germany, 2Pfizer Deutschland GmbH, Berlin, Germany, 3University Medicine Greifswald, Greifswald, Germany . . . 1Universitätsmedizin Objectives: For patients with rheumatoid arthritis (RA) and inadequate response to traditional therapy, treatment with biologics is recommended to reduce disease progression and improve HRQol. Nevertheless, up to 30% of patients stop treatment on their own initiative. Aim of the study is to develop and validate an instrument to assess reasons for treatment discontinuation of biologics in patients with RA. Methods: We conducted expert interviews with rheumatologists (n= 5) to develop a questioning route, which was used in 2 focus groups with a total of 15 RA patients who dropped out therapy (phase 1). Based on these results a draft questionnaire was developed and pre-tested (phase 2; n= 6), resulting in the pilot questionnaire. In the validation phase (phase 3) the questionnaire is completed by approx. 200 patients, with an interim analysis planned after recruitment of half the sample size. Quantitative data analyses will focus on psychometric properties: missing data, floor/ceiling effects, factorial validity, distribution of properties. Patients are recruited at 40 German office based rheumatologists, with 5 patients per center. Results: The final questionnaire consists of 82 items covering sociodemographic aspects, HRQoL, history of treatment and disease, treatment information, expectations and satisfaction and ”reasons for discontinuation”. Piloting showed that HRQoL, treatment and working-life aspects have impact on treatment discontinuation. Particularly, information about treatment options, patient-doctor relationship and financial burden were addressed. The questionnaire has proven to be feasible in field-test. Validation phase is currently ongoing; results from the interim analysis will be presented. Conclusions: Interviews with experts and patients demonstrated that treatment discontinuation is triggered by multiple reasons. A targeted developed questionnaire is necessary to identify latent reasons for treatment discontinuation. Furthermore, an “easy-to-use” questionnaire could be used in daily routine to identify patients likely to withdraw treatment and need special patient adherence programs.This research was funded by Pfizer GmbH. PMS82 Assessing Willigness to Pay Among Psoriasis and Psoriatic Arthritis Patients Dierick K 1, Rose A 2 1GfK Disease Atlas, Brussels, Belgium, 2GfK Disease Atlas, London, UK . . Objectives: It is estimated that 30% of psoriasis (PSO) patients also develops psoriatic arthritis (PSA). This potential disease evolution brings along new symptoms such as swollen and painful joints and mobility problems. Since both diseases are potentially degenerative, the aim of this study was to measure which group of patients (PSO or PSA) is willing to pay most out of pocket to avoid their health related quality of life (HRQoL) to worsen. Methods: 395 US patients diagnosed with either psoriasis (n= 151) or psoriatic arthritis (n= 247) completed a questionnaire as part of a broader survey of treatment of PSO/PSA. The questionnaire included the EQ-5D-5L instrument and accompanying VAS. Patients were additionally asked to indicate by reference to the EQ-5D VAS scale the amount of money per month they would be willing to pay for treatments that would prevent a decline in HRQoL by 10 points. Price sensitivity curves were created by means of linear regression analysis that predict the proportion of patients willing to pay a certain amount of $ out of pocket per month. Results: For both PSO (R² = 0,82) and PSA (R² = 0,86) monthly cost out of pocket (x-variable) was a good predictor of the proportion of patients that is willing to pay a certain amount out of pocket per month (y-variable). Regression models look as follows. For PSO: y = 0,77e-0,005x for PSA: y = 0,86e-0,007x. To give a specific example 42% of the PSA patients is willing to pay $100 per month out of pocket whereas this is 47% among PSO patients. Conclusions: PSO patients are prepared to pay more out of pocket on a monthly basis to avoid their HRQoL to worsen than PSA patients. Further research is required to understand what drives this difference. PMS83 Sustained Improvements in Workplace and Household Productivity and Social Participation With Certolizumab Pegol Over 96 Weeks in Patients With Axial Spondyloarthritis, Including Ankylosing Spondylitis and Non-Radiographic Axial Spondyloarthritis van der Heijde D 1, Braun J 2, Rudwaleit M 3, Purcaru O 4, Kavanaugh A 5 University Medical Centre, Leiden, The Netherlands, 2Rheumazentrum Ruhrgebiet, Herne, Germany, 3Endokrinologikum Berlin, Berlin, Germany, 4UCB Pharma, Brussels, Belgium, 5UCSD, San Diego, CA, USA . . . . . 1Leiden Objectives: To report the long-term effect of certolizumab pegol (CZP) on workplace and household productivity up to 96 weeks (wks) in patients with axial spondyloarthritis (axSpA), including ankylosing spondylitis (AS, meeting modified New York criteria) and non-radiographic axSpA (nr-axSpA). Methods: The ongoing RAPID-axSpA trial (NCT01087762), is double-blind and PBO-controlled to Wk24, dose-blind to Wk48 and open-label to Wk204. Patients had active axSpA, according to ASAS criteria, including AS and nr-axSpA patients. Patients originally randomized to CZP (200mg Q2W or 400mg Q4W, following 400mg loading dose [LD] at Wks 0, 2, 4) continued on their assigned dose in the OLE; PBO patients entering dose-blind phase were re-randomized to CZP LD followed by CZP 200mg Q2W or CZP 400mg Q4W after Wk24 or, for non-responders, after Wk16. The validated arthritis-specific Work Productivity Survey (WPS; administered Q4W) assessed the impact of axSpA on workplace and household productivity. WPS responses (LOCF imputation) in patients originally randomized to CZP are summarized descriptively over 96 wks. Results: 325 patients were randomized, of whom 218 received CZP (200mg Q2W or 400mg Q4W) from Wk0. Of patients randomized to CZP at baseline (BL), 93% completed Wk24, 88% Wk48 and 80% Wk96. At BL, 72% of CZP patients were employed outside of the home. Employed CZP patients reported reductions in workplace absenteeism and presenteeism to Wk24, with continued improvements to Wk96 (BL: mean 1.8 days missed/month, mean 5.2 days with reduced productivity/month vs Wk96: mean 0.6 days missed/month, mean 1.4 days with reduced productivity/month). CZP patients also reported continued improvements in household productivity and social participation to Wk96 in both dose regimens, and similar improvements were observed in AS and nr-axSpA. Conclusions: The initial improvements with CZP in workplace and household productivity and increased participation in social/leisure activities were continued to Wk96 in axSpA, AS and nr-axSpA patients. PMS84 Sustained Improvements in Workplace and Household Productivity and Social Participation With Certolizumab Pegol Over 96 Weeks in Patients With Psoriatic Arthritis Kavanaugh A 1, Gladman D 2, van der Heijde D 3, Purcaru O 4, Mease P J 5 1UCSD, San Diego, CA, USA, 2Toronto Western Research Institute, Toronto, Ontario, ON, Canada, 3Leiden University Medical Centre, Leiden, The Netherlands, 4UCB Pharma, Brussels, Belgium, 5Swedish Medical Center and University of Washington, Seattle, WA, USA . . . . . . Objectives: To examine the long-term effect of certolizumab pegol (CZP) on workplace and household productivity up to 96 weeks (wks) in patients with active psoriatic arthritis (PsA). Methods: The ongoing RAPID-PsA trial (NCT01087788) is double-blind and PBO-controlled to Wk24, dose-blind to Wk48 and open-label to Wk216. Patients had active PsA and had failed ≥ 1 DMARD. Patients originally randomized to CZP (200mg Q2W or 400mg Q4W, following 400mg loading dose [LD] at Wks 0, 2, 4) continued on their assigned dose in the OLE; PBO patients entering dose-blind phase were re-randomized to CZP LD followed by CZP 200mg Q2W or 400mg Q4W after Wk24 or, for non-responders, Wk16. The validated arthritis-specific Work Productivity Survey (WPS) administered Q4W from baseline (BL), assessed the impact of PsA on workplace and household productivity in the randomized set. WPS responses (LOCF imputation) in patients originally randomized to CZP groups are summarized descriptively over 96 wks. Results: 409 patients were randomized, of whom 273 received CZP 200mg Q2W or CZP 400mg Q4W. Of patients randomized to CZP, 91% completed Wk24, 87% Wk48 and 80% Wk96. In employed patients in both CZP groups (60.8% of all CZP patients at BL), decreases in absenteeism and presenteeism to Wk24 were continued up to Wk96 (BL: mean 2.0 and 1.6 days missed/ month in the CZP 200mg Q2W and 400mg Q4W groups, respectively; mean 5.2 and 5.1 days with reduced productivity/month vs Wk96: mean 0.3 and 0.4 days missed/ month; mean 0.7 and 1.5 days with reduced productivity/month). Improvements in household productivity and social participation reported in both CZP groups over 24 wks were also maintained to Wk96. Conclusions: The initial improvements A388 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 with CZP in workplace and household productivity, and social participation were sustained up to 96 wks in PsA patients. Muscular-Skeletal Disorders – Health Care Use & Policy Studies PMS85 Market Access of Implantable Medical Devices - Part Ii: Decision Drivers Across Global Markets Chawla A S 1, Tao C 2, Spinner D S 3, Faulkner E C 4, Doyle J J 5 1Quintiles Consulting, Durham, NC, USA, 2Quintiles Consulting, Cambridge, MA, USA, 3Quintiles, Durham, NC, USA, 4Institute for Pharmacogenomics and Individualized Therapy, Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, NC, USA, 5Quintiles, Hawthorne, NY, USA . . . . . . . . . Objectives: With rising pressures on health care budgets, health technology assessment (HTA) agencies are increasingly scrutinizing medical devices (MDs) for economic benefits in addition to clinical benefits. This level of scrutiny has resulted in many unfavorable recommendations from agencies and only a small proportion of unconditionally favorable reviews. As an extension of our work reported at 2013 ISPOR Annual Congress (Dublin, IR) this study aims to: 1) Identify key criteria cited by HTA agencies as major decision drivers, 2) Note common criteria among reviews that were positive, negative, or positive with reservations, and 3) Analyze temporal or geographic trends among decision drivers. Methods: A review of 68 HTAs and reimbursement decisions of implantable MD with a variety of indications was conducted, focusing on decisions published from 2008-2013 identified by Quintiles’ HTA Watch from North America, Europe, and Australia. Clinical, economic, and other factors noted as pivotal to HTA and reimbursement decisions were registered and compared. Importantly, care was exercised to note only the criteria that triggered a HTA to make a favorable or unfavorable decision, as opposed to criteria that were only correlative. Results: Key product attributes affecting HTA decisions include 1) sufficiency and quality of evidence, 2) cost offsets and budget impact, 3) adverse event profiles, and 4) comparison to existing alternatives where available. Notab ly, 33% of HTA decisions were negative, with many decisions citing insufficient evidence. Additionally, a majority of favorable HTA decisions were reserved in their recommendations, citing a need for additional evidence to uphold the initially favorable recommendation. The relative importance of economic considerations varied across countries. Conclusions: HTA agencies’ scrutiny of sufficiency of evidence, among others, may significantly impact market access of medical devices. As such, manufacturers need careful planning to align evidence development, pricing and access plans with HTA agency, payer and pricing authority requirements. PMS86 Anti-Tnf Biosimilars Indicated for Rheumatoid Arthritis are Increasingly Available in Europe: How Do Payers and Key Stakeholders Perceive Them? Sewak N P S , Jones C Double Helix Consulting, London, UK . . . . Objectives: The process of bringing a biosimilar to market in Europe is quicker, easier and cheaper than developing a new biologic. As a class, rheumatoid arthritis (RA) has the greatest number of anti-TNF biosimilar molecules in development, with more expected to follow. This research was focussed on the key issues reported by payers and Key Opinion Leaders (KOLs) in France, Germany and Italy. Methods: Supported by secondary research our study entailed conducting one hour telephone interviews with influential senior payers involved in budgetary decision making at the national and regional level in addition to KOLs. These structured interviews explored how stakeholders perceived the introduction of anti-TNFs biosimilars. Results: Payers see anti-TNF biosimilars as an opportunity to reduce the biologic budget but KOLs want to treat more patients within the same budget. Payers in Germany and France reported a greater perception of the efficacy of biosimilars than their counterparts in Italy. Treatment naïve patients are considered most suitable for anti-TNF biosimilars while automatic substitution was not favoured by any respondents. Nonetheless, price played a role and some KOLs stated they may attempt to switch existing patients who have a low risk of acute complications with very close monitoring. Conclusions: Biosimilars may be perceived unequally across markets. Manufacturers are likely to require the use of differentiated value stories when presenting their biosimilar products to payers and KOLs, with the latter more inclined to perceive them as an opportunity to treat more patients with the same expenditure instead of reducing budgets. Manufacturers will likely struggle to encourage the switching of existing patients onto biosimilars without offering a significant discount. In France and Germany, anti-TNF RA biosimilars are currently generating demand that closely matches their increasing prevalence. PMS87 Comparison of Clinical Characteristics of Patients With Rheumatoid Arthritis (Ra) Receiving Biologic Monotherapy and Biologic-Containing Combination Therapy in Europe Narayanan S 1, Lu Y 2, Hutchings R 2, Baynton E 2 1Ipsos Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, London, UK . . . . Objectives: To assess the clinical characteristics of patients with RA who received biologic monotherapy (“Mono”) or biologic-containing combination therapy (“Combo”) in Europe. Methods: A multi-country, multi-center medical chart review study of patients with RA was conducted in Q42012 among physicians in hospitals and private practices to collect de-identified data on patients who were recently treated with a biologic as part of usual care in France/Germany/Italy/Spain/UK. Physicians were screened for duration of practice (3-30yrs) and patient volume (≥2 RA biologic patients/week) and recruited from a large panel to be geographically representative in each country. Eligible patient charts (≥ 5) were randomly selected from among the patients visiting each center/practice during the screening period. Physicians abstracted date of diagnosis, treatment patterns/dynamics, and symptomatology/ disease status. Mono and Combo patients were compared used descriptive statis- tics. Results: 1534 eligible RA patients were assessed; Mono: 428 (28%), Combo: 1106 (72%). Patient characteristics (Mono/Combo) included: age 51.8/51.7; female 71%/75%; weight 68.6/68.5kg; top three comorbidities: dyslipidemia (16%/19%), depression/anxiety (9%/13%), obesity (8%/12%). Time since diagnosis: 68.6/78.2mo. Current line of biologic therapy: first-line 86%/75%, second-line 11%/18%, ≥ third-line 3%/6%. Top four biologics used across the two patient groups: etanercept (33%)/adalimumab (30%)/tocilizumab (9%)/certolizumab pegol (7%). Current lab values/disease severity measures: ESR (mm/h) 21.7/23.2; CRP (mg/l) 10.3/10.3; rheumatoid factor (positive) 84%/87%; anti-CCP (positive) 70%/80%; current disease stage per physician judgment: mild 65%/52%, moderate 32%/40%, severe 3%/8%; mean VAS 3.4/3.6; mean HAQ 1.4/1.1; mean DAS 28 3.6/3.3; mean tender joint count 3.5/4.1; mean swollen joint count 2.4/2.6. Conclusions: In this cohort of RA patients in Europe, the majority of patients on monotherapy and combination therapy had mild disease per physician judgment and were on first-line biologic therapy. Lab measures and joint counts indicated only slightly higher disease burden among combination therapy patients. The impact of specific biologic treatments on observed patterns and the need for therapeutic sequencing may warrant further research. PMS88 Coping With A New Biologic Paradigm: Payer Strategies for the Purchasing of Complex Biosimilars Ziai Buetas A , Vidal Pinheiro A , Storer M ICON, London, UK . . . Objectives: In Europe, biosimilars of complex molecules such as monoclonal antibodies have started to enter the market. In this research we aim to provide an understanding of the current expectations for the purchasing of these products as well as an overview of the tools that payers expect to employ to encourage biosimilar use. Methods: Qualitative survey of payers across national, regional and local levels in France, Germany, Italy, Spain, the UK and Netherlands. Collection and analysis of data on (1) current and future attitudes relative to expected biosimilar purchasing systems; and (2) the tools that payers expect to use to drive biosimilar use, assuming this is a payer goal. Results: (1) The method of biosimilar purchase in the short term will vary by country, with the majority of countries using tenders to procure these products; (2) While tenders restricting choice of product to a single winner will not be used extensively in the short term, these will be common across most countries in the future; (3) Payers will also use a number of other tools such as formal and informal recommendations, prescription incentives and auditing or prescription targets in order to encourage use of the product they have chosen; (4) New procurement pathways and tools are being developed to introduce biosimilars, as highlighted by the new biosimilar law in Italy which defines a level of discount at the national level. Conclusions: Payer strategies should maximize savings by introducing less expensive biosimilars, but also must consider physician preferences, especially when influenced by potentially valid concerns about lack of data. In order to do this, payers are creating novel purchasing frameworks and tools, and while they are currently reticent to use restrictive methods to encourage use of biosimilars, this is expected to rapidly change in a short timeframe. PMS89 An Assessment of the Association Between Rural Status and Health Service Resource Use Among Patients With Ankle Sprains in Ontario Lucas G H , Bielska I A , Fong R K , Johnson A P Queen’s University, Kingston, ON, Canada . . . . . . . . Objectives: Despite Ontario’s universal health care system, differences exist in health care accessibility and quality across the province. The objective of this study is to assess health care resource utilization for patients with ankle sprains based on rurality. Methods: Data on individuals who sought medical attention for ankle sprains between 2003 and 2011 in Ontario were obtained from multiple databases linked through the Institute for Clinical Evaluative Sciences (ICES). The Rurality Index of Ontario (RIO) was used to measure the rurality level of patients based on their population density and geographic distance to health care facilities. Demographic characteristics were obtained for each of five RIO categories. Health care utilization (number of visits to primary care physicians, specialists and ambulatory care) and physician billing costs were obtained and compared among the RIO categories. Results: Between 2003 and 2011, the Ontario Health Insurance Program was billed $64 million and $36 million (2013 CAD) by specialists and general practitioners, respectively, for the treatment of ankle sprains and dislocations. Approximately $116 million was spent on direct and indirect costs of emergency room visits for ankle sprains and dislocations. The largest proportion of rural injuries occurred in the top income quintile. Patients in the most rural RIO category saw specialists least often and had the highest number of ambulatory care visits. However, specialist visits constituted higher costs when compared to GP visits. The highest specialist costs found were for males, elderly patients, and those who sought medical attention during winter. The observed statistical differences in cost of GP visits across RIO categories were not clinically meaningful. Conclusions: The differences in health care utilization between RIO categories may indicate a lack of access to specialist care with those residing in rural areas relying on emergency departments for care. These results may be useful in allocating future resources to better serve rural patients. PMS90 Predicting the Burden of Knee Arthroplasty Revision Over A 20-Year Horizon Comas M 1, Guerrero-Ludueña R E 1, Espallargues M 2, Coll M 3, Pons M 4, Sabatés S 5, Allepuz A 2, Castells X 1 (Hospital del Mar Medical Research Institute; Red de Investigación en Servicios de Salud en Enfermedades Crónicas (REDISSEC), Barcelona, Spain, 2Agència de Qualitat i Avaluació Sanitàries de Catalunya (AQuAS); Red de Investigación en Servicios de Salud en Enfermedades Crónicas (REDISSEC), Barcelona, Spain, 3Hospital de Mataró, Mataró, Spain, 4Hospital de Sant Rafael, Barcelona, Spain, 5Hospital Mútua de Terrassa, Terrassa, Spain . . . 1IMIM . . . . . . A389 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To estimate future scenarios of utilization of knee arthroplasty (KA) revision in the Spanish National Health System at the short and long term and its impact on primary KA utilization. Methods: A discrete event simulation model was built to represent the utilization of KA for 20 years (2011–2031) in the Spanish National Health System, especially the burden of KA revision according to different scenarios of utilization and protheses survival. Data on KA utilization from 1997 to 2011 was obtained from the Spanish Minimum Data Set. Three scenarios of future utilization of primary KA were estimated: 1) fixed number since 2011; 2) fixed age and sex adjusted rates since 2011; and 3) projection using a linear regression model. These three scenarios were combined with two prostheses survival functions L) from a study including primary KA from 1995 to 2000; and H) from the Catalan Registry of Arthroplasty, including primary KA from 2005 to 2013. The model was programmed using ARENA. The simulation results were analyzed at the short (2015) and long-term (2030). Results: Variations in the number of revisions depended on both the primary utilization rate and the survival function applied, ranging from 8.3% to 31.6% increase at the short-term and from 38.3% to 176.9% at the long term, percentages corresponding to the combinations of scenario 1 (low primary utilization rate) and survival function H (better survival) versus scenario 3 (high primary utilization rate) and survival function L (worse survival), respectively. The prediction of increase on overall surgeries ranged from 0.1% to 22.3% at the short-term and from 3.7% to 98.2% at the long-term. Conclusions: Projections of the burden of knee arthroplasty provide a quantitative basis for future policy decisions relating to concentration of high complexity procedures, the number of orthopaedic surgeons required to perform these procedures and the number of resources needed. PMS91 Cachexia in the Us Health Care System Noone J 1, Blanchette C M 1, Roy D 2, Van Doren B 2, Arthur S 1 of North Carolina at Charlotte, Charlotte, NC, USA, 2University of North Carolina, Charlotte, Charlotte, NC, USA . . . . . . 1University Objectives: Cachexia is a medical syndrome associated with several chronic health conditions including many cancers, COPD, HIV, and kidney disease. Cachexia is a wasting type syndrome characterized as a loss in body mass or metabolic dysfunction. The loss in mass is associated with decreases in strength and functional capacity. Currently there is little research into cachexia and our objective is to characterize cachexia patients, their health care utilization and costs. Methods: For this study we utilized one year (2009) of the Nationwide Inpatient Sample (NIS). The NIS represents all inpatient stays at a random 20% sample of hospitals within the United States. We grouped cachexia individuals by primary or secondary diagnosis and then compared those with cachexia to all others in terms of length of stay (LOS) and total cost. Finally we looked into factor predicting increased LOS using a negative binomial model. Results: We estimated US prevalence for cachexia related admissions at 161,898 cases. Cachexia patients were older with an average age of 67.95 versus 48.10 in their non-cachexia peers. Hospitalizations associated with cachexia had an increased LOS compared to non-cachexia patients (6 days versus 3) with average costs per stay $4,641.30 greater. Differences were seen in loss of function (LOF) with cachexia patients mostly in the major LOF category (52.60%) whereas non-cachexia patients were spread between minor, moderate, and major LOF (36.28%, 36.11%, and 21.26%). Significant positive predictors of increased LOS among cachexia patients included urban hospital (IRR= 1.21 non-teaching urban, IRR= 1.23 teaching urban), having either major (IRR= 1.41) or extreme (IRR= 2.64) LOF, and having a primary diagnosis of pneumonia (IRR= 1.15). Conclusions: Cachexia is a diverse syndrome associated with a number of chronic diseases. We have characterized cachexia and seen it associated with increased length of stay, increased cost, and more severe loss of function compared to those without cachexia. PMS92 Increased Bone Mineral Density (Bmd) In Postmenopausal Women With Osteoporosis (Op) Receiving Two Denosumab Injections In Routine Clinical Practice In Bulgaria Boyanov M 1, Shinkov A 2, Psachoulia E 3, Intorcia M 3, Petkova R 4 1University Alexandrovska Hospital, Sofia, Bulgaria, 2University Hospital of Endocrinology, Sofia, Bulgaria, 3Amgen (Europe) GmbH, Zug, Switzerland, 4Amgen Bulgaria, Sofia, Bulgaria . . . . . Objectives: To describe baseline characteristics and changes in BMD T-scores at 1 year, in postmenopausal women with OP receiving 2 denosumab injections in routine clinical practice in Bulgaria. Methods: This retrospective observational study, conducted in 11 specialist (endocrinology or rheumatology) practices scattered geographically across Bulgaria, included postmenopausal women ≥50 years old with a clinical diagnosis of OP, who initiated denosumab 60 mg Q6M on or after Oct 2011 (regulatory approval of denosumab in Bulgaria) and received a follow-up injection within 7 months (until Aug 2013). All study outcomes were recorded as per routine clinical practice/reimbursement requirements, with BMD T-scores recorded (at ≥ 1 site) at first denosumab injection (baseline) and 1-year follow-up. Descriptive statistics were conducted. Results: 222 women met the inclusion criteria with a mean (SD) age of 64.2 (±8.54) years; approximately half (49.5%) were < 65 years old and13.1% ≥ 75 years. Mean (SD) age at menopause was 48.1 (±3.98) years. 26.6% reported a prior fragility fracture, with vertebral the most common site (71.2%) followed by hip (6.8%) and other sites (32.2%, excluding hip). At baseline, 2.7% were receiving vitamin D only, 5.9% calcium supplements only and 35.1% both; 31.5% had received prior OP therapy. At baseline, mean (SD) BMD T-score was -3.2 (±0.63) at the lumbar spine (LS; n= 189), -2.3 (±0.81) at the total hip (TH; n= 75) and -2.7 (±0.71) at the femoral neck (FN; n= 137). At 1-year follow-up, all women had BMD assessed at ≥ 1 site; T-scores increased to -2.7 (±0.57) at the LS (n= 187), -2.1 (±0.91) at the TH (n= 65) and -2.4 (±0.68) at the FN (n= 123). Conclusions: Postmenopausal women with OP receiving 2 denosumab injections in Bulgarian clinical practice had a mean age of 64.2 years and experienced improved mean BMD T-scores at the LS, TH, and FN after 1 year. Study funded by Amgen. PMS93 Baseline Patient Characteristics of A Prospective Observational Study to Evaluate the Care Map of Women With Postmenopausal Osteoporosis (Pmo) in Switzerland (Campos) Lippuner K 1, Lamy O 2, Theiler R 3, Merlin C 4, Pendl G 5, Del Ponte A 5, Murigande C 5, Schwenkglenks M 6 1University of Bern, Bern, Switzerland, 2Lausanne University Hospital, Lausanne, Switzerland, 3Triemli Hospital, Zurich, Switzerland, 4Rheumatology Practice and Osteoporosis Center, Baden, Switzerland, 5Amgen Switzerland AG, Zug, Switzerland, 6University of Basel, Basel, Switzerland . . . . . . . . Objectives: Report baseline patient characteristics of the CAMPOS study, which is evaluating the PMO care map in Swiss clinical practice. Methods: Between June 2012„ŸMay 2013, specialist osteoporosis centers operating a DXA machine enrolled women diagnosed with PMO who initiated intravenous (IV) ibandronate or zolendronate, or subcutaneous (SC) denosumab (index date) within 6 months prior to center initiation. Study outcomes are recorded at index date (baseline) and any visit in the 24-month observation period thereafter, as available from routine practice. Results: Twenty-one centers, mainly in urban regions (81%), of a non-academic nature (76%), and specializing in rheumatology (71%), recruited 275 women. Two-hundred-sixty-three met the inclusion criteria and were included in this baseline analysis; 180 (68%) of these received follow-up care at the specialist center and 83 (32%) were referred back to general practice. At baseline, mean (SD) age was 70.9 (9.7) years and mean time since diagnosis 5.4 (5.4) years (n= 248). The following diagnostic parameters and risk factors were assessed at the time of PMO diagnosis: prior osteoporotic fracture, 258 [94%] patients, with 60% reporting this risk factor; DXA bone mineral density (BMD) scan, 255 (97%) patients (hip and either lumbar spine or forearm BMD measured in 231 [88%] patients), mean (SD) BMD T-scores at the femoral neck and lumbar spine -2.34 (0.79; n=230) and -2.89 (1.28; n= 232), respectively; calcium and vitamin D status, 260 (99%) patients; fracture history, 258 [98%] patients. Conclusions: In the current study, prevalent fractures were assessed in almost all women initiating IV ibandronate or zolendronate, or SC denosumab, and the majority reported at least one prior fracture. Calcium and vitamin D status, DXA BMD scans and/or fracture history were also routinely assessed at PMO diagnosis. PMS94 What Could the Future Hold? Simulating the Demand for Osteoarthritis (oa) Care in Alberta to Plan a Sustainable Oa Care System Marshall D 1, Vanderby S 2, Carter M 3, Wasylak T 4, Mosher D P 1, Noseworthy T 5, Maxwell C 6, MacDonald K 1, Frank C 7 1University of Calgary, Calgary, AB, Canada, 2University of Saskatchewan, Saskatoon, SK, Canada, 3University of Toronto, Toronto, ON, Canada, 4Alberta Health Services, Calgary, AB, Canada, 5University of Calgary, Alberta Health Services, Calgary, AB, Canada, 6University of Waterloo, Waterloo, ON, Canada, 7Alberta Innovates Health Solutions, Calgary, AB, Canada . . . . . . . . . . Objectives: Osteoarthritis (OA) and the demand for OA care are increasing with the aging population. Policy-makers seek to identify policies to sustainably manage this growing demand, yet envisioning the short- and long-term effects of policy options is difficult within chronic care. We aimed to develop a decision-support tool enabling policy-makers to explore policies and their effects. Methods: We developed a system dynamics (SD) simulation of patient flow across the continuum of OA care in Alberta: from self-directed to primary and specialist care, through surgical interventions, post-surgical follow-up and subsequent re-operations. The simulation was developed using SD modeling principles and an iterative, integrated knowledge translation process, including multiple workshops with clinicians and administrators to define the problem, system boundaries and current patient flow. The resulting simulation was populated with data extracted from administrative databases (e. g. physician claims, inpatient records). Results: The model yields patient population, OA care resource requirements and associated cost results at each stage of care over 10 years by region and patient characteristics (e. g. sex). If current practices continue, annual hip and knee replacement surgery volumes are estimated to increase by more than 5,000 between 2015 and 2025. If a 14 week surgical wait-time is implemented in 2015, 600 additional surgeries must be performed in the first year to “catch-up” on the existing surgical queue, yet long-term surgery rates are similar to those without the wait-time target. The costs of the additional surgeries are partly offset by the savings achieved by fewer patients requiring care while awaiting surgery. Conclusions: This simulation can be used as a decisionsupport tool to estimate changes in patient populations, resource requirements and costs over time that may result from various OA management scenarios. Such results can equip policy makers with additional evidence to make more informed OA care policy decisions. PMS95 The Use of Clinical Data Repository for the Establishment of an Osteoporosis Registry in A Large Health Organization in Israel: Epidemiologic and Pharmaepidemiologic Findings Goldshtein I 1, Shalev V 2, Chodick G 3, Chandler J 4, Martin Nguyen A 5, ish Shalom S 6 1Maccabi Healthcare Services, Tel Aviv, Israel, 2Medical Division, Maccabi Healthcare Services, Tel Aviv, Israel, 3Maccabi Healthcare Services and Tel Aviv University, Tel Aviv, Israel, 4Merck, North Wales, PA, USA, 5Merck, North wales, PA, USA, 6Technion university, Haifa, Israel . . . . . . Objectives: Osteoporosis is an important public health issue due to its rising prevalence and excess morbidity and mortality among this population. The present study aimed to demonstrate the use of clinical data repository in Israel’s second largest health organization (Maccabi Healthcare Services) to establish a registry of osteoporosis patients and assess its early findings with respect to the epidemiology and burden of the disease, high risk populations, and quality of care. Methods: Included in the registry are patients with history of osteoporosis diagnosis, typical fractures (e. g. closed fractures of proximal femur, vertebral, Colles’ and proximal humerus) and or purchases of relevant medications, documented since 2000. In addition, we included patients with low bone density from over 140,000 measurements, using an A390 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 automated Optical Character Recognition (OCR) system. Patients younger than 18 years of age, diagnosed with Paget’s disease or purchasing pamidronic or zoledrnic acid due to cancer were excluded. Registry entry date was defined as the first medical event consistent with any of inclusion and all exclusion criteria. Results: We identified 118,141 patients with osteoporosis, with a point prevalence of 19% in 2013 among members aged 50+, and an average incidence of approximately 6,300 cases a year. Average age at registry entry was 61 y/o for women and 66 y/o for men. 66% of the patients had more than one qualifying criteria. 28% of the registry patients were never treated. Conclusions: The methods and results described may provide a model for the development of an important tool to assess the substantial burden of osteoporosis and to monitor the quality of care over time. PMS96 Concomitant Use of Non-Steroidal Anti-Inflammatory Drugs (Nsaids) And Proton Pump Inhibitors (Ppis) in Newly Diagnosed Patients With Osteoarthritis (Oa), Rheumatoid Arthritis (Ra) or Ankylosing Spondylitis (As) Dziarmaga A 1, Reidel K 2, White R 1, Tarride J E 1, Corner N 2 1AstraZeneca Canada Inc., Mississauga, ON, Canada, 2IMS Brogan Canada, Kirkland, QC, Canada . . . . . . Objectives: The use of PPIs to reduce the risk of upper GI events caused by NSAIDs is well documented. The aim of this study was to understand whether newly diagnosed patients with OA, RA or AS not previously exposed to PPIs, receive a gastro-protectant with their NSAID treatment at initiation or follow up in a primary care setting. Methods: We analyzed data from primary care Electronic Medical Records in Ontario, Canada. Patients ≥ 18 years, who were new users of NSAIDs, NSAID plus a PPI, a fixed combination of diclofenac and misoprostol, or Celecoxib between January 1st 2010 and May 31st 2012 were selected. Patients must have had a diagnosis for OA, RA, or AS within 6 months of treatment initiation, but no diagnosis in the year prior. Patients were followed for 6 months to assess for subsequent PPI prescription. Patients prescribed a PPI in the last 12 months were excluded. Results were compared to Canadian longitudinal pharmacy data. Results: There were 692 patients included, 56% females and a mean age of 44 years (SD 15). Naproxen was the most commonly used treatment (41%), followed by diclofenac (26%), celecoxib (14%) and diclofenac with misoprostol (19%). PPIs were added at treatment initiation in 2.5% of naproxen patients, 1.7% of diclofenac patients and 3.7% of patients with fixed combination of diclofenac and misoprostol. PPI prescription patterns within 6 months were similar and consistent with patterns observed in the Canadian longitudinal pharmacy data. Conclusions: Results suggest that primary care physicians in Canada do not prescribe PPIs with NSAIDs at or within 6 months after NSAID initiation. Limitations include small sample size and young patient cohort. Further studies are required to better understand the impact of low PPI concomitant use with NSAIDs, especially in older patients or those at a higher risk of GI events. PMS97 Gastrointestinal Risk Factors and Treatment Pattenrs of Rheumatoid Arthritis Versus Osteoarthritis Patients in Korea Lee S H 1, Lee E Y 2, Kim H J 3 1Konkuk University College of Medicine, Seoul, South Korea, 2Seoul National University College of Medicine, Seoul, South Korea, 3Pfizer Pharmaceuticals Korea Limited, Seoul, South Korea . . . . . . Objectives: This study aimed to investigate and compare the gastrointestinal (GI) risk factors and treatment patterns of rheumatoid arthritis (RA) and osteoarthritis (OA) patients in real clinical practice of Korea. Methods: This was a nationwide, cross-sectional study of patients taking non-steroidal anti-inflammatory drugs (NSAID, either non-selective or selective COX-2 inhibitor (COX2i)) from 20 hospitals between December 2012 and September 2013. Total 1,896 patients who were ≥ 20 years old (RA: 981 OA: 915) and were taking NSAID at least 1 month were enrolled. Data were collected through medical chart review and patients survey. The GI risk factors included NSAID duration (≥ 3 months), high-dose of NSIAD use, drinking, smoking, comorbid disease, aspirin use, anticoagulant (warfarin) use, steroid use, Helicobacter pylori infection, experience of GI event (i. e. GI bleeding or ulcer). The treatment patterns were identified as non-selective NSAID (ns-NSAID) or COX2i with/without gastroprotective agents respectively. Results: In RA, proportion of patients taking NSAID≥ 3 months, smoker and steroid users were higher than in OA patients (p< . 0001). In OA, proportion of patients who have comorbid disease and take aspirin were higher than in RA patients (p< . 0001). The rest of the GI risk factors were present as a similar proportion in both groups. The percentage of treatment with COX2i (RA: 54.3% vs OA: 44.2%, p< . 001) and gastroprotective agents (RA: 83.0% vs OA: 78.3%, p= . 009) in RA patients was higher than that in OA patients. In older aged patients (age≥ 60) in both groups, there was tendency to get more treatment of COX2i (RA: 60.9%, OA: 50.2%) compared to ns-NSAID. Interestingly, as patients get more numbers of GI risk factors, there seemed to get more proportions of ns-NSAIDs users in both RA and OA patients. Conclusions: The proportion of GI risk factors found in OA patients was comparable to that in RA patients. There was a tendency to show preferential ns-NSAID treatment pattern rather than COX2i especially in the presence of multiple GI risk factors in arthritis patients. Neurological Disorders – Clinical Outcomes Studies PND1 Modelling the Impact of Persistence Improvements with an Electronic Injection Device on Escalation to 2nd Line Treatment in Patients with Relapsing Remitting Multiple Sclerosis (RRMS) Meletiche D M 1, Rutkowski T 2, Martin de Bustamante M A 3, Chowdhury C A 4, Beckerman R 2 1EMD Serono, Inc., Rockland, MA, USA, 2CBPartners, New York City, NY, USA, 3CBPartners, San Francisco, CA, USA, 4CBPartners, New York, NY, USA . . . . . . . . Objectives: The purpose of this analysis was to estimate how higher persistence in RRMS patients treated with interferon beta 1-a and its electronic injection device (Rebismart®) affects the rate of escalation to 2nd line therapies. Methods: A 2-year, decision-analytic model was developed to track a hypothetical cohort of 1,000 RRMS patients initiating treatment with either interferon beta 1-a and Rebismart® or another 1st line self-injectable disease modifying drug (DMD). The model calculated the expected probability of escalating to 2nd line therapy, which was defined as either Gilenya® or Tysabri®. Persistence curves were estimated from a real-world dataset collected from the UK NHS where at 2 years Rebismart® users had 83% persistence vs. 60% in non- Rebismart® users. A market research study was utilized to determine the treatment pathways for non-persistent patients, where 39% transitioned to another 1st line DMD; 21% abandoned therapy and 40% escalated to 2nd line therapies. Oneway sensitivity analyses (OWSAs) varied the probability of non-persistent 1st line patients escalating to 2nd line therapies from 5% to 100%. Results: The model estimated that over two years 5.4% of patients on the Rebismart® device are expected to escalate to 2nd line therapy, while 18.3% of patients that initiated treatment on other DMDs are expected to escalate. At the lower bound of the OWSA, 0.68% of patients are expected to escalate on Rebismart®, vs. 2.3%; while at the upper bound of the OWSA, 13.3% of Rebismart® patients are expected to escalate vs. 42.7%. Conclusions: Higher persistence rates in RRMS patients initiating treatment with interferon beta 1-a and the Rebismart® electronic device are expected to result in fewer escalations to 2nd line therapies relative to other self-injectable DMDs. PND2 A Systematic Review and Network Meta-Analysis of Pharmacological Therapies Used for Patients with Advanced Parkinson’s Disease Ren S 1, Cooper K 2, Cooper J A 3, Smith H T 3, Shaikh S 3 1Sheffield University, Sheffield, UK, 2University of Sheffield, Sheffield, UK, 3GSK, Brentford, UK . . . . . . . Objectives: To assess the relative efficacy and safety of modified-release levodopa (IPX066), controlled-release levodopa and add-on therapy to immediate-release levodopa (including dopamine agonists, monoamine oxidase-B inhibitors (MAOBI), and catechol-O-methyl transferase inhibitors (COMTI)) using network meta-analysis (NMA). Methods: A systematic literature search was conducted in MEDLINE, MEDLINE In-Process, EMBASE, Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, DARE and HTA. A random effects NMA was used to determine the relative efficacy and safety of treatments on off-time reduction, three Unified Parkinson’s Disease Rating Scale (UPDRS) scores, patient withdrawals, and six adverse events in WinBUGS. An additional analysis was conducted to assess treatment class effects. Node-splitting approach was used to assess the assumption of consistency when direct and indirect evidence was combined. Results: Forty-three trials with 9,453 patients were identified. Immediate-release levodopa plus pramipexole produced the greatest reduction in off-time relative to immediate-release levodopa plus placebo (-1.71 hours a day; 95% CrI: -2.11, -1.35), followed by IPX066 (-1.40; 95% CrI: -2.19, -0.67). The greatest improvement on UPDRS ADL score was given by add-on ropinirole (-2.33 points; 95% CrI: -3.53, -1.06); on UPDRS motor and total score was given by add-on pramipexole (-5.88 points; 95% CrI: -7.22, -4.63 and -10.09 points; 95% CrI: -13.75, -6.56, respectively). Dyskinesia was increased with IPX066 and adjuvant therapy classes except for MAOBIs. Patient withdrawals were also increased with IPX066 and controlled-release levodopa. Conclusions: The NMA showed that all treatments except controlled-release levodopa were associated with a statistically significant reduction in off-time. Dopamine agonist class as adjunctive to levodopa therapy had the greatest reduction. IPX066 was broadly comparable with add-on dopamine agonists on off-time reduction, and comparable with add-on MAOBIs and COMTIs on UPDRS scores. However, the treatments were associated with an increase in the risk of having dopaminergic side effects, particularly dyskinesia. PND3 The Burden of Parkinson Disease Amongst Caregivers in Spain Over 4 Years Martinez P 1, Rodriguez-Blazquez C 1, Paz S 2, Lizán L 3, Forjaz M J 4, Frades B 5, Jimenez O L 6 Center for Epidemiology and CIBERNED, Carlos III Institute of Health, Madrid, Spain, 2Outcomes’10, Castellon, Spain, 3Outcomes 10, Castellon, Spain, 4National School of Public Health and REDISSEC, Carlos III Institute of Health, Madrid, Spain, 5Alzheimer Center Reina Sofia Foundation, Carlos III Institute of Health, Madrid, Spain, 6H. U. de Gran Canaria Dr. Negrín, Las Palmas de Gran Canaria, Spain . . . . . . . . . 1National Objectives: To describe caregivers’ Health Related Quality of Life (HRQoL), burden, anxiety and depression and their relation with patients’ symptoms in Parkinson disease (PD). Methods: A descriptive, observational, longitudinal design in PD patients and their caregivers of ELEP study (2006-2010), in Spain. Information collected during 3 month per year for 4 years, included sociodemographic characteristics, motor symptoms (SCOPA-Motor) and HRQoL (EQ-5D: index and VAS) from PD patients and burden (Zarit’s CBI), anxiety (HADS-A), depression (HADS-D) and HRQoL from their caregivers. Linear mixed models were performed to assess the relation between caregivers’ burden with the patients’ motor symptoms and HRQoL. Results: 174 PD patients, mean age: 63±11, 50% male, PD duration: 8±6 years; 173 caregivers, 66.5% female, mean number of years taking care of patients: 6±5, were included. Patients’ SCOPA-Motor varied from 39.3±8.4 in year 1 to 42.5±9.5 in year 4; EQ-5D index and VAS from 0.7±0.3 and 63.7±20.3 to 0.6±0.3 and 60.3±17.9, respectively. Caregivers’ ZCBI varied from 17.6±12.6 to 20.0±14.8; EQ-5D index remained unchanged at 0.8±0.2, while VAS varied from 73.7±18.6 to 75.0±16.2; HADS-D scores changed between 4.0±3.7 and 4.4±4.3, and HADS-A from 6.7± 3.8 to 6.9±4.3. Linear mixed models revealed that caregivers’ anxiety and depression negatively influenced their own perception of disease burden [HADS-A, Estimated coefficient (EC): 1.02 (SE: 0.15), p<0.000; HADS-D, EC: 1.19 (SE: 0.17), p<0.000]. Patients’ motor symptoms predicted worse caregiver burden [SCOPA-Motor EC: 0.22 (SE: 0.09), p= 0.014]. Decline in patients HRQoL implied a higher caregiver burden [EQ-5D index EC: -5.56 (SE: 2.70), p< 0.041]. Conclusions: PD patients’ moderate motor symptoms and fair HRQoL worsened over time as did caregivers’ anxiety and disease A391 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 burden. Improving the control of PD motor symptoms on patients may contribute to prevent anxiety and HRQoL deterioration in caregivers. PND4 A Comprehensive Literature Review of the Burden of Gaucher Disease Nalysnyk L 1, Hamed A 1, Hurwitz G 1, Simeone J 2, Rotella P 2 1Genzyme, a Sanofi Company, Cambridge, MA, USA, 2Evidera, Lexington, MA, USA . . . . . Objectives: Gaucher disease (GD) is an inherited, rare, lysosomal storage disorder caused by a genetic deficiency of glucocerebrosidase. The result is the accumulation of the substrate, glucosylceramide, in the lysosomes of macrophage cells in the liver, spleen, bones, lungs, and other vital tissues. Three subtypes of Gaucher disease are recognized: type 1 (GD1, non-neuropathic), type 2 (GD2, acute neuropathic), and type 3 (GD3, subacute/chronic neuropathic). Clinical manifestations of the disease are multisystemic, clinically heterogeneous and require lifelong management. Methods: To better understand the burden of GD, a comprehensive review of the published literature was conducted. MEDLINE, EMBASE, CENTRAL and “grey” literature sources published in English between January 1990 and February 2013 were searched for relevant publications. Results: A total of 97 publications focusing on the epidemiologic, clinical, and socioeconomic burden of GD, treatment options and guidelines were summarized. The standardized incidence and prevalence of GD in the general population varies from 0.30 to 5.80 per 100,000 and 0.33 to 1.75 per 100,000, respectively, and GD1 is the predominant type in most regions. The risk of mortality is highest in GD patients younger than age 5 years and generally increased after age 55; the life expectancy is lower than the general population. Common manifestations of GD such as anemia, thrombocytopenia, splenomegaly, hepatomegaly and bone disease lead to a decreased quality of life. Reported GD comorbidities include Parkinson’s disease and cancer. Current treatment options consist of enzyme replacement therapy (ERT, standard of care) and substrate reduction therapy (SRT). ERT is the standard of care, though unmet needs still exist, especially for GD2 and GD3. Conclusions: GD is a rare, chronic disease associated with significant burden to patients and caregivers. While ERT is an effective and well-established treatment for GD patients, several unmet needs exist and further research is needed in this area. PND5 Restless Leg Syndrome Detection in Hemodialysis Castejón N 1, Arenas M D 2, Rebollo P 1, Sellés Galiana F 3, Delgado Conde P 2, Gil González M T 2, Gutierrez Rivas P 2, Reichert García J 2 1LASER ANALYTICA, Oviedo, Spain, 2Hospital Vithas Perpetuo Socorro, Alicante, Spain, 3Hospital General Universitario de Alicante, Alicante, Spain . . . . . . . . AD dementia. The effectiveness of this test was evaluated for diagnostic accuracy. Diagnostic accuracy for identifying AD by ELISA was high (pooled sensitivity, 0.772 (95% CI 0.747–0.796); pooled specificity, 0.732 (95% CI 0.699–0.762). Conclusions: Reduced CSF Ab1–42 levels are of potential utility in the differential diagnosis of AD versus non-AD dementias and healthy controls. PND7 Prevalence of Cystic Fibrosis Among the U.S. National Medicaid Population Xie L 1, Dysinger A H 1, Wang Y 1, Kariburyo M F 1, Baser O 2 Research, Ann Arbor, MI, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . . . . 1STATinMED Objectives: Cystic fibrosis (CF) prevalence according to U. S. geographic region as well as patient age, gender and race was examined in the U. S. Medicaid population for patients younger than age 45. Methods: Patients ≤ 45 years from the Medicaid fee-for-service (FFS) population (2008-2009) were identified using International Classification of Disease 9th Revision Clinical Modification (ICD-9-CM) diagnosis code 277.0x. Patients were required to have continuous Medicaid FFS enrollment in both years and no evidence of managed care enrollment. CF prevalence was stratified by U. S. region, state, age group, gender and race, and was measured by number and percentage of patients in each category. Results: A total of 2,142 patients were diagnosed with CF among the Medicaid FFS population under age 45 years in 2008 and 2009. Prevalence was the highest (0.17%) for patients under age 17 years, followed by those age 18-35 (0.14%), and 36-45 years (0.06%). However, some states had the highest CF prevalence in the 18-35 age range (Colorado: 0.92%; North Dakota: 0.50%; Kentucky: 0.54%). CF prevalence by race was also examined with the following results: White (0.17%), Hispanic (0.10%), Asian (0.07%), Black (0.06%) and Native American (0.03%). Male patients had a relatively higher prevalence than female patients (0.14% vs. 0.12%). The highest CF prevalence was observed in Colorado (0.47%), followed by Maryland (0.46%), North Dakota (0.31%), Ohio (0.28%) and Pennsylvania (0.27%). Patients residing in the Midwest U. S. region had the highest prevalence rate (0.15%), compared to the South (0.15%), Northeast (0.12%) and West (0.04%) regions. Conclusions: CF prevalence was the highest in patients age < 17 years nationwide, however, certain states showed the highest prevalence among patients age 18 to 35. White and male patients residing in the Midwest U. S. region were found to be at higher risk of a CF diagnosis. . . Objectives: Restless leg syndrome (RLS) is a condition with possibly high prevalence in hemodialysis (6-60% according to the literature), and a specific treatment available. Thus it’s important to identify it among other conditions present in this population that might confound diagnosis (such as peripheral vascular disease or neuropathies). An approach based on a self-completed screening test will be assessed in this study, along with an estimation of RLS prevalence in hemodialysis. Methods: Patients from two hemodialysis units answered a RLS screening test. Those with a positive screening completed the International Restless Legs Syndrome Study Group Rating Scale (IRLS) that assesses symptom severity. A neurophysiologist performed a clinical interview to confirm the diagnosis, including a supervised administration of the IRLS. Results: 164 patients were recruited. Mean age was 65.7 years (range 33-87; Pct25-75: 55.5-77.5), 67% were male and mean time in dialysis was 64.16 months. Self-completed screening test identified 69 possible cases of RLS (42.07%). 44 (26.8%) patients had RLS symptoms according to the self-completed IRLS, and 79% of them were classified as having moderate to severe RLS symptoms. The clinician confirmed just 22 of those cases (13.4% of the total sample), with a demographic profile similar to the sample. The screening test had in this sample a sensitivity of 100%, specificity 66.43% and positive predictive value 31.88%. Conclusions: This study found RLS is a relatively common condition in hemodialysis patients. The screening test showed a high sensitivity to detect RLS, but very low specificity, so the confirmation of an expert neurologist or neurophysiologist is neccesary. PND6 Cerebrospinal Fluid f”-Amyloid1-42 Levels in the Differential Diagnosis of Alzheimer’s Disease - Systematic Review and Meta-Analysis Mo J National Evidence-based Health Care Collaborating Agency & Inha University, SEOUL, South Korea . Objectives: The purpose of this study was to carry out systematic review of the literature and meta-analysis to evaluate the diagnostic utility of cerebrospinal fluid (CSF) levels of the 42 amino acid form of amyloid-b (Ab1−42) as a biomarker for differentiating Alzheimer’s disease (AD) from non-AD dementia. Methods: Design Systematic literature review was used to evaluate the effectiveness of the Ab for the diagnosis of Alzheimer’s disease. The Scottish Intercollegiate Guidelines Network (SIGN) tool was used by two evaluators to evaluate independently the quality of the 15 studies. Data sources - The literature review covered from October 27, 1946, to October 22, 2013, and searched eight domestic databases including Korea Med and international databases including Ovid-MEDLINE, EMBASE, and Cochrane Library. Eligibility criteria for selecting studies - Primary criteria for inclusion were valid studies on (i) patients with mild cognitive impairment with confirmed or suspected AD and non-AD dementia, and (ii) assessment of Ab1−42levels using appropriate comparative tests. Results: A total of 15 studies (15 diagnostic evaluation studies) were identified in which levels of CSF Ab1−42 were assessed. Meta-analysis was performed on nine robust studies that compared confirmed AD with healthy individuals (n = 1587), 10 studies that compared AD with non-AD dementias (n = 860), and four studies that compared a-MCI (amnestic mild cognitive impairment) with na-MCI (non-amnestic mild cognitive impairment) subjects (n = 857). Overall, Ab1–42levels were reduced in CSF from AD patients versus healthy controls or non- PND8 Risk of Relapse Among Propensity Score Matched Multiple Sclerosis Patients Receiving Natalizumab or Platform Therapy in the US Watson C 1, Bonafede M M 2, Johnson B H 2 Idec Inc., Weston, MA, USA, 2Truven Health Analytics, Cambridge, MA, USA . . . . . 1Biogen Objectives: To examine claims-based relapse rates and time to relapse among multiple sclerosis (MS) patients treated with natalizumab or propensity score matched patients treated with platform therapy (interferon beta/glatiramer acetate) in the US. Methods: The Truven Health MarketScan Research Databases were used to identify adults with a MS (ICD-9-CM code 340) diagnosis treated with natalizumab or platform therapy; the first claim between January 1, 2009 and April 1, 2012 was the index. Patients had to have one year continuous enrollment pre- and post-index and remain on index therapy for 12 months. Patients were excluded if they used a nonindex therapy in the pre-index. Natalizumab and platform patients were propensity scored matched using nearest neighbor matching on demographic characteristics, selected comorbid conditions and medications, MS severity (using an adaptation of Kurtzke’s Functional System), pre-index relapse and pre-index expenditures. MS-relapse was defined as MS-related inpatient (IP) admission, IV or oral corticosteroid use. Cox Proportional Hazard models were used to evaluate time to relapse, controlling for demographic and pre-index clinical characteristics. Results: A total of 897 natalizumab patients met the study criteria, 882 of which were 1: 1 matched to 882 platform therapy patients (mean age 45 years, 70% female) with a standardized difference < 10 on all matching measures. Compared to platform patients, natalizumab patients were significantly less likely to have MS-relapse post-index (26.5% vs. 35.5%, p< 0.001), with lower post-index rates of MS-related IP admissions (1.0% vs. 2.6%), IV-corticosteroid use (15.6% vs. 19.0%) and oral corticosteroid use (15.4% vs. 23.1%) (all p<0.001). Natalizumab patients also had 25 more relapse-free days (308 vs. 283 days, p< 0.001). Post-index MS-relapse risk was lower for natalizumab patients (HR= 0.69, p< 0.001) after controlling for baseline characteristics. Conclusions: Natalizumab was associated with a significantly lower risk and rate of MS-relapse and had longer time to a MS-relapse compared to platform therapy. PND9 Image-Guided Navigation Systems (IGNS) Improve Accuracy of Catheter Placement in Shunted Hydrocephalus Patients Annoni E 1, Joedicke H 1, Birinyi-Strachan L 2 1Medtronic International, Tolochenaz, Switzerland, 2Medtronic Australasia, Sydney, Australia . . . Background: The most common surgical complication associated with shunt placement in Hydrocephalus patients is obstruction causing shunt malfunction. The primary cause of obstruction is incorrect placement of the catheter tip, most notably in the choroid plexus. Objectives: To investigate the clinical and economic value of IGNS use in the accurate placement of catheters in Hydrocephalus patients. Methods: A search of the Embase and PubMed electronic databases was conducted to identify studies evaluating the accuracy, effectiveness, quality-of-life (QOL) and economic aspects of IGNS in patients with Hydrocephalus. No language restrictions were applied. Results: We conducted a meta-analysis of studies reporting accuracy of ventricular catheter placement in patients with hydrocephalus undergoing shunt placement with stereotactic IGNS versus freehand technique. The definition of accurate catheter placement was similar in all studies. The meta-analysis showed the odds of achieving an accurate catheter placement for surgeons who utilize IGNS was almost 6 times higher (odds ratio 5.55, 95% CI [2.84, 10.85], P< 0.00001) than surgeons who used freehand placement techniques. Furthermore, accurate place- A392 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 ment of the catheter tip was associated with fewer shunt failures, as demonstrated in 9 clinical studies investigating accurate catheter placement using the AxiEM™ IGNS (Medtronic Inc). In addition to being costly, studies showed shunt revision surgery was associated with significant morbidity and lower long-term QOL. In a study of 80 paediatric Hydrocephalus patients, investigators found that patients with a history of two or more shunt revision surgeries had a significantly worse QOL (p< 0.02), as measured by the Hydrocephalus Outcomes Questionnaire (HOQ). Conclusions: The use of IGNS significantly increases the accuracy of ventricular catheter placement compared to freehand techniques in hydrocephalus patients undergoing ventricular shunt insertion. Clinical studies have shown the use of IGNS in shunt placement surgery results in lower shunt failure rates, which improve QOL and lowers the economic impact to payers. PND10 Assessing the Comparative Outcomes from Teriflunomide and Dimethyl Fumarate Studies in Relapsing Ms: Use of “Number Needed to Treat” Analysis Freedman M S 1, Montalban X 2, Miller A E 3, Dive-Pouletty C 4, Leist T P 5 1University of Ottawa and the Ottawa Hospital Research Institute, Ottawa, ON, Canada, 2Vall d’Hebron University Hospital, Barcelona, Spain, 3Icahn School of Medicine at Mount Sinai, New York, NY, USA, 4Genzyme, a Sanofi company, Chilly-Mazarin, France, 5Thomas Jefferson University Hospital, Philadelphia, PA, USA . . . . . . . . Objectives: Teriflunomide and dimethyl fumarate (DMF), oral therapies for relapsing–remitting multiple sclerosis (RRMS), have demonstrated efficacy in clinical trials. Despite challenges in comparing outcomes across studies, exploratory analyses of treatment effects can be compared informally using relative reductions in a specific endpoint. However, these outcomes do not account for differences in disease severity among study populations or differences on very low event rates. The number needed to treat (NNT) to prevent an event is an important outcome to consider for any comparisons within the field of MS. Methods: NNTs were derived using data from studies with teriflunomide 14 mg (TEMSO, NCT00134563; TOWER, NCT00751881) or DMF (DEFINE, NCT00420212; CONFIRM, NCT00451451) based on inverse of absolute differences between treatment and placebo groups. Results: Teriflunomide studies included patients with progressive disease; patients in DEFINE had slightly lower Expanded Disability Status Scale scores. Teriflunomide and DMF significantly reduced risk of relapse (all studies). NNTs to prevent one relapse were similar across studies (5.9 [TEMSO], 5.6 [TOWER], 5.3 [DEFINE], 5.6 [CONFIRM]). Risk of disability progression sustained for 12 weeks was significantly reduced in TEMSO, TOWER, and DEFINE but not CONFIRM. Corresponding NNTs to prevent disability progression were 13.8, 17.4, 10.8, and 30.2. Risk of relapse leading to hospitalization was significantly reduced in TEMSO and TOWER but not in DEFINE and CONFIRM. Corresponding NNTs were lower in TEMSO (12.5) and TOWER (20) than in DEFINE (50) and CONFIRM (50). Conclusions: Using the NNT approach, we demonstrate a comparable effect size for teriflunomide and DMF on relapse. NNTs to prevent disability progression with teriflunomide showed a consistent significant reduction in risk versus placebo in both TEMSO and TOWER, whereas for DMF, comparable NNTs were observed only in DEFINE, and not in CONFIRM. Reduction of risk for relapse leading to hospitalization was significant only for teriflunomide. PND11 The Clinical Evidence Base of Treatment Options in Alzheimer’s Disease: A Systematic Literature Search Droeschel D 1, Kaier K 2, Walzer S 1 1MArS Market Access & Pricing Strategy GmbH, Weil am Rhein, Germany, 2University of Freiburg, Freiburg, Germany . . . Objectives: Alzheimer’s Disease (AD) destroys brain cells, causing problems with memory, thinking, and behavior severe enough to affect work, family and social relationships, and basic activities of daily living. AD gets worse over time, it is incurable and fatal. Donepezil, galantamine, rivastigmine and memantine are the current treatment options but the latest evidence was not systematically reviewed recently. Methods: PubMed, Health Technology Assessment Database, NHS Economic Evaluation Database, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, DAHTA-Datenbank, PSYNDEX and PsycINFOwere searched systematically for randomized-controlled studies. For the abstracts that met the pre-defined inclusion criteria, full text articles were obtained. The abstracts that did not meet the search criteria were excluded. Based on these manuscripts it was evaluated whether each study meets the selection criteria. Results: After elimination of duplicates the search indicated above yielded 418 articles of which another 299 were excluded based on the title selection; after abstract review, 82 articles have been reviewed in full text which were also deemed to be relevant based on the research question. For donepezil 24 RCTs were available for which another 4 subgroup and exploratory analyses have been published. For galantamine 11 RCTs with 6 exploratory analysis are available. For rivastigmine 10 RCTs are available with 7 exploratory papers. For memantine 14 RCTs with 6 exploratory analysis were found. Out of those studies eleven head-to-head studies are available; 5 studies comparing donepezil vs memantine, 3 studies comparing donepezil vs rivastigmine and one study each comparing donepezil vs galatamine, rivastagmine vs memantine. There was one study comparing rivastagmine vs donepezil vs galantamine. In comparison the clinical evidence seems diverse dependent on the patient characteristics, study duration, and severity of disease. Conclusions: Appropriate evidence assessment for the approved AD treatments requires clinical expertise and close review of the study characteristics. PND12 Evaluation of Disability Progression as an Endpoint in Clinical Trials for Relapsing-Remitting Multiple Sclerosis (RRMS): Comparison of the Define and Confirm Studies Viglietta V , O’Gorman J , Yang M , Zhang R , Raghupathi K Biogen Idec Inc., Cambridge, MA, USA . . . . . Objectives: Time to 12-week confirmed disability progression, measured by the Expanded Disability Status Scale (EDSS), is a key endpoint in RRMS trials. However, the EDSS has widely discussed limitations, and several therapies have shown inconsistent results for this endpoint in terms of statistical significance. Here we contextualize differences in 12-week confirmed disability progression results in the Phase 3 studies of gastro-resistant dimethyl fumarate (DMF), DEFINE and CONFIRM. Methods: Time to 12-week confirmed disability progression at 2 years was a secondary endpoint in both studies; however, the studies were not powered to detect statistical significance for this endpoint. Patients had the option of discontinuing study treatment and initiating alternative therapy at any time due to 12-week confirmed disability progression or after completing 48 weeks of study treatment and experiencing one confirmed relapse after 24 weeks (DEFINE) or two confirmed relapses at any time (CONFIRM). Results: Although gastro-resistant DMF 240mg BID demonstrated consistent reductions on 12-week confirmed disability progression, statistical significance was achieved in DEFINE (p= 0.0050) but not CONFIRM (p= 0.2536). There was an apparent difference in the placebo rate of 12-week confirmed disability progression at 2 years (DEFINE, 27%; CONFIRM, 17%). In CONFIRM, a relatively higher percentage of placebo patients (4.1%) versus gastro-resistant DMF patients (1.7%) switched to alternative MS therapy or withdrew after the time of tentative disability progression without a subsequent EDSS assessment. Additionally, a relatively higher percentage of placebo patients who switched to alternative MS therapy had ≥ 2 relapses without 12-week confirmed disability progression prior to switch in CONFIRM (45%) compared with DEFINE (16%). Conclusions: Relapse-based criteria for switching to alternative therapy may have contributed to the lower placebo progression rate and decreased assay sensitivity for this particular endpoint in CONFIRM. The totality of evidence needs to be taken into account when assessing a therapy’s effect on disability progression. PND13 A Real-World Assessment of Annual Multiple Sclerosis Prevalence and Disease-Modifying Drug Treatment Rates Using an Administrative Claims Database Phillips A L 1, Munsell M J 2, Menzin J 2, Dangond F 1, Locklear J C 1 Serono, Inc., Rockland, MA, USA, 2Boston Health Economics, Inc., Waltham, MA, USA . . . . . . . . 1EMD Objectives: To examine annual prevalence and treatment rates of multiple sclerosis (MS) patients using a large US commercial administrative claims database. Methods: Random sample of 5 million lives from the IMS LifeLink Plus database was used for this analysis. Individuals with ≥ 1 month eligibility were included in the denominator; those with ≥ 1 month eligibility and an MS diagnosis (ICD-9-CM: 340. xx) were included in the numerator. Presence of a disease-modifying drug (DMD) was defined as ≥ 1 claim during the calendar year of interest. Baseline demographics and clinical characteristics were evaluated for each group. Annual prevalence (per 10,000) and treatment rates were reported for each calendar year (2006–2012) and were further stratified by age and gender. Results: MS patients were older than patients without MS (mean age range 46.8–47.7 vs. 34.4-35.1, respectively) and more likely to be female (73% vs. 51%, respectively). Comorbidities such as gastrointestinal disorders (42.8%), hypertension (43.5%), arthritis (24.8%) and anxiety (22.8%) were common among MS patients (2006 estimates). MS prevalence ranged from 16.4/10,000 (2006) to 17.8/10,000 (2010). Similar patterns over the years were observed when data were stratified by sex and age, with absolute rates being higher among women vs. men (24.4/10,000 vs. 8.2/10,000, respectively; 2012; P< 0.001) and patients aged 45–64 years (29.4/10,000 vs. 0.4/, 15.6/ and 18.2/10,000 patients aged < 18, 18–44 and ≥ 65, respectively; P< 0.001). The proportion of MS patients receiving a DMD increased from 2006 (42.5%) to 2012 (51.2%) (P< 0.001). Similar rates and trends in the proportion of patients with a DMD were observed when stratified by gender (42.8% (2006) and 51.3% (2012) [female]; 41.5% (2006) and 50.7% (2012) [male]; both P< 0.001). Conclusions: In a recent 5-year period, MS prevalence in a large US insured population increased slightly, with a greater increase in the likelihood of DMD use. PND14 The Characteristics of Multiple Sclerosis in Iran Khanizadeh H 1, Nikkhah K 2, Izham M 3 SAINS MALAYSIA, penang, Malaysia, 2Mashhad University of Medical Sciences, Mashhad, Khorasan, Iran, 3Universiti Sains Malaysia, Pinang, Palau Pinang, Malaysia . . . 1UNIVERSITY Objectives: Multiple Sclerosis (MS) is a chronic disease of the Central Nerve System. The aim of this paper is to characterize the various clinical and demographical feature of the MS population. Methods: In a 6-month cross-sectional study248 patients were investigated in Khorasan provinces. Data was collected by employing a 32- item self-administered questionnaire in a face to face interview. Results: A total of 248 patients were recruited (186 female, 75%; 62 male, 25%). The mean age was 31.9±8.7, the mean onset age was 26.3 (26.3±8: 1) and the median duration of illness was 3.8 years. The prevalence and incidence were estimated to be 25/100000 and 2.5/100,000 respectively. Significantly more patients had a Relapsing Remitting MS course. Self reported character of MS individuals were significantly more (193, 77.8%) regarding nervous character (p value= .000). A family history of ms was reported in 11%. However, there was no a significant difference between men and women with respect to age, age of onset, BMI, disease duration and gap between clinical onset and diagnosis. The education level was reported as 154 (62%) had a bachelor and greater degree and 94 (38%) had a diploma or under-diploma degree. Thirty six percents of the patients were born in the spring. Conclusions: In contrast to reports from Caucasians, the Iranian differs with respect to age, age of onset of illness, disease duration, family history, sex ratio. The sex ratio of 3: 1 in this study is somewhat higher than the usual 2: 1 in the standard text and seen in Asia or the neighboring Arab countries. This might reflect the role of hormone or genetic factors or much more visiting by women, or women stressful life. These included BMI, birth season, education stand. The educated people in developing country are probably likely to adopt a “western” lifestyle, therefore more probably to get the risk of developing Western disease. A393 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PND15 Epidemiology of Multiple Sclerosis in Latin America: Critical Analysis of the Literature Einarson T R 1, Bereza B 1, Machado M 2 1University of Toronto, Toronto, ON, Canada, 2Biogen Idec, São Paulo, Brazil . . . . Objectives: To summarize rates of prevalence, incidence and types of multiple sclerosis (MS) in Latin America. Methods: We searched Medline, Embase, Scielo and LILACS using key words “multiple sclerosis” and “esclerosis múltiple”, plus “Latin America” and all country names. Accepted were full articles or abstracts reporting original research in any language at any time, with MS diagnosed using any acceptable criteria. Clinically isolated syndrome was excluded. Results: 1482 articles were identified and 203 reviewed in full; 88 were rejected and 115 were analyzed (38 prevalence, 3 incidence, 2 both, 72 clinical epidemiology), including 68 (59%) full text and 47 (41%) abstracts. Studies originated in 14 countries, mostly larger, more affluent nations (47% Brazil, 15% Argentina, 8% Mexico). Prevalence studies examined 19,357 MS cases from 1968-2012; rates ranged from 0 in a group of Mexican natives to 52/100,000 in Puerto Rico, with a mean of 13.9/100,000. This range would be classified as low to medium, with 83,000 (CI95%: 62,000-104,000) prevalent cases in Latin America. Incidences ranged from 0.15/100,000 personyears in Panama to 1.76 in Argentina (median= 1.32, mean= 1.35), or 7,887-8,365 new cases/year in Latin America. Authors noted increasing prevalence and incidence rates over time. Clinical epidemiology MS data were obtained from 94 studies; 66% used Poser criteria, 48% McDonald (various versions), and 14% other, from 19,893 patients (72% females). The average age at assessment was 38.2 and 32.2 at disease onset with average EDSS of 3.1. The relapsing-remitting form was most prevalent (74% of cases), followed by secondary progressive (12%) and primary progressive (10%). Conclusions: MS prevalence and incidence rates vary widely in Latin America in a low to medium range, but are increasing as reported in the reviewed studies. Information is scant with many gaps. More research is needed to provide a basis for decision making and budget allocation. PND16 Entacapone did not Increase Prostate Cancer Risk or Mortality in Patients with Parkinson’s Disease Korhonen P 1, Kuoppamäki M 2, Prami T 1, Hoti F 1, Christopher S 1, Ellmén J 2, Aho V 2, Vahteristo M 2, Pukkala E 3, Haukka J 1 Research, Espoo, Finland, 2Orion Corporation, Espoo, Finland, 3Finnish Cancer Registry, Helsinki, Finland . . . . . . . . . . 1EPID Objectives: The association between Parkinson’s disease (PD) and prostate cancer (PCA) is under discussion. We investigated whether exposure to entacapone increases risk of PCA by comparing PCA incidence and mortality rates in PD patients treated with levodopa and dopa decarboxylase inhibitors (DDCIs) with or without entacapone as an adjuvant. Methods: This was a retrospective population-based register cohort study in Finland. Male PD patients exposed to levodopa/DDCI with entacapone and patients on levodopa/DDCI without entacapone between 1998 and 2009 were identified from the Finnish Prescription Register. Concomitant treatment with dopamine agonists (DA) and monoamine oxidase B (MAO-B) inhibitors was allowed. PCA cases and PCA deaths were identified from the Finnish Cancer Register and from the Cause of Death Register. PCA Incidence and mortality of PCA were modeled by Cox’s proportional hazards model with adjustments for relevant baseline and time-dependent variables. The robustness of the findings for PCA incidence was evaluated in sensitivity analyses with alternative definitions for exposure and by using a nested case-control study setting with 4-5 individually matched controls randomly selected for each incident prostate cancer case within the cohort. Results: The study cohort consisted of 11 396 male patients. During a mean follow-up time of 4.7 years 359 PCA cases and 89 PCA deaths were observed. Entacapone use with levodopa/DDCI did not increase the risk of PCA incidence [hazard ratio (HR) 1.05, 95% CI 0.76-1.44] or PCA mortality (HR 0.93,95% CI 0.43-1.98) when compared to levodopa/DDCI without entacapone. For cumulative entacapone use (exposure > 360 days vs. never use) the HR was 0.82 (95% CI 0.56-1.18) for PCA incidence and 1.27 (95% CI 0.60-2.72) for PCA mortality. The results for PCA incidence were found robust in the sensitivity analyses. Conclusions: No increased risk in PCA incidence or PCA mortality was seen in relation to entacapone use. Neurological Disorders – Cost Studies PND17 The Clinical and Economic Value of Antibiotic-Impregnated Shunt Catheters (AISC) in the Treatment of Hydrocephalus Annoni E 1, Joedicke H 1, Birinyi-Strachan L 2 1Medtronic International, Tolochenaz, Switzerland, 2Medtronic Australasia, Sydney, Australia . . . Background: Shunt infections are one of the most serious and costly complications associated with shunt therapy. They can have long-term consequences to the patient and impose significant burden on the family and health care system. Objectives: To investigate the clinical and economic value of AISC in the treatment of Hydrocephalus. Methods: A search of the Embase, PubMed, and NHS EED electronic databases was conducted to identify studies evaluating the effectiveness, efficacy and economic aspects of AISC’s in patients with Hydrocephalus. No language restrictions were applied. Results: Twenty-three clinical and 9 economic studies were included for review. The majority of published studies using AISC in the treatment of hydrocephalus demonstrated a statistically significant reduction in shunt-associated infections when AISC vs. standard shunt (SS) catheters were used. One meta-analysis of 14 studies that compared patients implanted with AISC to SS catheters showed AISC to be protective against shunt infection in both paediatric and adult Hydrocephalus patients. The pooled infection rate decreased from 7.0% in the SS catheter group to 3.5% in the AISC group. Economic studies conducted globally reported significant cost savings associated with using AISC. One study in paediatric Hydrocephalus patients found that ongoing management of shunt infections uses a disproportionate amount of hospital expenditure (71% of total management costs or € 57,650 per patient) when compared to shunt revision surgery (29% of total costs or € 6,720 per patient). Using shunt implant data from the UK Hospital Episode Statistics (HES), procedures and interventions database, we estimated the potential cost savings to the UK Health Care system in 2012-13 was £1,461,207 if all new lumboperitoneal or ventriculoperitoneal shunt surgeries were performed using AISC systems. Conclusions: Current clinical evidence has demonstrated AISC reduce infection rates in shunted Hydrocephalus patients. The potential health care savings associated with reduced infection rates are significant. PND18 Budget Impact Analysis of Fingolimod in Relapsing Remitting Multiple Sclerosis Ruggeri M 1, D’Ausilio A 2, Lo Muto R 3, Cottone S 4, Ghezzi A 5, Mecozzi A 6, Sacchini D 7, Mangone M 8 1Università Cattolica del Sacro Cuore, Rome, Italy, 2Creativ Ceutical, Milano, Italy, 3CreativCeutical, Milan, Italy, 4Azienda Ospedaliera Ospedali Riuniti Villa Sofia- Cervello, Palermo, Italy, 5Azienda Ospedaliera di Gallarate, Gallarate, Italy, 6Regione Lazio, Roma, Italy, 7Università Cattolica del Sacro Cuore, ROMA, Italy, 8Novartis Farma S.p.A., Origgio, Italy . . . . . . . . Objectives: In recent years, fingolimod and natalizumab - though expensive - have brought significant improvements in second-line treatment of Multiple Sclerosis (MS). In Italy, fingolimod was acknowledged in 2011 the therapeutic innovation criterion which exempts the manufacturer from a mandatory discount of 5% on the ex-factory price for a three-year period. Since the innovation period is soon expiring and in order to assess the affordability of this therapy for the National and Regional Healthcare Systems (HSs), we performed a Budget Impact Analysis (BIA) of fingolimod in second-line treatment of MS, enclosed an ethical assessment of justice issues. Methods: A model in MSExcel was built to perform the BIA on five Regions located in the Nord, Centre and South of Italy. Interviews with regional stakeholders/clinicians provided information about epidemiology, market share, patient journey and resource consumption in the respective regions. Costs were calculated using regional tariffs for outpatient/inpatient health care services. The time horizon is three years which is the timeframe mostly requested by Healthcare Authorities. Results: Preliminary results conducted for Lazio and Sicily show that, in second line, the use of fingolimod compared to natalizumab leads to savings of € 2,500 per patient in Lazio and € 2,700 in Sicily due to lower consumption of health care resources (different drug administration and patient management). The BIA shows that fingolimod can determine savings up to € 3.4 million approximately in Lazio and € 2.2 million in Sicily. Furthermore, if patients with frequent relapses in first line were early switched to second line, savings could be even higher: an additional € 2.0 million for Lazio and € 0.6 million for Sicily in the first year of treatment. Conclusions: The present analysis shows that, despite innovation expiry, spending remains sustainable for fingolimod in the Italian and Regional HSs. Analyses on the other Regions must be performed to confirm such results. PND19 The Cost Burden of Switching Patients with Relapsing-Remitting Multiple Sclerosis from Glatiramer Acetate To Newly-Approved Disease Modifying Therapies Blackney M 1, Kelly M 2, Zeidman R 2, Andreykiv M 3, Plich A 4 Inc., London, UK, 2Covance, Inc., London, UK, 3Teva Pharmaceuticals Europe B. V., Amsterdam, The Netherlands,, 4Teva Pharmaceuticals Europe B. V, Amsterdam, The Netherlands . . . . . 1Covance Objectives: Relapsing-remitting multiple sclerosis (RRMS), the most common form of MS, imposes a considerable cost burden on health care systems. Glatiramer acetate 20mg/mL, a once-daily injectable disease modifying therapy (DMT), has long-term real-world efficacy (more than two million patient/years exposure) for the treatment of RRMS. An economic model was developed to estimate the financial impact on the UK health care system of switching RRMS treatments from glatiramer acetate to emerging oral (teriflunomide and dimethyl fumarate) and infused (alemtuzumab) DMTs over five years. Methods: The eligible patient population in the model was estimated based on RRMS incidence rates, and the proportion of people receiving glatiramer acetate in the UK in 2013 (based on market share data). Medication costs were derived from British National Formulary list prices, while treatment administration, initiation and monitoring costs were calculated from UK health care unit and hospital tariff codes taking account of requirements specified in Summary of Product Characteristics documentation. Results: The model estimated that 5,900 people with RRMS were receiving glatiramer acetate in the UK in 2013, corresponding to 17.8% of those currently receiving DMTs. Assuming progressive switching within the eligible RRMS population from glatiramer acetate to teriflunomide (from 1–6% in year one to five), dimethyl fumarate (1–6%) and alemtuzumab (1–3%), additional spending of £17.0 million, £65.3 million and £147.6 million over one, three and five years, respectively, was incurred. This was primarily driven by higher acquisition costs compared with glatiramer acetate. Administration requirements associated with alemtuzumab, and the initiation and monitoring requirements of the oral DMTs also contributed to additional costs. Conclusions: No data exist supporting the clinical superiority – and therefore associated economic benefit – of the emerging DMTs over glatiramer acetate; switching patients from glatiramer acetate to new therapies may be associated with substantial costs for the UK health care system. PND20 Improving Clinical Outcomes and Health Care Resources Utilization in Multiple Sclerosis: A Portuguese Hospital Perspective Viriato D , Carrasco J , Fonseca J, Pacheco R Novartis, Porto Salvo, Portugal . . . Objectives: In Portugal there are about 2,694 patients treated with interferon beta (IFNB) and it is estimated that 30.4% are poor responders. Fingolimod treatment costs are an obstacle to treatment escalation, despite the fact that early initiation of fingolimod may impact favorably on long-term clinical outcomes. A394 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 This analysis aimed to assess if the early switch from IFNB to fingolimod impacts MS clinical outcomes and promotes better resource utilization in a Portuguese hospital perspective. Methods: This analysis was based on TRANSFORMS phase III trial extension data. A cost-effectiveness model was developed to calculate the cost per relapse avoided with 4.5 years of continuous treatment with fingolimod (early treatment) versus 1 year of treatment with IFNB followed by a 3.5 years of treatment with fingolimod (delayed treatment). A Portuguese hospital perspective was adopted addressing only direct costs: drug, monitoring and relapses’ treatment. Drug costs were based on Portuguese list prices, while the unit cost of each complication was obtained from the Diagnosis Related Groups tariff. The costs of relapses were derived from the Portuguese literature. Results: Assuming there are 819 patients treated with IFNB that are poor responders, the early treatment with fingolimod resulted in more relapses avoided when compared with delayed treatment with fingolimod (2,211 versus 1,843). The early treatment with fingolimod led to an increase of drug acquisition costs, but reduced costs associated to monitoring and relapses’ treatment. The total costs were 86,380,820€ for early treatment versus 79,257,091€ for delayed treatment. This represents an average incremental investment of 1,933€ per patient per year. The early strategy resulted an incremental cost effectiveness ratio of 19,358€ per relapse avoided when compared with the delayed strategy. Conclusions: Under the Portuguese hospital perspective, early treatment with fingolimod is expected to result in better clinical outcomes associated with a more efficient health care resources allocation. Embase, Scielo and LILACS using the key words “multiple sclerosis” and “esclerosis múltiple” plus “Latin America” and all country names. Full articles or abstracts from meetings reporting original research on cost or economic analyses, budget impact or resource utilization were obtained. No restrictions were placed on publication date or language. All work was done in duplicate by two independent reviewers with adjudication by consensus discussion. Results: We identified 1482 papers, of which 27 were considered for analysis. There were 7 economic analyses (5 costeffectiveness, 2 cost-utility), 5 budget impact analyses, 10 cost analyses (6 drug expenditures and 4 cost of illness), 4 on resource utilization and 1 on productivity loss. Studies were obtained from 5 countries (18 Brazil, 3 Argentina, 3 Colombia, 2 Mexico, 1 Chile). Mostly (22/27, 81%) were published as abstracts; 5 were published as full text articles (19%). Dates for these publications ranged from 2002 to 2013, with an exponential increase over time. The number of MS patients is increasing rapidly (71% increase in Brazil between 2006 and 2009). However, hospitalization rates (overall and per patient) have been decreasing, as newer more effective drugs have been increasingly used. Disease modulating therapies are predominantly used. Costs of care are quite high and have risen dramatically, e. g. > 200% in Brazil between 2007-2012, with beta-interferons mostly used (78%). Some high cost drugs such as fingolimod and natalizumab have been found cost-effective over older drugs such as beta-interferons or glatiramer acetate in Mexico, Brazil and Colombia, with modest impact on budgets. Conclusions: Very little evidence related to cost of MS has been produced in Latin America. More research is needed to better support decisions regarding care of MS patients. PND21 Cost Analysis of two Aftercare Strategies in Chronic Continuous Intrathecal Baclofen Therapy in Patients with Intractable Spasticity PND24 Alzheimer’s Disease: Medication Costs and Impact of Generic Substitution Burgers L T 1, Goslinga-van der Gaag S M E2, Delhaas E M 2, Redekop W K 3 University Rotterdam, Rotterdam, The Netherlands, 2Erasmus Medical Center, Rotterdam, The Netherlands, 3Erasmus University, Rotterdam, The Netherlands . . . . . . . . 1Erasmus Objectives: Intrathecal baclofen (ITB) therapy is indicated for use in the management of intractable spasticity. Patients treated with ITB are required to receive a pump refill at least once every three months in the hospital (standard care (SC)). Since SC can be very burdensome for both patients and informal caregivers, an alternative approach (Care4homecare) has been developed which enables patients to receive pump refills at home. Moreover, the use of specially trained nurse practitioners ensures that there is no reduction in effectiveness. We compared the costs of both strategies. Methods: Resource use in both strategies was estimated using observational data of 38 adult patients with spasticity (due to e. g. multiple sclerosis or spinal cord injury) that are currently living at home. We then combined this data with expert opinion and the Dutch costing manual to estimate the total one-year costs from a societal perspective. Results: Patients included in the analysis had on average an age of 52±14.4 years, 50% was men and patients scored on average 44±12.5 points on the Care Dependency Scale. The Care4homecare strategy involves care that is almost identical to SC and therefore can result in comparable direct medical costs. However, patients receiving Care4homecare do not incur any travel costs compared with SC patients (€ 489). In addition, the productivity costs of informal caregivers (SC € 195; Care4homecare € 40) and of patients treated with Care4homecare are less than the costs of patients receiving SC. From a societal perspective, the total costs of Care4homecare can be lower than that of SC. Conclusions: Care4homecare is an alternative approach to treat patients with intrathecal baclofen that can be cost-neutral from a health care sector perspective and cost-saving from a societal perspective. Moreover, it can be a welcome option for many patients and caregivers who want to avoid the burden of regular hospital visits. PND22 Cost Analysis of the Use of Glatiramer Acetate Compared to Interferon- in Patients with Relapsing-Remitting Multiple Sclerosis and Spasticity in Spain R 1, Garcia L 2, Meca J3 Sanchez de la Rosa Bujalance Lallana 1Teva Pharma, Madrid, Spain, 2Teva Pharmaceuticals, Madrid, Spain, 3Virgen de la Arrixaca Hospital, Murcia, Spain . . . Objectives: To analyze the costs associated with first-line use of glatiramer acetate (GA) compared to interferon-B (INF-β ) in patients with relapsing-remitting multiple sclerosis (RRMS) and spasticity from the perspective of the National Health System of Spain. Methods: A cost analysis of treatment and spasticity management with INF-β compared to GA for 6 months were analyzed. The clinical data were taken from the ESCALA study, which showed an improvement in spasticity in terms of spasm frequencies, muscle tone, and pain 3 and 6 months after the start of GA therapy. Unit costs for the resources used were taken from the BOTPLUS 2.0 database and available literature. The cost analysis is expressed in euros as of 2014, and a price discount of 7.5% was applied as set forth in Spanish Royal Decree 8/2010. Results: The costs associated with the management of RRMS, spasticity, and relapses using GA and INF-β were € 4,671.31 and € 7,078.02, respectively, generating a cost savings of € 2,406.72/patient, in favour of GA. Conclusions: The use of AG in the firstline treatment of patients with RRMS not only improves spasticity but it could be a strategy that offers savings cost after 6 months from the start of treatment. To initiate the treatment with AG and keep it in patients with optimal response would be a more efficient treatment option than INF-β . PND23 Systematic Review of the Economics of Multiple Sclerosis in Latin America Einarson T R 1, Bereza B 1, Machado M 2 1University of Toronto, Toronto, ON, Canada, 2Biogen Idec, São Paulo, Brazil . . . . Objectives: To summarize published articles dealing with economic issues related to multiple sclerosis (MS) in Latin America. Methods: We searched Medline, Truter I Nelson Mandela Metropolitan University, Port Elizabeth, South Africa . Objectives: Alzheimer’s disease has a substantial economic impact on patients, their caregivers and society. There are four cognitive enhancers commonly used in the treatment of Alzheimer’s disease: three cholinesterase inhibitors (donepezil, rivastigmine and galantamine) and one NMDA receptor antagonist (memantine). Studies have indicated that the cost of cholinesterase treatment may be offset by savings in other health care costs. Methods: The cost of medication on the South African market for Alzheimer’s disease was analysed using June 2014 retail prices with the Defined Daily Dose (DDD) as unit cost indicator. A retrospective drug utilisation study was conducted on prescription data of a medical insurance scheme administrator for 2012. Results: The cost per DDD for memantine was R26.20 (20 mg, two 10 mg tablets). For rivastigmine, the cost was R41.02 per DDD and for galantamine R27.72 per DDD (using the most convenient dosage strengths). These three products were all originator products. For donepezil, the originator and three branded generics were available. The cost of the originator was R27.86 per DDD, and for all three generics R16.29 per DDD. Only 32 patients were included in the drug utilisation study since not all medical aids reimburse these products. The average age of patients was 74.17 (SD= 9.54) years, with 50% females. Only memantine and donepezil were prescribed. Donepezil accounted for 77.48% of prescriptions (of which 60.93% were generic prescriptions). The average Prescribed Daily Doses (PDDs) were 16.30 (SD=4.92) mg for memantine and 8.73 (SD=2.84) mg for donepezil. The most frequent PDDs for memantine was 20 mg (62.96% of prescriptions) and 10 mg (37.04% of prescriptions), and for donepezil 10 mg (65.96% of prescriptions) and 5 mg (29.79% of prescriptions). Conclusions: More South African studies on Alzheimer’s disease treatment cost are needed that include the stage of the disease and adherence to treatment. PND25 Costs Associated with the Use of Enzyme-Inducing Anti-Epileptic Drugs Versus Non-Enzyme-Inducing Anti-Epileptic Drugs: A Systematic Review Xiong T 1, Gallagher E 2, MacGilchrist K S 1, Thieffry S 2 1Abacus International, Oxfordshire, UK, 2UCB Pharma, Brussels, Belgium . . . . . Objectives: Several commonly prescribed enzyme-inducing anti-epileptic drugs (EIAEDs) stimulate the synthesis of some hepatic enzymes responsible for drug metabolism. This synthesis can lead to complications by altering endogenous metabolic pathways or by affecting the elimination of concomitant drugs thus increasing health care costs. This study aimed to systematically review published estimates of direct and indirect costs associated with the use of EIAEDs compared with non-enzyme-inducing anti-epileptic drugs (nEIAEDs) in patients with focal and generalised seizures, and to evaluate methodological differences between the studies. Methods: Comprehensive electronic searches were undertaken using MEDLINE, EMBASE, Cochrane Library, EconLit, relevant conference proceedings and cost effectiveness analysis registries. All studies reporting any direct and indirect costs of AEDs for the treatment of patients with epileptic seizures were included. Study quality assessment was performed for every included study using a predesigned check list. Results: Thirty-seven full-length articles and two abstracts reporting costs were reviewed. Two studies reported AED costs, drug-specific adverse event costs and non-drug health care costs subsequent to the initiation of each individual AED (medical visits, MRI scans, etc.). Six studies reported specific AED costs and the overall subsequent non-drug health care cost without stratification by event. Eighteen studies reported AED acquisition costs but did not report any other subsequent AED-related health care costs stratified by treatment. Thirteen studies reported the whole cost of illness with only a list of AEDs included. To date, no study has been specifically designed to compare the total costs between EIAED and nEIAED use, although some studies compared direct and indirect costs of several newer AEDs versus older AEDs. Conclusions: Insufficient data and heterogeneity in methodology prevent valid comparisons being made between the total cost of EIAEDs and nEIAEDs. More research is required to identify if meaningful differences in the total cost of treatment exist between EIAEDs and nEIAEDs. A395 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PND26 Pharmacoeconomic Study of Botulinium Toxin Type a in Treatment of Post-Stroke Spasticity in the Russian Federation: Cost-Effectiveness Analysis Yagudina R , Kulikov A , Ugrekhelidze D I.M. Sechenov First Moscow State Medical University, Moscow, Russia . . . Objectives: To assess the cost-effectiveness of abobotulinumtoxinA, onabotulinumtoxinA, incobotulinumtoxinA and local standard therapy (oral muscle relaxers) in patients with post-stroke spasticity in Russia for 1-year period. Physical therapy was used in all therapy schemes. Methods: A decision tree was used to simulate the effects of abobotulinumtoxinA, onabotulinumtoxinA, incobotulinumtoxinA and standard therapy. The data on drugs efficacy (measured as decrease in the Modified Ashworth scale (MAS) score) was obtained from available clinical trials [1-3]. The following costs were taken into account:, the costs of BTA and other drugs, costs of inpatient and outpatient care in the Russian Federation, costs of adverse events, disability pensions, GDP loss due to post-stroke spasticity. Costs of BTA and other drugs were taken from the essential drug list and the database of drugs prices. Medical care costs were estimated from the Standard of treatment of stroke consequences developed by Ministry of Health of the Russian Federation. Costs of adverse events were calculated basing on Russian clinical guidelines and database of drugs prices. Disability pensions were taken from Russian Pension Fund database. GDP loss was based on the GDP information from World Bank. Cost-effectiveness ratio (CER) of BTA and standard therapy was calculated and compared in four treatment schemes. Results: Therapy with abobotulinumtoxinA showed most prominent decrease of Modified Ashworth score equal to 1,67, as for onabotulinumtoxinA – 1,17, incobotulinumtoxinA – 0,87, standard therapy – 0,67. The calculated CER in USD per 1 spasticity decrease point according to MAS was lowest for abobotulinumtoxinA (389524 RUB/11356 $) in comparison with onabotulinumtoxinA (635631 RUB/18532 $); incobotulinumtoxinA (798750 RUB/23287 $) and standard therapy (873312 RUB/25461 $). Conclusions: Therapy conversion to abobotulinumtoxinA comparing with standard therapy, onabotulinumtoxinA and incobotulinumtoxinA was associated with decrease of spasticity. Transfer to abobotulinumtoxinA is considered cost-effective in patients with post-stroke spasticity, given a cost-effectiveness ratio 389524 RUB/11356 $. PND27 Cost Comparison of Deep Drain Stimulation (DBS) and Continued Subcutaneous Apomorphine Infusion (CSAI) in Patients with Advanced Parkinson’s Disease Walleser Autiero S 1, Eggington S 1, Valyi A 2 1Medtronic International Trading Sarl, Tolochenaz, Switzerland, 2Medtronic Ltd UK & Ireland, Watford, UK . . . Objectives: Deep Brain Stimulation (DBS) for the treatment of advanced Parkinson’s Disease (PD) is a therapy supported with high level evidence, however no direct comparative studies exist of DBS against other therapy options, such as continued subcutaneous apomorphine infusion (CSAI) exists. The objective of this study was to evaluate the 5-year cost profiles of two therapies for advanced PD, DBS and CSAI, from a UK payer perspective. Methods: A Markov model, previously used to model cost-effectiveness of DBS+BMT vs BMT alone (Eggington 2013), served to evaluate the cost profile of DBS and CSAI over five-years. Equal efficacy of the two therapies was assumed. The cost analysis covered: device acquisition, implantation, adverse event management, concomitant drug use, device replacements and follow-up. Cost data were taken from UK national tariffs, combined with device/ drug price lists and data from previous economic studies of interventions for PD. Disease-related inputs were based on recent studies of DBS and CSAI in patients with advanced PD, plus long-term data from the literature. Costs were discounted at 3.5% per annum. Results: Total discounted costs over 5 years were £69,566 and £80,843 for DBS and CSAI, respectively, leading to cost savings of £11,277 of DBS compared to CSAI over 5 years. DBS is cost saving compared to CSAI from 3 years onwards, with the initial costs of DBS device acquisition shown to be offset by the on-going provision of CSAI. Conclusions: The results indicate that DBS requires less health care resources than CSAI over five-years. Comparative clinical data are needed to formally assess the relative cost-effectiveness of the two interventions. PND28 Retrospective Analysis of the Economic Burden of U. S. Long-Term Care Facility Residents Diagnosed with Parkinson’s Disease Huang A 1, Shrestha S 1, Baser O 2, Wang L 1 1STATinMED Research, Plano, TX, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . Objectives: To examine the health care utilizations and costs of long-term care facility patients diagnosed with Parkinson’s disease (PD). Methods: Patients diagnosed with PD (International Classification of Disease, 9thRevision, Clinical Modification [ICD-9-CM] diagnosis code 332) were identified using the Minimum Data Set (MDS) linked to 5% Medicare data from 01JAN2009 through 31DEC 2010. The initial diagnosis date was designated as the index date. A comparison cohort was created for patients without a PD diagnosis, using 1: 1 propensity score matching (PSM) to control for age, region, gender, index year and baseline Charlson Comorbidity Index score. The index date for the comparison group was randomly chosen to reduce selection bias. Patients in both cohorts were required to be at least age 65 years, have at least two consecutive quarterly assessments in MDS data in the 6 months prior to the index date, and have continuous medical and pharmacy benefits 1 year before and after index date. Study outcomes, (health care costs and utilizations) were compared between the disease and comparator cohorts, based on the matched sample. Results: After applying PSM, a total of 986 patients were included in each group (diseased and comparator cohorts), and baseline characteristics were balanced. A higher percentage of patients with PD had inpatient admissions (35.09% vs. 30.32%, p= 0.02), outpatient visits (93.91% vs. 89.45%, p< 0.001) and durable medical equipment (DME) utilization (27.69% vs. 21.91%, p< 0.01), compared to those without a PD diagnosis. The PD cohort also incurred significantly higher skilled nursing facility ($6,458 vs. $5,182, p= 0.03), DME ($344 vs. $206, p< 0.01) and pharmacy costs ($6,025 vs. $4,998, p< 0.0001) compared to the comparison cohort. Conclusions: Study results suggest that patients diagnosed with PD incurred significantly higher costs and had higher resource utilization than those without a PD diagnosis. PND29 Relationship Between the Direct Medical Costs and Direct Non-Medical Costs of Parkinson’s Disease According to Disease Severity During 4 Years of Follow-Up in Spain Martinez P 1, Rodriguez-Blazquez C 1, Paz S 2, Lizán L 3, Forjaz M J 4, Frades B 5, Moreno R 6 Center for Epidemiology and CIBERNED, Carlos III Institute of Health, Madrid, Spain, 2Outcomes’10, Castellon, Spain, 3Outcomes 10, Castellon, Spain, 4National School of Public Health and REDISSEC, Carlos III Institute of Health, Madrid, Spain, 5Alzheimer Center Reina Sofia Foundation, Carlos III Institute of Health, Madrid, Spain, 6Infanta Cristina Hospital, Madrid, Spain . . . . . . . . 1National Objectives: To describe the medical and non-medical direct costs of PD in relation to the total direct cost and its variation with disease severity during 4 years of follow-up. Methods: A descriptive, observational, longitudinal study in PD patients belonging to the ELEP study (2007-2010). Data on disease severity and use of resources were collected for three consecutive months, yearly, for 4 years. Direct medical (funded medical equipment, medications and medical assistance) and direct non-medical costs (alternative care, home assistance, non-funded medical equipment and medications) for 4 years depending on severity by Hoehn and Yahr (HY) was described. Costs were estimated by multiplying rates obtained from the database Oblikue (http://www.oblikue.com) and pharmacy costs from the BotPlus Web (https://botplusweb.portalfarma.com) by the number of resources used, updated to Spanish € , 2012. Results: 198 patients were included. Average age: 63±11 years, 50% male, mean PD duration of 8±6 years. Mild (HYI-II) and moderate (HYIII) PD varied from 76% and 21% to 64% and 29%, respectively during follow-up. Total direct cost was higher in severe stages, being in year 4, € 1,477.32 HYI (95%CI: 219.55-2,735.10) and € 3,606.66 HYIV (95%CI: 893.97-6,319.35) compared to year 1, € 1,093.32 HYI (95%CI: 624.99-1,561.65) to € 2,656.27 HYIV (95%CI: -53.14-5,365.68). Direct medical costs ranged from € 886.62 (95%CI: 475.13-1,298.11) HYI and € 2,376.30 HYIV (95%CI: -53.12-4,805.73) in year 1 to € 909.96 HYI (95%CI: 942.43-1,327.49) and € 2,768.49 HYIV (95%CI: 34.21-5,502.76) at year 4. Direct non-medical cost variation were determined by PD temporal evolution, increasing between year 1 and 4 within each stage, € 723.47 to € 4,255.20 HYI and € 653.27 to € 1,676.35 HYIV. Conclusions: The economic burden of PD rises with duration and severity of the disease, progressively increasing the direct, medical and non-medical costs. Besides to improve patients’ HRQoL, therapies aimed at controlling the symptoms severity will favor a more efficient management of the disease. PND30 Does Current Portuguese Financing Model for Multiple Sclerosis Covers for Estimated Needs? Montóia B 1, Guerreiro T 1, Viana R 2 1Novartis Farma – Produtos Farmaceuticos S. A., Porto Salvo, Portugal, 2Novartis Farma-Produtos Farmacêuticos S. A., Portugal, Porto Salvo, Portugal . . . Objectives: Portuguese financing model set for Multiple Sclerosis (MS) establishes a comprehensive price of 1.031,65€ per remitting patient/month. This study aimed to estimate global and individual direct costs of MS in Portugal and evaluate the coverage of MS Financing model for Relapsing-Remiting Multiple Sclerosis (RRMS) patients. Methods: A cost-of-illness (COI) approach was taken to assess direct costs/per RRMS patient by EDSS level, both for remission and relapse status using Delphi methodology for clinical practice measurement. Portuguese epidemiological data available for prevalence and relative distribution by EDSS level was used to estimate global MS costs and an average cost per RRMS patient. Results: The global cost for RRMS in Portugal is estimated to be approximately 83M€ /year for patients treated with 1st line Disease Modifying Therapies (DMT). Relapse treatment costs account for 8M€ . Direct costs estimated per disability level show that disability levels are directly related with expenditure in RRMS patients in remission (EDSS≤ 3: 10.754,90€ ; 3,5≤ EDSS≤ 4,5: 20.113,99€ ; 5≤ EDSS≤ 6: 21.170,36€ ; EDSS≥ 6,5: 24.945,94€ ). The same was observed for relapse cost (EDSS≤3: 4.952,23€; 3,5≤EDSS≤4,5: 5.568,49€; 5≤ EDSS≤ 6: 8.801,03€ ; EDSS≥ 6,5: 9.265,31€ ). An average cost of 14.563,48€ per RRMS patient was estimated based on Portuguese EDSS epidemiological distribution. Conclusions: Current comprehensive financing model per patient does not cover the costs associated with an average RRMS patient. We demonstrated that considering the epidemiological distribution per EDSS level there is a gap of 2.183,68€ (14.563,48€ vs. 12.379,80€ ). In addition, it was estimated that 2nd line high efficacy treatments (additional investment of 76% in treatment costs) may lead up to 4.3M€ savings in relapse treatment total costs by reducing annualized relapse rate. Further studies should be developed to assess the budget impact of 2nd line high efficacy treatments on disability progression since lower EDSS scales are associated with less expenditure per patient (EDSS≤ 3: 10.754,90€ vs EDSS ≥ 6,5: 24.945,94€ ). PND31 Pharmacoeconomic Aspects of Multiple Sclerosis Treatments in Iran Khanizadeh H 1, Izham M 2, Nikkhah K 3 SAINS MALAYSIA, penang, Malaysia, 2Universiti Sains Malaysia, Pinang, Palau Pinang, Malaysia, 3Mashhad University of Medical Sciences, Mashhad, Khorasan, Iran . . . 1UNIVERSITY Objectives: Multiple Sclerosis (MS) is a chronic and progressive which represents a catastrophic payment to patient, society and health care system. Iran, differing to the other countries in Middle- east, is considered to have a medium to high prevalence of MS. Although much is known about the MS cost in the world, there is a very paucity of the MS cost study in Iran. The aim of study was to estimate the costs and QOL in MS individuals and determine whether these costs increase as disability progress. Methods: We studied 160 MS patients who attended in the MS A396 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 association of Khorasan privence (the widest province in Iran ). Quality of life was measured with MSQOL-54 instrument. Data was collected by employing a 32- item self-administered questionnaire in a face to face interview. Parametric, nonparametric tests and descriptive statistics analysis were applied (p value<0.05). Patients were grouped into three disability stages according to their Expanded Disability Scale Score (EDSS). Results: The Patients mean age was 31.78 (SD: 9.67 ) years, % 73.8 were female and %26.3 were male, and their mean EDSS was 2.4 (SD; 1.26) whereas EDSS increases, the costs also increases, which is a positive correlation. The mean QOL was 0.54 that as QOL increases, the costs decreases, which is a negative correlation. The MS medications (Interferon) have a cost around $ 46625 per year for each patient that are subsidized about $ 24452 IR by governmental sector. Up to $ 17104 are paid by insurance and $ 5263 directly by patients. The costs per patient-year were calculated as $ 11560) - 27970.5591 (EDSS= 1-2.5), $ 29916.90930015.645 (EDSS= 3-4.5) and $34678.776- 34793.22 (EDSS= 5- 7.5). Conclusions: We conclude that the costs are relevant in MS, especially when disability increases. The catastrophic cost has a high burden to patients, society and health care system PND32 Whole Exome Sequencing as a Diagnostic Tool for Complex Neurological Disorders Frederix G W 1, Monroe G 2, Hövels A M 1, van Haaften G 2 University, Utrecht, The Netherlands, 2Utrecht University Medical Center, Utrecht, The Netherlands . . . . . . apy. Methods: Observational, non-interventional, prospective, single-center study of 1 year follow-up from ITB implant onward. Results: 20 consecutive patients with ITB indication were recruited; 13 received an ITB implant during the study period; 1 implant was rejected and thus explanted. 12 patients, of whom 10 had spasticity due to spinal-cord injury, 1 to multiple sclerosis and 1 to adrenoleukodistrophy, provided data for the study and 9 completed follow-up. After 12 months of ITB, mean changes from baseline were: -2.6 on the Penn scale (p= 0.011), -1.1 (p=0.059) and -2.8 (p=0.011) on the Ashworth upper and lower scale, respectively and +4.4 on the FIM scale (p=0.075). Mean utility gain, assessed with the HUI3 scale, was 0.054 (p= 0.091) after 1 year. Mean total ITB test and permanent implant cost per patient were € 528 and € 14,225, respectively. Mean quarterly spending on oral antispastics decreased by € 42 at month 12, while consumption of intrathecal baclofen stabilized after 6 months at € 39. At baseline, 4 patients received botulinum injections, while none did at the end of follow-up. Catheter-related adverse events occurred in 2 out of 12 patients, with a total mean cost per event of € 2.387. While waiting to receive ITB, spasticity-related hospitalizations due to urological complications occurred in 2 out of 20 patients, with a mean cost of € 9.044 per event; no such events were observed after ITB implant. Conclusions: Clinical outcomes of patients with N-FDS improved after ITB. Initial therapy costs were considerable, but were partially offset by savings in drugs and spasticity-related events after 1 year follow-up. Results should be interpreted cautiously because of the small number of observations. 1Utrecht Objectives: The primary objective of this study is to elucidate the effect of whole exome sequencing (WES) in diagnosing children with a developmental delay due to unexplained conditions presumed to be genetic. A secondary objective is to collect all resources used by these children to gain insight into the total costs over time for the traditional diagnostic pathway and the additional costs to diagnose a patient using WES. Methods: We included twenty children at the Sylvia Toth Centre (STC) in Utrecht, the Netherlands, who have underwent previously extensive clinical diagnostic workups and for whom no diagnosis was found after the last extensive workup. On all twenty children and parents WES will be performed, thereby obtaining a list of exonic candidate mutations for each patient. In parallel all resources used were collected by assessing the clinical records of patients. These resources were linked to unit costs to obtain the total cost per patient. Total cost per patient was then compared to the cost of care using WES, assessed for each individual patient. Results: The diagnostic yield from the 13 patients sequenced thus far is 23% indicating a 23% increase in number of diagnoses compared to the current diagnostic pathway. On average these patients have had numerous visits to the hospital, overnight stays and different diagnostic workups to unravel the genetic cause of their neurological disorder. Total cost of the current diagnostic pathway is therefore up to ten fold higher compared to the total cost of only providing WES. Conclusions: Comparing the diagnosis and costs with and without the use of WES gives a clear picture of the clinical and economic feasibility of putting WES into standard diagnostic practice at the STC and similar genetic centers over the world. PND33 Financial and Clinical Implications of Intramuscular Versus Subcutaneous Interferon Beta-1a in Portugal, Based on the Findings from the Cochrane Collaboration Review of First-Line Treatments for Relapsing-Remitting Multiple Sclerosis Silverio N 1, Sequeira L 1, Meletiche D 2 1Merck SA, Alges, Portugal, 2EMD Serono, rockland, ME, USA . . . Objectives: To estimate the clinical and financial impact of Interferon beta-1a intramuscular (IM) and subcutaneous (SC) formulations in Portugal, based on the findings from Cochrane review regarding first-line treatments for relapsing-remitting multiple sclerosis. Methods: An Excel-based model estimated the number of relapses and costs incurred by a hypothetical cohort of 1000 patients treated with two types of interferon beta-1a. The model evaluated the consequences of treatment with SC versus IM interferon beta-1a, as this was the only comparison whose data quality was assessed as ‘high’ by the Cochrane Review (Filippini 2013). Risk of relapse was based on the 2-year data from the Cochrane meta-analysis. The analysis was performed from a Portuguese National Health Service (NHS) perspective including only direct costs. Although efficacy was kept constant as Cochrane did not report outcomes based on Expanded Disability Status Scale (EDSS), costs of relapse were available for patients with different EDSS values, thus allowing estimation of cost impact for different types of population. Results: According to the model, treatment with IM interferon beta-1a is expected to result in a total of 743 episodes of relapse, whereas SC interferon beta-1a is expected to result in a total of 570 cases, less 173 cases, over a 2-year period. Use of the SC formulation in a population with EDSS ≤ 3 will result in savings of € 690,213.04 over the 2-year period due to avoided relapses. In a more severe population, with EDSS between 3.5 and 4.5, these savings are expected to be € 889,865.74 over the same 2-year period. Conclusions: When compared with the IM formulation, the use of SC interferon beta-1a seems to be associated with fewer cases of relapse, resulting over a 2-year period in considerable potential savings to the NHS in terms of relapses avoided. PND34 Clinical Outcomes and Health Care Resource Utilization in a 1-Year Observational Study of Patients with Non-focal Disabling Spasticity who are Resistant or Intolerant to Oral Therapy Treated with Intrathecal Baclofen Therapy at the Institut Guttmann (Spain). Epice Study PND35 Nuedexta for the Treatment of Pseudobulbar Affect: Estimating the Financial Impact to the Scottish NHS Kiff C 1, Mealing S 1, Singh M 1, Baculea S 1, Badhan A 1, Yonan C 2 1ICON Health Economics, Oxford, UK, 2Avanir Pharmaceuticals, Inc., Aliso Viejo, CA, USA . . . . . . Objectives: Pseudobulbar Affect (PBA) is a neurologic disorder of emotional expression, resulting in frequent and involuntary episodes of crying and/or laughing. Common neurological conditions associated with PBA include: Alzheimer’s disease, amyotrophic lateral sclerosis, multiple sclerosis, Parkinson’s disease, stroke and traumatic brain injury. Nuedexta® (Avanir Pharmaceuticals Inc.) is the only EMA-approved PBA treatment. The financial impact of introducing Nuedexta to a national health care system, including Scotland, has never been formally estimated. Methods: An Excel® based cost-calculator was developed. Prevalence, epidemiology and mortality estimates for causative neurological conditions as well as PBA prevalence in those conditions were sourced from the literature. Unit costs (drugs, hospitalisation etc.) were taken from national databases and standard care treatment mix and resource use were derived from a US claims database. A range of market uptake rates were used with further sensitivity analyses undertaken. Results: The estimated cost of standard care in Scotland for PBA is circa £32.4 million annually (circa 22,500 patients). In year 1 following introduction, 67 patients are expected to receive Nuedexta, resulting in a cost increase by £0.1 million to £32.5 million. By year five, 836 patients are estimated to receive Nuedexta, resulting in a projected total annual cost of £34.6 million. Therefore the estimated annual budget impact of Nuedexta ranges from £0.15 million (year 1) to £1.88 million (year 5). The incremental cost per patient is £2,246. The model was sensitive to changes in uptake rates, cost of background therapy and PBA symptom severity. When patients with moderate to severe PBA symptoms receive treatment, the projected cumulative year 5 budget impact estimate is £7.56 million. Conclusions: The estimated financial impact of introducing Nuedexta in Scotland is modest. Even if more patients are identified, the relatively small incremental cost per-patient of Nuedexta is unlikely to have a major impact on the Scottish NHS. PND36 Analysis of Expenditure in Multiple Sclerosis Disease Modifying Therapies Evolution Between 2004-2013 in Spain Villoro R 1, Hidalgo A 2 Max Weber, Madrid, Spain, 2University of Castilla La Mancha, Toledo, Spain . . 1Instituto Objectives: To analyze factors of recent evolution of Multiple Sclerosis (MS) Disease Modifying Therapies (DMT) budgets in Spain between 2004 and 2013. Methods: 2004-2013 single DMT monthly expenditure was provided by IMS Health. Monthly and annually evolution of number of patients, billing, drug cost per patient and cost per year of treatment were calculated. Two periods: 2004-2013 and 2007 (start marketing second lines DMT) -2013 period were analyzed for each DMT line. (First line: subcutaneous and intramuscular interferon (IFN) β -1a, subcutaneous IFN β -1b and glatiramer acetate injection. Second line: natalizumab and fingolimod). Results: During 2004-2013 DMT expenditure increased from €115.5M to €319.3M due to: A greater number of patients 147% (10.60 % annual growth per year) and a further growth of annual cost per patient: 12% (1.29 % annual growth per year). In December 2013 second lines correspond to a 29.61% of DMT expenditure. Annual cost per patient in second line represents 70% over cost per treated patient and 83% greater than first line DMT cost per year. If year 2007 is omitted from analysis (Only 68 second-line treatments and M1.44€ of associated expense) and is analyzed 2008-2013 period, second-line DMT represent 43% of new treatments causing a 60% increase in DMT expenditure. In 2013 second line DMT participation reaches 64% of new regimens causing the 79% of increase DMT expenditure. Conclusions: The growing incorporation of new therapies and the noticeable rise in the number of treated patients (10.60 % annual growth per year) are components to consider in the pharmaceutical budget management. PND37 Health Care Resource Use and Cost Of Multiple Sclerosis in Slovakia: Results from the National Cross-Sectional Study Slof J 1, Serrano D 2, Álvarez López-Dóriga M 3, Álvarez M 3, Marqués T 4, Benito J 4, Vidal J 4 1Universitat Autònoma de Barcelona, Bellaterra, Spain, 2Autonomous Consultant, Barbera del Valles, Spain, 3Medtronic Ibérica, S. A., Madrid, Spain, 4Institut Guttmann, Barcelona, Spain Ondrusova M1, Psenkova M1, Donath V2 1Pharm-In Ltd, Bratislava, Slovak Republic, 2F.D. Roosvelt University Hospital, Banska Bystrica, Slovak Republic Objectives: To assess clinical outcomes, health care resource utilization and associated costs of intrathecal baclofen therapy (ITB) for the treatment of non-focal disabling spasticity (N-FDS) in patients who are resistant or intolerant to oral ther- Objectives: Comprehensive economic costs of multiple sclerosis (MS) according to EDSS states can only be assessed by evaluating MS management in real clinical practice. The objective of this cross-sectional study was to measure the . . . . . . . A397 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 resource utilisation and the costs associated with health care management of MS in Slovakia and to provide a basis for cost-effectiveness evaluations. Methods: Descriptive epidemiological data of 2,552 MS patients from 27 MS-centers across Slovakia were collected electronically and analyzed. In 152 selected patients followed up in 2011-2012 in 34 MS centers, all types of health care services and costs were analyzed. These patients were randomly included in the study. Continuous variables were calculated using standard descriptive statistical methods. Results: 77% of patients had the relapsing-remitting form of MS (RRMS); 60 % of patients were in EDSS 1-3, mean age of patients at the time of diagnosis was 32.3 (± 9.7) years, and 68% of patients were females. Total direct health care costs, DMT excluded, ranged from € 752 to EDSS 1 to € 2,839 to EDSS 7, being the lowest for EDSS9 (at € 963). Costs for DMT ranged from € 8,584 for EDSS1 to € 13,026 in EDSS4, being lower for EDSS6 (€ 1,668) and none for EDSS7-9.67% patients were receiving 1st line DMT, and 14 % receiving 2nd line treatment. DMT was mostly applied in EDSS 2 (97%). The most frequently used DMTs were glatiramer acetate (20%), interferon beta1a IM (15%), natalizumab (9%) and fingolimod (5%). The most expensive grade 1/2 adverse events were abdominal pain (46.62 €), pain in joints, back and arms (39.35 €). Conclusions: This cross-sectional study determined the average annual direct cost per MS patient to be € 1,640; DMT excluded. As the EDSS increases, DMT costs decrease (except of EDSS1-2) and the costs of medical devices rise. PND38 Cost of Health Care Services Offered by Parkinson Disease Associations in Spain Galvez M 1, Prades M 2, Paz S 2, Lizán L 2 Española de Párkinson, Madrid, Spain, 2Outcomes’10, Castellon, Spain . . . . 1Federación Objectives: In Spain, Parkinson Disease Associations (PDA) offers a wide range of care services with partial and variable financial support from the government. The study objective was to estimate the costs of PDA services and to calculate the potential savings that they represent to the National Health System (NHS). Methods: A survey conducted by the Federation of PDA collected information on their location, number of members, type and use by patients of the offered services. Services were classified according to whether or not they were financed by the NHS, based on the existing national portfolio for reimbursed services. Weekly use was recorded and costs were calculated upon official rates (updated to € , 2014). Potential savings for the NHS were estimated by calculating the weekly cost associated to unfinanced services that were provided by the PDA. Results: 42 Spanish PDA that embraced a total of 11,420 patients participated in the study. From the 26 services offered, speech therapy (n=41), physiotherapy (n=39), cognitive stimulation (n=23) and occupational therapy (n=23) were the most frequently offered and used. The weekly cost associated to the provided services was €655,219.87 [mean: €15,980.97 (SD: 22,662.98)] per PDA. 53.8% of services were classified as potentially refundable by the NHS. Costs assaignable to potentially financed services represented 78.29%, reaching savings for the NHS of €512,971.60/week. The cost of physiotherapy (27.3%) and of the adapted transport (21.3%) were the main components. Costs attributable to services not usually financed by the NHS accounted for 21.71% (€142,248.23), mostly attributable to cognitive stimulation costs (63.39%). Conclusions: PDA offer valuable services to patients and carers and afford a major proportion of the costs of the supportive and complementary care of the disease. Their economic efforts imply great savings to the NHS in Spain. PND39 The Impact of Adherence and Development of Neutralizing Antibodies to Interferons β on Treatment of Multiple Sclerosis in the Czech Republic Kruntoradova K 1, Klimes J 1, Dolezal T 1, Mandelikova M 1, Pavlikova P 2 OUTCOMES, Prague, Czech Republic,, 2Biogen Idec, Prague, Czech Republic . . . . . 1VALUE Objectives: To compare clinical outcomes (reduction in relapse number) and costs associated with multiple sclerosis (MS) treatment with one of the interferon β in the Czech Republic in five-year horizon based on development of neutralizing antibodies (NAbs) and patient non-adherence. Intramuscular (IM) interferon β -1a is characterized by very high adherence rate and low rate of NAbs production. Methods: Markov cohort model was developed with one-year cycle length. In the Czech Republic, patients with MS initiate treatment with one of the interferon β. NAbs-positive patients (in the model, NAbs are detected during the second year of treatment and thereafter) are switched/escalated to a different disease modifying drugs; DMD (glatiramer acetate, fingolimod, natalizumab). If patients experience two or more relapses during one year of treatment, they are escalated to fingolimod or natalizumab. Adherence data, development of NAbs, relapse rate and costs were sourced from the literature. Results: One hundred patients, who initiated treatment with IM interferon β -1a, experienced 287 relapses over 5 years. Those, who started treatment with subcutaneous (SC) interferon β -1a and interferon β -1b, experienced by 15 and 19 relapses more. In one hundred patients, total cost of treatment with IM interferon β -1a was 6.4 million € . This was about 139-200 thousand € less compared to SC interferon β -1a and interferon β -1b. Hence incremental cost-effectiveness ratio was -262 thousand € /relapse avoided and -285 thousand € /relapse avoided. Conclusions: Intramuscular interferon β -1a represents dominant intervention in MS treatment, i. e. cost-saving treatment from payer’s perspective and simultaneously more efficacy intervention in terms of reduction in number of relapses due to better patient adherence and lower incidence of NAbs compare to the other interferons β in the Czech Republic. The one-way sensitivity analyses showed that results are the most sensitive to DMD costs and relapse rate. PND40 An Economic Evaluation of Subcutaneous and Intramuscular Interferon Beta-1a in Multiple Sclerosis Using a Direct Head-ToHead Study Phillips A L 1, Edwards N C 2, Locklear J C 1 1EMD Serono, Inc., Rockland, MA, USA, 2Health Services Consulting Corporation, Boxborough, MA, USA . . . . . . Objectives: To use health economic modeling techniques to quantify and compare the clinical and economic outcomes associated with the use of subcutaneous interferon beta-1a (scIFNβ1a) vs. intramuscular interferon beta-1a (imIFNβ1a) over 2 years in the management of relapsing forms of multiple sclerosis (MS) from a US health care payer perspective. Methods: The 2-year decision analytic model was populated with IMS LifeLink Plus prevalence and treatment data, and clinical data from the EVIDENCE (EVidence of Interferon Dose-response: European North American Comparative Efficacy) study, a direct head-to-head comparison of 44 mcg scIFNβ 1a three times a week vs. 30 mcg imIFNβ 1a once a week. Relapse data from 16-month results were extrapolated for the 2-year model. Disease-modifying drug (DMD) costs were based on 2014 wholesale average cost with consideration of patient copayment in the base case. The model was created with the ability to customize the rate of copayment as well as to incorporate contractual discounts, if desired. One-way sensitivity analyses were conducted on key parameters using alternate plausible values, including the rates of real-world DMD adherence. Results: For a hypothetical health plan with 1 million members, it is estimated that 911 patients with MS would be treated with DMDs. More relapses were avoided with scIFNβ1a over 2 years (979) than with imIFNβ1a over 2 years (778). The average cost-effectiveness of 44 mcg scIFNβ 1a was lower (more favourable) than the average cost-effectiveness of 30 mcg imIFNβ 1a ($123,854 vs. $148,749 per relapse avoided). Sensitivity analyses around model input values showed the model was robust and cost-effectiveness results were consistent. The model results are most sensitive to drug cost. Conclusions: Cost-effectiveness assessment may facilitate the decision-making process in selecting MS treatments. Using the highest-quality clinical data (Level 1, head-to-head study, EVIDENCE), the cost-effectiveness of 44 mcg scIFNβ 1a was shown to be favourable compared with 30 mcg imIFNβ 1a. PND41 Cost-Effectiveness Evaluation of Data from the Evidence (Evidence of Interferon Dose-Response: European North American Comparative Efficacy) Study Phillips A L 1, Edwards N C 2, Locklear J C 1 1EMD Serono, Inc., Rockland, MA, USA, 2Health Services Consulting Corporation, Boxborough, MA, USA . . . . . . Objectives: To evaluate the cost-effectiveness of 44 mcg subcutaneous interferon beta-1a (scIFNβ1a) and intramuscular interferon beta-1a (imIFNβ1a) during the comparative and open-label extension phases of the EVIDENCE (EVidence of Interferon Dose-response: European North American Comparative Efficacy) study. Methods: A decision analytic model from a US health care payer perspective was populated with 2-year data from the EVIDENCE study of imIFNβ 1a and 44 mcg scIFNβ 1a (comparative and open-label extension phases). EVIDENCE results showed that the annualized relapse rate of 44 mcg scIFNβ 1a patients was 0.46 during the comparative phase and 0.34 during the open-label extension phase. The annualized relapse rate of imIFNβ1a patients during the comparative phase was 0.64. imIFNβ1a patients who switched to 44 mcg scIFNβ 1a for the open-label extension phase had an annualized relapse rate of 0.32. These data were used to model the cost-effectiveness of 44 mcg scIFNβ 1a and imIFNβ 1a patients during the comparative phase as well as for the combined comparative and open-label extension phases (for patients remaining on scIFNβ 1a throughout and for those switching from imIFNβ 1a in the comparative phase to scIFNβ 1a in the open-label extension). Disease-modifying drug (DMD) cost was based on 2014 wholesale average cost with consideration of patient copayment in the base case. Results: The cost-effectiveness for 44 mcg scIFNβ 1a and imIFNβ1a during the comparative phase was $123,854 and $148,749 per relapse avoided, respectively. The cost-effectiveness of patients who remained on 44 mcg scIFNβ 1a throughout the study was $99,398 per relapse avoided, while the cost-effectiveness of imIFNβ 1a patients who switched to 44 mcg scIFNβ1a for the open-label extension was $116,404 per relapse avoided. Sensitivity analyses showed that the model was robust and was most sensitive to DMD cost. Conclusions: This decision analytic model evaluation shows that remaining on 44 mcg scIFNβ 1a and switching from imIFNβ 1a to 44 mcg scIFNβ 1a were cost-effective treatment strategies. PND42 Economic Evaluation of Lacosamide in the Management of Epileptic Partial Onset Seizures in Greece Geitona M 1, Kousoulakou H 1, Carayianni V 2, Baltogiannis C 3, Garganis K 4, Gatzonis S 5, Giannakodimos S 6, Kodounis A 7, Nasios G 8, Polychronopoulos P 9, Christou P 10, Theodoratou D 11, Karachalios G 11, Palaistis S 11 1University of Peloponnese, Corinth, Greece, 2Technological Educational Institution of Athens, Athens, Greece, 3IASO General Hospital, Athens, Greece, 4Agios Loukas Hospital, Thessaloniki, Greece, 5National and Kapodistrian University of Athens, General Hospital of Athens, Athens, Greece, 6General Hospital of Athens, Athens, Greece, 7251 Hellenic Air Force Hospital, Athens, Greece, 8Technological Educational Institute of Epirus, Ioannina, Greece, 9University Hospital of Patras, Patra, Greece, 10UCB Pharma, Athens, Greece, 11UCB AE, Athens, Greece . . . . . . . . . . . . . . Objectives: To assess the cost-effectiveness of Lacosamide (LCM) in the management of epileptic partial onset seizures (POS) versus standard AED therapy in Greece, as well as its impact on the health care budget. Methods: A cost-effectiveness model was developed simulating the treatment pathway of a hypothetical cohort of 1000 patients over two years. A comprehensive literature search was conducted to identify local resource use data for medical, pharmaceutical and hospital treatment. Due to lack of relevant data, an expert panel with 8 neurologists was convened. The perspective was that of the Social Insurance Fund, and unit costs were taken from officially published sources (Ministry of Health and Social Insurance Fund). Primary and secondary analyses were carried out, in which the treatment algorithm was based on trial data and was adapted to the Greek setting, respectively. Deterministic and probabilistic sensitivity analyses were conducted to test the model’ results. In addition, a budget impact analysis (BIA) was carried out to estimate the annual cost of treating uncontrolled epileptic patients in Greece. Results: Treatment with LCM was shown to be dominant compared to standard therapy, as it is associated with 38 additional QALYs and reduced cost by € 410,024 and € 754,684 in the primary A398 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 and secondary analysis, respectively. Extensive sensitivity analyses indicated that results were robust. The most influential parameters were the utility estimates, probability of hospitalization per seizure and unit cost of hospitalization. The BIA showed that the annual cost of treating uncontrolled epileptic patients with LCM in Greece ranges between € 274.9 and € 271.5 million, in the primary and secondary analyses, respectively, compared to the respective costs with standard AED (€ 277.3 and € 279.9 million). Conclusions: LCM appears to be both less costly and more effective compared with standard AED treatment in Greece and results in cost savings ranging between € 2.4 and € 8.3 million. Study funded by UCB Pharma. PND43 Cost-Effectiveness of Subcutaneous VersUS Intramuscular Interferon Beta-1A In Portugal Based on the Findings of Cochrane Collaboration Review of First-Line Treatments for RelapsingRemitting Multiple Sclerosis Silverio N 1, Sequeira L 1, Meletiche D 2 1Merck SA, Alges, Portugal, 2EMD Serono, rockland, ME, USA . . . with 44 mcg scIFNβ 1a in patients with EDSS > 3.5–5.0 was 1.21 per patient over 2 years. The mean number of relapses avoided for the overall study population was 0.74 per patient over 2 years. The average cost-effectiveness of 44 mcg scIFNβ 1a was estimated to be $107,861 per relapse avoided for the EDSS > 3.5–5.0 cohort. The average cost-effectiveness for the overall study population was estimated to be $181,208 per relapse avoided. Sensitivity analyses showed that results were robust to changes in key input parameters such as DMD costs, the number of relapses in untreated patients, the relative risk reduction in clinical relapse rates, the rate of adherence, and the average cost of relapse. Conclusions: Based on model results, the average cost-effectiveness of 44 mcg scIFNβ 1a was favorable for both the overall study population and the EDSS > 3.5–5.0 cohort. PND46 A Cost-Utility Analysis of Sacral Anterior Root Stimulation (SARS) Compared to Medical Treatment in Complete Spinal Cord Injured Patients with a Neurological Bladder Bénard A 1, Morlière C 1, Verpillot E 2, Donon L 1, Salmi L R 1, Joseph P A 1, Vignes J R 1 Bordeaux, Bordeaux, France, 2Université de Bordeaux, Bordeaux, France . Objectives: To estimate the cost-effectiveness of Interferon beta-1a subcutaneous (SC) when compared with Interferon beta-1a intramuscular (IM) in Portugal, based on the findings published by the Cochrane review of first-line treatments for relapse-remitting multiple sclerosis. Methods: An Excel-based model estimated the number of relapses and costs incurred by a cohort of 3,000 patients treated with two types of interferon beta-1a. The model evaluated the consequences of each treatment based on the findings of a Cochrane meta-analysis (Filippini 2013). The analysis was performed from a Portuguese NHS perspective, including only direct costs. Costs of relapse were obtained from a local publication (Mateus C, 2000) whereas costs of both drugs were obtained from local official databases (Cat@ logo). Although efficacy was kept constant as Cochrane did not report outcomes based on EDSS, costs of relapse were available for patients with different EDSS values, thus allowing estimation of cost-effectiveness for different types of population. Results: According to the model, over a 2 year period and in a population with EDSS ≤ 3, treatment with IM interferon beta-1a will result in a total of 2,228 relapses, and a total cost of 69,572,717€ , whereas treatment with SC interferon beta1a will result in 1,709 relapses and a total cost of 70,480,835€ . For a population with EDSS between 3.5 and 4.5 cost values for IM and SC are 72,141,975€ and 72,451,135€ , respectively. Cost-effectiveness ratios were 1,748€ per relapse avoided when EDSS ≤ 3 and 595€ per relapse avoided when EDSS was between 3.5 and 4.5. Conclusions: Considering that the cost of a relapse varies between 3,896€ (EDSS ≤ 3) and 5,139€ (EDSS between 3.5 and 4.5) the incremental cost-effectiveness ratios found for interferon beta-1a SC seem to indicate that that SC interferon beta-1a is a costeffective alternative to the use of IM interferon beta-1a. PND44 Treating VersUS Non-Treating Obsturictive Sleep Apnea in Italy and France: A Markov Model-Based Cost-Effectiveness Analysis . . . . . . . . . 1CHU Objectives: To estimate the cost-utility of sacral anterior root stimulation (SARS, using the Finetech-Brindley device) compared to medical treatment (anticholinergics + catheterization) in complete spinal cord injured patients with a neurological bladder. Methods: A probabilistic Markov model was elaborated with a 10-year time-horizon, one-year cycles and a 2.5% discount rate. Three irreversible states were defined: 1) treatment without urinary complication; 2) surgery for urinary complication (sphincterotomy, urinary derivation); 3) death. Reversible states (urinary calculi; Finetech-Brindley device failures) were integrated in the two first irreversible states. A systematic review and meta-analysis were performed to estimate transition probabilities and Quality Ajusted Life Years (QALYs). In the perspective of the French Healthcare System, costs were estimated from a published comparative cost-effectiveness research (Neurosurgery 2014; 73: 600), and through simulations using the 2013 French prospective payment system (PMSI) classification. Results: In the primary analysis, the cost-utility ratio was 10,647€ /QALY gained. At a 30,000€ ceiling ratio the probability of SARS being cost-effective compared to medical treatment was 63%. If the French Healthcare System reimbursed SARS for 200 patients/ year the two first years and 50 patients/year during 8 years (anticipated target population) the expected incremental net health benefit would be 222 QALYs, and the expected value of perfect information (EVPI) would be 4,570,000€ . The highest partial EVPI is reached for transition probabilities toward urinary calculi (4,420,000€ ). With discount rates of 1% and 6% the cost-utility ratios were 6,951 and 19,770€ / QALY gained, and the probabilities of SARS being cost-effective were 66% and 58%, respectively. Conclusions: Our model shows that SARS using Finetech-Brindley device offers the most important benefit and should be considered cost-effective at a 30,000€ ceiling ratio. Despite a high uncertainty, EVPI and partial EVPI may indicate that further research would not be profitable to inform decision making. Whitehouse J T 1, Da Deppo L 2, Lazzaro C 3, Pedretti R F E 4, La Rovere M T 5, Pepin J L 6, Defaye P 6 1GfK Market Access, Melton Mowbray, UK, 2Boston Scientific, Milan, Italy, 3Studio di Economia Sanitaria, Milan, Italy, 4IRCCS Fondazione Salvatore Maugeri Tradate, Tradate Milano, Italy, 5IRCCS Fondazione Salvatore Maugeri Montescano, Montescano, Italy, 6Université Grenoble Alpes, Grenoble, France PND47 Comparison of a Markov Cohort Model and a Discrete-Event Simulation for Economic Analyses of Treatments for Multiple Sclerosis Objectives: To investigate the cost-effectiveness of treating vs. not treating obstructive sleep apnea (OSA) in Italy and France Methods: A 5-year, 10-state Markov model with disease states including; disease and event-free with OSA, diabetes, hypertension, myocardial infarction [MI], post-MI, stroke, post-stroke, atrial fibrillation [AF], heart failure [HF], and death; was developed to compare costs, outcomes, and event-free life-years (LYS) of treating vs not treating OSA from the Italian National Health Service (INHS) and French National Health System (FNSH) viewpoint. Health care resources included those related to diagnosis, treatment (CPAP only) and follow-up of OSA; management of hypertension, diabetes, HT, AF, post MI, and post-stroke (per year); MI, and stroke (per episode). Health care resources were valued at Euro (€) 2012 using published sources. Results: After 5 years the number of incremental event-free LYS per patient treated for OSA reaches 0.31 (Italy: 4.15 vs 3.84; France: 4.07 vs 3.76). Treating OSA incurs an incremental cost of €1011.01 and € 2998.45per patient for Italy and France, respectively. The ICER of treating OSA is € 3212.39 for Italy and € 9777.09 for France, respectively. Conclusions: Treating OSA can be considered highly cost-effective for both the INHS and the FNSH when compared to the acceptability range for incremental cost effectiveness proposed for Italy (€ 25,000-€ 40,000) and for Europe (€ 50,000). Objectives: Multiple sclerosis (MS) is a disease with lifelong impact, making the cost-effectiveness (CE) of its treatments particularly sensitive to assumptions embedded in model designs. Traditional sensitivity analysis (SA) can test many assumptions, but it is not designed to investigate sensitivity to structural assumptions. The aim of this study was to compare a Markov cohort model (MM) and a discrete-event simulation (DES) model of MS that were based on common clinical data but developed independently to understand the impact of their structural differences on model predictions. Methods: A similar population was simulated in the MM and the DES model; aggregated cost and utility estimates were compared over varying time horizons. The average expanded disability status scale (EDSS) and the distribution of EDSS were also compared over time to study the dynamics of disease progression and treatment effects. Results: The two modeling approaches led to different natural history behavior over longer time horizons, even after shortterm behaviors were well-aligned, with the DES model predicting slightly fewer life-years (25.9 vs. 26.2 in the MM) but more quality-adjusted life-years (9.6 vs. 8.1 in the MM). These differences reflect slower progression of EDSS in the DES model, particularly to higher EDSS states. When disease history (including a baseline EDSS term) was excluded from the DES model, the natural history simulations of the two models agreed more closely. Conclusions: Structural SA can help quantify the impact of key modeling decisions. In this study, a comparison of an MM and a DES model showed that natural history predictions diverge over long time horizons, in part due to the consideration of disease history in the DES model. A better understanding of the differences between the two model designs helps ensure interpretation of the model results while taking into consideration the assumptions embedded in those designs. . . . . . . . . . . . . PND45 Cost-Effectiveness of Subcutaneous Interferon Beta-1A in a SubPopulation of Multiple Sclerosis Patients (Kurtzke Expanded Disability Status Scale [EDSS]: > 3.5–5.0) Phillips A L 1, Edwards N C 2, Locklear J C 1 1EMD Serono, Inc., Rockland, MA, USA, 2Health Services Consulting Corporation, Boxborough, MA, USA . . . . . . Objectives: To evaluate the cost-effectiveness of 44 mcg subcutaneous interferon beta-1a (scIFNβ 1a) in patients with multiple sclerosis (MS) with Kurtzke Expanded Disability Status Scale (EDSS) score > 3.5–5.0. Methods: The analysis was performed from a US payer perspective. The time horizon of the analysis was 2 years. The decision analytic model was populated with real-world inputs and related assumptions, as well as pivotal placebo-controlled clinical trial data for 44 mcg scIFNβ1a 3 times a week (PRISMS Study). Clinical inputs were obtained for the overall study population as well as a subpopulation of patients with Kurtzke EDSS score > 3.5–5.0. Disease-modifying drug (DMD) cost was based on 2014 wholesale average cost with consideration of patient copayment in the base case. Sensitivity analyses were conducted on key input variables to assess their impact on cost per relapse avoided. Results: Model results showed that the mean number of relapses avoided Kansal A 1, Tafazzoli A 1, Leipold R 1, Sarda S 2 Idec, Weston, MA, USA . . . . 1Evidera, Bethesda, MD, USA, 2Biogen PND48 The Long-Term Value of Glatiramer Acetate for the Treatment of Relapsing Remitting Multiple Sclerosis in the Netherlands Wilson,L E 1,2, Prüfert A 2 . . . 1Quintiles, Reading, UK, 2Quintiles, Hoofddorp, The Netherlands Objectives: To evaluate the cost-effectiveness of glatiramer acetate (Copaxone®) as a disease-modifying treatment (DMT) for relapsing-remitting multiple sclerosis (RRMS) compared to intravenous [natalizumab (Tysabri®), alemtuzumab (Lemtrada®) ] or subcutaneous injectables [interferon-beta-1b (Betaferon®), interferon-beta-1a 44mcg, 22mcg, 30mcg (Rebif-44®, Rebif-22®, Avonex®) ] and oral DMTs [fingolimod (Gilenya®), dimethyl fumarate (Tecfidera®), teriflunomide (Aubagio®)]. Methods: A Markov model followed patients over 50 years through 21 health A399 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 states: Expanded Disability Status Scale (EDSS) 0-9 for patients with RRMS and secondary progressive multiple sclerosis (SPMS), respectively, and death (EDSS10). Total treatment costs, quality-adjusted life-years (QALYs) gained and incremental cost-effectiveness ratios (ICERs) were calculated from the Dutch societal perspective. Baseline demographics; transition probabilities; treatment-specific relative risks; and utility values were obtained from published literature. Health resource use was based on the products’ SmPCs. Unit costs were based on national tariffs and published data. A base case analysis considered the direct and indirect costs of treatment. To test the robustness of the results, univariate and probabilistic sensitivity analyses (PSA) were performed. Results: Total treatment costs of glatiramer acetate were lower versus teriflunomide, interferon-beta-1a 44mcg, 30mcg, fingolimod, natalizumab, dimethyl fumarate and alemtuzumab; and higher versus interferon-beta-1a 22mcg, interferon-beta-1b, and BSC. Glatiramer acetate dominated interferon-beta-1a 44mcg and 30mcg, and was cost-effective versus other comparators at a willingness-to-pay threshold of € 20,000/QALY. The PSA indicated these results were robust. Univariate analysis showed that relative risks of disease progression and drug costs were the most influential model parameters. Conclusions: In the Netherlands, glatiramer acetate is a cost-effective treatment versus a number of other treatments indicated in multiple sclerosis, and results proved robust. Pnd49 Expected Value of Partial Perfect Information for the Disability Progression Efficacy of Teriflunomide and Fingolimod in the Treatment of Relapsing-Remitting Multiple Sclerosis Maruszczak M 1, Kusel J 1, Adlard N 2 1Costello Medical Consulting Ltd., Cambridge, UK, 2Novartis Pharmaceuticals UK Limited, Surrey, UK . . . Objectives: To calculate the expected value of partial perfect information (EVPPI) of the disability progression efficacy for teriflunomide and fingolimod in relapsingremitting multiple sclerosis (RRMS) and investigate the potential differences between these two disease modifying therapies (DMTs). Methods: A cohort Markov model based on the ScHARR model structure was developed from the UK NHS perspective, which allows the cost-effectiveness of the DMTs to be calculated and also has the additional ability to calculate the EVPPIs for progression efficacies. The evaluated intervention consists of beta-interferons followed by fingolimod. The comparator treatment includes beta-interferon therapy followed by teriflunomide. The EVPPIs for both drugs were calculated and compared. Results: The EVPPIs varied substantially depending on the evaluated drug and the specific willingness-to-pay (WTP). Although both EVPPI curves had a similar shape, the curve for teriflunomide had an EVPPI equal or larger compared to fingolimod’s curve for every WTP value. At a WTP of £30,000 per QALY, the EVPPI per patient of teriflunomide and fingolimod were £252 and £50, respectively. In the sensitivity analysis, where the uncertainty around the estimate of teriflunomide progression efficacy was artificially lowered to that of fingolimod, the divergence between the EVPPI curves was reduced substantially, confirming that the uncertainty around teriflunomide progression efficacy is driving its higher EVPPI. Conclusions: Higher EVPPI estimates of teriflunomide indicate that the precise knowledge of the disability progression efficacy of teriflunomide would be considerably more valuable to the decision maker than that of fingolimod. One of the main factors affecting larger EVPPI of teriflunomide’s progression efficacy is a higher uncertainty concerning effects of teriflunomide. There are, however, many data limitations and uncertainties within the DMT modelling. Additionally, the expected value of sample information analysis (EVSI) would be required in order to evaluate more precisely the cost-effectiveness of additional clinical trials. PND50 Cost-Minimisation Analysis of Colistimethate Sodium in Serbia- off Label Use Approach Djordjevic J 1, Mitrovic M 1, Marinkovic V 1, Tasic L 2, Krajnovic D 2 Pharma doo Serbia, Belgrade, Serbia and Montenegro, 2University of Belgrade – Faculty of Pharmacy, Belgrade, Serbia and Montenegro . . . . Objectives: Colistimethate sodium (Colistin) is an old, “forgotten” antibiotics revived due to multidrug-resistant Gram-negative bacteria in nosocomial infections. In many countries, both – intravenous and inhalational administrations of Colistin have marketing authorisation. In Serbia, Colistin is registered only for intravenous administration. Therefore, inhalational administration of Colistin in treatment for cystic fibrosis is considered as off label use. AIM: The purpose of this study was to assess benefits of Colistin off label use, currently available in Serbia, for treatment for cystic fibrosis. Methods: Data collected from a review of the literature (Pubmed searching key word off label, Colistin, Tobramycine, cost minimisation) and data from Health Insurance Fond of Republic Serbia (2014) were used to perform a cost-minimisation analysis comparing DDD in inhalation administration of Colistin with Tobramycin (official methodology in Serbia for drug reimbursement Listing). Results: Both, Colistin and Tobramycine are effective in cystic fibrosis treatment. The cost of treatment per patient per year in Serbia is 3 616,00 Euro for Colistin and 5 950,00 Euro for Tobramycin. Total savings per year for 200 registered patients in Serbia could be 466 857,00 Euro. Conclusions: Given cost-minimisation analysis justified the treatment with Colistin as cost saving therapeutic alternative to Tobramycin for cystic fibrosis patients in Serbia. The lack of guidelines and principles regarding off label medicine use in health care policy and decision making in Serbia was evidently an obstacle for better patients care. PND51 Health Care Utilization and Costs of Medicaid Program Services for Patients Diagnosed with Multiple Sclerosis Li L 1, Shrestha S 1, Baser O 2, Wang L 1 Research, Plano, TX, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . PND52 Cost-Utility Analysis (cua) Of First-Line Disease-Modifying Treatments (DMT) Versus Best Supportive Care (Bsc) In Finnish Relapsing-Remitting Multiple Sclerosis (RRMS) Patients Soini E 1, Asseburg C 1, Sumelahti M L 2 1ESiOR Oy, Kuopio, Finland, 2School of Medicine, University of Tampere, Finland . . . . Objectives: CUA of teriflunomide and first-line DMTs (glatiramer acetate (GA), interferon-β -1a (IFNβ -1a: intramuscular IM, or subcutaneous SC) or interferon-β -1b (IFNβ -1b)), compared to BSC in Finnish RRMS-patients. Methods: Register study and Markov cohort modelling. During the 50-year time horizon, patients could stay or progress to another Expanded Disability Status Scale (EDSS) state, experience relapse (with/without hospitalisation), adverse events (AEs) or death. EQ-5D utilities were derived from literature, and Finnish costs (health care 2013; official drug costs 4/2014 without VAT) were associated with EDSS, relapses, and AEs. Indirect treatment comparison informed treatment effects. EDSS-related standardised mortality ratios (SMR) and RRMS-progression were analysed from Finnish MS-register (Tampere, Vaasa, Seinäjoki regions; 1359 patients). Other health risks were based on literature. Cohort characteristics were estimated from a subgroup (713 patients, clinically confirmed RRMS, age, gender and EDSS 0–6.5 at first DMT initiation). Caregiver’s disutility, Finnish productivity losses, and informal care costs were included in sensitivity analysis (societal perspective). Costs and quality-adjusted life-years (QALYs) were discounted with 3%/year. Results: Expected lifetime health care [societal] costs per patient were: teriflunomide: € 798,864 [€ 1,596,570], IFNβ -1aSC: € 805,069 [€ 1,618,344], IFNβ -1b: € 826,633 [€ 1,647,745], GA: € 823,438 [€ 1,649,961], IFNβ -1a-IM: € 820,516 [€ 1,645,472], and BSC: € 788,932 [€ 1,626,554]. Respective QALYs were 6.55 [5.44], 6.27 [5.13], 6.15 [4.99], 6.05 [4.88], 6.04 [4.87], and 5.79 [4.60]. From the payer [societal] perspective, incremental cost per QALY gained of teriflunomide was € 13,089 [dominant] relative to BSC, and teriflunomide dominated other firstline DMTs. Teriflunomide had 83% [per-patient expected value of perfect information, EVPI € 241], 98% [EVPI € 22], and 100% [EVPI € 0] cost-effectiveness probability with € 0, € 10,000, and € 20,000 per additional QALY, respectively. Conclusions: Cost-effectiveness analyses of first-line DMTs compared to BSC showed that teriflunomide provided lowest cost and highest number of QALYs and was therefore dominant over IFNβ s and GA for Finnish RRMS patients. . 1Alvogen 1STATinMED Objectives: Health care resource utilization and costs were evaluated for patients diagnosed with multiple sclerosis (MS) in the U. S. Medicaid program. Methods: Patients diagnosed with MS (International Classification of Disease, 9thRevision, Clinical Modification [ICD-9-CM] diagnosis code 340) were identified using Medicaid data from 01JAN2008 through 31DEC2010. The initial diagnosis date was designated as the index date. Patients of the same age, race, and gender but without an MS diagnosis were identified and matched as the comparison cohort, with a randomly chosen index date to minimize selection bias. Patients in both groups were required to be at least age 18 years, and have continuous medical and pharmacy benefits 1 year before and 1 year post-index date. One-to-one propensity score matching (PSM) was used to compare health care costs and utilizations during the follow-up period, between the diseased and the comparison cohorts, and adjusted for baseline demographic and clinical characteristics. Results: After risk adjustment by PSM, a total of 14,179 patients in each cohort were matched. Significantly more MS patients had inpatient admissions (23.75% vs. 10.87%, p<0.0001) and long-term care (22.64% vs. 4.13%, p< 0.0001), other service (99.70% vs. 89.00%, p< 0.0001) and pharmacy visits (73.08% vs. 67.71%, p< 0.0001) compared to those without an MS diagnosis. Higher health care utilization by MS patients led to significantly higher inpatient ($1,688 vs. $725, p< 0.0001), long-term care ($14,189 vs. $2,778, p< 0.0001), other service visit ($22,981 vs. $9,977, p< 0.0001) and pharmacy costs ($5,284 vs. $1,785, p< 0.0001) compared to those without MS. Conclusions: Compared to patients in the comparator cohort, MS patients in the Medicaid program incurred substantially higher health care resource utilization and costs. PND53 Cost-Effectiveness Model Results of Intrathecal Baclofen Therapy Compared to Conventional Medical Management in Patients with Non-Focal Disabling Spasticity Who are Resistant or Intolerant to Oral Therapy at the Institut Guttmann Slof J 1, Serrano D 2, Álvarez M 3, Álvarez López-Dóriga M 3, Marqués T 4, Benito J 4, Vidal J 4 1Universitat Autònoma de Barcelona, Bellaterra, Spain, 2Autonomous Consultant, Barbera del Valles, Spain, 3Medtronic Ibérica, S.A., Madrid, Spain, 4Institut Guttmann, Barcelona, Spain . . . . . . . Objectives: To estimate the cost-effectiveness of intrathecal baclofen therapy (ITB) against conventional medical management (CMM) in non-focal disabling spasticity (N-FDS) patients who are resistant or intolerant to oral therapy. Methods: A markov model was developed to estimate clinical and economic outcomes for patients treated with ITB or CMM. Treatment effects, patients´ baseline characteristics, resource utilization and health utility values were taken from the EPICE study. The model was built in accordance with the Institut Guttmann´s clinical practice, a reference center in Spain. Unit costs were obtained from the cited center. The analysis was conducted from the Institut Guttmann´s perspective over a lifetime horizon with direct medical costs (2013) and outcomes discounted at 3%. Uncertainty was assessed through univariate and multivariate sensitivity analysis (SA). Results: When comparing ITB with CMM, the model estimates ITB would increase remaining lifetime costs by € 35,605 and result in a QALY gain of 1.06; thus showing an incremental cost-effectiveness ratio (ICER) of € 33,619 per QALY gained. SA reflecting the most current clinical practice at the Institut Guttmann (where now a new catheter, associated with an 80% reduction in adverse events, is used) showed an ICER of € 27,805 per QALY gained. ICER results were also sensitive to changes in the post-operation hospitalization period, baclofen dose titration management, pump battery life, and health utility values. Conclusions: The present evaluation results in an ICER of ITB against CMM, in the treatment of N-FDS patients who are resistant or intolerant to oral therapy at the Institut Guttmann, close to a willingness-to-pay threshold of € 30,000 per A400 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 QALY gained. However, SA results indicate that the ICER drops below this threshold when most current clinical practice at the cited center is considered. Moreover, SA results suggest potential ways to optimize the current clinical pathway in order to reduce procedure costs even further. PND54 Long-Term Costs and Consequences of Patients with Familial Chylomicronemia Syndrome – A Simulation Model Approach Lin F , Thomas S , Calado F , Clegg J Novartis, East Hanover, NJ, USA . . . . Background: Familial chylomicronemia syndrome (FCS) is a rare genetic disorder characterized by deficiency of lipoprotein lipase, causing accumulation of chylomicrons. An estimated 0.1-0.2 per 100,000 people has FCS worldwide. FCS patients present massively elevated triglyceride levels (typically > 2,000 mg/dL), resulting in increased risk of recurring acute pancreatitis. Standard triglyceride lowering medications are ineffective for FCS patients, who rely on restrictive low fat diet to control their triglyceride. There is limited literature about longterm progression, the burden of illness or consequences of acute pancreatitis for FCS. Objectives: To estimate long term disease progression, costs and consequences of FCS. Methods: An individual Monte Carlo simulation model was built to track disease progression of a cohort of FCS patients with a mean age of 37.8 years, 60% male, and a mean triglyceride level of 2,741 mg/dL. The model projected the number of acute pancreatitis events, mortality and medical costs. Benefits of a hypothetical triglyceride reduction intervention were assessed. Results: With standard diet control, the average life expectancy of the studied cohort was estimated to be 16.45 years. These patients were expected to experience 10.16 episodes of acute pancreatitis during their lifetime, resulting in 80.7 inpatient days. The discounted lifetime cost of acute pancreatitis was projected to be $154,126 per patient. The cumulative mortality due to acute pancreatitis was estimated to be 54.3%. Should an intervention reduce triglyceride levels by 50% in FCS patients, the life expectancy would be increased by 3.16 years and 7.72 fewer episodes of acute pancreatitis would occur, preventing 61.21 inpatient days and saving $118,594 in medical cost. Conclusions: FCS patients are at high risk of lifethreatening and costly acute pancreatitis. Reduction in triglyceride levels has a significant impact of morbidity and mortality associated with acute pancreatitis. An effective triglyceride lowering intervention could mitigate the consequences of FCS significantly. PND55 Working Ability and Monetarily Valued Productivity of Patients with Multiple Sclerosis Treated with Natalizumab Kreimendahl F 1, Rychlik R P T 1, Patel S 2, Gleissner E 2, Becker V 3 1Institute of Empirical Health Economics, Burscheid, Germany, 2Biogen Idec GmbH, Ismaning, Germany, 3Neurologische Praxis Eppendorf, Hamburg, Germany . . . . . . . Objectives: Relapsing remitting multiple sclerosis (RRMS) is a chronic inflammatory disease that represents the most common chronic neurological disorder in young adults. RRMS often leads to disability and is a major cause of reduced working capacity due to neurological diseases. Aim of this study is to investigate patients’ working productivity during treatment with natalizumab. Methods: RRMS-Patients treated with natalizumab for a maximum of three months prior to baseline were eligible for study participation. Patients completed the EQ-5D and a questionnaire focused on occupational status, working ability, and days absent from work at study start study, after 6 months and 12 months. Socio-demographic and clinical data were collected. Primary endpoint was work productivity, which is defined as hours worked. To estimate costs and cost offsets due to the therapy with natalizumab, an average monetary value per working hour has been calculated to value productivity monetarily. Results: Preliminary results including 95 patients after 6 months of therapy show a significant increase in work productivity of 84 hours (p= 0.014) for the whole study population compared to baseline. The average reduction of days absent from work was 4.8 days compared to baseline. The increase in working hours leads to an average change of monetarily valued productivity per patient and half-year from 6,550.40€ at baseline to 7,600.35€ after 6 months for the entire study population. A subgroup analysis has been run on the group of employed patients. The working hours increased significantly by 63h per half-year (p= 0.025). Conclusions: In this study, the preliminary results show a significant increase in work productivity, which leads to a significant increase in the monetary value of productivity, for RRMS-patients treated with natalizumab. An increase in working hours as well as a decrease in days absent from work led to an increase in monetarily valued productivity. Study was funded by Biogen-Idec. Neurological Disorders – Patient-Reported Outcomes & Patient Preference Studies PND56 Using a Panel Survey to Identify Predictors of Disease-Modifying Drug Adherence in Patients with Multiple Sclerosis Phillips A L 1, Kozma C M 2, Locklear J C 1 Serono, Inc., Rockland, MA, USA, 2Independent Research Consultant, Saint Helena Island, SC, USA . . . . . . 1EMD Objectives: Identify predictors of multiple sclerosis (MS) disease-modifying drug (DMD) adherence. Methods: A random sample of adult MS patients from the US National Health and Wellness Survey or Lightspeed Research panel completed an internet survey in Nov/Dec2012. Clinical trial–naïve subjects with relapsing-remitting MS who were on their current DMD for ≥ 4months were included. Adherence was measured using the 4-item Morisky Scale. Patients were assigned to ‘high’ (negative response to all 4 questions: forget to take, careless at times about taking, stop if better, stop if worse) or ‘intermediate/low’ (any positive response). The survey contained demographic, disease characteristic and health care experience variables, which were evaluated as potential predictors of DMD adherence using logistic regression, controlling for age and gender. No adjustment was made for multiplicity. Results: 969 patients completed the survey; 579 met analysis criteria. High vs. intermediate/low adherers represented 47.7% and 52.3%, respectively. Average age for high adherers was 49.1 (SD: 10.9), 81.5% female; intermediate/ low adherers had an average age of 47.3 (SD: 11.4), 87.7% female. Of 149 variables, the following were associated with greater odds of high adherence: detailed discussions with health care professional about “how long the treatment had been available” (p= 0.02; OR: 1.47; high 56.8% vs. intermediate/low 47.1%) or “long-term safety” (p= 0.01; OR: 1.54; high 54.8% vs. intermediate/low 44.2%); satisfaction with DMD (p= 0.04; OR: 1.14; high 5.4 [SD: 1.4] vs. intermediate/low 5.2 [SD: 1.3] ); or insurance covered entire cost of DMD (p= 0.01; OR: 1.71; high 23.1% vs. intermediate/low 15.2%). Results indicating lower odds of high adherence included longer duration on current therapy (p< 0.01; OR: 0.99; high 63.2 [SD: 53.5] months vs. intermediate/low 75.9 [SD: 60.0] months) and cost made them “skip a dose” (p< 0.01; OR: 0.37; high 5.3% vs. intermediate/low 13.4%) or “not fill or refill a prescription” (p< 0.01; OR: 0.43; high 5.3% vs. intermediate/low 12.0%). Conclusions: Patient and provider dialogue, patient satisfaction with treatment and health plan benefit design aspects may affect DMD adherence. PND57 Modelling the Persistence of Disease-Modifying Drug Treatment (DMT) and its Independent Drivers in Finnish Multiple Sclerosis (MS) Patients: Parametric Survival Modelling Soini E 1, Holmberg M 2, Asseburg C 1, Sumelahti M L 2 1ESiOR Oy, Kuopio, Finland, 2School of Medicine, University of Tampere, Finland . . . . . Objectives: Explore how MS DMT-persistence can be modelled, compare model’s performance and assess the independent drivers for DMT-persistence. Methods: Analysis was based on 1638 DMT uses (1.1. 1991–31.12.2010, time at risk 4009 years) in incident MS-patient register from Tampere, Vaasa and Seinäjoki regions, Finland. DMT-persistence = DMT end-day - DMT initiation day. Cox, exponential, generalized gamma, Gompertz, log-logistic, log-normal, and Weibull regression survival models were used to model DMT-persistence. Models were compared based on goodness-of-fit statistics (Akaike and Bayesian information criteria). Results: Mean follow-up from first MS-symptoms and age at first DMT-initiation were 13 and 36 years, respectively. 73% of patients were female. Based on the data exploration of all known covariates, three DMT-persistence approaches with different interpretations, selected covariates and data needs were modelled: 1] sex, birth year, time from symptoms to DMT, age, DMT line (1st, 2nd, 3rd, 4th, 5-7th), DMT (interferon-β -1a and -1b, glatiramer acetate, other) at DMT initiation; 2] approach 1] + Expanded Disability Status Scale (EDSS) at DMT initiation; and 3] approach 2] + DMT-discontinuation reason (pregnancy plan, flu-like symptoms, injection-site reactions, ineffectiveness, antibodies, other/unknown). There was no gold standard survival model for DMT-persistence, and some models accommodated higher number of covariates and associated dependencies better. For approaches 1] and 2] Weibull and for 3] Gompertz model provided the best goodness-of-fit. Based on all three models (one per approach 1] –3]) with best goodness-of-fit, higher EDSS or higher age at DMT initiation, treatment line (3rd and later), and incidence of intolerable adverse events (AE) or ineffectiveness were independently associated with shorter DMT persistence. In the approach 3], flu-like symptom and injection site AEs had the highest hazard ratio for shorter DMT-persistence. Conclusions: AEs, EDSS, age, treatment line and ineffectiveness were strong predictors for DMTpersistence. Flu-like symptoms and injection site AEs showed the highest hazard for DMT-discontinuation. PND58 Persistence with Fingolimod Versus Dimethyl Fumarate in Patients with Multiple Sclerosis: Retrospective Analysis of us Open-Source Pharmacy Data Bergvall N 1, Lahoz R 1, Nazareth T 2, Korn J R 3 1Novartis Pharma AG, Basel, Switzerland, 2Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA, 3IMS Health, Waltham, MA, USA . . . . . Objectives: To compare 6-month persistence rates among patients initiating the oral multiple sclerosis (MS) disease-modifying therapies (DMTs) fingolimod and dimethyl fumarate (DMF). Methods: Our retrospective analysis used mail-order pharmacy claims from the US open-source LRx™ database (IMS). Patients with ≥ 1 fingolimod or DMF prescription (index DMT) between 01-April-2013 and 31-July2013 were included. Patients were ≥ 18 years old, naive to fingolimod and DMF, and had not received multiple DMTs on the date of the first index DMT claim (index date). Prescription records were collected from pharmacies supplying ≥ 1 index DMT claim between the index date and the last month of follow-up. Persistence was assessed as time from initiating index DMT until discontinuation (gap of ≥ 60 days), receipt of another DMT or the end of the 6-month follow-up period. The risk of and time to index DMT discontinuation was assessed using a Cox proportional hazards model (controlling for age, gender and region) and Kaplan-Meier analysis, respectively. Results: The study included 9546 patients (fingolimod: n= 1390; DMF: n= 8156). The proportion of patients discontinuing index DMT was significantly lower for patients receiving fingolimod (23.3%) versus DMF (36.6%; p< 0.0001). The risk of discontinuation was 1.6-fold higher in the DMF cohort versus the fingolimod cohort (hazard ratio, 95% confidence intervals: 1.58, 1.41-1.77; p< 0.0001). Time to discontinuation was significantly longer with fingolimod than with DMF (p< 0.0001), resulting in a longer duration of therapy persistence for fingolimod versus DMF (mean ± standard deviation: 152±53 days versus 135±62 days, respectively). Results were similar when discontinuation was defined as a gap of ≥ 30 days (p< 0.0001 for all outcomes). Conclusions: This analysis provides the first insight into short-term persistence rates with oral DMTs. In a real-world setting, the risk of discontinuation over 6 months was lower for patients initiating fingolimod versus DMF. A401 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PND59 Persistence in Open and Closed Data Sources: A Study of Fingolimod VersUS Interferons/Glatiramer Acetate in Patients with Multiple Sclerosis Lahoz R 1, Bergvall N 1, Nazareth T 2, Korn J R 3 1Novartis Pharma AG, Basel, Switzerland, 2Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA, 3IMS Health, Waltham, MA, USA . . . . . Objectives: To compare 6-month persistence rates among patients receiving the multiple sclerosis (MS) disease-modifying therapies (DMTs) fingolimod or interferon/ glatiramer acetate (IFN/GA) (index DMT), using open- and closed-source data that reflect unrestricted or continuous health care coverage, respectively. Methods: Retrospective analyses used administrative claims and mail-order pharmacy databases (IMS PharMetrics Plus™ [closed]) and LRx™ [open], respectively). All patients were ≥ 18 years old and naive to fingolimod and index DMT, had ≥ 1 prescription for index DMT between 01-Oct-2010 and 31-Mar-2013 and had not received multiple DMTs at index date. An additional PharMetrics cohort was selected using more stringent criteria (continuous enrolment pre-/post-index; MS diagnosis code). LRx prescriptions were collected from pharmacies supplying ≥ 1 claim for index DMT between the index date and the last month of follow-up. Persistence was defined as time from initiating index DMT until discontinuation (gap ≥ 60 days), receipt of another DMT or end of the 6-month follow-up period. Risk of and time to discontinuation were assessed by a Cox proportional hazards model (controlling for age, gender and region) and Kaplan-Meier analysis, respectively. Results: Using identical criteria, 22,467 PharMetrics patients (fingolimod: n= 1968; IFN/GA: n= 20,499) and 49,803 LRx patients (8325 and 41,478, respectively) were selected. Proportions of patients discontinuing index DMT were significantly lower for fingolimod vs IFN/GA (PharMetrics: 23.1% vs 27.2%; LRx: 26.9% vs 33.4%; p<0.0001). Risk of discontinuation was higher for IFN/GA vs fingolimod (PharMetrics: hazard ratio, HR= 1.18; 95%CI: 1.07–1.30, p= 0.0008; LRx: HR= 1.23; 95%CI: 1.17–1.29, p< 0.0001). Time to discontinuation was significantly longer for fingolimod vs IFN/GA (PharMetrics: p= 0.0005; LRx: p< 0.0001). With more stringent criteria used in PharMetrics, risk of discontinuation of IFN/GA vs fingolimod increased (HR= 1.47; 95%CI: 1.24–1.74, p< 0.0001). Conclusions: Both data sources provided similar results, supporting use of open-source LRx data, which allows access to up-to-date information that can improve sample size and statistical power. PND60 Comparison of the Valuation of Treatment Alternatives in Parkinson’s Disease with Best-Worst Scaling, Time Trade-off and Visual Analogue Scales Weernink M G M 1, Groothuis-Oudshoorn C G M 1, IJzerman M J 2, van Til J A 1 1University of Twente, Enschede, The Netherlands, 2University of Twente and MIRA institute for Biomedical Technology & Technical Medicine, Enschede, The Netherlands . . . . . . . . . . Objectives: Traditional valuation methods are insensitive to small improvements in process and outcome of care. Best-Worst scaling (BWS) was proposed as a sensitive and efficient method to determine the relative value of different treatments for the same disease, which would be desirable to estimate cost-effectiveness. The study objective was to compare the ability of BWS to differentiate between different treatment alternatives to that of Time Trade Off (TTO) and Visual Analogue Scales (VAS). Methods: An online survey was conducted to estimate individual values for six different treatments reflecting the real-life options in the treatment of Parkinson’s Disease with BWS2, BWS3, TTO and VAS (n= 592). Pearson correlation coefficient was used to examine the strength of linear dependence between estimated utility scores. Results: Twenty-seven percent of respondents was not willing to trade life years in TTO. Only two percent of the respondent does not differentiate between the value of health states with VAS. When non-traders were excluded from the analysis, the best case scenario was valued significantly higher than the worst case scenario with all methods. Rank reversals among intermediate alternatives were common. The correlation between utility scores was very strong (VAS-BWS2 1,0; VAS-BWS3 0.98; TTO-BWS2 0.99; TTO-BWS3 0.98, BWS2BWS3 0.96; P< 0.000, n= 434). Conclusions: The results demonstrate that BWS, TTO and VAS can be used to elicit incremental utility gain of small improvements in care. However, all methods have limitations. VAS does not result in utilities and some respondents do not trade with TTO. While the use of BWS is attractive because of its ability to estimate utilities for many different treatment alternatives, its applicability in CEA is limited because BWS utilities are not anchored on a 0-1 utility scale. We propose to use TTO to estimate utility for extreme health states, and to use BWS to value intermediate health states which differ on process characteristics. PND61 Health-Related Quality of Life in Migraine Without Aura Based on Attack Frequency: A time Trade-Off Study Rencz F 1, Brodszky V 1, Péntek M 1, Bereczki D 2, Gulacsi L 1 University of Budapest, Budapest, Hungary, 2Semmelweis University, Budapest, Hungary . . . . . 1Corvinus Objectives: To evaluate health-related quality of life (HRQOL) in migraine based on attack frequency by time trade-off method (TTO) in a mixed population sample consisting of migraneurs and non-migraneurs. Methods: A cross-sectional questionnaire survey was designed to measure HRQOL in migraine without aura by TTO. A convenience sample was recruited from university students and staff regardless of having ever experienced migraine or not. Participants were asked to elicit two hypothetical health states characterised by different migraine frequency (‘m’: two migraines lasting 4 hours each month and ‘w’: each week) within two hypothetical lifetime frames (20 years left to live/lives until the age of 80 years). Utilities were calculated for the four tasks (U20m, U80m, U20w, U20w) and compared between subgroups. Results: Altogether 180 respondents were included in the analysis. Mean age was 25.6 years (SD 6.4), 71% were female and 61% were self-reported migraneurs. Proportion of non-traders varied between 14% and 43% within the four questions. Respondents were willing to sacrifice median 5 and 8 years of their remaining lifespan until the age of 80 years for avoiding two migraines per month and per week, respectively. Mean utilities for two migraines each month were U20m= 0.84 (SD 0.26) and U80m= 0.89 (SD 0.14), and for each week were U20w= 0.79 (SD 0.27) and U80w= 0.83 (SD 0.17), respectively. Self-reported migraneurs elicited higher mean utilities compared to non-migraneurs for all migraine health states but this difference was statistically significant only for U80m (p= 0.039). Also, males attached lower mean utilities for all health states than females but this was significant only for U80m (p= 0.018). Besides, older respondents valued higher mean utilities for U20m compared to the younger ones (p= 0.040). Conclusions: Our findings provide the first time trade-off utilities on migraine associated HRQOL impairment. Disutility caused by migraine ranged between 0.1 and 0.2 depending on attack frequency. PND62 Humanistic Research Outcomes in Multiple Sclerosis: Review of the Literature from Latin America Einarson T R 1, Bereza B 1, Machado M 2 1University of Toronto, Toronto, ON, Canada, 2Biogen Idec, São Paulo, Brazil . . . . Objectives: This research reviews the research literature reporting humanistic outcomes related to multiple sclerosis (MS) in Latin America. Methods: We conducted a systematic search of Medline, Embase, LILACS and Scielo from inception through 2013 for articles reporting original research on quality of life (QoL), utility scores for states of MS, patient preference, mental health, social and emotional wellbeing in people with MS in Latin America. Adherence and related issues were not included. Outcomes were categorized into: mental domain (cognitive function; mental health), physical domain (mobility/independence; fatigue; restless legs syndrome), employment, QoL, caregiver burden, and patient preference. Results: A total of 38 studies were selected for analysis. Among them, 23 addressed issues in the mental domain (9 cognitive functions and 14 mental health), 41 in the physical domain (24 mobility/physical function, 15 fatigue and 2 restless legs syndrome). One addressed impact of MS on employment, 16 QoL, 2 caregiver burden and 1 patient preference. Researchers used 56 different instruments to collect their data from 2286 patients. Compared with controls, MS patients had significantly (P< 0.05) lower levels of functioning, cognition and increased presence of mental illness. All of these factors were significantly associated with decreased QoL in patients (odds ratios ranged from 4.2-10.1; P< 0.05). Similarly, fatigue and restless legs syndrome correlated significantly with anxiety, depression and level of mobility/functioning as well as QoL. Conclusions: As in other parts of the world, MS exerts a substantial negative impact on the lives of people with MS in Latin America. It lowers their QoL and interferes with their ability to move about, care for themselves and work. Their social life is also negatively affected. The amount of literature on this subject is quite limited. More research in Latin America is needed to understand humanistic outcomes in these patients and management of their MS. PND63 The Effect of Insomnia and Insomnia Treatment Side Effects on Health Status, Work Productivity, and Health Care Resource Use DiBonaventura M 1, Richard L 2, Kumar M 1, Forsythe A 3, Moline M 4, Flores N 1 1Kantar Health, New York, NY, USA, 2Eisai Europe Ltd, Hatfield, UK, 3Eisai, Inc., Woodcliff Lake, NJ, USA, 4Eisai Inc, Woodcliff Lake, NJ, USA . . . . . . Objectives: The aims of this study were to quantify the burden of insomnia and to quantify the association between side effects of insomnia medications and health outcomes. Methods: Data from the 2013 US (N= 75,000) and 5EU (N= 62,000) National Health and Wellness Survey (NHWS) were used. The NHWS is a patientreported survey administered to a demographically representative sample of adults (with respect to age, sex, race/ethnicity, and region). Those who met DSM-V criteria for insomnia and, separately, those treated with insomnia were compared with their respective propensity score-matched control groups on health status (SF-36v2), work productivity (WPAI-GH), and health care resource use using ANOVA tests. Among those with treated insomnia, those with and without side effects were compared on health outcomes using general linear models controlling for demographics, health history, and comorbidities. Results: Compared with their respective matched control groups, patients with insomnia (n= 4147) and treated insomnia (n= 2860) in the 5EU reported significantly worse mean health utilities (0.60 vs. 0.74; 0.60 vs. 0.74, respectively), greater overall work impairment (38.74% vs. 14.86%; 39.50% vs. 15.66%), and more annual physician visits (9.10 vs. 4.08; 9.58 vs. 4.11). Similar findings were observed in the US cohort. Among those treated for insomnia, 13.56% and 24.55% in the US and 5EU, respectively, were non-adherent due to side effects. In the US, this behavior was associated with significantly worse health utilities (0.60 vs. 0.64) and greater overall work impairment (37.71% vs. 29.08%), among other variables (all p< .05). These relationships were not significant in the 5EU. Conclusions: A significant humanistic and economic burden of insomnia was observed in both the US and 5EU, and the burden remains even after treatment. Non-adherence due to side effects was common and, in the case of the US, associated with significantly poorer health outcomes. PND64 Quality of Life Among Patients with Multiple Sclerosis Treated with Prolonged-Release Fampridine 10 Mg Tablets for Walking Impairment Liu Y 1, McNeill M 2, Lee A 1, Zhong J 1, Mehta L R 1 1Biogen Idec, Cambridge, MA, USA, 2Biogen Idec, Maidenhead, UK . . . . . . Objectives: To evaluate the effect of prolonged-release (PR) fampridine 10 mg tablet on generic quality of life (QoL) as measured by the EQ-5D in patients with multiple sclerosis (MS) with walking impairment. Methods: The study population included 132 patients who enrolled in a 24-week randomized, double-blind, and placebo-controlled phase 2 trial (NCT01597297) of PR-fampridine 10 mg tablets or placebo twice daily in multiple sites in Europe and Canada. Patients were A402 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 categorized into three groups: placebo, PR-fampridine responders (those with a mean improvement from baseline in the 12-item MS walking scale [MSWS12] of ≥ 8 points over 24 weeks), and PR-fampridine nonresponders (those with worsening, no change, or < 8 points improvement in MSWS-12). Changes from baseline were calculated for the EQ-5D utility index and visual analogue scale (VAS) by visit and over the 24 weeks. Within-group and pairwise comparisons were assessed based on the least square (LS) means using analysis of covariance (ANCOVA) models adjusting for baseline EQ-5D scores. Results: The placebo patients (n= 64), responders (n= 33), and nonresponders (n= 35) were similar in baseline age, race, weight, and number of relapses in the past 1 and 2 years. The responders had higher mean baseline EQ-5D utility (0.56 vs. 0.51 and 0.52, respectively) and VAS (62.9 vs. 59.1 and 60.5, respectively) scores than the placebo patients and nonresponders. Over the 24 weeks, the EQ-5D utility score improved in the responders (mean change: 0.06, 95% CI: [0.01, 0.12]) but worsened in the placebo patients (mean change: -0.03,95% CI: [-0.07, 0.01]) and nonresponders (mean change: -0.07,95% CI: [-0.12, -0.02] ). Similar trends were observed in the VAS scores although the differences within groups were generally not statistically significant. Conclusions: PR-fampridine 10 mg tablets demonstrated significantly greater improvement in QoL among responders, despite starting from higher baseline EQ-5D scores, than the placebo or nonresponder groups. PND65 The Burden of Primary Generalized Tonic-Clonic Seizures in Europe and the United States: An Analysis of The National Health and Wellness Survey Gupta S 1, Forsythe A 2, Pomerantz D 1, Tsong W 2 Health, Princeton, NJ, USA, 2Eisai Inc., Woodcliff Lake, NJ, USA . . . . 1Kantar Objectives: Many prescription medications are available to treat Primary Generalized Tonic-Clonic (PGTC) seizures. The objective is to understand the impact of PGTC on health outcomes. Methods: Data from the 2011-2013 U.S. and 2010 & 2013 5E.U. National Health and Wellness Survey (NHWS) were analyzed. The NHWS is self-administered, internet-based survey of a nationwide sample of adults (18+ years) stratified to represent the demographic composition of each country. Patients self-reported a diagnosis of epilepsy with PGTC and were grouped as ≥ 1 seizure per week, 1-3 seizures per month, 1-4 seizures per year, < 1 seizure per year (reference). Patients provided information on health status (mental (MCS), physical component summary (PCS), and SF-6D (health utility) from the SF-36v2), resource utilization in the past six months and productivity loss (Work Productivity and Activity Impairment Questionnaire). Costs were estimated from the literature. Regression analyses were conducted controlling for covariates (e. g., age, gender, marital status, comorbidities, years diagnosed with epilepsy). Results: There were 782 patients in the U.S. and 418 in the 5EU. In both geographies the proportion of employed respondents decreased as frequency of seizures increased. Over 75% of patients in both geographies were taking a prescription medication for epilepsy. After adjustments, patients reporting < 1 seizure per year had greater PCS, health utilities, lower activity impairment, fewer emergency room visits and lower total direct costs compared to the other three seizure groups (p< 0.05 both geographies). Among employed patients, patients reporting < 1 seizure per year had the lowest overall work impairment (p< 0.05 both geographies). For all employable patients (18-60 years) indirect costs were highest for the ≥ 1 seizure per week group (p< 0.05 both geographies). Conclusions: Results suggest a significantly higher economic and humanistic burden in patients with more frequent seizures. PGTC patients are very much in need of newer treatment options. PND66 Factors Associated with Caregiver’s Burden in Relapsing-Remitting Multiple Sclerosis and Satisfaction with Current Therapies. MS-Feeling Study Balaña M 1, Fabregas M 2, Meca-Lallana J 3, Mendibe M 4, Garcia E 1 Farmaceutica, Barcelona, Spain, 2TFS Develop S.A., Barcelona, Spain, 3Hospital Virgen de la Arrixaca, Murcia, Spain, 4Hospital de Cruces, San Vicente de Barakaldo, Spain . . . . . 1Novartis Objectives: To describe the characteristics and burden of caregivers and patients affected by relapsing-remitting multiple sclerosis (RRMS), and to describe their satisfaction with treatment. Methods: Multicenter, observational, cross-sectional study in RRMS patients≥ 18 years (y), treated for ≥ 1y. Burden was assessed with the Zarit Burden Interview. Other measures were: Center for Epidemiologic Studies Depression Scale, short form (CESD-7); and treatment satisfaction of caregivers (adhoc questionnaire) and patients (Treatment Satisfaction Questionnaire for Medication (TSQM)). Results: We included 180 patients (mean (SD) age 41 (11) y, 66% female) and caregivers (47 (12) y, 56% female, 67% economically active). Most caregivers were relatives (59% partner; 25% parent; 9% brother or sister) and lived with the patient (86%); 37% helped with the medication. Median (Q1, Q3) time since diagnosis was 7 (4,10) y and between diagnosis and first treatment, 0.3 (0.2, 1.3) y. Most patients received monotherapy with interferon beta (51%), glatiramer acetate (20%), natalizumab (14%) or fingolimod (7%); 15.6% received ≥ 2 drugs. Median EDSS was 2.5 (1,4). According to the Zarit Interview, 19% of caregivers had some degree of burden (median: 10 (7,15)). Factors associated with burden were: EDSS (mean (SD) in caregivers with burden vs without burden, respectively: 4.0 (2.1) vs 2.6 (1.9), p<0.0005), years of caregiving (mean: 4.0 (4.5) vs 2.7 (4.5) y, p= 0.005), daily hours dedicated to patient (10.8 (12.1) vs 5.8 (4.6) hours, p=0.038), and >1 drug (39% of burden when patients take≥2 drugs vs 16% with monotherapy, p=0.004). Other caregiver’s characteristics (age, gender, professional activity, relationship with patient or cohabitation) were not significantly associated. 28.2% of caregivers had depression (CESD-7≥ 15). Both patients (90% satisfied or very satisfied according to the TSQM) and caregivers (mean of 7.6 (2.3) on a scale from 0-10 [maximum satisfaction]) were quite satisfied with treatment. Conclusions: In the RRMS population with moderate disability, around 2 in 10 caregivers have some degree of burden, and 1 in 4 suffer depression. The burden increases proportionately with disease severity and number of medications administered. Overall, patients and caregivers are satisfied with treatment. PND67 Impact of Caregiving for Patients with Alzheimer’s Disease and Dementia on Psychiatric and Clinical Comorbidities in Brazil Goren A 1, Novick D 2, Barros B R 3, Laks J 4, Dueñas H 5, Kahle-Wrobleski K 6 1Kantar Health, New York, NY, USA, 2Eli Lilly and Company, Windlesham, UK, 3Eli Lilly do Brasil Ltda, São Paulo, Brazil, 4Federal University of Rio de Janeiro, Rio de Janeiro, Brazil, 5Eli Lilly de Mexico, Mexico City, Mexico, 6Eli Lilly and Company, Indianapolis, IN, USA . . . . . . . Objectives: Patients with dementia due to Alzheimer’s disease (AD) are often cared for by family members who experience physical, psychological, social, and financial burdens associated with provision of care. This study quantifies the impact of caregiving in Brazil and helps identify characteristics associated with caregiving. Methods: Data were analyzed from the 2012 National Health and Wellness Survey (NHWS) in Brazil (n = 12,000), an Internet-based survey of adults (aged 18+), using stratified random sampling (by sex and age) to ensure demographic representation of the Brazil adult population. Caregivers were compared with non-caregivers on select comorbidities considered potential outcomes of caregiving, plus sociodemographic characteristics, health characteristics and behaviors, and Charlson comorbidity index (CCI) scores. Binary logistic regression models assessed comorbidities associated with caregiving, adjusting for potential confounds (CCI, age, gender, education, income, insurance, and marital status). Results: Among 10,853 respondents (caregivers [n=209]; non-caregivers [n= 10,644]), caregivers were on average 42.1 years old, and 53% were female, 52% married/living with a partner, 87% insured, and 42% living with 1+ children in the household. Caregivers vs. non-caregivers were more frequently obese, smokers, insured, employed, and with college education or above, and they had higher CCI scores and higher income, all p <.05. Adjusting for covariates, caregiving was associated with significantly increased risk of depressive symptoms (odds ratio [OR] =2.008), major depressive disorder (OR=1.483), anxiety (OR=1.714), insomnia (OR= 1.644), hypertension (OR=1.584), pain (OR=1.704), and diabetes (OR= 2.103), all p<.015. Conclusions: This is the first study on caregivers for persons with dementia in Brazil using a large dataset (NHWS). Being a caregiver (compared with noncaregiver) is a predictor of overall psychiatric and clinical disorders in this sample. The online survey format provides certain sampling advantages but may under-represent caregivers without access/comfort with online technology. Direct treatment and policies to help caregivers are needed in Brazil. PND68 The Value of Diagnostic Tests for Alzheimer’s Disease: DiscreteChoice Experiment and Contingent-Valuation Mühlbacher A C 1, Johnson F R 2, Yang J C 3, Happich M 4 of Applied Sciences Neubrandenburg, Neubrandenburg, Germany, 2Triangle Health Preference Research, Chapel Hill,, NC, USA, 3RTI Health Solutions, Research Triangle Park, NC, USA, 4Eli Lilly and Company, Bad Homburg, Germany . . . . . . . 1University Objectives: Despite the existence of standardized medical criteria, clinical diagnosis of Alzheimer’s disease remains difficult. Lack of diagnostic certainty or possible distress related to positive results could limit application of new testing technologies. Independent of the therapeutic value of prevention or cure, however, diagnostic information could have value in informing contingency planning or have intrinsic value: the value of “just knowing”. This paper aims to quantify respondents’ preferences for obtaining AD diagnostic tests and to estimate the perceived value of AD test information. Methods: Discrete-choice experiment (DCE) and contingentvaluation (CV) questions were administered to N= 1301 respondents aged 60 years or older in Germany and the United Kingdom. 12 pairs of virtual AD diagnostic tests were presented (defined by test type, test precision defined by false-positive or falsenegative test results, and test cost) and a no-test alternative. DCE questions were based on a predetermined experimental design. A double-bounded, dichotomouschoice CV question was used to further elicit willingness to take an AD test and pay for it. Choice data of respondents interested in taking a test were analyzed using random-parameters logit. A probit model characterized respondents who were not willing to take a test. Results: Most respondents in both countries had a positive value for AD diagnostic test information. Most respondents who indicated an interest in testing in Germany (N= 631) and the UK (N= 670) preferred brain imaging without radioactive marker, followed by brain imaging with radioactive marker and then spinal tap. Diagnostic tests with better precision were preferred. German respondents had relatively greater heterogeneous preferences and lower moneyequivalent values for test features compared to UK respondents. Conclusions: Respondents preferred less invasive diagnostic procedures and tests with higher accuracy. Diagnostic test accuracy was more important than test type. Respondents expressed a willingness to pay up to € 700 to receive a less invasive test with the highest accuracy. PND69 The Humanistic and Economic Burden of Partial Onset Seizures in the Europe Five and Brazil Using a Patient Survey Gupta S 1, Forsythe A 2, Pomerantz D 1, Tsong W 2 Health, Princeton, NJ, USA, 2Eisai Inc., Woodcliff Lake, NJ, USA . . . . 1Kantar Objectives: The aim of this study was to understand the current impact of partial onset seizures (POS) on health outcomes and costs. Methods: Patients were identified from the 2010 & 2013 5EU and 2011-2012 Brazil National Health and Wellness Survey, a nationally represented, internet-based survey of adults (18+ years). Patients self-reported a diagnosis of epilepsy with POS and were categorized into ≥ 1 seizure per week, 1-3 seizures per month, 1-4 seizures per year, < 1 seizure per year (reference). Patients completed the SF-36v2 (mental (MCS), physical component summary (PCS)), SF-6D (health utility), Work Productivity and Activity Impairment Questionnaire and reported on resource utilization in the past six months. Costs were estimated from the literature. Generalized linear regression analyses were conducted controlling for covariates (e. g., age, gender, marital status, comorbidities, years diagnosed with epilepsy). Results: There were 175 patients in the 5EU and 32 in Brazil, 11.1% reported ≥ 1 seizures/week, 7.7% reported 1-3 seizures/month, 26.1% 1-4 seizures/year, and 55.1% < 1seizures/year. MCS, PCS, utilities, productivity A403 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 loss and resource were similar in Brazil and the 5EU (all p> 0.05). After adjustments (combined 5EU and Brazil), as the number of seizures decreased, PCS (≥ 1 seizures/ week: 40.9; 1-3 seizures/month: 41.7; 1-4 seizures/year: 46.9; < 1seizures/year: 48.4) and health utilities (≥ 1 seizures/week: 0.58; 1-3 seizures/month: 0.63; 1-4 seizures/ year: 0.63; < 1seizures/year: 0.69) increased. As the number of seizures decreased, activity impairment (≥ 1 seizures/week: 58.6%; 1-3 seizures/month: 51.1%; 1-4 seizures/year: 35.7%; < 1seizures/year: 25.5%), emergency room visits and hospitalizations decreased. In Europe direct and indirect costs were highest for the ≥ 1 seizure/ week group. Conclusions: POS burden was similar in Europe and Brazil. As the number of seizures increased, economic and humanistic burden increased. Even with the large number of treatment options available, patients and the health care system need additional choices to reduce the frequency of seizures. PND70 Psychometric Analyses to Inform Item Reduction and Evaluate Sensitivity of the Early Mobility Impairment Questionnaire for Multiple Sclerosis Phillips G A 1, Mathias A 2, Coon C D 2, Agarwal S S 1, Sen R 2, Shah R 3, Ziemssen T 4 Idec Inc., Cambridge, MA, USA, 2Adelphi Values, Boston, MA, USA, 3University of Mississippi, University, MS, USA, 4University Clinic Carl Gustav Carus, Dresden, Germany . . . . . . . . . . 1Biogen Objectives: Mobility impairments affect most multiple sclerosis (MS) patients (93%), yet many patients (~40%) “rarely or never” discuss trouble walking with a health care provider (HCP). To facilitate patient-HCP dialogue and early identification of mobility impairments, the Early Mobility Impairment Questionnaire (EMIQ) was developed based on qualitative research with patients diagnosed with MS and in collaboration with key opinion leaders. To minimize the burden associated with administration of the EMIQ, psychometric analyses were applied to the initial draft to produce a parsimonious yet robust instrument. Methods: The EMIQ’s psychometric performance was evaluated in a multi-center, prospective, observational study in subjects with an Expanded Disability Status Scale (EDSS) score of 2.0 to 6.0. Iterative exploratory factor analysis (EFA) was used to guide item reduction; subsequently, an item response theory (IRT) graded response model was applied to confirm the robustness of the instrument. Logistic regression analyses (using EDSS scores) were used to confirm discriminatory power. Results: In total, 124 subjects with moderate MS (EDSS mean score = 4.2) were included in the study. Discriminant validity, factor analysis, and IRT modeling identified six items for reduction. The resulting 9-item scale had one strong underlying domain (first eigenvalue explained 60% of variance). The information curve showed sufficient sensitivity (I> 3.3 for -2.6≤ θ ≤ 2.4) throughout the scale (mild to severe), with the most precise information provided at ½ SD above the mean (I= 12 at θ = 0.4). Logistic regression showed that the EMIQ was able to discriminate between patients with and without walking impairment (p< 0.05). Conclusions: The psychometric analyses resulted in a streamlined scale that is appropriate for the screening of subjects experiencing early mobility impairments due to MS. In addition to a potential role in guiding clinical assessment and intervention, the EMIQ may also prove useful in the longitudinal assessment of mobility, as patients’ conditions decline. PND71 Cognitive Impairment and Health Related Quality of Life in Relapsing Remitting Multiple Sclerosis Balaña M 1, Galera J 1, López-Góngora M 2, Escartín A 2, Izquierdo G 3, Borges M 3, Garcia E 1 1Novartis Farmaceutica, Barcelona, Spain, 2Hospital Santa Creu i Sant Pau, Barcelona, Spain, 3Hospital Universitario Virgen Macarena, Sevilla, Spain . . . . . . . Objectives: The impact of Cognitive Impairment (CI) on Health Related Quality of Life (HRQoL) in patients with Relapsing Remitting Multiple Sclerosis (RRMS) is becoming a field of study increasingly relevant due to its implications on the effectiveness and quality research. Since the studies reviewed highlight contradictory results, the aim of this study was to determine the predictive value of CI and other correlated factors on HRQoL in a sample of RRMS patients. Methods: Observational, cross-sectional and multicenter study at 21 Neurology Departments in Spain. Multiple linear regression analysis (stepwise) was carried out to assess if HRQoL (EQ-5D scores) was predicted by CI (Brief Repeatable Battery of Neuropsychological Tests, BRB-N), physical disability (Expanded Disability Status Scale -EDSS), depression (Beck Depression Inventory-BDI-II) and disease duration (years diagnosed). Bivariate, partial and semi-partial correlations and multicollinearity analysis were performed to control confounding factors. Results: We included 291 RRMS patients (71.50% female), mean age 41.65 years (SD= 10.18). The mild disability group (EDSS 0-3) included 152 (52.20%) patients and the moderate disability group (EDSS 3.5-5.5) included 139 patients (47.80%). All correlations between EQ-5D scores and BRB-N, BDI-II, EDSS and disease duration variables were statistically significant (p< 0.01), not multicollinearity detected. The results of the regression analysis indicated that two predictors explained 56.9% of the HRQoL variance (Adjusted R squared= 0.569, F= 187,251, p< 0.01). It was found that depression significantly predicted HRQoL (Beta= -0.587, p< 0.01), as did physical disability (Beta= -0.315, p< 0.01). Conclusions: The results showed a weak predictive value of CI (measured with the BRB-N battery) in HRQOL scores while depression and psychical disability were important predictors. Future research is needed in order to clarify the relationship between CI and HRQoL. PND72 Evaluating Working Ability and Quality of Life of Patients with Multiple Sclerosis Kreimendahl F 1, Rychlik R P T 1, Patel S 2, Gleissner E 2, Becker V 3 1Institute of Empirical Health Economics, Burscheid, Germany, 2Biogen Idec GmbH, Ismaning, Germany, 3Neurologische Praxis Eppendorf, Hamburg, Germany . . . . . . . Objectives: Relapsing remitting multiple sclerosis (RRMS) is a chronic inflammatory disease representing the most common chronic neurological disorder in young adults. RRMS usually leads to permanent disability and as a result is a major cause of reduced working capacity due to neurological diseases. This abstract pre- sents methods to evaluate working ability and quality of life in natalizumab-treated RRMS patients. Methods: A non-interventional study was performed in Germany. Patients treated with natalizumab for a maximum of three months prior to baseline were eligible for study participation. Demographic and occupational information was collected; DSS and EQ-5D-questionnaires were used. Data were collected at baseline, after 6 months and after 12 months. Main objective was to evaluate productivity per year, by using the formula: (working hours per day*5 days per week*46 working weeks a year) - (days absent from work*working hours per year). To assess health economic cost savings, productivity has been valued monetarily by calculating a daily rate of working costs. Subgroup analysis divides the study population into employed, unemployed, patients with statutory sick pay and students. Analyses of variance and subsequent post hoc tests will be performed to identify subgroup differences. Results: Of 95 patients included by January 2014, 46.3% were employed. Average hours worked/day were 5.3 hours (employed: 6.5 hours); average number of days absent from work during the past half-year was 17.4 days (employed: 16.5 days). Calculated productivity per 6 months results in 569.6 hours (employed: 661.0), meaning 6,5540.40€ (employed: 7,634.55€ ) in monetarily valued productivity. Conclusions: This non-interventional study aimed at providing new insights in the therapy of RRMS patients treated with natalizumab. The study’s intention was to show how an increase in working hours in employed patients as well as a decrease in days absent from work can lead to an increase in productivity. Study was funded by Biogen Idec. PND73 Significant and Meaningful Improvement in Treatment Satisfaction with Teriflunomide Versus Subcutaneous IFNB-1A in Patients with Relapsing MS Results from Tenere Mäurer M 1, Van Wijmeersch B 2, de Seze J 3, Meca-Lallana J 4, Bozzi S 5, Vermersch P 6 Krankenhaus Bad Mergentheim, Bad Mergentheim, Germany, 2Hasselt University, Diepenbeek, Belgium, 3Strasbourg University, Strasbourg, France, 4Universidad Católica San Antonio de Murcia, Murcia, Spain, 5Sanofi, Chilly-Mazarin, France, 6University of Lille Nord de France, Lille, France . . . . . . 1Caritas Objectives: Teriflunomide is a once-daily oral immunomodulator approved for relapsing–remitting MS. The phase 3 TENERE (NCT00883337) study comparing teriflunomide with subcutaneous interferon beta-1a (sc IFNβ -1a) did not meet its primary endpoint (superiority of teriflunomide vs sc IFNβ -1a on time to treatment failure) although there was no difference in annualized relapse rate between teriflunomide 14 mg and sc IFNβ -1a. The objective of the current analysis was to compare patient treatment satisfaction of teriflunomide with that of sc IFNβ-1a. Methods: Randomized patients (n= 324) received once-daily teriflunomide 14 mg or 7 mg or sc IFNβ -1a three times per week; the study was completed 48 weeks after the last patient was randomized. Patient satisfaction with treatment was assessed as a secondary endpoint using the Treatment Satisfaction Questionnaire for Medication (TSQM) version 1.4, a validated generic questionnaire with scores for effectiveness, side-effects, convenience, and global satisfaction. A mixed-effect model with repeated measures was used to analyze TSQM scores at Week 48. Magnitude of effects was assessed using effect size (ES), defined as the difference in treatment effect divided by standard deviation. The ES differences were ranked as follows: <0.2, negligible; ≥ 0.2–< 0.5, small; ≥ 0.5–≤ 0.8, moderate; > 0.8, high. Results: At Week 48, TSQM values showed significantly better patient satisfaction in the teriflunomide 14-mg group vs the IFNβ -1a group in three domains (side-effects, P< 0.0001; convenience, P< 0.0001; global satisfaction, P= 0.02), with no perceived difference on effectiveness (P= 0.28). High ES values favoring teriflunomide 14 mg vs IFNβ-1a were seen with side-effects (1.09) and convenience (1.81), with a smaller ES for global satisfaction (0.39). Conclusions: A significant and meaningful improvement in treatment satisfaction for teriflunomide 14 mg vs IFNβ-1a was observed with regards to side-effects, convenience, and global satisfaction, which may potentially improve treatment adherence and outcomes in clinical practice. Neurological Disorders – Health Care Use & Policy Studies PND74 Antiepileptic Drug (AED) Treatment Sequencing in the Uk in Patients with Epilepsy: Real-Life Practice Data Using CPRD Borghs S 1, Logan J 2, Hong L S 3, Butt T 4 Pharma, Slough, UK, 2Stats4Pharma, Blarney, Ireland, 3Redsen Limited, Bournemouth, UK, 4UCB Pharma, Brussels, Belgium . . . . . 1UCB Objectives: Analyze real-life AED treatment sequencing in the UK using prescription data from the Clinical Practice Research Datalink (CPRD), and compare it to the 2012 National Institute for Health and Care Excellence (NICE) clinical guidelines. Methods: Patients were included if they had an epilepsy diagnosis and an AED prescription between January, 2009 and January, 2014. Patients who entered the database untreated were followed from first AED prescription following an epilepsy diagnosis, for up to five lines of treatment until being censored at the end of registration, death, or end of data coverage. Follow-up time could differ substantially between patients. AED treatment changes were classified as add, switch, or stop. The number of patients per treatment line was calculated for each individual AED combination and each regimen, grouped as Monotherapy, Polytherapy, and “No AED”. Results: Overall, 8931 patients went through 2469 unique AED treatment sequences; 97.1% were started on Monotherapy. 30.8% of patients stayed on their first Monotherapy until censored. 16.5% of initial Monotherapy patients switched to a second Monotherapy; 27% went to “No AED”; 25.7% progressed to Polytherapy. The first treatment line was consistent with NICE guidelines for 70.4% of patients. Thereafter, 14.7% of patients were treated within guidelines in the first 2 lines. The main divergence from guidelines involved prescribing Polytherapy or “No AED” rather than a second monotherapy in line 2. Largely consistent with NICE, the most frequent (87.1%) initial monotherapy AEDs were valproic acid (30.1%), lamotrigine (22.8%), carbamazepine (17.5%), levetiracetam (10.3%), and phenytoin A404 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 (6.4%). Conclusions: These data suggest that, after initial monotherapy, a majority of epilepsy treatment decisions do not follow 2012 NICE guidelines, with the degree of divergence increasing through lines of therapy. These results call for more detailed investigation into treatment patterns and the reasons for divergence from clinical guidelines in epilepsy. PND75 Patterns of Use of Tests to Monitor Disease Activity Among Patients With Relapsing Remitting Multiple Sclerosis in the United States and Europe Narayanan S 1, O’Meara P 2, White J 3, Chan J 3, Gabriele S 3 1Ipsos Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, New York, NY, USA, 3Ipsos Healthcare, London, UK . . . . . Objectives: Progression of symptoms among Relapsing Remitting Multiple Sclerosis (RRMS) patients may necessitate active disease management. The objective of the study was to assess the patterns of use of tests to monitor disease activity/ progression in RRMS in EU and US, based on disease severity. Methods: A multicenter retrospective chart-review study of MS patients was conducted in EU (UK/ France/Germany/Italy/Spain) and US to collect de-identified data on diagnosis, clinical status, and disease management approaches. Health care providers (HCPs; 94% neurologists) were screened for duration of practice (≥3yrs) and patient volume (≥15 MS patients/mo) and recruited from a large panel to be geographically representative in each country. Medical charts of next 10 consecutive MS patients were abstracted by each HCP. RRMS patient data was analyzed. Results: 2433 eligible RRMS patient charts were abstracted by 365 HCPs (EU/US: N= 265/100; mean neurology practice duration= 17/21yrs; mean number of patients seen per month= 66/62). In EU/US, patient mean age: 37/41yrs, female: 67%/70%, currently treated: 85%/88%, currently with mild: moderate: severe disease-60%: 34%: 5%/67%: 29%: 3%. Top-5 tests used to monitor RRMS varied between the regions overall as well as by patient disease severity (EU/US: %patients): Neurological exam: overall-83/88, mild-80/89, moderate-87/86, severe-91/90; Gadolinium MRI: overall-62/71, mild-57/71, moderate-69/71, severe-77/70; T2 MRI: overall-62/73, mild-59/74, moderate-66/70, severe-64/75; T1 MRI: overall-53/62, mild-51/64, moderate-55/58, severe-59/70; Expanded Disability Status Scale (EDSS): overall-86/29, mild-87/29, moderate-86/30, severe-80/35. Percent with unknown Gd-enhanced lesions were (EU/US): overall-33/40, mild37/42, moderate-29/37, severe-17/25. Percent with unknown T2 lesions were (EU/ US): overall-33/35, mild-34/36, moderate-32/34, severe-17/20. UK HCPs reported substantially more patients with unknown Gd-enhanced (70%) and T2 lesions (66%). Conclusions: Besides neurological exam and EDSS, MRI testing is widely used and its use increased with increasing disease severity (EU only); the percent of patients with unknown lesions decreased correspondingly. Still, one-third of patients had unknown lesions (with UK reporting two-thirds), warranting further scrutiny on practice patterns and diagnostics to optimize available therapeutic choices and alleviate disease burden. PND76 Frequency of Outpatient Physiotherapy Services in Neurology Diseases in Hungary Molics B 1, Hanzel A 1, Járomi M 1, Csákvári T 2, Danku N 1, Sebestyén A 3, Boncz I 4 of Pécs, Pécs, Hungary, 2University of Pécs, Zalaegerszeg, Hungary, 3National Health Insurance Fund Administration, Pécs, Hungary, 4Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . . 1University Objectives: The aim of our study is to assess amount and frequency of the physiotherapy services in the different neurology diseases within out-patient care in Hungary. Methods: The data come from the nationwide, financial data base of the National Health Insurance Fund Administration covering the official reports of outpatient care institutes in 2009. The total numbers of different physiotherapy services were determined by selecting the reported specific diagnoses codes and counting the number treatments provided for that specific diagnosis code. The diseases of the nervous system are listed in the International Classification of Diseases (ICD) with code of G00-G99. Results: The total number of the 151 different types WHO-classified physiotherapy procedures was 32318413; and 1331675 (4.12%) of them related to neurology care with the ICD code group G00-G99. The amount and frequency of the physiotherapy services are in the different neurology diseases are the following: Inflammatory diseases of the central nervous system 3584 (0.26%), Systemic atrophies primarily affecting the central nervous system 7265 (0.52%), Extrapyramidal and movement disorders 82923 (5.96%), Other degenerative diseases of the nervous system 15717 (1.13%), Demyelinating diseases of the central nervous system 30300 (2.18%), Episodic and paroxysmal disorders 191909 (13.80%), Nerve, nerve root and plexus disorders 438780 (31.56%), Polyneuropathies and other disorders of the peripheral nervous system 96871 (6.97%), Diseases of myoneural junction and muscle 21276 (1.53%), Cerebral palsy and other paralytic syndromes 434550 (31.26%), Other disorders of the nervous system 67024 (4.82%). Conclusions: The diseases of the nervous system showed different utilization frequency in physiotherapy services and represent a significant burden for the health care system. PND77 Analysis of the Recent HTA Decision Making Landscape for Multiple Sclerosis Therapies – Trends and Future Opportunities Mardiguian S , Walsh S C M PAREXEL, London, UK . . . . Objectives: In recent years, a number of new DMTs for the treatment of MS have been approved by the FDA and EMA, specifically fingolimod, dimethyl fumarate, teriflunomide, and alemtuzumab. This study evaluated the HTA appraisals for these DMTs to identify any trends or future opportunities in MS. Methods: HTA appraisals for fingolimod, dimethyl fumarate, teriflunomide, and alemtuzumab from NICE (England), SMC (Scotland), CADTH (Canada), PBAC (Australia), HAS (France), TLV (Sweden) and IQWiG (Germany) were identified. Recommendations, reasoning, and the supporting clinical and economic evidence were extracted. Results: Our search identified 19 submissions, all of which included a placebo-controlled trial, and all except dimethyl fumarate also included a head-to-head trial against one of the beta-interferons but not any other therapies. Indirect comparisons against old and newer DMTs were also included in the submissions. Approximately 90% of appraisals received positive recommendations; reasons for positive recommendations included recognition that there is a need for additional treatment options in MS, particularly for patients who do not have an adequate response to beta-interferons, and that oral DMTs represent an innovative step in the treatment of MS. However, submissions were generally criticised for trial design, scarcity of head-to-head trials, use of EDSS to measure disability, the uncertainty regarding long-term clinical and economic outcomes, and the heterogeneity of trial populations included in indirect comparisons. Conclusions: Due to the changing treatment landscape and the population included in MS trials, it is clear that placebo-controlled trials and the inclusion of beta-interferon trials in indirect comparisons are becoming less suitable for the evaluation of new MS treatments. Future HTA submissions will need to take into account the need to demonstrate superiority over comparators, and manufacturers and regulators will need to work together to develop new and reliable measures that can evaluate MS symptoms and impairments in the context of DMTs. PND78 Comparing Characteristics of Hungarian “Real Life” Patients With Participants of RCTS: First-Line Disease Modifying Therapies in Multiple Sclerosis Foldesi C 1, Penzes M 2, Szabo M 2 1National Institute of Quality and Organizational Development in Healthcare and Medicines, Budapest, Hungary,, 2National Health Insurance Fund, Budapest, Hungary . . . Objectives: Health technology assessment and decision making process lean on evidences provided by RCTs. Our aim was to compare the characteristics of Hungarian patients with multiple sclerosis to the participants enrolled in pivotal trials of first-line DMTs. Methods: The database of the National Health Insurance Fund was used for analyze the attributes of all MS patients starting first-line DMTs between 2008 and 2013. We investigated the following characteristics: 1. age, 2. gender ratio, 3. time since first MS symptoms, 4. time since first MS diagnosis, 5. number of relapses in 2 years prior to therapy. For the comparison we summarized phase III RCTs of interferon beta-1a and 1b, glatiramer-acetate, teriflunomide and dimethyl-fumarate. Results: A total of 3877 Hungarian MS patients were involved and 1670 were selected for analysis after excluding the population of the year of DMT inclusion. The mean age was 36.18±9.96 years which resembles to the participants of old immunomodulatory drugs (≈36 years) and slightly younger than the patients of the trials with new therapies (≈38 years). There was no difference in gender ratio (≈70% female). While in the RCTs the mean time since first symptoms was longer than the mean time since first diagnosis (≈7–8 vs ≈5.5 years), in the Hungarian patients these two values did not differ significantly (3.30±3.80 vs 3.17±3.77 years). We suspect that the reason for this unexpected difference can be explained by the imprecise use of ICD codes in Hungarian clinical practice. The relapse rate per 2 years seemed to decline in trials of the last two decades from 3 to 2-2.5 which is equal to the Hungarian patients. Conclusions: The population of RCTs represents the Hungarian MS patients in certain aspects but not completely. Our analysis can help the adaptation of international models to local circumstances in HTA of forthcoming MS therapies. PND79 Characteristics of Patients with Relapsing Remitting Multiple Sclerosis Taking Injectable and Oral Disease Modifying Treatments in the United States Narayanan S 1, O’Meara P 2, White J 3, Chan J 3, Gabriele S 3 1Ipsos Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, New York, NY, USA, 3Ipsos Healthcare, London, UK . . . . . Objectives: Disease Modifying Treatments (DMTs) to manage Relapsing Remitting Multiple Sclerosis (RRMS) patients have traditionally been injectables. With the introduction of new oral DMTs, a better understanding of the shift in treatment patterns and their impact on patient disease burden is needed. The objective of this study was to assess the characteristics of RRMS patients taking injectable and oral DMTs in the United States (US). Methods: A multi-center retrospective chart-review study of RRMS patients was conducted in the US in 4Q2013 to collect de-identified data on diagnosis, clinical status, and treatment patterns. Physicians were screened for duration of practice (≥ 3yrs) and patient volume (≥ 15 MS patients/ mo and recruited from a large panel to be geographically representative. Medical charts of the next 10 consecutive MS patients were abstracted. RRMS patients taking an injectable (INJ) or oral DMT were analyzed. Results: 561 eligible RRMS patients were included in the analysis (INJ: 66%; Oral: 34%). Patient characteristics were (INJ/ Oral): age: 41/40yrs, female: 71%/70%, time since diagnosis: 62/53mo, time between diagnosis and initiation of first DMT: 5/3mo, percent on first, second, and third line DMT were (INJ/Oral): 78%/30%, 18%/45%, and 4%/25%; time on current DMT (INJ/ Oral): 47mo/6mo. Reasons for current DMT initiation included (INJ/Oral): efficacy against relapses (32%/27%)/efficacy in slowing progression (24%/23%)/efficacy in early MS (21%/4%)/patient decision (7%/17%)/oral administration (0%/17%). Patient JCV status (INJ/Oral): positive-3%/23% & not tested-75%/46%; number of gadoliniumenhanced & T2 lesions: 0.8/1.3 & 7.7/10.3; percent with ≥ 1 relapse in past 12mo: 42%/66%; percent with Expanded Disability Status Scale score of ≥ 4.5: 15%/31%; percent with moderate-to-severe disability per physician judgment: 22%/40%; percent with active or highly active disease per physician judgment: 37%/60%; percent with moderate-to-severe disease per physician judgment: 22%/44%. Conclusions: In this cohort of RRMS patients, physicians initiated oral DMTs predominantly as second or third-line treatment, and these patients may have higher disease burden compared to those receiving injectable DMTs. Further research is needed to evaluate the observed treatment patterns and treatment sequencing strategies to alleviate patient burden. A405 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PND80 Adoption of Oral Disease Modifying Treatments to Manage Patients with Relapsing Remitting Multiple Sclerosis from 2011-2013 in the United States Narayanan S 1, O’Meara P 2, White J 3, Chan J 3, Gabriele S 3, Hautamaki E 1 Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, New York, NY, USA, 3Ipsos Healthcare, London, UK . . . . . . 1Ipsos Objectives: New oral Disease Modifying Treatments (DMTs) for Relapsing Remitting Multiple Sclerosis (RRMS) were recently introduced in the US. The objective of this study was to assess trends in adoption of oral DMTs among RRMS patients in the US. Methods: A multi-center retrospective chart-review study of RRMS patients was conducted in the US in 4Q2011, 4Q2012, 2Q2013, and 4Q2013 to collect de-identified data on diagnosis, clinical status, and treatment patterns. Neurologists were screened for duration of practice (≥ 3yrs) and patient volume (≥ 15 MS patients/mo) and recruited from a large panel to be geographically representative. Medical charts of next 10 consecutive MS patients were abstracted by each neurologist. RRMS patients taking oral, injectable, or infusible DMTs were analyzed. Results: 2362 eligible RRMS patient charts were evaluated (4Q2011:23%, 4Q2012:25%, 2Q2013:25% & 4Q2013:26%). Use of oral DMTs increased four-fold (4Q2011:7%, 4Q2012:9%, 2Q2013:21% & 4Q2013:31%); use of injectable DMTs decreased (4Q2011:82%, 4Q2012:78%, 2Q2013:71% & 4Q2013:60%), as did use of infusible DMTs (4Q2011:10%, 4Q2012:12%, 2Q2013:7% & 4Q2013:8%). Across timeframes, 61%, 27% & 12% of patients were on 1st-line, 2nd-line, and 3rd-line or subsequent treatment, respectively. Among 3rd-line or subsequent patients, oral DMT use increased from 23% to 59%; use of injectable and infusible DMTs decreased from 34% to 16% and 37% to 25%, respectively. Among 2nd-line patients, oral DMT use increased from 10% to 46%; use of injectable and infusible DMTs decreased from 66% to 37% and 21% to 15%, respectively. Among 1st-line patients, oral DMT use increased from 3% to 16%; use of injectable DMTs decreased from 96% to 83% and infusible DMT use remained at 1-2%. Conclusions: Oral DMT use increased between 4Q2011-4Q2013, predominantly in 2nd or subsequent lines. The impact of this observed pattern of reserving new treatment options for later lines warrants scrutiny to optimize patient management and alleviate disease burden. RESEARCH POSTER PRESENTATIONS – SESSION II HEALTH CARE USE & POLICY STUDIES HEALTH CARE USE & POLICY STUDIES – Consumer Role in Health Care PHP1 Awareness and Interest in the United States Health Insurance Marketplace Bias T K , Fitzgerald P M , Agarwal P, Vasile E West Virginia University, Morgantown, WV, USA . . . . . Objectives: To examine the level of awareness and interest in the newly setup Health Insurance Marketplace under Affordable Care Act in West Virginia state of the United States of America. Methods: Primary survey data were collected in July/August 2013 from a stratified sampling of West Virginians. A mail survey was completed by respondents in a cross-sectional study. Key variables included general awareness of the Health Insurance Marketplace and the availability of subsidies, the individual mandate, interest in using the Marketplace, and perceptions of respondents’ ability to qualify for financial assistance. Results: Six thousand surveys were mailed with a nine-page questionnaire. A total of 1,198 completed surveys were returned. Two months prior to launch, awareness of the Health Insurance Marketplace was low in WV; yet interest in the Marketplace was higher among those most likely to benefit—the insured and residents likely to qualify for financial subsidies. West Virginians reported being familiar with the individual mandate. Conclusions: Efforts should be increased at the federal, state, and local levels among government and non-government organizations to heighten awareness of the Health Insurance Marketplace in WV and, particularly, the availability of subsidies. Many, once made aware, expressed interest in learning more. PHP2 Attitudes of Patients Toward Generic Substitution and Implication for Practice in Slovakia: First Results From Adopting The Law in 2012 Babela R , Sajdikova K St. Elizabeth University, BRATISLAVA, Slovak Republic . among patients. Conclusions: Based on our research among patients we conclude that generic substitution is most likely effective drug policy tool, but since there are also many negative experiences (own or shared) with generic drugs, GS is still considered as alternative treatment. . Objectives: Slovakia has from December 2011 new law that defines list of molecules for mandatory generic substitution. It was one of the cost-containment measures applied in the same time at the field of drug policy. We provided research on statistically selected sample of patients in Slovakia to detect current attitude, knowledge and believes toward generic substitution (GS). Methods: We created special questionnaire for patients and distributed it in selected regions in Slovakia. From 600 questionaires, we evaluated 432. Questionares were distributed among patients visiting pharmacies in 2013/2014 and all patients were older than 20 years. Results: There were 57% of women in our sample and only 11% from all patients had lower than college education. 71% of all patients were using prescription drugs regularly and 58% of all patients were familiar with term GS. Suprisingly, only 16% from all patients selected co-payment as the key factor influencing their decision toward GS. Recommendation of GS in pharmacy or by doctor was selected as key factor that influenced patients in choosing GS or generic molecule (40%). Second most influential factor was own experience with generic (27%). Only 8% of patients believed that GS can decrease overall consumption of drugs. We also found out that 36% of patients consider GS as “risky” because of extended number of generics available from various unknown companies. There was also a strong statistical relationship between gender and positives as well as negatives of GS PHP3 The Simultaneous Effects of Pharmaceutical Policies from Payers’ and Patients’ Perspectives. Italy as a Case Study Armeni P 1, Otto M H 2, Jommi C 1 University, Milano, Italy, 2CERGAS Bocconi University, Milano, Italy . . . . 1Cergas, Bocconi Objectives: The research analyses (i) the individual and interactive effects of three pharmaceutical policies (cost-sharing, prescription quotas, therapeutic reference pricing) on drugs public and private expenditure and volumes, using Italian regional policies as a case-study; (ii) the extent to which the long-run effect of policies on expenditure is mediated by prescribers’/patients’ behaviours. Methods: An enriched difference-in-differences model is used. Firstly, policies impact on public and private expenditure and volume is separately estimated. Then, the hypothesis that the effects of policies on public expenditure are mediated by behaviours (transmission mechanism) is tested. As robustness check, a possible reverse causality and feedback mechanisms is tested, by switching the mediator and the independent variable. Results: The analysis shows (i) that combined policies do not necessarily produce a higher impact than policies alone; (ii) a larger impact of policies in the short-run, whereas in the long-run the trend is often reversed, but not enough to compensate the final impact, which is usually in the expected direction; (iii) as for cost-sharing, that its negative impact on public expenditure is mainly due to a decrease in volumes than to a shift from public to private expenditure. Conclusions: Despite its limitation, this study has shined a light on the impact of policies which are implemented in different time and places, thus covering an information gap and supporting policy-makers. Some empirical findings show also that pharmaceutical policies may have an unintended impact on health: e.g. the volumes decrease due to cost-sharing may imply patients under-treatment. PHP4 Patient, Insuree and Public Participation in Health Technology Assessment: An International Comparison Mühlbacher A C , Juhnke C Hochschule Neubrandenburg, Neubrandenburg, Germany . . . Objectives: There is a general consensus on the need for a stronger patient-centeredness, even in HTA processes. In international comparison different ways of public participation (citizens, insured and patients) in the decision-making process are discussed and tested. The need was recognized, but not yet fully reflected in practice. This study describes how preferences can be taken into account in different decision situations and shows how methods of preference measurement/ citizens’ councils are used in an international context to support decision-making and understand the importance of various decision-criteria that influence these decisions. Methods: A systematic literature review in PubMed/Medline revealed 95 articles and showed that methods of patients, citizens and policyholder participation are manifold. In order to structure the international approaches further, international HTA-organizations worldwide were questioned via e-mail in the end of 2013 on patients and public participation in their countries. Results: 17 out of 126 contacted organizations answered to these questions. In general, the participation efforts extend from qualitative survey of patients’ needs up to the science-based documentaries of quantitative patient preferences. The review and the survey of the HTA-agencies show that internationally three mechanisms are used to involve the public in decision-making bodies: membership of at least one patient representative (e.g., Australia, France, Germany), presentation of oral/written comments from patients (e.g., Australia, The Netherlands, Great Britain) and the possibility to check the HTAreports and the corresponding draft recommendations before publication (e.g., France, Germany, Great Britain, New Zealand, USA). Conclusions: The role of the patients or citizens seems to be limited to an informal or ad-hoc basis and is mostly restricted to consulting activities. In order to achieve a patient-centered health technology assessment two ways to sharing information are relevant: the public needs information on medical and health policy/economic issues and decisionmaker need information on the patient perspective. HEALTH CARE USE & POLICY STUDIES – Diagnosis Related Group PHP5 The Effect of Degressive Financing Method on the Hungarian Drg Based Hospital Reimbursement Between 2011-2013 Endrei D 1, Zemplényi A 1, Ágoston I 1, Molics B 1, Csákvári T 2, Danku N 1, Vajda R 1, Boncz I 3 of Pécs, Pécs, Hungary, 2University of Pécs, Zalaegerszeg, Hungary, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . . . 1University Objectives: Diagnosis Related Groups (DRG) like financing method was introduced in Hungary in 1993. In addition to DRG based reimbursement, an degressive upper ceiling (financial cap) was introduced for hospital reimbursement. The aim of our study was to investigate the financial effects of degressive financing method on the Hungarian DRG-based hospital financing. Methods: The data in our analysis were derived from the nationwide administrative dataset of the National Health Insurance Fund Administration (OEP), the only health care financing agency. We examined the period between 2011 and 2013. In 2011 and 2012 hospital activity over financial cap was reimbursed up to 110% by 30% of DRG base rate, while in 2013 hospital activity over financial cap was reimbursed up to 104% by 25% of DRG base rate. Results: In 2011 hospital activity exceeded the financial cap by 4.6% with a monthly variation of 1.5% - 6.6%. In 2012 hospital activity exceeded the financial cap by 4.2% with a monthly variation of 1.1% - 6.5%. In 2013 hospital activity exceeded the financial cap by 1.9% with a monthly variation of 1.1% - 2.6%. Between 2011 and 2013 the DRG base rate remained the same (150000 A406 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Hungarian Forint / DRG costweight). Conclusions: Introduction of degressive financing method – in addition to DRG reimbursement – managed to control the activity of hospitals. The soft regulation in 2011 and 2012 resulted in a 4.2-4.6% excess activity of hospitals, while the more rigorous regulation in 2013 managed to decrease the excess hospital activity to 1.9%. Degressive regulation can serve as a cost containment tool for health policy decision makers. PHP6 Nub Status - A 2014 Situation Analysis for Drugs: Oncology As Leading Therapeutic Area Friedmann B 1, Keck E 2, Schalk E 3, Schmitz D 2 1Quintiles Commercial, Mannheim, Germany, 2Healthcare Manufaktur, Cologne, Germany, 3Quintiles, Mannheim, Germany . . . . Objectives: In the German hospital landscape NuB’s (Neue Untersuchungs- und Behandlungsmethoden) are essentially the precursor for cost-intensive drugs to be reimbursed within the G-DRG system. Hospitals can only start SHI negotiations for reimbursement once drugs have been given a NuB 1 status. The objective of this research was to provide an overview on the proportion of drugs (vs. methods, medicinal products) and their respective indications, which submitted NuB applications for 2014. Methods: The German DRG database issued by the InEK (Institute for the Hospital Remuneration System) was used to analyse NuB subgroups sorted according to key therapeutic indications. In parallel, the number of NuB 1 status products that went through the AMNOG process was analysed. Results: Out of 618 NuB submissions, only 133 (22%) were classified as drugs. In total, 114 (18%) of all NuB applications received a NuB 1 status, out of these 43 (38%) were drugs. The leading therapeutic area of the NuB 1 status drugs was oncology with 28 drugs (65%), followed by 5 ophthalmic products (12%). NuB 2 status was given to 465 (75%) procedures out of which 82 (18%) were drugs. The analysis reveals that, the success rate to receive the essential NuB 1 status is relatively low. Chances to receive a successfull NuB 1 status approval is one in three for drugs, however only one in six for other procedures. Out of the 43 drugs that were given NuB 1 status, already 24 (55%) passed through the AMNOG process and were given equal proportions from important additional benefit to no additional benefit. Conclusions: Drug applications are more likely than procedures to be given NuB 1 status and thereby initiate reimbursement negotiations with the SHI. The application quality and support by the scientific societies and treatment centres is essential to make a NuB application successful. HEALTH CARE USE & POLICY STUDIES – Disease Management PHP8 Multicriteria Decision Analysis and Cost Analysis in Health Care Decision Making: A Literature Review Ivlev I , Landova M Czech Technical University in Prague, Kladno, Czech Republic . . Objectives: The purpose this literature review is to investigate the application of multicriteria decision analysis and cost analysis methods within health care decision-making. Methods: A search of the literature was conducted using scintific databases. A combination of the following key words and phrases were inputted into these databases: Analytic Hierarchy Process (AHP), Analytic Network Process (ANP), ELimination and Choice Expressing REality (ELECTRE), Goal programming, Grey relation analysis, Markov process, Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), CBA, CEA, and related words. The located articles were divided into the following twelve health care topics: evaluation of health information services; evaluation of the product development process; project and technology selection; pharmaceutics; health care management; therapy/treatment; management of medical waste; human resource planning in health care; organ transplantation; evaluation of health care policy; diagnostics; and shared decision-making with the patient. Results: Ninety research articles were retrieved and determined relevant. The pertinent articles were published between 1981 and 2013. It was found that the AHP is the most commonly used method in health care decision making (65 articles). AHP is mainly exploited in project and technology selection (22). The ANP method is utilized in the evaluation of health information services, project and technology selection, pharmaceutics and therapy/treatment. For the evaluation of health care policy AHP (11), CBA (1), CEA (1) and Grey relation analysis (1) were used. The TOPSIS, VIKOR, Markov process methods were utilized once in human resource planning in health care, health care management and therapy/treatment respectively. The CBA (4) and CEA (2) methods were especially useful for solving therapy/treatment tasks. Conclusions: Multicriteria decision analysis and cost analysis offers a scientifically sound evaluation framework for health care management, where stakeholder interests are of crucial concern and complex criteria that cannot easily be reduced to simple monetary expressions, can be assessed in resource limited settings. PHP9 Trends in Physical and Occupational Therapy Utilziation in the Us And Western Europe Eichmann F 1, DiBonaventura M 2, Schoefer A 3 1Kantar Health, Munich, Germany, 2Kantar Health, New York, NY, USA, 3Private Practice, Marquartstein, Germany . . . Objectives: Allied health care (AHC) disciplines, such as physical (PT) and occupational therapy (OT), are primarily performed by non-medical health care professionals. Although the budget impact of AHC is generally low, reimbursements are often scrutinized for their financial impact and benefit/risk ratios. To better inform the health care decision making regarding AHC, the aim of this study was to examine trends and utilization of PT and OT. Methods: Data from the 2013 US (N= 75,000) and 2013 5EU (France, Germany, Italy, Spain, and UK; N= 62,000) National Health and Wellness Survey (NHWS). The NHWS is a patient-reported survey administered to a demographically representative sample of adults (with respect to age, sex, and region) in each country. Overall rates of OT/PT visits were reported. Patients who reported an OT/PT visit in the past six months were compared with those who did not with respect to demographics, health history, and comorbidity variables. Logistic regression models were then conducted to predict OT/PT visits from these variables. Results: Rates of OT/PT visits did not change from 2010 to 2013 but significant differences among countries was observed (p< .05). In 2013, France (0.54%) and the US (4.51%) had the most infrequent visits while Spain (11.13%) and Germany (11.92%) had the most frequent. Being in Germany (OR= 3.46), being in Spain (OR= 3.24), and having an above the median income (OR= 1.14) were the strongest demographic predictors of an OT/PT visit (all p<.05). Although most comorbidities were associated with an increased probability of an OT/PT visit, pain (OR= 2.30), arthritis conditions (OR= 1.73), and psychiatric conditions (OR= 1.73) were most strongly associated (all p< .05). Conclusions: PT and OT utilization varies significantly across countries, being highest in Germany and Spain where over 10% of adults reported a visit in the past six months. Pain-related (pain, arthritis) and psychiatric comorbidities were among the strongest predictors of PT/OT use. PHP10 A Gender Medicine Post-Hoc Analysis: Background and Methods of the Metagem Project Simoni L 1, Colombo D 2, Bellia G 2, Vassellatti D 2, Zagni E 2, Rizzoli S 1, Sgarbi S 1 1Medidata srl, Modena, Italy, 2Novartis Farma Italy, Saronno (VA), Italy . . . . . . . Objectives: Gender is a social construct, which is defined by the way people perceive themselves and how they expect others to behave. Gender medicine is the field of medicine that studies the biological and physiological differences between the human sexes and how that affects differences in disease. The progress of research has confirmed that men and women differ not only sexually but also in relationship to factors such as liver enzymes, sex hormones and to variables determined by the environment, education, culture and psychology of the individual [Soldin and Mattison, 2009; Regitz-Zagrosek and Seeland 2012]. The Italian Drug Agency has recognized the importance of gender-specific analysis when evaluating new drug efficacy. The gender-medicine METAGEM project aims to describe gender differences in clinical outcomes, therapeutic approach and safety parameters in real world data. Methods: Areas of interest were defined regarding Dermatology, Central nervous system, Infectivology, Rheumatology, and Transplantation; data were considered which were collected in ten observational studies conducted between 2002 and 2013 in Italy in routine clinical practice. A post-hoc subgroup analysis is performed by study, during which males are compared with females by statistical tests. A merge of different study data will be performed in order to evaluate safety. As post-hoc analysis all p-values are exploratory. Results: The number of enrolled patients range between 238 to 1746 considering Rheumatology and Dermatology areas respectively, for a total of 3743 male and 3018 female patients. Conclusions: The papers and congress communications which will arise from METAGEM project will make the scientific community more aware of the importance of a gender-dedicated approach in the care of patients. HEALTH CARE USE & POLICY STUDIES – Drug/Device/Diagnostic Use & Policy PHP11 Health-Economics in Czech Republic: Can Formal Health-Economic Methodology Improve The Quality Of Submitted Analysis? Hambalek J , Spurna M , Vocelka M , Ballokova A State Institute for Drug Control, Prague, Czech Republic . . . . Objectives: Health economic analysis (HEA) has been required since 2008 as a part of drug reimbursement applications submitted to the State Institute for Drug Control (SUKL). In 2013, SUKL introduced a formal health economic guideline, mostly based on the Czech Pharmacoenomic Society´s guideline published in 2011. The aim of this study was to assess the impact of guideline implementation on the quality of submitted HEAs. Methods: We reviewed all (18) applications for new innovative drugs with prescription limited to specialized centers, in which the final decision was issued and came into force between 1/2013-6/2014. The HEAs were described in terms of type of analysis and further confronted with a ‘HEA checklist‘ to identify common deficiencies in submitted HEAs. All check-lists were peer-reviewed to ensure objectivity. Our results were compared with previously published research (assessing HEAs submitted in 2008-2009 before the guideline release). Results: All investigated dossiers contained HEA. Nine of them (50%) were cost-effectiveness analysis (outcome expressed mostly in LYGs) and nine (50%) were cost-utility analyses (outcome in QALYs). In general, the quality of HEAs was higher compared to the previous research, with ‘evidence completeness’ and ‘uncertainty’ being the most commonly marked as ‘unsatisfactory’ in the HEA checklist (less than 50%). Our results showed that 1/3 HEAs fell into the ‘best’ category (80100% ‘satisfactory’ questions), while none of HEAs fell into this category in previous research. Conclusions: The present analysis showed a trend in higher quality of submitted analyses (1/2013-6/2014) compared with years 2008-2009. PHP12 Price Comparison in the European Pharmaceutical Market Ressl S , Walter E Institute for Pharmaeconomic Research, Vienna, Austria . . Objectives: Due to rising health care costs, stricter governmental cost containment measures like international price comparison and external reference pricing, the European pharmaceutical markets will lead to future challenges in pricing. Despite of these prices of pharmaceuticals vary across European countries. Thus, the aim of the analysis was to draw a comprehensive picture of the pharmaceutical price levels and consumption (expenditure per capita) of pharmaceuticals in Europe. Methods: The study analyzed drug prices of the outpatient setting (Rx and OTC) of all countries of the European Union except Denmark plus Switzerland. A407 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Based on ex-factory prices per Standard Unit SU (tablet, vial etc.), price indices were calculated. Study results showed international pharmaceutical price levels (in Euro and purchasing power parities) and expenditure per capita. Observation period was the year 2012. Data were collected via international literature review and based on IMS Health data on file. Results: The international comparison of prices indices based on SU prices showed (European average price = 100), France (138) and UK (126) had the highest price indices and in contrast to Netherlands (69) and Romania (69) were at the lower end. In the international country sample, the Netherlands had the highest health expenditure (11.91% of GDP) in 2011, closely following by France with 11.33% of GDP. Subsequently, Switzerland occupied a leading position based on the determined price index per capita (187) per ex-factory price, followed by Sweden (167). The rear was Romania (45). Conclusions: In case of converting drug prices with purchasing power parities differences between drug prices and affordability in different countries were found. Low income countries reported less affordability of drugs, leaving room for potential problems with drug access, and consequently, a negative impact on health. PHP13 How is Current Pharmaceuticals Pricing Policy on Generics Performing in Turkey Regarding Price Erosion? Gürsoy K , Köselerli R , Dogru S T , Köksal I Social Security Institution of Turkey, Ankara, Turkey . . . . . Objectives: Generics are commonly accepted to contribute significantly to treating disease by improving the affordability of pharmaceuticals. Once the patent expires for an originator brand, generics erode prices through creating fierce competition. The objective of this study is to investigate on Turkish pharmaceutical policies over generics through measuring how much they lower the prices, and then develop alternative strategies to maintain higher level of price reduction and hence saving. Methods: Claims data from Turkish Social Security Institution for all ambulatory care drugs reimbursed were collected monthly in the period of January 2009 to December 2013. First, the share of generic drugs both in terms of sales volume and value were calculated. Then, selected 12 equivalent groups each relying on same molecule were evaluated and saving impacts due to generic entry were computed. Finally, the price erosion for 15 generics entering the reimbursement list in 2012 as first generics was measured. Results: Only 48% of all units sold was prescribed as generic drugs, whereas it was only 31% when it comes to value. For 12 equivalent groups constituting nearly 7% of SSI drug spending, the price erosion was nearly 41% ranging from 8% to 74%. Interestingly, when a first generic having an originator brand whose any form were marketed before 1987 in any country entered the market, new discounted public price was higher than the price with no generic. Finally, in the first year of the first generic entry, on average the prices were only shrunk by 39% with an increase of 41% in units sold. Conclusions: Turkey is not maximizing its full potential with respect to generic medicines. Therefore, it is of great importance that policies such as therapeutic equivalence, tendering, and aggressive generic pricing policy to stimulate higher savings need to be introduced. PHP14 Cost-Effectiveness of Targeted Pharmacotherapy – A Systematic Review of the Literature Amler N , Becker T , Bierbaum M , Schöffski O Friedrich-Alexander-Universität Erlangen-Nürnberg, Nuremberg, Germany . . . . Objectives: Targeted therapies (e.g. trastuzumab, cetuximab) are often said to constitute a new era of medicine. The new agents promise to be more effective and less toxic than standard care. The consequences on costs are still a matter of debate. While proponents refer to the enormous cost saving potential due to the avoidance of side effects or hospitalizations for example, opponents expect a massive increase in costs of care. However, the impact of targeted therapies on cost as well as their benefits are not yet clear. This review assesses the cost-effectiveness of targeted therapies. Methods: We searched Medline, the Cochrane Library, Scopus, ISI Web of Knowledge and ScienceDirect for relevant articles published between 2001 and 2013. Internet search and scanning reference lists complemented our search. Authors independently reviewed titles, assessed articles’ eligibility for inclusion, determined methodological quality and extracted relevant data. Results: In total 487 studies were identified. 50 of them met the inclusion criteria. 39 studies target either trastuzumab, or one of the aromatase inhibitors anastrozole, letrozole or exemestane. Quality of the studies was assessed using the checklist provided by Drummond/Jefferson (1996). The majority of the studies scored 80% or higher. With few exceptions studies are based on Markov models. Effectiveness as well as cost data were mostly obtained from clinical trials or literature respectively. Stated cost-effectiveness ratios ranged from 4.370€ /QALY to 220.658€ /QALY, whereas the majority of studies report favorable incremental cost-effectiveness ratios (≤50.000€/ QALY). Models often did not incorporate the respective companion test adequately. Almost all studies were funded by the pharmaceutical industry. Studies without financial support report higher cost-effectiveness ratios. Conclusions: Results of the review indicate that targeted therapies seem to be cost-effective compared to standard care. As most of the studies have been funded by pharmaceutical companies, authors suggest that results must be interpreted with care. PHP15 Impact Of Pharmacoeconomic Guidelines on the Reimbursement Process of Medicines In Serbia In 2014 Baltezarevic D 1, Novakovic T 1, Parker M 2, Samardzic J 3 1Pharmacoeconomics Section of the Pharmaceutical Association of Serbia, Belgrade, Serbia and Montenegro, 2University of Liverpool Management School, Liverpool, UK, 3University of Belgrade, Medical Faculty, Belgrade, Serbia and Montenegro . . . . Objectives: To assess the impact of new rules for pharmacoeconomic evaluations in Serbia, after the publishing of a new rulebook covering the criteria, methods and conditions for listing medicines for reimbursement. Methods: We investigated the reimbursement submissions to the National Health Insurance Fund (NHIF) in Serbia, in the period September 2013 to June 2014. Quality was assessed using the criteria for pharmacoeconomic evaluations from the in the most recent ISPOR guidelines and the Serbian pharmacoeconomic guidelines. The impact of the introduction of the new rulebook was assessed by comparing the period following its introduction with the number and quality of submissions in the preceding period. Results: Between September 2013 and April 2014 there were 268 submissions to the NHIF and none were accompanied by the necessary pharmacoeconomic analysis. The new rulebook was published in April 2014, making budget impact analysis an obligatory part of every submission, and cost-effectiveness analysis obligatory for all new INNs. 127 new and renewal submissions were made between April and June 2014, 47 for new INNs, 31 original and 16 for innovative medicines. All 47 completed the obligatory pharmacoeconomic analysis. There was a widespread failure to reach the requirements laid out in either the Guidelines for Pharmacoeconomic Evaluations for Serbia or ISPOR guidelines. Conclusions: The introduction of compulsory pharmacoeconomic component to reimbursement submissions in Serbia has raised the standard of submissions to the NHIF. This standard is still too low to fully facilitate transparent evidence based decision making, however the new rulebook is expected to force both the NHIF and the pharmaceutical industry to increase their skills in evidence based decision making. However, Serbia still suffers from a lack of the information required to construct quality pharmacoeconomic analysis, with limited local cost estimates available and limited supply of health economic skills in the region. PHP16 Are Prices of Patent-Protected Pharamceuticals in the Top 5 Eu Countries Converging? Reinaud F 1, Ando G 2 1IHS, Paris, France, 2IHS, London, UK . . Objectives: The goal of this study is to assess whether prices of innovative drugs in the top 5 EU markets converge over time. Methods: Price trends of a sample of 50 innovative drugs, approved by the European Medicines Agency (EMA) in 2009 or earlier, that are still patent-protected and marketed in each of the top 5 EU markets were analysed over the period 2009-2014. The focus was on ex-manufacturing prices, which were extracted from IHS’s international pricing database POLI. Discounts and rebates were excluded from the analysis, and current exchange rates into Euros were used. Given the timeframe, none of the products included in the sample was assessed under the AMNOG reform in Germany. Results: While the data indicates that there are still price differences across countries in 2014, these differences have declined over time. In 2009, the price difference between the highest priced country (Germany) and the lowest one (United Kingdom) was 24%; this figure dropped down to 15% in 2014. The average price of the 50 products declined in all countries and the overall price mean (all countries) slightly decreased - by around 5% - over the period studied: from € 303.3 in 2009 to € 289.0 in 2014. This demonstrates a global trend leading innovative drug prices to decrease over time while they are still patent protected. Conclusions: Our data shows that prices slightly decrease over time, and also that prices have started to decline in Germany. German prices in the sample are still comparatively higher than the other major European markets, but with most prices in Italy and Spain remaining constant, and those in France converging to the mean, there are clear trends of price corridors shrinking across Europe. The divergence is commonly larger at launch, and converges over time. PHP17 Letting The Data Speak: A Shift of the Pharmaceutical Spending to Households in Greece Kyriopoulos D 1, Ollandezos M 1, Papadopoulos K 1, Kyriopoulos I I 2, Pentafragas V 1 1Pan Hellenic Union of Pharmaceutical Industry, Athens, Greece, 2National School of Public Health, Athens, Greece . . . . . . Objectives: Since 2009, Greece has implemented a fiscal consolidation programme that includes several reforms and budget cuts in health sector. Indicatively, health expenditure has been reduced by more than 20% between 2009 and 2012. Pharmaceutical sector can be characterized as one of the main sources of these cuts. The objective of this study is to draw some early conclusions regarding the impact of the current pharmaceutical policy on the structure of expenditure and the access to medicines. Methods: The data are obtained from the Hellenic Statistical Authority, the national accounts and the household surveys for the period 20092012, in order to examine potential changes of the public/private mix of the pharmaceutical expenditure. Results: Public pharmaceutical expenditure fell from €5.1 bn in 2009 to € 3 bn in 2012 (-42.5%). Private pharmaceutical expenditure was € 1.3 bn in 2009, while it was approximately € 1.5 bn in 2012 (+12.4%). The total expenditure dropped by 31.2%. The share of private expenditure on pharmaceuticals has sharply increased during the study period. Specifically, the private expenditure was 20.6% of the total pharmaceutical expenditure in 2009, while it approximated 34% in 2012. Conclusions: The aforementioned imply that there is a shift from public to private spending for pharmaceuticals between 2009 and 2012. Moreover, preliminary estimates for 2013 suggest that this trend is increasing. Therefore, apart from the existing financial and income constraints, households are in front of an additional financial burden. This change is associated with potential catastrophic expenditures and reduction in patient compliance. The study findings suggest that individuals’ degrees of freedom are constrained and that patients express their preferences by increased out of pocket payments. In the light of the above, the pharmaceutical policy implemented so far can be associated to a sharp “indirect increase” of copayment and coinsurance rates. PHP18 Economic Burden of Intravenous Iron Products In Public Hospitals of Paris and Impact of Their New Hospital-Restricted Status Therasse C 1, Bocquet F 2, Fusier I 1, Cordonnier A L 1, Sinègre M 1 1General Agency of Equipment and Health Products (AGEPS), AP-HP, Paris, France, 2Dauphine University, Paris, France . . . . . . A408 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To assess the global economic consequences for French hospitals of the European Commission (EC) decision to strengthen administration conditions of intravenous iron products (IIP) (iron sucrose (IS) and ferric carboxymaltose (FC)) due to safety concerns. Following this EC decision, in February 2014, French Health Authorities decided to give a hospital-restricted status (HRS) to IIP. Methods: We compared, in acute care (medicine, -surgery-obstetrics) hospitals, IIP consumption and expenditure (extracted from SAP software) before (2012 vs. 2013) and before-after (02/2013-04/2013 vs. 02/2014-04/2014) they had a HRS, and the number of diagnosisrelated group “sessions of chemotherapy for non-tumor disease” (DRG-CNTD) linked to anemia (Technical Agency of Information on Hospitals data). Results: 20 hospitals were included. Before they had a HRS, IIP global consumption in volume increased by 9.6% (i. e. €171,022 spent more) in a year. After IIP had a HRS, the increase was 16.7% (i. e. €164,338 spent more) in 3 months (02/2013-04/2013 vs. 02/2014-04/2014). 23.7% of the increase was attributable to day hospital admissions (DHA) and 4.4% to dialysis units. FC consumption was 23.9% higher (i. e. €188,838 spent more) in 2014 compared to 2013, whereas IS consumption was 16.4% lower (i. e. € 24,500 spent less). DHA in DRG-CNTD linked to anemia increased by a factor of 1.8. Conclusions: DHA as part of the DRG-CNTD is the only way to finance this additional hospital activity (no additional funding for traditional hospital care). FC’s cost of daily treatment (€120 to €150) is about 20 times IS’s (€ 5.2 to € 7.8), while a single tariff is allocated for the DRG-CNTD (drug case-mix: €62, whatever the drug). Costs of the FC increasing use to manage anemia may be offset by administration of other drugs, or the DRG-tariff should be adjusted. Another alternative would be to implement a tender between IIP, taking into account the cost-effective ratio of each drug. PHP19 The Impact of Cost Containment Reforms to the Pharmaceutical Benefits Scheme (Pbs) On Prescribing Volumes And Expenditure In Australia: 1992 To 2011 Lee K S , Hendrie D , Sunderland V B , Moorin R Curtin University, Perth, Australia . . . . . . Objectives: The increase in PBS expenditure over the past two decades led to the Australian Government implementing several cost containment measures. The objective of the present study was to investigate the impact of these reforms on PBS prescribing volumes and expenditure. Methods: Data retrieved from Medicare Australia’s PBS Statistics database provided monthly government expenditure (benefit) and prescription volume (service) data. Segmented linear regression models were used to analyse the time series data starting from 1 January 1992 to 31 March 2012. In each segment, between implementing the cost containment measures, two parameters, the level and trend were used to estimate the impact of the intervention. A lag of 12 months was applied to adjust for seasonality and forward stepwise eliminations were applied to obtain a parsimonious model. Seven cost containment measures were investigated: the economic evaluation requirement for drug listing (1/01/1993), re-supply limits (1/11/1994), two co-payment increases (1/01/1997 & 1/01/2005), therapeutic group premium (TGP) policy (1/02/1998), safety net 20-day rule (1/01/2006), and price reductions in multiple brand drugs in Formulary 2 (F2) (1/08/2008). Results: All interventions except the re-supply limits policy were found to have a significant impact on PBS services or benefits. Reductions in the services and benefits trend were observed in two measures, the TGP policy and safety net 20-day rule while a reduction in the services trend was observed for the economic evaluation requirement measure. No significant trend changes were observed in the post F2 price reductions while higher co-payments resulted in a reduction in the level of services followed by an increase in the trend of services and benefits. Conclusions: Many of the cost containment measures implemented in Australia have been effective in containing cost. Among these measures, the safety net 20-days rule was estimated to be the most effective in reducing drugs utilisation and expenditure. PHP20 The Effects of Reforms, Price Cuts and Global Budget Implementation on Biotechnological Medicine Sales Which Have Annual Average Highest Amount Of Sales Between 2008-2013 In Turkey Vural E H 1, Vural I M 1, Yigit O 2, Uman N 1, Tolun C 1, Babacan S 1, Dogan E 1, Akbulat A 1, Kahveci R 3, Malhan S 4, Artiran G 1, Kerman S 1 1Turkish Medicines and Medical Devices Agency, Ankara, Turkey, 2Turkish Statistical Institute, Ankara, Turkey, 3Ankara Numune Education and Research Hospital, Ankara, Turkey, 4Baskent University, Ankara, Turkey . . . . . . . . . . . . . . Objectives: In 2003 Health Transformation Program and in 2006 Social Security Reform were launched in Turkey. At the end of the years 2009 and 2011 price cuts were done by the Government and between the years 2010-2012, there was a global budget implementation in Turkey. The top 100 medicines, annual average highest amount of sales of between 2008-2013, had one to four of the total pharmaceutical market value in 2013. We aimed to examine the status of biotech-medicines in the first 100 medicine’s and evaluate the effects of policy interventions on these medicine sales. Methods: While pharmaceutical sales data were obtained from the IMS Health-Turkey data base, prices and characteristics of medicines were obtained from the Turkish Medicine and Medical Devices Agency data bases. Each group (biotech/ nonbiotech-medicines) was analyzed using TRAMO and SEATS method. Results: 19 medicines are biotechnological. In 2009 compared to 2008 while biotech-medicines amount increased by 43.8%, nonbiotech-medicines amount increased by 14.2%. In subsequent years, an increase over the previous year is observed at lower rates for biotech-medicines. This total amount has decreased compared to previous years for nonbiotech-medicines. 2014 and 2015 for the consumption of biotechnological medicines was estimated to be around 9 million boxes, expenditure for 2014 of 1360 million Turkish Liras (TL) and 2015 million TL for the year 2015. On the other hand, non-biotechnological medicines consumption would be approximately 185 million boxes and for the year 2014 2225 million TL, for the year 2015 2280 million TL are projected. Conclusions: Biotechnological medicines have high unit costs and these costs are increasing compared to year to year. Policy interventions did not effected biotechnological medicine sales negatively. While big differences between the biotechnological and non-biotechnological medicine box sales will continue, the gap between the biotechnological and non-biotechnological medicines total amount will be closer. PHP21 Trends In Clinical Drug Development Timeframes, 1981-2013 – An Example From Virology Ward D J 1, Hammond E 2, Linden-Phillips L 1, Stevens A 2 1NIHR Horizon Scanning Centre, Birmingham, UK, 2University of Birmingham, Birmingham, UK . . . . . Objectives: There are concerns that the time taken to develop innovative new drugs is increasing, delaying access to and increasing the costs of new medicines. Both the clinical trial stage and regulatory approval stage for new drugs are important components of overall development time, and we sought to determine whether the length of either of these stages is increasing, using new anti-viral drugs launched in the UK as a case-study. Methods: New anti-viral drugs launched in the UK between 1981 and 2013 were identified from the British National Formulary. Initiation of clinical development was determined from Pharmaprojects (Informa Healthcare) and Medline searches; the Investigational New Drug Application or first report of phase I trials was taken as the start of clinical development, whichever was earlier. Regulatory submission and approval dates were obtained from the European Medicines Agency and UK Medicines and Healthcare Regulatory Agency. Results: 43 new anti-viral drugs were launched in the UK between 1981 and 2013, with a mean time from start of clinical trials to approval of 6.4 years. This period increased from 4.8 years for drugs launched 1981-93, to 6.0 years for those launched 1994-2003, and 7.9 years for those launched 2004-2013; and a statistically significant positive linear trend was observed (r= 0.47). The mean time from regulatory submission to approval was 1.1 years, but no significant linear trend was seen for this time period (r= 0.20). New drugs for HIV typically spent less time in clinical development than other anti-viral drugs (mean 5.7 vs 7.3 years, p=0.049). However clinical development times for HIV and other antiviral drugs increased at a similar rate. Conclusions: For anti-viral drugs launched in the UK, the time spent in clinical development has increased markedly since the 1980s, and this increase is not related to increasing time taken for regulatory approval. PHP22 Impact of Health Policy Changes on Trends of Pharmaceutical Market in Turkey Safak Yilmaz E 1, Kockaya G 2, Yenilmez F B 3, Saylan M 2, Tatar M 3, Hilal Vural E 1, Vural I M 1, Akbulat A 1, Gursoz H 1, Artiran G 1, Kerman S 1 1Turkish Medicines and Medical Devices Agency, Ankara, Turkey, 2Health Economics and Policy Association, Ankra, Turkey, 3Hacettepe University, Ankara, Turkey . . . . . . . . . . . . . Objectives: Turkish Ministry of Health (MoH) initiated Health Transformation Program (HTP) in 2002. HTP impacted all clinical and economic outcomes of health including pharmaceutical sales by improving access to health services. Total pharmaceutical market reached US $ 8 billion in last 10 years. HTP improved health coverage and access to health services, additional policies are implemented by MoH for controlling this increasing trend on pharmaceutical consumption. The aim of this analysis is to understand the impact of selected 5 major policy changes to total pharmaceutical consumption in between 2002-2012. Methods: 132 months sales data with segmented regression analysis for interrupted time series were used. International reference pricing of pharmaceuticals (RF), mandatory reimbursement dossier submission for new molecules, new indications and line extensions with medical and economic evaluations (MRDS), auditing for good manufacturing practice (GMP), family physician system (FP) and compulsory medical service for physicians (CMS) were selected as five major policies that may affect cost, demand and supply of pharmaceuticals. We analyzed possible breaks in trends prior and after the implementation of 5 selected policies of the HTP. The analysis was conducted for cost (CS) and unit sales (US) for all pharmaceutical products. The Durbin-Watson d statistics of SPSS version 20.0 was used as a test for serial correlation of error terms. Shift in slope with p< 0.05 was considered as statistically significant. Results: All, except for RF policy changes, had a negative impact on the trends for CS. RF policy did not reverse the trends in CS and US however it slowed positive trend in CS significantly down. All policies hindering impact on the increasing trend in US. However, only RF and CMS policies reached statistically significant level. Conclusions: Policies within the HTP framework were successful to control pharmaceutical expenditures while improving access to health care. International reference pricing has a limited impact on controlling growth of pharmaceutical market. PHP23 Boosting Biosimilars Uptake in European Countries Young K E 1, Rémuzat C 2, Urbinati D 3, Toumi M 4 1Creativ-Ceutical, Milan, Italy, 2Creativ-Ceutical, Paris, France, 3Creativ-Ceutical, Luxembourg, Luxembourg, 4University Aix-Marseille, Marseille, France . . . . . Objectives: To provide an overview of biosimilars uptake in Europe and assess policies and initiatives that might boost uptakes of biosimilars in European countries. Methods: A literature review was conducted from European and national health authorities websites, Generics and Biosimilars Initiative (GaBi) website, Medline® database, and available grey literature. Results: To date, 17 biosimilars were granted marketing authorization throughout the European Union and the European Medicines Agency released 9 product-specific biosimilar guidelines. Recent approval of the first monoclonal antibody biosimilar of infliximab demonstrates the evolution of the European regulatory framework over time allowing approval of structurally complex molecules. Despite Europe has been a pioneer in this regulatory path, biosimilars uptake still remains limited (about 30% to 50% volume uptake) with modest price discounts from the originators (15% to 30%). If Germany authorized substitution of biological products produced by a same manufacturer in 2011, none of the Members States had allowed substitution of biological products from different manufacturers so far. With adoption of the 2014 social A409 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 security funding law, France is becoming the first country to allow biosimilars substitution when initiating treatment course. To encourage biosimilars substitution, the Norwegian Medicines Agency announced in 2013 the funding of clinical studies with infliximab originator and biosimilars in which patients would be switched from originator to biosimilars forth and back. Moreover, increased price discounts of about 40% were applied for infliximab in Norway. By 2020, expected savings for biosimilars are estimated around € 12-33 billion in big EU5, Poland, Romania, Sweden. Conclusions: While patent cliff of major biologic drugs is expected on the next 5 years, initiatives to reassure physicians to prescribe biosimilars and implementation of substitution rules, even if still raising some reluctance, might contribute to boost biosimilar uptake in Europe. Price competition will impose manufacturers of branded biologics to adopt new pricing strategies. PHP24 Effects of Reference Price System on Medicines Which Have Annual Average Highest Amount Of Sales Of Between Years 2008-2013 Uman N 1, Tolun C 1, Babacan S 1, Dogan E 1, Vural I M 1, Vural E H 1, Dasdag M M 1, Safak Yilmaz E 1, Malhan S 2, Kahveci R 3, Akbulat A 1, Artiran G 1, Kerman S 1 1Turkish Medicines and Medical Devices Agency, Ankara, Turkey, 2Baskent University, Ankara, Turkey, 3Ankara Numune Education and Research Hospital, Ankara, Turkey . . . . . . . . . . . . . . . . Objectives: In Turkey, a medicine reference pricing system has been in use since 2004. The price of pharmaceuticals is determined by the acceptance of the lowest ex-factory price in the reference countries (Greece, France, Italy, Portugal, Spain). We aimed to examine the first 100 medicine’s, having the annual maximum amount on the average Turkish Lira (TL) based medicine sales between the years 2008-2013 which have 25.1% value in the total pharmaceutical market, reference price changes in these period. Methods: While pharmaceutical sales data were obtained from the IMS Health-Turkey data base, medicine prices were obtained from the Medicine Price List published by Turkish Medicines and Medical Devices Agency. Results: In 2008, the top 100 medicines in determining the price of most of Italy were taken as the reference country, followed by Spain and Greece. In 2013 while Greece is taken as the reference country more common; France and Italy are to follow. In 2008, only one medicine’s price increased, and only one medicine’s price decreased because of the reference price. In 2009 and 2010 price increases did not seen. In 2009, 8 and in 2010, 30 medicines prices had decreased. In 2011, 2012 and 2013 totally 27 medicines reference prices had increased. 21 medicine’s price in 2011, 20 medicine’s price in 2012 and 24 medicine’s price in 2013 had decreased. While mostly France (13 medicines) connected to the reference price increases, mostly Greece (58 medicines) has been based reference price drop in the analyzed period. Conclusions: The application of reference prices, medicine prices to be reduced to a large extent. Greece based reference price decreases are observed to be mainly from the year 2010. The reason for this is considered as the economic crisis in Greece and the agreement with the IMF, EU and ECB, the Greek government at cost-containment. PHP25 Implications Of External Price Referencing of Pharmaceuticals in Middle East Countries Kalo Z 1, Alabbadi I 2, Al Ahdab O G 3, Al-Badriyeh D 4, Alowayesh M 5, Al-Saggabi A H 6, Elmahdawy M 7, Tanzi V L 8, Alsultan H 9, Akhras K 10, Kanavos P 11 1Eötvös Loránd University (ELTE), Budapest, Hungary, 2University of Jordan, Amman, Jordan, 3Ministry of Health, Abu Dhabi, United Arab Emirates, 4College of Pharmacy, Qatar University, Doha, Qatar, 5Kuwait University, Yarmook, Kuwait, 6Ministry of National Guard Health Affairs, Riyadh, Saudi Arabia, 7Ministry of Health, Cairo, Egypt, 8American University of Beirut, Beirut, Lebanon, 9Dasman Diabetes Institute, Yarmook, Kuwait, 10Novartis, Dubai, United Arab Emirates, 11London School of Economics and Political Science, London, UK . . . . . . . . . . . . . . Objectives: External price referencing (EPR) is a method to control pharmaceutical prices, wherein decision-makers specify a basket of countries, whose prices they use to inform their national target price. EPR is practiced frequently worldwide, though countries vary substantially in how they execute EPR. The objective of our pioneer collaborative health policy research initiative was to analyse how EPR is used in Middle East countries, and assess whether EPR results in narrower price corridor for innovative pharmaceuticals compared to non-pharmaceutical services not subjected to EPR. Methods: In Q1 2014, we conducted a survey on EPR regulations, and collected prices of 16 innovative pharmaceuticals and 10 non-pharmaceutical services in six Middle East countries (Egypt, Jordan, Kuwait, Lebanon, Saudi Arabia, UAE). Prices in local currency were converted to USD by using market exchange rates. Maximum and minimum prices of each pharmaceutical and non-pharmaceutical technology were compared to mean prices in the study countries. Results: EPR regulations are most stringent in Egypt and Saudi Arabia (largest study countries), mandating the lowest pharmaceutical price out of a basket comprising more than 25 countries each. By contrast, Kuwait references the country of origin only. The average price corridor is +/-38% for pharmaceuticals and +/-78% for outpatient and hospital services compared to mean prices. Conclusions: EPR results in narrower price corridor for innovative pharmaceuticals compared to other health care services. Prices of innovative pharmaceuticals are the lowest in Egypt and Saudi Arabia, and the highest in Kuwait, indicating the importance of population size and EPR implementation on drug price levels. However, EPR results in higher pharmaceutical prices in lower income countries compared to non-pharmaceutical services, which may limit timely access of patients to new medicines in these countries compared to global markets. Stakeholders should understand the implications of EPR and develop solutions to prevent its negative consequences. PHP26 Quantification of Switching Trends in the Greek Pharmaceutical Market During the Period Of Crisis Ollandezos M 1, Charitonidis S 2, Papadopoulos K 1, Katsoulakis M 2, Kofinas K 2, Kyriopoulos J 3 1Pan Hellenic Union of Pharmaceutical Industry, Athens, Greece, 2IMS Health Hellas, Athens, Greece, 3National School of Public Health, Athens, Greece . . . . . . Objectives: In 2009, Greece entered into one of the most serious economic downturns in its modern history. In May 2010, the country was put under the supervision of Troika (EU, ECB and IMF). Retail pharmaceutical market with size c. a. € 6.5 bn in retail prices for 2009 (public pharmaceutical expenditure € 5.2 bn), was one of the main targets for change through the implementation of new policies and drastic spending cuts. The purpose of this study was to measure the impact that these new policies for pharmaceutical spending had and how this was attributed to the levers of price, market volume and product mix. Methods: An economic model was used, based on IMS Hellas’ and Hellenic Statistical Authority data, to measure the contribution that (a) price; (b) market volume; and (c) product mix had on the reduction in the size of the retail market for pharmaceuticals in the period 2009-2014. The detailed approach decomposes the market change, as measured in values, into those attributed to (a) price (assuming market volume and product mix are constant); (b) market volume (assuming prices and product mix are unchanged); (c) product mix. Results: The analysis indicated that, the major contributor in the reduction of the pharmaceutical market size was the price of medicines. It has been estimated that 87% of the reduction (> € 1.2 bn) is coming from price. Volume contributed an additional c. a. 38% (> € 0.5 bn) in the reduction of the size but, at the same time, it was partially offset from substitution with pricier medicines. Conclusions: Price was the major driver for the reduction in the size of the retail pharmaceutical market during the period of crisis. Volume had an impact as well but it was partially offset by switch towards more expensive medicines. PHP27 Is the French Liste–En–Sus Still Supporting Access To Innovative Medicines? Ortiz I 1, Conti C 1, Furniss S J 2 . . . . 1GfK, London, UK, 2GfK, Melton Mowbray, UK Objectives: The “liste-en-sus” was implemented within the framework of the HPST [1] law. One of its objectives was to ensure access to highly-priced innovative medicines in hospital settings without distorting the DiagnosticRelated Groups (DRG). The objective of this research is to analyse the degree of innovation of the medicines included in the “liste-en-sus”. Methods: Starting from the ATIH [2] database, an initial analysis consisted of identifying all the Health Technology Assessments (HTAs) for each product included in the “listeen-sus” available on the HAS website [3] . Then, for each HTA, the following information was collected: assessment date, SMR [4] and ASMR [5] scores. Results: The liste-en sus includes 123 medicines. 21% have no HTA available. Another 19% were last evaluated before 2004. Among the medicines which had undergone an HTA since 2004, 7% were granted an ASMR I, 27% an ASMR II, 22% an ASMR III, 8% an ASMR IV, 36% an ASMR V. In other terms, amongst the medicines which have undergone an HTA in the last 10 years, about 45% of them were deemed non-innovative (ASMR IV/V). Those medicines mainly consist in antihemorrhagics (27%), antianaemics (18%), antineoplastics (15%) and immune sera and immunoglobulins (15%)[6] . Although they are not innovative, those medicines are only used in a proportion of patients and are thus likely to distort DRG. To put these results into perspective, since 2005, 92% of evaluated medicines were granted an ASMR IV/V [7]. Conclusions: Looking at the list-en-sus’ objectives, the most decisive criterion seems to be more stability of the DRG rather than access to innovative medicines; however a higher proportion of the medicines in the listeen-sus are innovative. PHP28 A Multi-Stakeholder (Physician, Payer, Patient, and Industry) Qualitative Analysis of the Policies That Would Support A Sustainable European Biosimilars Medicines Market Combined With A Quantitative Analysis of the Multi-Stakeholder Benefits A Sustainable Medicines Market Would Deliver Whitehouse J T 1, Teale C W 1, Glover J C 1, Taylor C 1, Lino Mendonca V 2 1GfK Market Access, Melton Mowbray, UK, 2European Generic Medicines Association (EGA), Brussels, Belgium . . . . . . . . Objectives: To establish the key policy areas that will drive the establishment of a sustainable biosimilar medicines market. To outline the benefits that these will bring to Physicians, Payers, Patients, and Industry, with particular focus on the benefits for European National Health systems. Methods: 71 qualitative in-depth interviews were conducted across 7 European markets: France, Germany, Hungary, Italy, Poland, Spain and the UK, collecting insight from experts and policy influencers at pan-European, National and Regional levels, Physicians, Payers, Pharmacists, Patients, and Industry. Quantitative modelling used a systems dynamics approach with in-depth analysis of 3 representative biologic products: trastuzumab, bevacizumab, and adalimumab. Dynamics were based on a delphi panel of expert opinions. The five forces of supplier power, buyer power, impact of new entrants, impact of substitutes, and competitive rivalry were addressed. A ranking of the attractiveness of policy combinations from a sustainability and benefit perspective was made based on a biosimilar medicines market “Sustainability Index” and the calculation of the magnitude of the benefits (cost savings, additional patients treated) that the policy combination was likely to produce. Results: The qualitative analysis has shown that a European biosimilars medicines market based on stakeholder and policy alignment in four key areas (1. education and understanding, 2. experience and use, 3. sustainable pricing, 4. rational decision making) will be sustainable and deliver benefits to all stakeholders. The quantitative analysis demonstrated that the most efficient policy combination, measured in terms of the sustainability index, was the same for all 3 molecules and would deliver cumulative 10 year cost savings of between 24% and 26%. Conclusions: Greater stakeholder alignment and the combination of specific policies will increase the sustainability of the European biosimilar medicines market. A sustainable biosimilar medicines market will deliver significant benefits to all stakeholders. A410 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PHP29 Impact of Health Policy Changes on the Cost Sales of 5 Top Selling ATC1 Phramaceutical Groups in Turkey PHP32 Impact of Health Policy Changes on the Growth Locally Manufactured and Imported Pharmaceutical Markets In Turkey Yenilmez F B 1, Kockaya G 2, Saylan M 2, Safak Yilmaz E 3, Tatar M 1, Hilal Vural E 3, Vural I M 3, Akbulat A 3, Gursoz H 3, Artiran G 3, Kerman S 3 1Hacettepe University, Ankara, Turkey, 2Health Economics and Policy Association, Ankra, Turkey, 3Turkish Medicines and Medical Devices Agency, Ankara, Turkey Hilal Vural E 1, Kockaya G 2, Yenilmez F B 3, Saylan M 2, Safak Yilmaz E 1, Tatar M 3, Vural I M 1, Akbulat A 1, Gursoz H 1, Artiran G 1, Kerman S 1 1Turkish Medicines and Medical Devices Agency, Ankara, Turkey, 2Health Economics and Policy Association, Ankra, Turkey, 3Hacettepe University, Ankara, Turkey Objectives: Turkish Ministry of Health (MoH) initiated Health Transformation Program (HTP) in 2002. HTP impacted all clinical and economic outcomes of health including pharmaceutical sales by improving access to health services. Total pharmaceutical market reached US $ 8 billion in last 10 years. The objective of this study is to understand the differences in the impact of selected 5 policies on 5 top selling ATC1 groups in terms of cost sales (CS) in the respective periods. Methods: 132 months sales data with segmented regression analysis for interrupted time series were used. International reference pricing of pharmaceuticals (RF), mandatory reimbursement dossier submission for new molecules, new indications and line extensions with medical and economic evaluations (MRDS), auditing for good manufacturing practice (GMP), family physician system (FP) and compulsory medical service for physicians (CMS) were selected as five major policies that may affect cost, demand and supply of pharmaceuticals. We analyzed possible breaks in trends prior and after the implementation of 5 selected policies of the HTP. The Durbin-Watson d statistics of SPSS version 20.0 was used as a test for serial correlation of error terms. Shift in slope with p< 0.05 was considered as statistically significant. Results: There was an increasing trend for all ATC1 groups prior the implementation of policies. The trends in systemic antienfectives (J0), alimentary and metabolism (A0) and Respiratory system (R0) Central Nerveous System (N0) groups were negatively impacted from all policies except for RF. All policies impacted negatively the trend in cardiovascular system (C0) group. Implementation of RF had a significant positive impact on A0 and R0 group. Impact of RF was positive on C0 however it did not reach signifcance level. Conclusions: Policy changes were very successful to control growth of top selling pharmaceutical groups while improving access to health. RF and CMS policies were the least effective cost containment measures Objectives: Turkish Ministry of Health (MoH) initiated Health Transformation Program (HTP) in 2002. HTP impacted all clinical and economic outcomes of health including pharmaceutical sales by improving access to health services. The objective of this analysis is to to understand the impact of selected 5 major policy changes by MoH to sales of locally manufactured and imported pharmaceutical products in the respective periods. Methods: 132 months sales data with segmented regression analysis for interrupted time series were used. International reference pricing of pharmaceuticals (RF), mandatory reimbursement dossier submission for new molecules, new indications and line extensions with medical and economic evaluations (MRDS), auditing for good manufacturing practice (GMP), family physician system (FP) and compulsory medical service for physicians (CMS) were selected as five major policies that may affect cost, demand and supply of pharmaceuticals. The analysis was conducted for total imported pharmaceutical (IP) sales and total locally manufactured pharmaceutical (LMP) sales. The Durbin-Watson d statistics of SPSS version 20.0 was used as serial correlation. Shift in slope with p< 0.05 was considered as statistically significant. Results: The negative effect of RF policy change on CS trends was more prominent for IP than LMP sales. However, the shift in CS due to other 4 policy changes was lower for IP when compared with LMP sales. The differences reached statistical significance level except for CMS policy. Although not significant, positive shift of US due to RF policy change was higher for LMP than IP sales. There was a decreasing slope of LMP unit sales following MRDS and GMP policies but an increasing slope of IP unit sales. Conclusions: Policy changes may effect at differently direction and amount the unit and cost sales of LMPs and IPs. Cost control mechnism such as RF has a more negative effect on imported product as expected. . . . . . . . . . . . . . . . Objectives: To compare the savings achieved by appraisal of the clinical and cost effectiveness of individual drugs by NICE’S Technology Appraisals for England & Wales with those due to price cuts under the Pharmaceutical Pricing Regulation Scheme 2000-2014. Methods: Maximum and best estimates of savings attributable to the 512 technologies appraised and published by NICE to end 2013 with published reports of the PPRS. Published estimates of the cost of the Multiple Risk Sharing Scheme, the Cancer Drugs Fund and the End-of-Life criteria. Results: Savings attributable to NICE were relatively low. Few drugs were not recommended and of these, special schemes, funds and exceptions required by the governments of the day, reduced savings that might otherwise have resulted. The bulk of the savings due to NICE resulted from price cuts under Patient Access Schemes from 2009. Higher savings resulted from cross the board price cuts in the PPRS. Conclusions: Health systems aiming to control pharmaceutical expenditure should consider both individual level appraisal of drugs as overall budget control of the branded prescription drugs budget. PHP31 The Effects of Reforms, Price Cuts And Global Budget Implementation on Original/Generic Medicine Sales Which Have Annual Average Highest Amount Of Sales Between 2008-2013 In Turkey Vural I M 1, Vural E H 1, Yigit O 2, Dogan E 1, Babacan S 1, Tolun C 1, Uman N 1, Akbulat A 1, Malhan S 3, Kahveci R 4, Artiran G 1, Kerman S 1 1Turkish Medicines and Medical Devices Agency, Ankara, Turkey, 2Turkish Statistical Institute, Ankara, Turkey, 3Baskent University, Ankara, Turkey, 4Ankara Numune Education and Research Hospital, Ankara, Turkey . . . . . . . . . . . . . . . . . . . . Young K E 1, Rémuzat C 2, Urbinati D 3, Toumi M 4 1Creativ-Ceutical, Milan, Italy, 2Creativ-Ceutical, Paris, France, 3Creativ-Ceutical, Luxembourg, Luxembourg, 4University Aix-Marseille, Marseille, France Raftery J P University of Southampton, Southampton, UK . . . PHP33 An Overview of the Biosimilar Market in the Us PHP30 Saving Money in Health Care: Cost Effectiveness of Individual Drugs (As By Nice) Or Budget Cuts (As Under Pprs)? . . . . . Objectives: In 2003 Health Transformation Program and in 2006 Social Security Reform were launched in Turkey. At the end of the years 2009 and 2011 price cuts were done by the Government and between the years 2010-2012, there was a global budget implementation in Turkey. The top 100 medicines, annual average highest amount of sales of between 2008-2013, had one to four of the total pharmaceutical market value in 2013. We aimed to examine the status of original and generic medicines in the first 100 medicine’s and evaluate the effects of policy interventions on these medicine sales. Methods: While pharmaceutical sales data were obtained from the IMS Health-Turkey data base, prices and characteristics of medicines were obtained from the Turkish Medicine and Medical Devices Agency data bases. Each group (original/generic medicines) was analyzed using TRAMO and SEATS method. Results: 78 medicines are original, 22 medicines are generic. In 2009 compared to the previous year both generic and original medicine spending increased by respectively 16% and 20,9%. Between the years 2010-2012 compared to the previous year both generic and original medicines spending decreased by different ratios. In 2013 according to 2012, generic medicine spending decreased by 8.3% but original medicine spending increased by 7.9%. In 2013 total original medicine spending was 2888 million Turkish Liras (TL). This amount estimated to be 3050 million TL in 2014 and 3145 million TL in 2015. Generic medicines total spend was 501 million TL in 2013. This amount projected to be 510 million TL in 2014 and 515 million TL in 2015. Conclusions: Original medicines dominated the top 100 medicines. The total effect of the intervention is more negatively in generic medicines. At the end of 2013, it is understood that original medicines total amount have a tendency to increase. . . . . Objectives: This study intends to provide an overview of the US biosimilar policies and its impact on the development of the biosimilar market in this country. Methods: A literature review was conducted from the US Food and Drug Administration (FDA) website, Generics and Biosimilars Initiative (GaBi) websites, Medline® database, and available grey literature. Results: The Biologics Price Competition and Innovation Act (2009) established an abbreviated Biologic License Application (aBLA) pathway/351 (k) for biosimilars in addition to the 1. Non-abbreviated biologic license application (BLA) /351 (a); 2. New Drug Application (NDA) /505 (b) (2); or 3. Abbreviated New Drug Application (ANDA). 10 follow-on biologics under NDA/ANDA and one under BLA were previously approved. Product identity and therapeutic equivalence of some biologics (i. e. Lovenox, Copaxone) led to important debates for defining the application pathway. Currently, the FDA has no 351 (k) approvals. The lack of clear FDA guidance on data requirements for biosimilarity in aBLA was a limiting factor for manufacturers to go through aBLA pathway. They preferably opted for a classical BLA pathway due to its longer exclusivity period and nearly the same amount of data required (i. e. Neutroval). The impact of the recently released FDA draft guidance on designing clinical studies for biosimilarity (5/13/14) is yet to be seen but should address issues on proving biosimilarity in aBLA. Further debate is anticipated but an increase in applications is expected and biosimilars savings are projected at $250 billion by 2024. Conclusions: The biosimilars market is still lagging, specifically compared to the EU, with no 351 (k) approvals despite the US’ leading position in the biopharmaceuticals market. However, the US’ high prices for innovative products and history of generic utilization can signify a positive market projection after a transition period, as was seen in Germany and Sweden. Further, the new FDA guidance by addressing biosimilarity issues may ease biosimilar market entry. PHP34 Impact of Health Policy Changes on the Growth Locally Manufactured and Imported Pharmaceutical Markets of Top Selling Atc1 Phramaceutical Group (Alimentary and Metabolism (A0) In Turkey Kockaya G 1, Yenilmez F B 2, Saylan M 1, Safak Yilmaz E 3, Tatar M 2, Hilal Vural E 3, Vural I M 3, Akbulat A 3, Gursoz H 3, Artiran G 3, Kerman S 3 1Health Economics and Policy Association, Ankra, Turkey, 2Hacettepe University, Ankara, Turkey, 3Turkish Medicines and Medical Devices Agency, Ankara, Turkey . . . . . . . . . . . . . Objectives: Turkish Ministry of Health (MoH) initiated Health Transformation Program (HTP) in 2002. HTP impacted all clinical and economic outcomes of health including pharmaceutical sales by improving access to health services. Total pharmaceutical market reached US $ 8 billion in last 10 years. The objective of this analysis is to understand the impact of selected 5 major policy changes by MoH on the growth locally manufactured and imported pharmaceutical markets of top selling ATC1 phramaceutical group, which was Alimentary and Metabolism (A0) with US $ 1.1 billion sales in 2012, in the respective periods in Turkey. Methods: 132 months sales data with segmented regression analysis for interrupted time series were used. International reference pricing of pharmaceuticals (RF), mandatory reimbursement dossier submission for new molecules, new indications and line extensions with medical and economic evaluations (MRDS), auditing for good manufacturing practice (GMP), family physician system (FP) and compulsory medical service for physicians (CMS) were selected as five major policies that may affect cost, demand and supply of pharmaceuticals. We analyzed possible breaks in trends A411 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 prior and after the implementation of 5 selected policies of the HTP. The analysis was conducted for total imported pharmaceutical (IP) sales and total locally manufactured pharmaceutical (LMP) sales in the A0. The Durbin-Watson d statistics of SPSS version 20.0 was used as a test for serial correlation of error terms. Shift in slope with p< 0.05 was considered as statistically significant. Results: All policies effected the LMP sales more positively than IP sales except FP. However, the difference of impact was moderately positive for LMP sales, there was not any statistically significant change. Conclusions: Policy changes may effect at differently direction and amount the cost sales of LMPs and IPs. Non significant effect of these policy changes may partly explained by limited oberservation time and by other market dynamics. PHP35 Characteristics of The Medicines Which Have Annual Average Highest Amount Of Sales Of Between Years 2008-2013 Akbulat A 1, Dogan E 1, Babacan S 1, Uman N 1, Tolun C 1, Vural I M 1, Vural E H 1, Safak Yilmaz E 1, Dasdag M M 1, Kahveci R 2, Malhan S 3, Artiran G 1, Kerman S 1 1Turkish Medicines and Medical Devices Agency, Ankara, Turkey, 2Ankara Numune Education and Research Hospital, Ankara, Turkey, 3Baskent University, Ankara, Turkey . . . . . . . . . . . . . . . . Objectives: In 2003 Health Transformation Program and in 2006 Social Security Reform were launched in Turkey. At the end of the years 2009 and 2011 price cuts were done by the Government and between the years 2010-2012, there was a global budget implementation in Turkey. Health implications of these developments in the medicine market have not been analyzed in a comprehensive manner. The top 100 medicines, annual average highest amount of sales of between 2008-2013, had one to four of the total pharmaceutical market value in 2013. In this study we aimed to determine these first 100 medicine’s, which have higher total sales amount, defining characteristics. Methods: While pharmaceutical sales data were obtained from the IMS Health-Turkey data base, characteristics of medicines were obtained from the Turkish Medicine And Medical Devices Agency and the Social Security Agency data bases. Results: While 78 medicines are original, 22 medicines are generic. 60 medicines are imported medicines, 40 medicines are manufactured medicines. 19 medicines are biotechnological medicines and all of these biotechnological medicines are original and imported. 96 medicines covered by Social Security payments. Equivalent of 65 medicines are available (each equivalent group from 1-30, an average of 13 generics available). 19 medicines and 15 medicines are respectively systemic anti-infectives and antineoplastics and immunomodulating agents. In this study, the license holders of the medicines are 40 firms in total. 15 firms have a market share of 75% and the medicines which, have 77,7% portion of total amount, are created by multinational firms. Conclusions: In the years which were the effects of reforms, price cuts and global budget implementation seen, the medicines which have higher total sales amount were mostly original, imported, covered by Social Security payments and created by multinational firms. PHP37 Impact Of Health Policy Changes on Unit Sales of 5 Top Selling Atc1 Phramaceutical Groups In Turkey Saylan M 1, Safak Yilmaz E 2, Yenilmez F B 3, Kockaya G 1, Tatar M 3, Hilal Vural E 2, Vural I M 2, Akbulat A 2, Gursoz H 2, Artiran G 2, Kerman S 2 1Health Economics and Policy Association, Ankra, Turkey, 2Turkish Medicines and Medical Devices Agency, Ankara, Turkey, 3Hacettepe University, Ankara, Turkey . . . . . . . . . . . . PHP39 The Grass is Always Greener on the Other Side or Why There Is Little Meaning in International Pharmaceutical Price Comparison Bierbaum M 1, Düttmann S 2, Amler N 1, Döpfer S 2 Erlangen-Nürnberg, Nuremberg, Germany, 2FAU ErlangenNürnberg, Nuremberg, Germany . . . . 1Friedrich-Alexander-Universität Objectives: In many countries there is an ongoing debate about pharmaceutical pricing. Especially in Germany there is common belief that pharmaceuticals are more expensive than in other countries. Payers are making use of this belief when negotiating prices with pharmaceutical companies. We wanted to know if there is scientific evidence for this conception. Methods: We conducted a systematic literature review in order to identify price comparison studies comparing Germany with at least three other countries. We searched Pudmed and six other databases to identify relevant articles published between 1998 and today. Furthermore we developed a quality rating tool based on the approaches from Andersson (1993) and Danzon/Kim (1998). Results: Our review delivered 4.927 articles from which 28 met our inclusion criteria. Study quality was quite heterogeneous, ranging from 3 to 13 points with an average score of 8.8 out of 15. Some studies use old data back from 1992 and no study considers the recent changes in German legislation (AMNOG). In addition no study includes rebates and selection of compared pharmaceuticals is often arbitrary. Reviewed studies report German pharmaceutical prices slightly above international average. High quality studies (upper quartile, quality score: > 9) find German prices below international average, whereas low quality studies (lower quartile, quality score: < 7) find German prices above international average. Conclusions: Results of the review suggest that there is a misconception of pharmaceutical pricing in Germany. Within a price comparison study any desired result can be achieved by deliberately choosing different approaches. At the end of the day payers and policy makers should stop comparing prices with other countries. Instead resources should better be spent on making value based reimbursment decisions in the respective health care setting. . Objectives: Turkish Ministry of Health (MoH) initiated Health Transformation Program (HTP) in 2002. HTP impacted all clinical and economic outcomes of health including pharmaceutical sales by improving access to health services. The objective of this study is to understand the differences in the impact of selected 5 policies on 5 top selling ATC1 groups in terms of unit sales (US) in the respective periods. Methods: 132 months sales data with segmented regression analysis for interrupted time series were used. International reference pricing of pharmaceuticals (RF), mandatory reimbursement dossier submission for new molecules, new indications and line extensions with medical and economic evaluations (MRDS), auditing for good manufacturing practice (GMP), family physician system (FP) and compulsory medical service for physicians (CMS) were selected as five major policies that may affect cost, demand and supply of pharmaceuticals. We analyzed possible breaks in trends prior and after the implementation of 5 selected policies of the HTP. The top 5 selling ATC1 groups were Systemic Antienfectives (J0), Cardiovascular System (C0), Alimentary and Metabolism (A0), Respiratory (R0) and Central Nerveous System (C0). The Durbin-Watson d statistics of SPSS version 20.0 was used as a test for serial correlation of error terms. Shift in slope with p< 0.05 was considered as statistically significant. Results: There was an increasing trend for all ATC1 groups prior the implementation of policies. The trends in J0 were negatively impacted from all policies except for RF. The C0 group was negatively impacted from all policies except for RF and CMS. The A0 group was positively impacted from all policies. The N0 was positively impacted from all policies except MRDS and FP. The R0 group was positively impacted from all policies except GMP and FP. Conclusions: Policy changes were not sufficient to control unit growth of top selling pharmaceutical groups. The effect of other policies to control unit sales of these group should also be evaluated. PHP38 Applicability of Turkish Pricing Policy on Price Increases Beykoz V , Saylan M Novartis Pharma, Istanbul, Turkey . increase. We excluded “plasma-derived blood products” that has different pricing schemes for the cases they get an exchange rate related increase and also excluded the price corrections. Price increase reasons were grouped as defined in the current decree. We calculated mean percentage of price increases in overall and in original vs. generic, import vs. locally manufactured, products below and above 6.79 TL pack price (local definition of cheap products) and ATC1 level subgroups. Results: 606 products and 274 molecules had price increase in 2013. The most frequent reasons for price increase were related critical product status (110 increases), increase in reference price (105 increases) and due to the rule of getting the highest reference for delisted or non-reimbursed products (89 increases). The average rate of price increases was 23.1%. 57% of the price increases were applied to imported products compared with 43% locally manufactured ones. Average rate of price increase in original and generic products were 24% and 21% respectively. Products with exfactory price above 6.79 TL had more price increase compared with cheap products. The highest rate of price increases were in ATC1 groups A and B (81 increases each) Conclusions: With the pricing policy change in 2013, price increases were possible in Turkey for products from different groups. . Objectives: Turkey is using international reference pricing for the pharmaceuticals. An update for Turkish pricing decree is published in April 2012 and price increases became more applicable due to various reasons. We analyzed price lists to identify pharmaceutical products that had any price increase in 2013 and defined causes in the decree Methods: We reviewed weekly cumulative price lists published in Turkish Medicines and Medical Device Institution and compared each list with the list published in the previous week to identify products that had price PHP40 Transformation of Green Card Program for the Poor: One Step Further to Universal Health Care Coverage in Turkey Seyhun O , Erdol S , Can H , Erdogan E Medtronic, Inc., Istanbul, Turkey . . . . Objectives: Since January 2012, Turkish government started to implement a mandatory general health insurance law. The Social Security Institution (SSI) coverage which has been 86% in 2012 was expected to be 100% after this reform instead has declined to 82% in 2013. On the other hand, as Turkish Green Card Program - a state social scheme to ensure the provision of health services for the poor - was abolished and transferred from Ministry of Health to SSI, approximately 9 million people have been subject to income audit in order to be classified as eligible to pay premiums themselves and non-eligible ones for which the state will pay the premiums. In this regard; this poster presents this transformation and gives its current status in terms of contribution to Universal Health Coverage in Turkey. Methods: Publications of SSI, World Bank Reports and online articles are utilized. Results: The recent available data shows that 82% of the population is under the coverage of SSI. %2 of the population corresponds to the groups which are out of SSI coverage according to Law 5510, Article 60. Remaining population corresponds to 12 million people subject to income audit. Approximately 62% of those could not pass income audit and their premiums are paid by the state. Conclusions: Since 2012 there are increasing number of people taking income audit and have the capacity to pay premiums. It is also noteworthy to state that there were 8,865,470 people under Green Card scheme whereas after the transfer of Green Card to SSI; 11,357,306 people applied for the income audit in 2012, this reached to 12,266,043 people at the beginning of 2014. Findings show that as Green Card Scheme brought under SSI, coverage has been made available to all eligible people in a systematic and just way thus contributed the extent of coverage in Turkey. PHP41 Generic Penetration Within Top-10 Genericized Molecules – Greece Versus Major European Countries Charitonidis S 1, Kofinas K 1, Katsoulakis M 1, Papadopoulos K 2, Ollandezos M 2, Kyriopoulos J 3 1IMS Health Hellas, Athens, Greece, 2Pan Hellenic Union of Pharmaceutical Industry, Athens, Greece, 3National School of Public Health, Athens, Greece . . . . . . Objectives: In early 2010, Greece was placed under International Supervision (EU, ECB and IMF), as a result of a growing public deficit and its non-sustainable state expenditure. At the time, the retail pharmaceutical market had a size of c. a. €6.5 bn in retail prices (public pharmaceutical expenditure €5.2 bn). Within that framework, A412 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 the reduction of pharmaceutical expenditure was one of the main targets for fiscal adjustment. The purpose of this study was to assess the usage of generic medicines in Greece, as a key driver for savings from the pharmaceutical market, and compare it with that of other major European countries. Methods: IMS data from several European countries was collected in terms of the overall retail pharmaceutical market and the consumption of generic medicines in each of them. Focus was placed on the top-10 genericized molecules for Greece including alendronic acid, atorvastatin, carvedilol, clarithromycin, clopidogrel, donepezil, fluconazole, olanzapine, omeprazole and simvastatin. To ensure an “apples-to-apples” comparison across countries, the study focused on the retail market only (excluding hospitals) given that, in Greece, there is no publicly-available data for pharmaceutical consumption within the hospital setting. Results: The analysis indicated that, in Greece, the penetration of generic medicines within the top-10 genericized molecules was 65% in terms of volume, with off-patent medicines holding the remaining 35%. According to IMS data, Greece had a lower penetration of 21 percentage points (-24%) versus the weighted European average and -8.2 percentage points (-11%) versus the cluster of Southern European countries plus Ireland for calendar year 2013. Conclusions: Generic penetration within the 10 largest genericized molecules, in Greek retail-pharmacy setting, is significantly lower versus the weighted average of major European countries but also compared to that of countries in similar economic situation with Greece. PHP42 Promoting Quality Use of Generic Medicines: Knowledge, Attitudes And Practices of Community Pharmacists in A Middle Eastern Country Awaisu A , Mohamed Ibrahim M I , Kheir N , ElHajj M Qatar University, Doha, Qatar . . . . . Objectives: The practice of generic medicines prescribing, dispensing and substitution in developing countries has been controversial among health care professionals, particularly due to issues on quality, safety and efficacy. These controversies are as a result of inter-country differences in policies and laws as well as individualized knowledge and attitudes of pharmacists pertaining to generic medicines. This study primarily aims to assess the knowledge, attitudes, and practices of community pharmacists in Qatar towards generic medicines. Methods: A cross-sectional study using a pretested paper-based survey was conducted among a random sample of community pharmacists in Qatar. The data were analyzed using IBM–SPSS® version 20. Both descriptive and inferential statistical analyses were applied. Results: A total of 160 surveys were distributed to community pharmacists of which 118 were returned (response rate, 74%). The mean total score of generic medicines knowledge among the pharmacists was 6.8 ± 1.6 (maximum possible score was 10). Years of practice as well as place of obtaining academic degree did not influence knowledge score. Approximately 72% of the pharmacists supported generic substitution for brand name drugs in all cases where a generic medicine is available and the majority (93%) agreed that pharmacists should be given generic substitution right. Nearly 61% of the pharmacists considered lack of proven bioequivalence to original brands as an important barrier for selecting generic medicines and 55% rated “lack of policy for directing the practice of generic medicine” as an important barrier. Conclusions: In order to enhance the quality use of and to promote the practice of generic medicines in Qatar, an educational program should be implemented. A national generic medicine policy and guidelines are warranted in the State of Qatar. PHP43 Evaluating Rates of Potential Generic Substitution for Prescription Drugs: Can We Improve On Existing Incentive Schemes? Walton S 1, Rash C 1, Lambert B L 2, Galanter W L 3 1University of Illinois at Chicago, Chicago, IL, USA, 2Northwestern, Chicago, IL, USA, 3University of Illinois Hospital and Health Sciences System, Chicago, IL, USA . . . . . . Objectives: Encouraging generic drug use has reduced health care costs for payers and consumers, but the rate of branded medications therapeutically interchangeable to generic is not equal across medications disease states. The extent to which different systems of care are able to substitute towards generics is not well understood. This study defined and measured the maximum generic rate (MGR) of currently prescribed drugs and illustrated differences across drugs associated with selected underlying diseases. Methods: Using information in prescription claims data, drugs were classified into “potentially generic substitutable” or not based on clinical consensus regarding the following algorithm: 1. They did not having have a narrow therapeutic index (NTI) as defined by the Food and Drug Administration (FDA); 2. Did not belong to one of six protected classes of drugs in Medicare D; 3. They were substitutable with a generic medication of the same chemical entity; 4. They were therapeutically interchangeable with a generic in the same class which was thought clinically to have class effect benefits and minimal risk of harm in switching. A maximum generic rate (MGR) was defined as the percent of prescriptions that could be generic. This rate was examined overall and across drugs known to be associated with illustrative diseases including hypertension, diabetes mellitus, and obstructive lung disease. Results: The MGR ranged from 100% for drugs used in hypertension to 26.7% for drugs used in obstructive lung disease. The overall maximum possible generic rate (MGR) was 83.6%. Conclusions: Payers wishing to promote generic substitution should incorporate the potential for substitution toward clinically appropriate generic medications as part of incentives for generic utilization to avoid unintended consequences of using a fixed target rate. A practical methodology for determining a MGR is offered here. Objectives: Medicine shortages are a global phenomenon. A growing number of reports indicate the problem is increasingly affecting the European pharmaceutical market. The present study aims to investigate the characteristics, determinants, legal aspects and management of medicine shortages in Belgium, France and from the perspective of the European Union. Methods: A review of scientific and grey literature was performed. The legal framework on European and national level was reviewed. Primary qualitative data was collected through 22 semi-structured interviews with key representatives of health care systems’ stakeholders on the national and European level. Results: France reported three times more shortages than Belgium. However, the main therapy area, the major cause and the dynamics of medicine shortages were analogous between the two countries. Determinants of medicine shortages were categorised in manufacturing problems, distribution and supply issues, and economic-related challenges. Manufacturing problems were most frequently reported as the primary cause of medicine shortages. Laws and regulations related to medicine shortages are more extensive in France than Belgium. Several preventive and responsive measures were identified to address such shortages. Conclusions: Although medicine shortages are country-specific, the underlying mechanisms of medicine shortages appear to be similar in Belgium and France. Economic aspects seem to play a central role in the phenomenon of medicine shortages, as it influences stakeholders’ business decisions. The impact of the legal framework around medicines on the occurrence of medicine shortages may be limited. Collaboration, communication and coordination are key to any effective approach to address medicine shortages. PHP45 Expedited Regulatory Review and Authorisation of Medicines and Their Subsequent Appraisal By Hta Agencies in the European Union McCarron C 1, Griebsch I 2 1Boehringer Ingelheim, Berkshire, UK, 2Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany . . Objectives: Firstly, to review whether medicines that have received expedited regulatory approval in the European Union (EU) subsequently received positive recommendations from selected Health Technology Assessment (HTA) agencies in the EU. A second objective was to understand the reasons for negative appraisals of these medicines by EU HTA agencies. Methods: Medicines that have received Conditional Marketing Authorisation (CMA) by the European Medicines Agency (EMA) were identified from the EMA website. Websites of HTA agencies in the United Kingdom (NICE, SMC); Ireland (NCPE), France (HAS) and Germany (IQWiG) were searched to ascertain whether these medicines had been appraised. Appraisal reports were reviewed. For medicines with at least one negative appraisal, reason (s) for the negative appraisal were identified. Results: 13 medicines have received CMA in the EU. 10 of these medicines have received a negative appraisal by at least one HTA agency (either ‘not recommended’ or deemed to offer ‘no additional benefit’ over existing treatment). Of the 10 medicines with at least one negative appraisal, 8 are oncology medicines. The most common reason for a negative appraisal (relevant to 9 of the 10 medicines) was insufficient or uncertain clinical effectiveness. Uncertain cost-effectiveness in relation to the proposed treatment cost was mentioned as a source of uncertainty for 5 of the 10 medicines. Conclusions: Despite receiving expedited regulatory approval, patient access to innovative medicines in the EU may be delayed due to negative appraisal of these medicines by HTA agencies (most often due to a relative lack of data on clinical effectiveness). Such delays obviate the intention of expedited regulatory review of medicines. PHP47 Challenges: in Implementing HTA In The Reimbursement Decisions In Algeria / A Compartive Analysis Aissaoui A 1, Soualmi R 2, Kaddar M 3 1Paris dauphine University, Paris, France, 2Algiers University, Algiers, Algeria, 3Independant Consultant, Oran, Algeria . . . Objectives: To provide a comprehensive description of the current Drug Reimbursement Systems in Algeria and to compare it to two archetypes drug reimbursement systems in France and UK and to a system in a middle income country: Turkey where the HTA has been recently implemented. Methods: We collected and reviewed relevant information to describe the health care and drug reimbursement systems in these countries; we reviewed the legal framework and procedure documents. For Algeria, in addition to the data and information collected, we conducted informal interviews supplemented by a survey among key stakeholders. Results: Compared to the UK, no similarities were found. This is probably due to the cultural differences and the lack of expertise in the use of cost-effectiveness approaches. Compared to the France, we didn’t find similarities, except the final decision which is taken at the Ministry level. This is due to the administrative nature and the lack of transparency of the assessment in Algeria especially where the methods as well as the results of the assessment are not explicitly expressed. Compared to Turkey, we found some similarities in terms of process, but not in terms of methods as this country is now more familiar with the HTA approach. Conclusions: Our study shows that the implementation of HTA differs according to cultural and financial factors and to expertise capacity in data collection, analysis and use in the decision making process. The use of HTA in the drug evaluation and reimbursement system in Algeria is underestimated and underdeveloped. That’s why before adopting HTA approaches in the pharmaceutical sector, the Algerian authorities should consider these factors and improve the data quality and decision process transparency. This is becoming vital as cost of drugs is increasing and the fiscal space will be more constrained in the near future. PHP44 Prevention and Management of Medicine Shortages in Belgium, France and from The Perspective of the European Union PHP49 An Analysis of the 2014 Moroccan Drug Price Cuts Bogaert P , Prokop A , Bochenek T Jagiellonian University Medical College, Krakow, Poland Hollis M , Ando G , Izmirlieva M IHS, London, UK . . . . . . A413 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: The Moroccan government announced new public and hospital prices for 5308 drugs in April 2014. These price cuts came into force in June 2014. The study examines the price changes and analyzed the potential impact of the price cuts on pharmaceutical companies operating in the market. Methods: Standard statistical methods were used to analyze government provided data in order to identify general trends within the pricing notification. Secondary sources were then examined to determine the general pricing rationale. Results: The price notification included 5308 drugs (note duplication occurs where a brand name has multiple formulations). Examining public prices: 390 of the drugs were newly listed on the list for public purchase, a further 2968 saw no price increase and two drugs saw modest price increases. Public price declines were seen for 1948 drugs with the level of decline covering a range of 1635 percentages between 0.006% and 78.6%. Average public price, excluding drugs with no decline, saw an average decline of 13.6% and median decline of 6.5%. All the 5308 drugs already had an existing hospital price, however 114 of these drugs saw no pricing change. Of the 5194 drugs which did see a price decline this varied across a range of 2559 different percentages between 0.001% and 79.8%. The average hospital price decline, excluding drugs which saw no decline, was 9.3% and the median was 5.8%. Conclusions: The large range of pricing declines suggests that the prices were not indicative of a formulaic event such as currency fluctuation changes. This also means that there is a differential impact on pharma companies, with certain parts of their portfolio being more protected than others. PHP51 Trends in Time To Market Access In Europe - Is It Getting Better? Flostrand S J A 1, Lor S 1, Anon Y 2 1IMS Consulting Group, La Défense Cedex, France, 2IMS Health, La Défense Cedex, France . . . . . Objectives: Unequal access due to delays or non-availability of medicines denies patients treatments they need and contributes to inequality across Europe. The EFPIA previously monitored access delays via the “Patients Waiting to Access Innovative Therapies (WAIT)” indicator, however; this has not been updated since 2011. Following the 2012 EC Transparency Directive 89/105/EEC mandating no more than 120 days of delay for national pricing and reimbursement decisions, an updated evaluation is needed. Methods: IMS MIDAS data was used to identify the first sale for all EMA-approved non-generic, non-biosimilar products since 2008 in 15 EU markets. The delay (from approval to first sale) was noted for each product and market and compared to the EFPIA Patients WAIT indicator for 2008-2010. Annual average delays were calculated over the 2008-2013 period and for available products, an analysis of minimum and maximum delay performed for each market. Additional analyses were performed for trends within and across markets, and to identify progress and variations. Results: Improvements across the 2008-2012 period give cause for optimism in most European markets, as delays have fallen sharply in almost all markets, from an average of 524 days in 2008 to 281 days in 2012, with only Greece showing no real improvement. However; absolute availability of new medicines has varied greatly by market, with same-year availability of EMA-approved medicines declining to just 41% on average in 2013. Notable declines were seen in Greece, Spain, Romania and Portugal, suggesting that economic difficulties may lead some governments to refuse reimbursement, rather than simply delay decisions. Conclusions: This updated access analysis shows that while delays are shortening, absolute access for patients is not always improving. Some new access rules such as “class C” drugs in Italy (access without funding), may send misleading signals, and underscore the need for monitoring of true access across Europe. PHP54 Market Access for Medical Devices: Adapting to Change Charter R 1, Bottegoni G 2, Stengel C 1, Pistacchio L 1, Lombardi L 3, Jommi C 4, Tarricone R 5 1SDA Bocconi (MIHMEP), Milano, Italy, 2D3-Computation, Italian Institute of Technology, Genova, Italy, 3Bocconi University, Milano, Italy, 4Cergas, Bocconi University, Milano, Italy, 5Bocconi University, Milan, Italy . . . . . . . Objectives: Market Access (MA) is key for medical devices given recent changes in regulation from one side, and pressures on costs from the other side. Nevertheless, there is scant evidence on how MA works in the medical device industry and whether it works effectively. The aims of the present study are to illustrate current gaps in MA for medical devices and to recommend future actions to make this function more effective. Methods: We conducted a literature review aimed at gathering a clear picture of current state of the art on MA for medical devices and at highlighting major gaps vis-à-vis other industries (e.g. pharmaceuticals). Based upon the review’s findings, we carried out an e-survey to small, medium, and large device companies operating in the EU in order to understand: i) how MA is perceived and actually organised, ii) which are the main perceived obstacles to MA, iii) how companies invest in market access and at which stage of medical device development, and iv) what the main challenges are to a more efficient market access process. Results: The findings of the literature review and preliminary results of the e-survey are presented. Conclusions: MA is an under-developed function in the medical device industry and in certain cases its role is not fully exploited. Although the pharmaceutical industry doesn’t represent a fully appropriate benchmark, MA for medical devices is even less developed when compared to its full potential. A different approach is needed to deal with new changes in the regulation and policy arenas aimed at synergistically encompassing knowledge, expertise and operating tools from public affairs, medical departments, pricing & reimbursement, health economics and marketing. PHP55 Qualitative Research on the Impact Of European Hospital Procurement Practices on Market Access For Disposable Medical Devices Hertz D , DiPaolo A , Armstrong S GfK Market Access, Wayland, MA, USA . . . Objectives: In European hospitals, budgetary pressure is driving centralized purchasing and increasing the importance of procurement in market access. Centralized purchasing is thought to lead to cost savings through stronger negotiating power, economies of scale and reduced inventory. The objective of this study was to understand the procurement methods for disposable medical devices employed by European hospitals and their impact on market access. Methods: A review of hospital purchasing practices was conducted in France, Germany and England, including published articles and government websites. Twelve 45-60 minute telephone interviews were conducted with hospital decision makers (purchasers, pharmacists and clinicians) in 2013. Results: Most disposable medical devices are evaluated at the facility level by committees made up of clinicians, procurement specialists, pharmacists and management. France and Germany require listing on the facility formulary and typically procure via tendering and group purchasing. England has a national procurement body but hospitals regularly purchase products directly from the manufacturer. All three countries report that austerity measures are causing many facilities to limit the number of products per category. Product trials and clinician preference are seen as the main drivers of adoption however clinicians are under increasing pressure to justify costs. Hospital purchase decisions are increasingly likely to be based on non-product specific attributes such as total cost of purchasing including volume discounts, broad portfolio of products, and administrative costs. Conclusions: In the current environment, market access for disposable medical devices will require a strong economic argument in addition to evidence of superior quality. Although clinicians still drive the adoption of consumable medical devices, in the absence of a strong clinician preference, non-product specific attributes may determine market access. Manufacturers can support their economic messaging along categories of inventory reduction and a broad portfolio of products that allows for administrative efficiencies and reduced shipping costs. PHP56 The Impact of the Information System of Medical Devices in Portugal Silva E , Inacio P , Figueira J INFARMED_ National Authority for Medicines and health Products, Lisboa, Portugal . . . Objectives: The Portuguese National Health System (NHS) identified the need to easily access information that allowed identification of all medical devices (MDs) with their respective manufacturers, distributors, characteristics, and purposes. The objective of the creation of an information system is to collect, store and analyze data of each MD and to provide NHS resources management, health technology assessment (HTA), as well as data about the Portuguese market. Methods: A system consisting of a database that is a repository, and an application system that stores the information was built. For the registry of information all MDs should disclose their respective instructions of use, labels and price. The concept of unique medical device and a process to assign a code to each device was designed, and tested by collecting data from NHS acquisitions. The prescription of medical devices is made by code, with all information online. A model for communication between the health regulatory agency and market operators was also devised, as well as rules and codes of procedure. Results: The creation of this system in 2011 allowed for the coding of 20% of all registered MDs (n= 838000) up to June 2014. This has allowed for the elimination or withdrawal of 16% (n= 136000) of references, thus keeping the registries up to date. The coding permitted the comparison of devices and the management of total expenditure in MDs by hospitals of the NHS. Conclusions: The system provides solid information about available MDs in the market, their respective manufacturer, distributor and pricing to which they are sold to the NHS. The system is a support for decision makers, payers, and HTA processes. It facilitates the management and traceability of medical devices in hospitals and has impact on reimbursement, access and pricing varies by market. PHP57 Pricing of Medicines In Poland – Two-Year Overview Of How The New Law Affected Costs Of Reimbursement Szmurlo D , Wladysiuk M , Plisko R HTA Consulting, Krakow, Poland . . . Objectives: Rapid increase of costs of reimbursement in Poland led to redesign of the drug pricing mechanisms. As a result a new law which brought revolutionary changes to the health care system was introduced in 2012. The aim of this work is to summarize the two years of new drug policy in Poland and to present mechanisms that resulted in substantial savings in reimbursement costs. Methods: Few savings-generating mechanisms were implemented in the new reimbursement law: negotiations of prices, new method of calculation of reimbursement limit (resulting in higher co-payment) and also formal obstacles in obtaining prescriptions for reimbursed products (some reimbursed drugs, i. e. antibiotics, due to formal issues are prescribed with 100% co-payment). The first and the third reason result in lower overall costs of reimbursed medicines, while the second reason influences the proportion of costs borne by the National Health Fund. We have analyzed data on prices of reimbursed drugs before 2012 and after implementation of the new law (currently the prices may change every two months). We have then compared the amount of sales of those drugs, costs of reimbursement and patients’ co-payment. For some reference groups (i. e. oral aromatase inhibitors) we made in-depth analysis of pricing mechanisms. Results: In 2012 total public expenditures amounted to 2.3 billion EUR, 83% of the planned budget, which gave 460 million EUR savings as compared to 2011. In 2013 the drug budget was executed in 87%, resulting in 430 million EUR savings. Saving were driven by price negotiations, but also by mechanisms of lowering the reimbursement limit, i. e. in case of oral aromatase inhibitors the reimbursement limit was lowered by half during 2012-2013. Conclusions: The new reimbursement law resulted in substantial savings in costs of drugs. The success is however relative, as costs of drugs were partially shifted towards patients. A414 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PHP58 Therapeutic Positioning Report: New Collaborative Network of Drug Assessment in Spain - The Start Of P&R Based On Relative Effectiveness? Paladio N , Cortés I , Gil A , Crespo C Boehringer Ingelheim, Sant Cugat del Vallès (Barcelona), Spain . . . . Objectives: Therapeutic Positioning Reports (TPRs) were introduced to the Spanish P&R process in May 2013. TPRs evaluate comparative efficacy and safety and introduce usage and monitoring criteria for new drugs and existing drugs seeking reimbursement. The procedure was set-up by the Spanish Medicines Agency (AEMPS) and the Ministry of Health with the aim to accelerate the P&R process and to generate a single, national relative effectiveness report avoiding additional regional evaluations, contributing to reducing the long-delays in market access timelines experienced in the last few years. The objective of this ongoing work is to describe and analyze both the metrics of the process and the contents and results of TPRs. Methods: Descriptive study based on public information available from the European and Spanish Medicines Agencies (May 2013-ongoing). Results: From June 2013 to May 2014 the Therapeutic Positioning Coordinating Group (TPCG) has officially launched 60 TPR plus 7 use protocols (UP) at the request of the Interministerial Price Council (IMPC). 75% of TPR involve new chemical entities (82% of those with a positive opinion from the CHMP between May 2013-April 2014) and assess a single entity (48% specialty care) indicated, mainly, for patients with neoplastic (28%) and endocrine (18%) diseases. All UP affect targeted therapies with approval between Jan2010-Dec2012. Since its establishment in May 2013, the TPCG has released 26 TPR (57% of those expected according to the pre-established procedure -3 months working time) with a median release time of 7,1 months. Only 1 has been published. The rest await for P&R decisions. Other 5 pilots have been published in parallel. In total 5 TPR recommend to further restrict the European label. Conclusions: TPR are not being released at the expected rate. Follow up is necessary to predict its impact in P&R and market access in regions across Spain. PHP59 Are There Any Commonalities In Payer Requirements and Reimbursement Pathways for Medical Devices in the Dach (Germany, Austria, Switzerland) Region? Walzer S 1, Droeschel D 1, Hentschel D 2, Waaga F 3, Portegies W 3, Schwander B 4 Market Access & Pricing Strategy GmbH, Weil am Rhein, Germany, 2Johnson & Johnson, Vienna, Austria, 3MedicalWriters. com GmbH, Zurich, Switzerland, 4AHEAD GmbH, Loerrach, Germany . . . . . . 1MArS Objectives: Medical devices constitute a set of important health technologies for the care of patients. While there are similarities between some reimbursement systems, each market has its own unique characteristics. This article focuses on the reimbursement procedures for medical devices in the DACH countries (Austria, Germany, Switzerland), and aims at finding commonalities in payer requirements and reimbursement pathways. Methods: Reimbursement application pathways for inpatient and outpatient medical devices were evaluated for Austria, Germany and Switzerland. The key items being analysed for similarities and differences in each setting were transparency, clinical and health economic evidence requirements, submission timelines and the length of the whole reimbursement application process. Results: In the inpatient setting, the evidence requirements for clinical data are different between the analysed countries: The lowest clinical evidence requirements are seen in Germany, while the highest are given in Switzerland (in some scenarios). In terms of health economics the requirements are medium to low. A medium rating was given for Austria and Switzerland (in some scenarios) as a health economic analysis is required (e.g. cost comparison), and a low rating was applied to Germany as limited economic information (cost assessment/ comparison) needs to be submitted. The length of the application process is well defined in Austria and Germany and vague in Switzerland. In the outpatient setting the requirements for clinical and health economic data are significantly increasing. Clinical requirements are getting close to pharmaceutical methods whereas health economic evidence is requested in all DACH countries. The length of the reimbursement process is not clearly defined in all three countries. Conclusions: Despite varying reimbursement processes within the DACH region, there are some important similarities between the evidence requirements which may help manufacturers to guide market access and reimbursement strategy decisions in order to drive successful submissions and applications of innovative medical devices. PHP60 The Relevance of Health Services Research for the Pharmaceutical Industry in Germany –Results of A Representative Online Survey Anton V 1, Gehrke K 1, Hessel F P 2 1German Pharmaceutical Industry Association, Berlin, Germany, 2SRH University Berlin, Berlin, Germany . . . . Objectives: The necessity for manufacturer of pharmaceuticals and medical devices seem to deal with health services research (HSR). Health care reform acts and revisions of HTA method guidance documents in many European countries continue to point out the importance of real-life consequences. The objective of the study was to evaluate the self-reported importance and the own activities of pharmaceutical companies in Germany concerning health services research. Methods: Between December 2013 and February 2014 an online survey among members of the German Pharmaceutical Industry Association (BPI) was accomplished. Similar surveys were conducted in 2009 and 2011. Results: 59% of 109 addressed companies took part in the survey. The participants were representative for Germany concerning research-orientation, size and type of products. For 88% of the participating companies HSR is of importance. This high rate remained constant compared to the last surveys. For most of the companies HRS yields valuable contributions for the negotiations with payers (96%), in the AMNOG evaluation process (89%) and for optimization of placement of marketed products (94%). 50% of the companies conducted HSR studies over the last years, respectively are currently conducting HSR studies or participate in joint projects, whereas 69% plan to conduct HSR studies in the near future. All these numbers increased since the last surveys. Conclusions: The results of the survey clearly pointed out the high and still increasing importance of HSR and real-life studies also in the health care industry. Although a considerable amount of uncertainty concerning the specific methodological requirements remain it is widely accepted that the quality of data and analyses is crucial for acceptance by payers. As HSR projects require considerable human and financial resources alliances and joint projects between industry, academia and payer are aspired. PHP61 Integration of Cost–Effectiveness Assessment in the Market Access Scheme of Drugs and Medical Devices In France Rumeau-Pichon C , Harousseau J L Health Economics and Public Health Department S, Health Economics and Public Health Committee C Haute Autorité de santé, Saint-Denis La Plaine, France . . . Objectives: In France, drugs and medical devices costs concern an important part of health care expenditure. Several reforms have been put in place over the past years in order to limit these expenditures, in particular price cutting policy. Costeffectiveness assessment was integrated in France in the market access scheme of health products by the Social Security Financing Act for 2012 and is required since October 3rd 2013. This new mission was assigned to the Health Economics and Public Health Committee (CEESP) of the French National Authority for Health (HAS) and is become compulsory for innovative health products and are likely to have a significant impact on the health insurance expenditures. The objectives of this work were to see how the cost-effectiveness assessment has been integrated in France and to discuss the impact of this assessment on the health products market access. Methods: The work consists in analyzing the process of the economic evaluation achieved by the HAS since October 3rd 2013. Results: Economic evaluation is a new step in the market access scheme. In order to respect the legislated timeframe of 90 days, it is simultaneously conducted with the medical assessment by the Transparency Commission of the HAS. For each health products, an efficiency opinion is delivered by the CEESP. It presents the methodological conformity according to HAS guidelines and a conclusion about the efficiency, on the basis of the Incremental Cost-Effectiveness Ratio (ICER). Currently, 15 dossiers were eligible for cost-effectiveness assessment. First assessments permit to confirm that the procedure is operational. The average processing time was 94,8 days. Conclusions: In France without efficiency threshold value, the CEESP can’t conclude absolutely on the efficiency of health product. We don’t know how the ICER will be considered by the French Healthcare Products pricing Committee (CEPS) at the time of pricing negotiation with the pharmaceutical industry. PHP62 Analysis of Cost-Effectiveness Assessments In France By The French National Authority For Health (Has) Rumeau-Pichon C , Harousseau J L Health Economics and Public Health Department S, Health Economics and Public Health Committee C. Haute Autorité de santé, Saint-Denis La Plaine, France . . . Objectives: In France, cost-effectiveness assessment is required since October 3rd 2013 for innovative health products and are likely to have a significant impact on the health insurance expenditures. The objective of this work is to report first costeffectiveness assessments, achieved by the Health Economics and Public Health Committee (CEESP) of the French National Authority for Health (HAS). Methods: The investigation consists in analyzing medico-economic dossiers submitted at the CEESP between October 2013 and October 2014. This analysis is based on the opinion delivered by the CEESP, the methodology used in the model and the process of the assessment achieved by the HAS. Results: At the time of writing the abstract, 15 dossiers were eligible for cost-effectiveness assessment. All dossiers were based on cost-utility models. Two dossiers presented a negative opinion due to the methodological conformity according to HAS guidelines. For dossiers with a positive opinion, the Incremental Cost-Effectiveness Ratio (ICER) were between 5 866€ /QALY (for a subgroup analysis) and 194 531 € /QALY. Main methodological reserves made by the CEESP concerned comparators, time horizon, robustness of clinical data, utility measures. The average processing time was 94,8 days. For the moment, no price has been published in the Official Journal. Conclusions: The first assessments permit to confirm that the procedure is operational. In light of these first assessments, an update of the HAS methodological guidelines should be done. In France, without efficiency threshold value, the CEESP can’t conclude absolutely on the efficiency of health products. Thus, we don’t know what will be the place of the ICER in the pricing negotiation between the pharmaceutical industry and the French Healthcare Products pricing Committee (CEPS). PHP63 Accessing the Medical Device Market in the People’s Republic of China--Policy Changes Since the Restructuring of the China Fda Zhang S X , Kriza C , Kolominsky-Rabas P L Centre for Health Technology Assessment (HTA) and Public Health (IZPH), Friedrich-AlexanderUniversity Erlangen-Nürnberg, Erlangen, Germany . . . . . Objectives: The objective of this research is to provide an overview of the regulatory process of medical devices in China. Potential challenges related to medical device registration and major policy changes are highlighted especially since the restructuring of the China FDA in March 2013. The results of this research are aimed at informing regulatory bodies, health policy decision makers, national and international Health Technology Assessment networks, as well as medical devices manufacturers. Methods: A systematic review was conducted from 2009-2013 to identify the challenges and opportunities in the Chinese medical device regulatory process searching the PubMed, Science Direct, Scopus databases and Zhongguozhiwang. The PRISMA guidelines were applied for the search. In addition, an analysis of A415 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 previous and current policies and regulations in the grey literature was conducted for this research. Results: 17 articles were included in the systematic review. The three most crucial issues in the regulatory process have been analyzed in detail and relate to products technical requirements, the type test report and the clinical trials evaluation. Concerning the ‘products technical requirements’ and the ‘type test report’, the main challenges are related to the lack of sufficient legal status and openness for interpretation. A new Regulation governing the Supervision and Administration of medical devices introduced in June 2014 now includes improvements regarding specification of requirements. In addition, if a medical device fulfills certain requirements, it should be exempted from clinical trials, which signifies a major simplification for manufacturers. Conclusions: Getting a medical device registration certificate in China is a complicated, time-consuming process. The new regulatory policies in place focus more strongly on the clinical effectiveness and safety of medical devices. However, some major structural problems remain, i. e. the lack of technical standards for manufacturers, with major policy planning and changes currently underway in order to address these problems. PHP64 The Benefits and Challenges of Submitting to the Nice Medical Technologies Evaluation Programme-Manufacturer Perceptions And Experiences Satherley A W 1, Entwistle J 1, Panciera D 2 Market Access, Melton Mowbray, UK, 2University of Nottingham, Nottingham, UK . . . . 1GfK Objectives: The Medical Technologies Evaluation Programme (MTEP) was set up by NICE in 2009 to identify new medical devices and diagnostics that could improve patient experience and outcomes whilst driving health care efficiencies. Innovative products are submitted by manufacturers for evaluation, requiring confidence in the programme and its methods. The purpose of the study was to examine the experiences of manufacturers whose products have been assessed by MTEP, to review the challenges of the submission process, the resources required for the submission, NICE support for further evidence generation, and the impact of NICE recommendations on product uptake in the UK and elsewhere. Methods: A systematic review of published literature on MTEP and products/companies that have been through this process was completed. Qualitative interviews were then undertaken with a representative sample of senior team members from manufacturers whose products have undergone MTEP review to elicit insight on their perceptions and experiences of the programme. Findings were collated and analysed thematically. Results: Manufacturers have identified the benefits of submitting to MTEP in terms of value communication in the NHS, however key issues are: the significant time and resource burden of completing the submission, the lack of funding for additional data generation where this is required, the challenges of adhering to the stringent timescale, and the questionable benefits of submitting for larger companies with extensive sales teams. There is also a recognised need for greater efforts to influence practice and prescribing decisions in the NHS based on the MTEP guidance. Conclusions: Manufacturer experience of the MTEP process is largely positive, however as MTEP is dependent on manufacturers to initiate submissions, key actions can be taken to incentivise future submissions. Priorities will be ensuring that MTEP recommendations have a tangible and proven impact on prescribing decisions and that support is available for the submission development. PHP65 Availability of Risk Sharing Agreements in the Turkish Pharmaceutical Sector E 1, Tatar M 2, Ergin G 1, Senturk A 1, Atikeler . . . PHP67 Economic Impact of Drugs De-Funding and Pharmaceutical Co-Payment Buigues-Pastor L 1, Sánchez-Chorro J L 2, Saurí-Ferrer M I 1, Muñoz-Fernández J I 2, Uso-Talamantes R 1, Ortiz-Rodríguez O 2 1Conselleria de Sanidad. Generalitat de Valencia, Valencia, Spain, 2Servicio Extremeño de Salud, Mérida (Badajoz), Spain . . . . . . . . . Objectives: to assess the consequences in terms of outpatient pharmaceutical expenditure, that accounted for changes in state law on de-funding of medicines and the introduction of pharmaceutical co-payment in 2012 in Spain. Methods: Cross-sectional study during 2012. Two health regions of Spain (Comunitat Valenciana with 5,129,266 inhabitants and Extremadura with 1,108,130. Total population: 6,237,396) where analysed to determine the evolution of outpatient pharmaceutical expenditure after a change in state law (July 1,2012). We compared the first over the second half of 2012 by age and gender distinction. Results: The total spending was reduced by 16.80% after the new law. By gender, the reduction was greater in women (17.47% versus 15.25%). By age, the decrease was seen most sharply in the stretch of paediatrics (0 to 14 years). In older people, the stretch mostly affected was the one from 66 to 70 years old. The stretches with fewer changes in terms of outpatient pharmaceutical expenditure were 21-25 years (reduced by 5.42%) and 46-50 years (reduced by 6.07%). Conclusions: After the de-funding of 400 drugs and the increase of pharmaceutical co-payment (40% previously and now variable, 40%, 50% and 60% depending on income levels), a drastic drop in the amount of pharmaceutical outpatient expenditure is observed. Further analysis is required in order to assess whether that decline is sustained over time or, on the contrary, a new phase of positive growth begins. PHP68 When Gender Matters: Impact of Fda Safety Warning on Dispensed Zolpidem Dose Irwin D E 1, Palmer L A 2 1Truven Health Analytics, Durham, NC, USA, 2Truven Health Analytics, Bethesda, MD, USA . K2 Tuna 1Polar Polar Health Economics & Policy, Ankara, Turkey, 2Hacettepe University, Ankara, Turkey . Region to legislate on the matter, supporting the prescription of the biosimilar to the naive patient. The aim of this study was to evaluate the impact of cost-containment policies about biosimilars in Campania between 2009 and 2013. Methods: IMS Health regional database was used to carry out this descriptive retrospective drugutilization for the years 2009 through 2013. Information was retrieved about different distribution channels (retail, direct distribution, hospital). Consumptions are expressed in Counting Units (CU) and trends have been calculated using Compound Average Grow Rate (CAGR). Results: In 2013, a total consumption of 930.859 CU of biosimilar drugs was registered in Italy (CAGR 13/09 68,8%). The penetration rate of biosimilar was 23,9% with regard to expenditure and 25,3% with regard to consumption. Analyzing the consumption of biosimilars by therapeutic category in Italy, the consumption of erythropoietin and filgrastim has an index of annual growth respectively equal to 153.4% and 174.7%. In Campania, the penetration rate of all biosimilars was 31.4% of total expenditure and 35.9% of consumption. Analyzing the trend in consumption for single therapeutic category, it can be seen that penetration is due to erythropoietin (297.5% CAGR); with regard to filgrastim, however, the average annual growth rate is around 54%. Conclusions: The results suggest that Campania Region records an index of market penetration of biosimilars higher than the national. This trend could be attributable to the cost-containment policies implemented in the region. . Objectives: Risk sharing agreements are among the new trends in pharmaceutical sector as a tool for reducing the drug expenditures and increasing patient access to innovative drugs. The objective of this study is to elaborate the availability of risk sharing agreements in the Turkish pharmaceutical sector. Methods: A literature review was undertaken to identify the existing risk sharing schemes in the European countries using “risk sharing agreements”, and risk sharing schemes” as key words. Results: Risk sharing agreements are mainly classified as financial based and performance based schemes. The vast majority of those agreements are implemented in oncology area, others are mostly implemented in ophthalmology, blood diseases and multiple sclerosis areas. Countries mostly prefer financial based schemes as they are easier to implement and track. Performance based agreements are relatively rare as they are more complicated due to the long length of follow up, lack of reliability of data generation/registration, administrative burden for all stakeholders. Conclusions: Taking into account that original drugs to be reimbursed in Turkey need to grant a compulsory statutory discount which is generally 41%, it can be stated that Turkey is already implementing a financial based scheme. On the other hand, when availability of performance based risk sharing agreements is assessed, there is a lack of infrastructure to track drugs or non-responders and also it seems to be difficult to make this kind of agreements in Turkey without making any amendments in current regulations. In addition to these regulation issues, transaction and administrative costs should also be taken into account. On the other hand despite these difficulties, these schemes could provide budget control and patient access to highly innovative and expensive treatments. . . . Objectives: The Food and Drug Administration (FDA) issued a safety announcement on January 10,2013 requiring manufacturers to lower the recommended dose of zolpidem for women. Lower dose recommendations for elderly patients were already labeled. This analysis describes the impact of the FDA safety announcement on the initial dose dispensed to new zolpidem users. Methods: Patients > = 18 years old with an immediate- (IR) or extended-release (ER) zolpidem prescription between July 2012 and July 2013 were evaluated using MarketScan Commercial and Medicare Supplemental Databases. The first zolpidem prescription during the pre-warning (July 1, 2012 – Jan. 9, 2013) and post-warning (Jan. 10, 2012 – July 31, 2013) time periods was evaluated. Chi square tests were used to examine differences in the proportion of new users dispensed low dose zolpidem during each of the time periods. Results: Overall a significantly higher proportion of patients in all age gender groups received low dose IR zolpidem during the post-warning compared to the pre-warning period. Similar results were found for the ER formulation except among males 65 years or older. Before and after the FDA safety warning, the highest proportion of patients receiving low dose zolpidem (any formulation) for their first prescription was among elderly women (46% pre-warning; 53% post-warning). Even after the FDA warning, a large proportion of women 18 to 64 years old (63% IR; 65% ER) and women 65 years or older (46% IR, 55% ER) received high dose zolpidem for their first prescription. Conclusions: For all age and gender categories, the proportion of patients receiving low dose zolpidem increased after the FDA safety warning issued in 2013. The highest proportion of low dose zolpidem use was seen in elderly females. However, the proportion of women and elderly who received low dose zolpidem even after the FDA safety warning was still relatively low. PHP69 Non-Steroidal Anti-Inflammatory Drugs in Household Drug Supplies in Serbia PHP66 Impact of Cost- Containment Policies on Biosimilars Market 1Federico Putignano D 1, Fiorentino D 1, Guerriero F 1, Monetti V M 1, Orlando V 2, Menditto E 2 II University of Naples, Naples, Italy, 2CIRFF- Center of Pharmacoeconomics, Naples, Italy 1Faculty Objectives: The expiration of biotech drugs patents has led to the creation of drugs copies of originator products, defined biosimilars. The automatic substitutability between the originator and the biosimilar is not allowed in Europe with the exception of France. In Italy, in the absence of a national legislation, some Regions have issued directives to the spread of the biosimilar use. Campania was the first Objectives: Medications in households offer a variety of opportunities for irrational consumption and development of adverse drug reactions (ADR). One of the medicines most commonly associated with ADR are non-steroidal anti-inflammatory drugs (NSAIDs), one of the most frequently used medications in Serbia. Therefore, the aim of this study was to analyze the amount and structure of . . . . . . . Kusturica M P 1, Tomas A 1, Horvat O 1, Bukumiric Z 2, Tomic Z 1, Sabo A 1 of Medicine, University of Novi Sad, Novi Sad, Serbia and Montenegro, 2Faculty of Medicine, University of Belgrade, Belgrade, Serbia and Montenegro . . . . . . . A416 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 NSAIDs present in household drug supplies, and address the issue of possible inadequate use. Methods: This was an observational, cross-sectional study of drug storage and self-medication practice in households in the city of Novi Sad, Serbia over the 8 months period. Study consisted of personal insight into the drug inventory, and drugs were classified according the Anatomical Therapeutic Chemical (ATC) classification system. Results: Out of 383 surveyed households, 280/383 (73%) households held at least one box of NSAID in their home-pharmacy and a total of 473 packages of NSAID have been inventored. Most commonly encountered NSAIDs were ibuprofen (52.64%), diclofenac (31.92%), nimesulide (8.03%) and meloxicam (3.38%). Other NSAIDs accounted for less than 4% altogether. Over 70% of all NSAIDs present have been bought without prescription. Majority of ibuprfoen (91.3%) and diclofenac (65.74%) were obtained without a prescription even though in Serbia all NSAIDs (except ibuprofen 200mg) are prescription only medication (POM). According to European Medicine Agency, ibuprofen is classified as OTC, but diclofenac is strictly POM. Large amount of diclofenac used without consulting a physician presents a serious issue. Nimesulide and meloxicam were almost exclusively bought with prescription (> 80%). Conclusions: NSAIDs were present in most of Serbian home-pharmacies and were usually bought without prescription. This present a serious problem, especially for unsupervized diclofenac use. Financial Sources: This work was supported by the Ministry of Science and Technological Development, Republic of Serbia, project No. 41012. PHP70 Pharmacoeconomic Analysis of Antifungal Drug Use in the Intensive Care Unit Selimzyanova L , Namazova-Baranova L , Vishneva E , Lazareva A , Katosova L , Kryzanovskaya O , Tepaev R Federal State Budgetary Institution “Scientific Centre of Children Health” under the Russian Academy of Medical Sciences, Moscov, Russia . . . . . . . Objectives: to investigate changing of Candida species (C. spp) distribution and expenditure for antimicrobial drugs in the intensive care unit (ICU). Methods: from 1.01.11 to 30.07.13 there were 200 patients with Candida spp positive samples in the ICU. We followed up antifungal drugs susceptibility of different C. spp. and evaluated expenditure for antifungal drugs. Results: C spp was detected in 200 patients (12,2% of ICU microbiota). In 2011: 47 (19,3%); in 2012: 50 (9,3%); in 2013: 103 (12,1%) patients. The predominant species, detected during period of investigation was Candida albicans (C.a.), which was isolated in 21 (44,7%), 21 (42%) and 52 (50,5%) patients in 2011,2012 and 2013 accordingly. Candida parapsilosis (C.p.) was on second place and frequency of this fungus detecting increased: 1 (2,1%); 12 (24%); 31 (30,1%) patients. Candida tropicalis (C.t.) was detected: in 5 (10,6%); 6 (12%); 10 (9,7%) patients. Candida glabrata (C.g.): in 9 (19,1%); 3 (6%); 2 (1,9%). Candida krusei (C.k.): in 6 (12,8%); 1 (2%); 1 (1%). Candida lusitaniae (C.l.): in 2 (4,3%); 2 (4%); 3 (2,9%). Fluconazole susceptibility was: for C.a.: 100%; 89%; 97%; C.t. 82%; 94%, 92%, C.g 18%; 46%; 0% in 2011, 2012 and 2013 accordingly; for C.p.: in 2012; 29%: in 2013: 23%. Changing of C. spp species distribution led to increasing of expenditure for antifungal drugs in ICU: 52% of expenditure for antimicrobial drugs in 2011; 72% in 2012; 78% in 2013. Echinocandines determined 44%; 65% and 70% of ICU expenditure for antimicrobial drugs in 2011; 2012; 2013 accordingly. Conclusions: Our data coordinate with modern tendency in C. spp changing in ICUs. Increasing of C. spp proportion, resistant to fluconazole requires more frequent echinocandines use, that leads to rise of hospital expenditure. PHP71 Impact of the Early Benefit Assessment on A Pharmaceutical’s Real Consumption in Germany Kotowa W 1, Häussler B 2, Reindl S 1, Höer A 2 1IGES Institut GmbH, Nuremberg, Germany, 2IGES Institut GmbH, Berlin, Germany . . . . Objectives: The Act on the Reform of the Market for Medicinal Products (“AMNOG”) introduced an early benefit assessment (EBA) of pharmaceuticals in Germany in 2011. Since the impact of the results of the EBA on a pharmaceutical’s consumption is unclear, those effects were quantitatively analyzed. Methods: All pharmaceuticals which were assessed by an EBA in 2011 or 2012 and were still available in 2012 and 2013, and gliptins launched before 2011 with an EBA in 2013 were considered. A pharmaceutical’s real consumption (measured by defined daily doses) in 2012 and 2013 was compared to the expected consumption (EC) based on the highest possible number of patients defined by the Federal Joint Committee according to a pharmaceutical’s label. For the EC only subpopulations with an acknowledged additional benefit (AB) were taken into account. Results are presented in terms of „shares“ (= real consumption / EC). Results: Pharmaceuticals with a low share (< 10%) were cabazitaxel (0.22% in 2012), saxagliptin, protease inhibitors against chronic hepatitis C, eribulin, vandetanib and rilpivirin. Sitagliptin, ticagrelor, antineoplastic agents and tafamidis achieved a midsize share (10-50%). Ivacaftor, abiraterone, emtricitabin/rilpivirin/tenofovir, apixaban, fingolimod and axitinib reached shares of > 50% and up to 398%, respectively (fingolimod in 2013). There was no correlation between the extent of AB and a pharmaceutical‘s share. For most pharmaceuticals consumption was higher in 2013 than 2012. Conclusions: The shares of pharmaceuticals considered in this analysis vary between 0.22% and 398%. No correlation between share and extent of AB could be identified. It is assumed that label restrictions regarding the use of a pharmaceutical, competing alternatives in the same therapeutic indication, a pharmaceutical’s life-cycle or overestimation of the EC influence a pharmaceutical’s share. Shares > 100% indicate that also subpopulations without an acknowledged AB receive treatment with the respective pharmaceutical. Further research is needed to confirm these assumptions. PHP72 Using Data Envelopment Analysis (Dea) To Rank Gp’s According To Their P4p Quality and Efficiency Scores Perronnin M , Pichetti S , Sermet C IRDES, Paris, France . . . Objectives: Payments for Performance (P4P) schemes are commonly used in many OECD countries in order to incite physicians to improve both quality and efficiency of prescribing. In France, a set of 29 indicators has been selected as the P4P perimeter by National Public Health Insurance. Each indicator provides useful information of GPs’ performance, but it is not that easy to combine them all at the same time in order to provide a synthetic view of GPs’ performance. Instead, some of these indicators can be grouped together according to a therapeutic rationale (diabetes indicators, high blood pressure indicators) or an efficiency rationale (multiple-sourced drug prescribing). Data Envelopment Analysis (DEA) can be used to calculate an aggregated score for each GP and for each dimension (therapeutic or efficiency). Methods: Database used in this study were taken from the IMS-Health Disease Analyzer (DA) database. It contains information on pharmaceutical prescriptions administered by 693 French General Practitioners and related diagnoses. Quality and efficiency indicators are calculated both inside and outside P4P perimeter. Indicators reflecting efficiency are considered separately from those reflecting quality and the latter are grouped according to therapeutic rationale (diabetes, high blood pressure). Data Envelopment Analysis (DEA) is used to produce performance scores while aggregating several efficiency or quality scores and neutralizing at the same time each GP’s characteristics. Results: DEA scores are calculated on each dimension (quality or efficiency) and for each GP. Comparing the scores on each dimension enables to better understand if GPs’ are faced with a trade-off between these two Objectives. Conclusions: Results from this analysis may have implications for public policy design. A better understanding of GPs’ strategies relatively to P4P schemes could enable to adapt incentives in order to improve the system. PHP73 Prescribing Pattern of Antibiotics In Neonates in a Tertiary Care Hospital Thakur S 1, Acharya L D 1, Lewis L E S 2, Kumar R 1, Kumar S 1 1Manipal College of Pharmaceutical Sciences Manipal University, Manipal, Karnataka,, India, 2Kasturba Medical College Manipal Univeristy, Manipal, Karnataka, India . . . . . . . . Objectives: Neonates are among special population group who are more susceptible to nosocomial infections due to underdeveloped immunity level at their age (0-28 days of life) and it is a common practice of administering antibiotics, pending bacterial culture results, to sick neonates and to neonates who are more prone to develop infectious diseases; which in turn is gradually progressing towards a global problem of multidrug antibiotic resistance in Neonatal Intensive Care Units. Objective of the present study was to study prescribing patterns of antibiotics in hospitalized neonates, and to assess duration of therapy, antibiotic regimen and outcome of therapy. Methods: All neonates prescribed with antibiotics during 6 months period from November 2014 to April 2014 were selected. A prospective observational study was carried on selected cases by using sources of data such as patient treatment charts, progress sheet, laboratory investigation reports, and patient discharge summary; collected datas were reported in case record forms. Data analysis was done with SPSS (18.0), Microsoft Excel (2010). Results: Of the total of 404 admitted neonates, 30% (n = 122) were prescribed with antibiotics; among which 76 (62.3%) males and 46 (37.7%) were females. Among n= 122 cases, 71 (58.2%) empirically and 51 (41.8%) were prescribed in indication of infections. Most commonly prescribed antibiotic class were Aminoglycoside (97.54%), broad spectrum penicillin (66.39%), extended spectrum penicillin (28.68%), 2ndgeneration cephalosporin (17.21%). Empirical use of ampicillin and amikacin were most frequently encountered. Our study showed that Dual antibiotic regimen produced a better outcome (50%) than single/triple/more antibiotic regimen. Conclusions: The broader outcome of this study would be the potential utility of this data in designing strategies both at level of physicians and administrators for rational prescribing and policy decisions respectively; so that a continuous surveillance for rational use of antibiotics and strategy of antibiotic cycling can provide some better answers for difficulties coming across in management of neonatal infections. PHP74 Do Quality or Efficiency Indicators Inside French P4p Perimeter Influence Quality Or Efficiency Indicators Outside P4p? Pichetti S, Perronnin M, Sermet C IRDES, Paris, France Objectives: Payments for Performance (P4P) schemes are commonly used in many OECD countries in order to incite physicians to improve both quality and efficiency of prescribing. In France, a set of 29 indicators has been selected as the P4P perimeter by National Public Health Insurance. P4P schemes may induce undesirable behaviors from GPs leading to a decrease both in quality and efficiency outside the P4P perimeter (De Pouvourville, 2013). On the contrary, other studies (Campbell et al., 2009) tend to show that respecting P4P targets may imply a virtuous effect outside P4P. Methods: Database used in this study were taken from the IMS-Health Disease Analyzer (DA) database. It contains information on pharmaceutical prescriptions administered by 693 French General Practitioners and related diagnoses. Quality and efficiency indicators are calculated both inside and outside P4P perimeter. In a first step, each category –quality or efficiency- is considered separately. We calculate an aggregated score for each category, by averaging indicators pertaining to it. In a second step, each indicator outside the P4P perimeter is regressed on quality or efficiency aggregated scores, controlling for other characteristics of the physician. Results: At first sight, there is no obvious link between quality and efficiency scores inside and outside P4P perimeter. As a matter of fact, descriptive statistics, correlation analysis and Principal Component Analysis do not show any straightforward relation. All these first results call for further analyses. Conclusions: Results from this analysis may have implications for public policy design. If respecting P4P targets imply a virtuous effect outside P4P, there is no need to extend the P4P perimeter. If not, regulatory agency has to increase the number of indicators in order to improve quality and efficiency. A417 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 HEALTH CARE USE & POLICY STUDIES – Equity and Access PHP75 Trend Comparison of the Colombian Multidimensional Poverty Index, Inequities in Maternal Mortality, Neonatal Mortality and Gini Coefficient, 1997-2011 Romero Moreno LF, Eslava Schmalbach JH, Sandoval Vargas G, Saboya Romero DM Universidad Nacional de Colombia, Bogotá, Colombia Objectives: To describe and to compare the Colombian multidimensional poverty index (MPI) ’s trends against Gini coefficient (GC) and inequities in maternal mortality and neonatal mortality since 1997 to 2011. Methods: An ecological study was performed. MPI and Gini coefficient were obtained from National Statistics Department´s (DANE) databases. The Maternal Mortality Rate and Neonatal Mortality Rate were estimated and standardized by age and sex respectively. The Attributable Fraction (AF) was estimated as the inequity indicator for these two variables estimating the excess of risk by geographical exposure. We estimate the trend and behavior among MPI’s Colombian version and health inequities and disparities indicators over time from 1997 until 2011. Results: A substantial change was evident in the MPI 51% decrease (1997-2011) and 40%, (2003-2011) decreasing from 0.6 to 0.3 (2003-2011), whilst Maternal Mortality attributable Fraction (MMAF) showed a 3% increase, rising from 83.4% to 86.2% (2000-2008), and a slight reduction for Neonatal Mortality attributable fraction (NMAF) 1.6%, (2000-2008) decreasing from 88.9% to 88.4%. At the same time. GC evidenced a 1% decrease between 2000-2011 decreasing from 0.57 to 0.54. Conclusions: The established MPI for Colombia in the last decade had a descending trend and did not resemble the stationary behavior of the major inequity indicators calculated for the country in the same time span. Consequently, there was an undervalued perception over the issues where affected population were not target of requested interventions. It is therefore important to question the validity of measures used to quantify the poverty (MPI’s Colombian version) in order to reconsider the strategies of interventions where health inequity is an important referent to create control and intervention measures. PHP76 Nurses VERSUS Other Health Professionals Perceptions on Quality and Safety Culture Elements in Greek Hospitals Dreliozi A 1, Siskou O 2, Bouchoris P 3, Galanis P 4, Kaitelidou D 4, Prezerakos P 5 Regional Health Authority of Piraeus and Aegean islands, ATHENS, Greece, 2Center for Health Services Management and Evaluation, National and Kapodistrian University of Athens, Athens, Greece, 3IASO GROUP, ATHENS, Greece, 4National and Kapodistrian University of Athens, Athens, Greece, 5University of Peloponnese, Sparta, Greece . . . . . . 12nd Quality and risk management together, are prerequisites of patient safety which forms the basis of health care quality management activity. Objectives: To explore the prevailing organizational climate-in terms of clinical governance’ factors “Climate of blame and Punishment” and “A planned and integrated QI program and proactive risk management”, in Greek hospitals and to compare nurses’ perceptions with those of the rest health professionals on the particular factors. Methods: It is a cross-sectional study, including a representative sample of all specialties of employees working in a public and a private Greek hospital. The validated Clinical Governance Climate Questionnaire (CGCQ) was filled by Nn= 261 nurses and Noth prof= 198 other professionals (response rate: 79%). A lower score signifies greater satisfaction in a particular concept. Data mining took place from May to August 2012. Data analysis was performed with the SPSS 19.0 and included factor analysis, t-test, X2test and regression analysis The two-tailed significance level was set ≤ 0.05. Results: The results for the factors “Climate of blame and Punishment” and “A planned and integrated QI program and proactive risk management”, demonstrated a slightly positive trend (Means: 2.73 and 2.28 respectively) in the total population. Nurses, appear to perceive more negatively the climate related to: i) protected time for Quality Improvement initiatives (p< 0.05), ii) systematic assessment of clinical risks (p< 0.001), iii) sharing of a common vision (p< 0.05), iv) dissemination of Risk prevention policies (p< 0.001), v) proactive risk management (p<0.05), vi) systematic evaluation of Human Resources development needs (p< 0.001), vii) equal employee’s valuation regardless of professional background (p< 0.05). Conclusions: The views of nurses are essential, as they are important and direct factors of care provision. The assessment of climate produces conclusions which if exploited properly, can mark the beginning and support the effort of continuous improvement of patient safety. PHP77 Pharmaceutical Market Access in Emerging Markets Through Innovative Patient Access Schemes Leibfried M J , Arsiwalla M Y , Richardson S K GfK Market Access, New York, NY, USA . . . . . . Objectives: Emerging markets are a major priority for pharmaceutical manufacturers, and continue to grow as areas of interest for many firms. However, coverage from 3rd party payers, whether public or private, is often limited, and as a result the patient is the primary payer. Pharmaceutical companies must devise innovative strategies in order to provide access to patients while driving sales. In this study, we aimed to assess pharmaceutical company strategies to address patient affordability in emerging markets. Methods: We undertook a secondary research horizon scan utilizing public resources to examine a variety of strategies, including innovative patient access schemes across emerging markets. We then filtered a total of n= 14 schemes in emerging markets into several categories in order to create archetypes and to identify common themes and trends. Examples from a number of countries and regions were considered, including Brazil, China, India, sub-Saharan Africa, and others. Results: We categorized n= 14 patient access schemes into three main categories/archetypes: Licensing and Technology Transfer, Differential Pricing and Cost Offset Programs, and Value Added Services based on the primary objectives and characteristics of each scheme. A majority of programs fell under the Differential Pricing and Cost Offset programs (n= 8), with the other two categories splitting up evenly at n= 3 each. Conclusions: Pharmaceutical manufacturers are using innovative patient access schemes to gain access in emerging markets and are doing so by finding ways to drive down overall patient spend. While we were able to archetype innovative patient access schemes into three categories, it is important to note that many schemes incorporate elements from across the categories identified. Finally, when designing a scheme, pharmaceutical companies must take into account a host of factors including country level dynamics, company assets and strategy, and particulars of scheme design. PHP78 Patient Envolvement in Reimbursment of Drugs in Slovakia Tomek D 1, Tomekova K 2 1Slovak Medical University, Bratislava, Slovak Republic, 2Slovak Society for Pharmacoeconomics, Bratislava, Slovak Republic . . Objectives: To define the possible ways of participation of patients in the decision-making processes in Slovakia. To analyze the current status of patients in the reimbursement process in Slovakia and to define barriers to the participation. Methods: To identify the relevant literature, a survey was carried out using a search engine to the literature in PubMed 2000-2013. The survey was carried out using the following key words: “Patient participation”, “Patient choice”, in combination with the terms “Priority Setting”, “Innovation” or “Pharmaceutical Innovation”. Materials with those keywords were found, as well as through a Web browser. We performed analysis of the Slovak legislation, which regulates the status of patients in the reimbursement process and the comparison of this with other legislative standards, governing the status of citizens in public processes. There has also been evaluated legislative consultation process, which took place during the years 2010 and 2011. Results: Actually the patients can (as the whole public) only view all documents relevant to reimbursement, since they are not a “registered” participant of the process. They are 3 possibilities to comment the process according to relevant legislation: send a writing comment to the MoH, draw up a petition, file a complaint. In the period 2010 – 2011, before the legislation change, the MoH received 318 comments, from that 140 were from public. Conclusions: The whole process is totally transparent and visible via internet site of MoH. The status of patients (e.g. patient representatives) in the process declined since the last change of legislation from 1.12.2011. Due to the introduction of WTP threshold for ICER/QALY is also the introduction of innovative drugs more restrictive as before, what makes the possibilities of patient participation in reimbursement process even more important. PHP79 Advantages of Extemporaneous Dosage Forms in Ukraine Zalis’ka O , Maksymovych N , Kacheray J , Sichkoriz O Danylo Halytsky Lviv National Medical University, Lviv, Ukraine . . . . Objectives: We evaluated the data of license registry of Ukraine on 01.01.2014. Extemporaneous medicines are presented in 426 pharmacies, about 3% of the total number in Ukraine. The range of extemporaneous formulations in Ukraine is significant, it is complemented by modern specifications physicians. Methods: We analyzed the 1570 prescriptions on extemporaneous dosage forms of various doctors in few Lviv pharmacies during 2013-2014. Results: We found that 53 names of substances used active substances and excipients. That should be noted that different formulations are made: medicine, tinctures, solutions, suspensions, drops, ointments, pastes, suppositories, powders. Among the most commonly prescribed by doctors’ prescriptions are liquid dosage forms, topical solution (solution furacillin 0.02% - 500.0; rivanol 0.05%, 0.1% - 500.0; hydrogen peroxide 3% - 500 0, a solution of methylene blue 1% - 10.0), solutions for electrophoresis, nasal drops (solution collargolum 3% - 10.0 and protargolum 2% - 10.0), thymi water, soothing medicine quater, Pavlova, Ravkin. These pharmacies offer 26 solutions to supply hospitals. Also popular are antiflu powder (antigrypin) and vitamins, care products for face and body. Among soft medicines dominate ointment for wound healing (Lesyuk ointment, kalanhoie ointment) and for the treatment of sinusitis, Unna paste (for healing venous wounds), suppositories for hemorrhoid treatment (by prof. Maslyak) and antifungals (solution Kastelyani and various ointment). Currently, there are difficulties in ensuring substances, including missing: dibazol, diphenhydramine, collodion 4% lithium chloride, paracetamol, pepsin, yellow mercury, sulfur purified, diluted hydrochloric acid 8.4%, zinc sulfate, extract althey root, aminophylline, 1% citral alcohol solution. Conclusions: Extemporaneous medicines are economically available, the average costs of solutions are 4-12 UAH (1 Euro= 16.0UAH), eye drops- 6-18 UAH, ointments - 5 - 15 UAH, which is 40-85% less than the manufactory drugs. That’s cost-effective therapy and provide individualized drug treatment for children, elderly, and chronically patients in Ukraine. PHP80 Towards Universal Health Care- A Review of the Basic Basket of Care Associated With Universal Health Care Delivery Models Burns R M 1, Dooley B 2, Armstrong J 3 1University of Oxford, Oxford, UK, 2AXIS Healthcare Consulting Ltd., Galway, Ireland, 3Erasmus University Rotterdam, Rotterdam, The Netherlands . . . . Objectives: In 2014, the Irish Department of Health and Children (DoHC) published a policy document outlining the introduction of Universal healthcare by 2019. The impetus for adopting Universal healthcare is multi-faceted; the main driver being the current two-tier, private/public delivery system limits access to care for subgroups of the population. Inequality in access to care and clinical outcomes has been extensively published in Ireland; the implementation of Universal health care will aid in reducing this inequality as well as stabilise variation in resource use across socio-economic groups providing more efficient and equitable delivery of health care. The aim of this review was to catalogue what basic care is offered at the population level across Universal delivery models in Europe and document disease-specific resource use and health outcomes as well as organisation around reimbursement structures for purposes of informing policy in Ireland. Methods: EU-27 countries were classified by health care delivery systems according to data available from A418 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 each country’s department of health. Countries were reviewed under a range of headings including: current delivery models in place, institutions responsible for delivery and organisation of reimbursement system, incentive structures in place, basic bundle of health care covered, additional options for coverage, disease-specific resource use and health outcomes, government contribution to cost of health care and overall health care expenditure. Results: Generally, the basic bundle of health care across countries with Universal entitlement ensures comprehensive medical care for everyone including GP services, access to tertiary care, post-natal care and medications. Co-payments, excess payments and retention fees apply in some countries dependent upon the Universal delivery model in place. Conclusions: This review presents characteristics of Universal health care delivery systems across Europe. Basic bundles of health care provision and organisation of reimbursement across delivery models have been outlined, thus providing further clarity on the characteristics of and variation across Universal health care models. PHP81 National Rare Disease Strategies: The Current State for Orphan Drug Market Access in European Union (Eu) Member States Scholten J , Lie X, Kalbasko A , Maervoet J Quintiles Consulting, Hoofddorp, The Netherlands . . . Objectives: By 2013 all European Union (EU) member states were recommended to elaborate and adopt a national strategy for rare diseases. This study provides insights into the national rare disease strategies, in particular about the status of the programmes, recent developments, and the congruencies and differences between the programmes regarding market access. Methods: A literature and Internet search was performed to identify national strategies for rare diseases published by EU member states. Both policies and recent orphan drug introductions have been analysed to compare the rare disease strategies. Results: All member states had a variety of approaches already in place before developing a national strategy. France is the frontrunner in implementing rare disease policies as well as the member state with the highest number of marketed orphan drugs. Over the past five years, most member states finalised their national plans with a peak in publications late 2013. Plans include similar methods on increasing patient access such as off-label use, compassionate drug use and utilising cross-border health care. Member states with a decentralised market access model (e.g., Spain and Italy) commonly use national funds and decision-making to provide equity in treatment levels across the nation for rare diseases. There are few orphan drug specific pricing policies; however special reimbursement criteria are common especially in countries with cost-effectiveness criteria. Conclusions: Increasing patient access to orphan drugs has been a focus point in the national plans for rare diseases. Congruencies in methods will aid the EU’s ambition to align policies at European level. However the implementation phase has only recently been initiated for most plans and actual policies are yet to be developed. As member states such as France are introducing cost-effectiveness criteria in their health technology assessments, one can expect that tailored criteria need to be developed for orphan drug assessments. PHP82 Measuring the Efficiency of Hungarian Hospitals by Data Envelopment Analysis Csákvári T 1, Turcsanyi K 1, Vajda R 2, Danku N 2, Ágoston I 2, Boncz I 3 of Pécs, Zalaegerszeg, Hungary, 2University of Pécs, Pécs, Hungary, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . 1University Objectives: Hospitals are important cost elements of the Hungarian health care system. During the past decade, several health care reform affected the number of hospital beds in Hungary. The aim of our research is to analyse the efficiency of the Hungarian acute inpatient-care system. Methods: Data derived from the Hungarian nationwide health insurance database. We analyzed the technical- (TE) and scale efficiency (SE) of the Hungarian acute inpatient-care system (2003, 2006, 2010). The number of hospitals included into the study was 133 in 2003,125 in 2006 and 93 in 2010. We chose four inputs and two outputs: the number of active hospital beds, the number of discharged patients, the number of oneday cases, completed days of nursing (inputs), average length of stay, DRG cost weights (outputs). The method we used for our calculations was Data Envelopment Analysis. Results: In 2003 both the technical and scale efficiency were high (TE: 96.9%; SE: 92.9%). To 2006 the situation deteriorated by some degree (TE: 96.6%; SE: 80.3%). By 2010 technical efficiency still did not show improvement (TE: 94.0%), but scale efficiency increased (SE: 88.2%). Usually the hospitals with higher number of beds are more efficient than the smaller units. Conclusions: The effects of the performance volume limit did not improve the two values; however, the capacity decrease of 2007 did improve the scale efficiency to some extent. The Hungarian health care system needs to reduce the numbers of hospitals and rethink their functions, but needs to improve the size of them. PHP83 Experiences With Price Competition of Biosimilar Drugs in Hungary Z 2, Tálos Z 1, Ágoston I 2, Endrei D 2, Csákvári T 3, Boncz I 4 Hornyák 1Csolnoky Ferenc Hospital, Veszprém, Hungary, 2University of Pécs, Pécs, Hungary, 3University of Pécs, Zalaegerszeg, Hungary, 4Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . PHP84 Impact of Prior Authorization Restrictions on Resource Utilization and Costs In US Health Plans: A Review Of Literature Shah D , Tongbram V , Paly V ICON Plc, Morristown, NJ, USA . . . Objectives: Prior authorization (PA) restrictions have been implemented by US health plans in an effort to ensure safety, manage appropriate utilization and control costs. A review of published peer-reviewed literature was conducted to evaluate the impact of such PA restrictions on resource utilization and costs. Methods: A targeted review of literature was conducted in Medline from 2009 onwards using search terms including “prior authorization”, “resource”, “use”, “utilization”, “cost”, “impact”, “economic”. Review articles, non-English language studies, non-US studies, kins, and studies evaluating the effectiveness of formulary policies of which PA may be a component were excluded. Impact of PA policies on health care utilization and costs was qualitatively assessed. Results: Fourteen studies were identified which met our inclusion criteria. Majority (57%) of the studies were conducted on Medicaid plans (Medicaid: 8, commercial: 4, Medicare: 1, not clear: 1). Majority (57%) of studies evaluated the impact of mental health medications (anti-convulsants, anti-depressants bipolar medications, antipsychotics), two studies were conducted on anti-diabetics, one on a multiple sclerosis drug, one for a lipid-lowering drug, one on an anti-hypertensive and one on a vaccine. Four studies were industrysponsored. 12 studies were retrospective data analyses and only 2 studies were decision-analytic models. Overall, the trend showed that PA restrictions were effective in reducing pharmacy utilization and health care costs, but few studies also raised concerns on patient safety and quality of care outcomes due to PA policies. Conclusions: Although PA restrictions may result in cost-savings, patient safety and quality of life concerns must also be evaluated while imposing these restrictions. Rigorously designed studies including assesment of PA administration costs as well as indirect costs due to lost productivity should be conducted to better assess the overall economic impact of such restrictions. PHP86 Do Nice Decisions Affect Decisions in Other Countries? Hernández-Villafuerte K , Garau M , Devlin N Office of Health Economics, London, UK . . . Objectives: The objective is to test the hypothesis whether NICE recommendations on the use of a new drug affect recommendations from other bodies in countries outside England and Wales. To our knowledge, this is the first attempt to approach this topic quantitatively. Therefore, a sub-objective is to determine the feasibility of gathering a high quality database with sufficient number of observations to test our hypothesis. Methods: A basket of 29 drug/indication pairs and a group of 15 countries were included (Australia, Canada, Denmark, France, Italy, Korea, The Netherlands, New Zealand, Poland, Portugal, Spain, Bosnia, Ecuador, Egypt and Ghana). Information regarding NICE HTA recommendations was extracted from NICE’s website and HTAinSite. com. Moreover, an online-survey of key opinion leaders was carried out to collect information regarding the HTA decision in 10 countries. For the remaining five countries, we used the information from their official webpages. A descriptive analysis was conducted, including an examination of the position of each decision of NICE in comparison with that of other HTA agencies in the timeline of decision making about the 29 medicines. Results: There is a lack of comparability between the publically available information. The findings suggest that after the publication of a NICE appraisal there is a higher probability that an HTA is undertaken for the same drug in other countries. Furthermore, when NICE has published a negative decision, the tendency of not recommending the drug by another HTA body is much larger after than before NICE’s decision. Conclusions: Issues encountered in the collection of information made it difficult to quantify the effect of NICE recommendations on HTA decisions in other countries. The results suggest that the selected agencies are considering NICE decisions as a factor for rejecting or restricting the use of drugs which in other case would be recommended or reimbursed. HEALTH CARE USE & POLICY STUDIES – Health Care Costs & Management PHP87 Complications, costs And Resource Utilization in Real-World Complex Abdominal Wall Reconstruction Patients HEALTH CARE USE & POLICY STUDIES – Formulary Development L 1, Nagy erythropoietin treatment, while in the first 12 months after the bid 3647 patients, resulting in a 12.5% decline. In the 12 months before biosimilar bid 13974 patients received colony-stimulating factor treatment, while in the first 12 months after the bid 13352 patients, resulting in a 4.5% decline. Conclusions: The analyses of the Hungarian price competition bid of biosimilar products showed a minimal decline in the number of patients under treatment by both colony-stimulating factor and erythropoietin products while the health insurance reimbursement of these drugs significantly decreased. . Objectives: The aim of our study is to analyse the biosimilar bids of the Hungarian National Health Insurance Fund Administration in case of colony-stimulating factor and erythropoietin products. Methods: Data derived from the nationwide pharmaceutical database of Hungarian National Health Insurance Fund Administration. We analysed how the number of patients treated by colony-stimulating factor and erythropoietin products changed before (01.07.2011.-30.06.2012.) and after (01.07.2012.-30.06.2013.) the first biosimilar bid performed in March 2012 in Hungary. Results: In the 12 months before biosimilar bid 4167 patients received Mencer M 1, Reaven N 2, Funk S 2, Franz M G 1, Macarios D 1, DeNoto I I I G 3 1LifeCell Corporation, Bridgewater, NJ, USA, 2Strategic Health Resources, La Canada, CA, USA, 3Hofstra North Shore-LIJ School of Medicine, Manhasset, NJ, USA . . . . . . . . . . Objectives: Little information is available on complication-related resource utilization and costs over time in patients with complex abdominal wall reconstruction. Under pay-for-performance requirements financial decision-makers need better information to allocate health care resources and budget dollars. This analysis reports complication-related resource utilization and costs over time in a real-world patient population undergoing complex abdominal wall reconstruction. Methods: A cohort of patients with complex abdominal wall reconstructions during inpatient stays between1/1/08 and 6/30/11 (Index event) were followed for 12 months. Related complications, returns for facility-based care and related costs were evaluated for 30-60-90-365 days after discharge. Insurance claims from the Truven Health A419 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Analytics MarketScan® database, inpatient costs from the Healthcare Cost and Utilization Project (HCUP) and costs reported for Ambulatory Patient Classifications (APC) were used to estimate costs from the hospital perspective. Results: 13,463 patients were evaluated. Rates of patients experiencing any complication were 17.3% within 30 days, 11.0% within 31-60 days, 8.1% within 61-90 days, and 21.4% within 91 days – 12 months. In total, 37.8% of patients experienced a complication over 12 months. The most frequent complications over 12 months were infection (16.6%), bowel obstruction/ other GI complication (12.6%), skin/connective tissuerelated complications (10.7%), and wound complications (8.1%). Complicationrelated cost over time followed a similar trend; average 12 month cost for patients experiencing an infection was $20,679, $21,558 for bowel-related complications, $14,950 for skin/ connective tissue related complications and $19,230 for wound complications. The index event average length of stay for patients with no complications and patients with complications was 3.9 (sd 4.4) and 17.0 (sd 19.6), respectively; p< 0.0001. Conclusions: Health care resource utilization, costs and complications for complex abdominal wall reconstruction patients increase over time. Resource utilization is exacerbated when complications occur. Further study may be required to validate these findings. PHP88 The Growing Financial and Quality-of-Life Burden Associated With Atrial Fibrillation (Af), Diabetes, Chronic Obstructive Pulmonary Disease (Copd) And Asthma in Ireland Daacke I M Boehringer Ingelheim UK, Bracknell, UK . . Many people in Ireland suffer from chronic diseases including AF, diabetes, COPD and asthma. With the prevalence of these conditions expected to rise, general wellbeing and quality-of-life will be increasingly affected. Chronic conditions also account for most of the health care resources used, and represent a significant economic burden for Ireland in the future. Objectives: Estimate the number of preventable disease associated events and costs associated with poor management of patients with AF, diabetes, COPD and asthma. Methods: For each province, calculate the number of patients diagnosed with AF, diabetes, COPD or asthma, based on disease prevalence and 2011 Census data. Estimate the number of patients not achieving target management of their condition and the associated number of preventable events and total costs, using publically available information. Results: Of the approximately 59,647 patients diagnosed with AF in Ireland, 23,561 patients are not receiving appropriate anticoagulation treatment. This results in 531 patients experiencing an avoidable stroke each year, costing the health care system around € 9.3m. Amongst the 238,589 patients diagnosed with type 2 diabetes, 77% will not achieve a target HbA1c of 6.5% or less, resulting in an expense of € 886m and 12,493 avoidable deaths each year. In addition, 30% of diagnosed COPD patients, and as many as 60% of asthma patients, are not managing their condition effectively, costing the Irish health care system € 899m per year in hospital admission costs alone. Conclusions: Much of the chronic disease burden is caused by preventable risk factors. This is intended as a key policy lever, to elevate chronic diseases on the health agenda of key policymakers, providing them with better evidence about risk factor control, and persuading them of the need for health systems change. Unless steps are taken now to effectively deal with chronic diseases, Ireland is headed for serious financial and quality-of-life crises. PHP89 What Is Working Well In Louisiana For Us Employers: A Descriptive Analysis of Employers Actively Engaged In Promoting Employee Health Copher R 1, Tolbert C 2, Li X 1, Wang Z 3 Lake, NJ, USA, 2Louisiana Business Group on Health, Baton Rouge, LA, USA, 3Eisai Inc., Woodcliff Lake, NJ, USA . . . . 1Eisai, Inc., Woodcliff Objectives: As health care costs continue to rise, employers seek options to improve the health and wellness of their employees. This study examined the practices of employers that are actively engaged in promoting employee health. Methods: A study of 18 employers who applied for the Working Well award provided by the Louisiana Business Group on Health in 2013, recognizing employers who are exemplar in their employee health and wellness activities. Applicants completed a survey detailing business policies and programs intended to promote wellness. De-identified data derive from 2013 applications. Results: Over half of the companies had fewer than 500 employees (55.6%), 4 had more than 2000. Health plan coverage was fully (8; 44.4%) or self (7; 38.9%) funded. Almost half implemented wellness programs within the last 3 years (44.4%), whereas 4 (22.2%) had programs more than 10 years; all were company funded. Annual spend on wellness was split across participating employers with 55.6% spending < $50,000 and the rest > $50,000 (8; 44.4%). Rationale most cited for programs: improve employee wellbeing (18; 100%), contain health care costs (17; 94.4%), increase productivity (13; 72.2%), and reduce absenteeism (12; 66.7%). Most employers incentivized program participation (16; 88.9%) through premium reductions (8; 44.4%), cash (8; 44.4%), or PTO (3; 16.7%). Information most reported to help with wellness planning were health risk assessments (HRAs) (15; 83.3%), health care claims and utilization (14; 77.8%), and worker’s compensation claims (8; 44.4%). Conclusions: In the US, employers are responsible for a significant portion of health care spend. Though a small self-selected sample, this analysis reveals that employers actively engaging their employees, using prevention and incentives to promote wellness are a more recent occurrence. The trend suggests increasing awareness that efforts to improve employee health and wellness can help attract and retain staff, as well as potentially reducing health care costs. PHP90 A Quantification of Expenditure on Hospital Stays In 5 European Countries Waterman J , Gavaghan M GfK Market Access, Wayland, MA, USA . . Objectives: Throughout Europe, economic conditions are forcing health care systems to reduce costs. One primary driver of health care costs is hospital length of stay (LOS). This study sought to determine which European countries have been most successful at reducing their average LOS for five inpatient admissions. This research also sought to quantify the potential savings for countries that have not been as successful in reducing their average LOS if they can align with their peers. Methods: A review of hospital LOS and cost per day of hospital stay data was conducted in five European countries (France, Germany, Italy, Spain and the United Kingdom), utilizing data published by the World Health Organization (WHO). Additionally, hospital payment systems were assessed in each country through published research to understand systemic motivations of health care providers with regards to LOS. Results: Substantial variability exists in average LOS for the studied admissions. The greatest variability was in breast cancer, with average stays ranging from 4.36 days in the UK to 11.01 days in Germany. The average LOS for three admissions (single spontaneous delivery, cataracts, and pneumonia) are relatively similar across countries. However, the average LOS in Germany for malignant neoplasm of the breast and acute myocardial infarction are significantly higher than the other four countries. There is little variability, however, in average costs per bed-day in the target countries. A review of payment mechanisms for inpatient stays revealed that hospitals are financially incentivized to minimize LOS in all five countries. Conclusions: Additional research is needed to understand the reason for the discrepancy between German stays and the other four countries. While there are many potential reasons for the differences, should Germany align their average LOS for malignant neoplasm of the breast and acute myocardial infarction with the other four countries, they could save € 744 million per year. PHP91 R&D Investments, Intangible Capital and Profitability in the Pharmaceutical Industry Goncharov I 1, Mahlich J C 2, Yurtoglu B 1 Beisheim School of Management, Vallendar, Germany, 2Janssen-Pharmaceutical companies of Johnson & Johnson, Neuss, Germany . . . . 1WHU-Otto Objectives: The pharmaceutical industry is in the center of political debate due to their high profitability. In this study, we argue that abnormal profitability in the pharmaceutical industry is a kind of optical illusion created by accounting standards and their influence on reported accounting profit and book equity – the two components of ROE. The internationally accepted accounting frameworks either do not permit capitalizing R&D investments as U. S. GAAP or limit capitalizing R&D investments as International Financial Reporting Standards (IFRS) applicable in the E. U. and most countries. This treatment understates assets and equity, and can overstate reported profit because relevant cost components (amortization of R&D) are not deducted from revenues they generate. We empirically aim to estimate the magnitude of this accounting bias. Methods: Based on international financial data of 413 pharmaceutical firms between 1972 and 2012, we assessed the “true” profitability of pharmaceutical firms by capitalizing R&D and amortizing it over the shelf-life of developed products. We use three amortization approaches (linear amortization, declining-balance amortization and amortization based on the empirical amortization rates). Results: Corrected profit and equity figures lead to substantially lower long-term profitability of pharmaceutical firms. Over the three proposed amortization approaches, the corrected ROE of 14.1% is comparable to profitability reported by U. S. firms from other industries (ROE = 11.1%). Non-U. S. pharmaceutical firms also have an adjusted ROE that is comparable to firms from other industries (7.6% pharma vs. 9.6% nonpharma). Conclusions: The policy implication of our study is that price regulation or rate of return regulation in the pharmaceutical market should be reviewed and applied with caution when it is solely motivated by the allegedly high profitability of the industry. This is especially true since such a policy also impedes R&D investments and innovation in the long run because profits serve as a major source of R&D investments. PHP92 Do Specialty Drugs Offer Greater Value for Money Than Traditional Drugs? Chambers J D , Thorat T , Chenoweth M , Pyo J , Neumann P J Tufts Medical Center, Boston, MA, USA . . . . . . . Objectives: Specialty drugs are often many times more expensive than traditional drugs, raising questions of affordability, and whether their clinical benefits are worth their added costs. The objective of this study was to consider new molecular entities (NME) (i. e., drugs that had not previously been approved by the FDA or marketed in the US) approved by the FDA from 1999 through 2011 to compare the value of specialty and traditional drugs. Methods: We searched the FDA website to identify all NMEs approved from 1999 through 2011. We identified published estimates of additional health gains (measured in quality adjusted life years (QALYs)) and costs (drug costs, hospitalization costs, etc) associated with specialty drugs compared to existing standard of care at their time of approval, and compared findings with traditional drugs. We compared incremental QALY gains, incremental costs, and the incremental cost-effectiveness ratio, for specialty vs. traditional drugs using a Mann Whitney U test. Results: We identified relevant estimates of additional health gains and costs for 101 (36%) of NMEs, including 59 specialty drugs. We found specialty drugs offered greater QALY gains than traditional drugs (0.19 vs. 0.01, p< 0.01), but were associated with greater additional costs ($10,460 vs. $906, p< 0.01). We found the cost-effectiveness of the different drug types to be broadly similar (p= 0.58). Conclusions: This research suggests specialty drugs may offer greater health benefits over existing care than traditional drugs, and despite specialty drugs being associated with greater costs, specialty and traditional drugs were comparable in terms of cost-effectiveness. As payers search for ways to control health care costs it is important to recognize the relative benefits as well as the costs of specialty drugs, and to mitigate inappropriate use and waste to ensure that effective treatments are affordable to patients. A420 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PHP93 Assessing Assessment: Does Health Technology Assessment Do Its Job of Controlling Costs Without Compromising Quality? Armstrong S 1, Garfield S 2 1GfK, Wayland, MA, USA, 2GfK Market Access, Wayland, MA, USA . . Objectives: The pace of innovation in health care has facilitated gains in life expectancy, but at a tremendous cost. Rational use of limited health care resources remains one of the greatest challenges in health care worldwide. This analysis sought to compare health care spending and health outcomes in countries with and without formal health technology assessment programs. Methods: Data was collected on health care spending from 1990-2012 in European countries with and without HTA and compared to the United States, which eschews any formal evaluation of comparative or cost effectiveness. Measures of health outcomes were considered, including life expectancy and overall mortality. The impact of population age, tobacco use and obesity rates on health outcomes were considered. Lastly, the number, content and methodology of HTA publications in countries with HTA were assessed. Results: The majority of HTA evaluations have focused on pharmaceuticals. Countries with established HTA had lower spending on pharmaceuticals, but not always lower overall health care spending, while maintaining or improving health outcomes relative to non-HTA countries. The United States had the highest rate and annual increase in rate in pharmaceutical spending as well as overall health care spending, with the worst health care outcomes in countries analyzed. Conclusions: The incorporation of formal health technology assessment as part of overall health care decision making appears to assist in controlling the rise in health care spending, particularly drug spending. While better health outcomes in life expectancy and mortality were found in HTA versus non-HTA countries, correlation versus causality can only be considered at this point. Systematic use of HTA on both pharmaceutical and non-pharmaceutical interventions may help to control overall health care spending. PHP94 The Trend of the Prices Of New Marketed Drugs in Turkey Yenilmez F B 1, Kockaya G 2 1Hacettepe University, Ankara, Turkey, 2Health Economics and Policy Manager, Ankara, Turkey . that setting arbitrary benchmarks based on empirical observation of comparative EU countries is incorrect due to wide inconsistencies over pharmaceutical distribution. . PHP96 HTA Informed Price Negotiations: Cost Savings To The Health Payer in Ireland McCullagh L M 1, Tilson L 1, Adams R 1, Barry M 2 Centre for Pharmacoeconomics, Dublin, Ireland, 2HSE Medicines Management Programme, Dublin, Ireland . . . . . 1National Objectives: In Ireland, the reimbursement decision process involves the Marketing Authorisation Holder (MAH) submitting a Health Technology Assessment (HTA) dossier to the National Centre for Pharmacoeconomics (NCPE). The NCPE evaluates this dossier and submits a reimbursement recommendation to the Health Service Executive (HSE) (the decision maker). Along with Cost-Effectiveness and Budget Impact analyses, the decision maker considers other criteria in determining the health benefit of an intervention. Those interventions which are not deemed cost effective at a threshold of € 45,000/QALY may proceed to HTA informed HSE-MAH price negotiations. The aim of this study was to estimate the potential annual cost savings (reduced potential budget impacts) to the HSE as a consequence of HTA informed price negotiations. Methods: All NCPE assessments received over a 2 year period were reviewed. Interventions were included if initially they had not been recommended for reimbursement, but had subsequently been reimbursed after HTA informed price negotiations. The potential total annual cost saving to the HSE (difference between the original Gross BI (submitted price) and the revised Gross BI (negotiated price) was estimated. The respective MAH BI models were used for these analyses. Results: Eight interventions fitted the inclusion criteria; all were classified as either oncology drugs or drugs for orphan diseases. When all such drugs are considered, it is estimated that the cost savings to the HSE, as a result of HTA informed price negotiations is over € 19 million per annum. Conclusions: In Ireland, HTA informed price negotiations lead to considerable cost savings to the Health Payer. . Objectives: Health Transformation Program was started at year 2002 by Ministry of Health (MoH) in Turkey. Today, Turkey has a well established health system mostly dominated by government on health care provider and payer. The program allows the patient access to all drugs. There was an increasing trend for the pharmaceutical consumption in years depending on the taken policies as other health care services. The aim of this study is to review an analysis for the new drug release prices for Turkey in the recent years. Methods: The data was obtained from the last price list of each calender year published by MoH for the years between 20092014. The year 2014 year data limited with the first quarter (Q1). The data classified in three groups which were; over 100 TL, over 1000 TL and over 10000 TL. All the result were based on ex-factory prices. A descriptive analysis was conducted on the price lists with the number of the products, maximum and average ex-factory prices. Results: The numbers of new released products over 100 TL were; 1, 7, 58, 36 and 52 in the years 2009, 2010, 2011, 2012 and 2013, respectively. The numbers of new released products over 1000 TL were; 0, 1, 10, 11 and 13 in the years 2009, 2010, 2011, 2012 and 2013, respectively. The highest ex-factory prices were 739 TL, 1117 TL, 4066, 72 TL, 5781, 05 TL, 22791, 55 TLand 36577, 33 in the years 2009, 2010, 2011, 2012 and 2013, respectively. The average ex-factory prices were 178, 98 TL, 329, 28 TL, 214, 36 TL, 312, 93, 497, 44 TL and 612, 85 TL in the years 2009, 2010, 2011, 2012, 2013 and 2014Q1 respectively. Conclusions: It was shown that the number of high prices drugs, the highest price and the average price of the new drugs were increased year by year. Increased patient access may encoruage pharmaceutical companies to enter the Turkish market with their innovative drugs which have high price tag compared to regular drugs. PHP95 Setting Targets For Public Spending Under Eu-Imf Assistance To Portugal - The Case Of Health Care and Pharmaceuticals PHP97 Access To Non Communicable Disease Medicines in India: A Comparative Analysis of State Level Public Procurement Data Chokshi M 1, Farooqui H H 2 1Public Health Foundation of India, New Delhi, India, 2Public Health Foundation of India, Gurgaon, India . . . Objectives: - To assess access to NCD medicine in India using public drug procurement data. Methods: - State procurement data was used to calculate, total and per capita value of procurement towards purchase of essential medicines. ABC analysis was performed to identify priority medicines that together account for 80% of total expenditure (Category A items). The proportion of NCD medicines among category A items was then estimated. Finally, the efficiency of the system was captured through comparative price analysis of procurement price to International Reference Price (IRP). Results: - Out of the 91 NCD medicines on WHO list of essential medicine, states were procuring 40 to 87 NCD medicines. In addition, the share of NCD procurement value to total procurement value was in range of 1.32% to 37%. Similarly, the annual per capita spending was in range of 0.1 to 103 Indian Rupees (INR). The procurement prices across all the states were lower than the international reference price. The retail market prices were between 45%-3600% higher than the state procurement prices, and were higher on average by 6 times of international reference price. The defined daily dose estimates were not uniform for all ATC. The most neglected category among all ATC’s was mental health, in the range of 0.02056.688 DDD per 1000. Conclusions: - There is limited access to NCD medicines in public health system in selected Indian states. PHP98 Constructing A Comorbidity Index According To Iso-Resource Consumption Sicras-Mainar A 1, Navarro-Artieda R 2 Serveis Assistencials, Badalona. Barcelona, Spain, 2Hospital Universitari Germans Trias i Pujol, Badalona, Spain . . Rocha L , Fernandes C , Viana R Novartis Farma-Produtos Farmacêuticos S.A., Portugal, Porto Salvo, Portugal 1Badalona Objectives: The 2008 global financial crisis hit Portugal strongly culminating in request financial assistance from European Union and International Monetary Fund on 17th May of 2011. In addition, Portuguese economy had already before 2008 a decreasing real growth rate of GDP, stagnant from early in the decade, and an increasing external deficit. Health care measures represented a key area within the assistance program. Specific targets were set for public spending in pharmaceuticals in order to align with EU average at 1.25 and 1.0 percent of GPD for 2012 and 2013, respectively. The aim of our study was to demonstrate that GDP targets was a political decision not supported by evidence, technically inaccurate, created a hurdle and equity asymmetries for patients access to medications, as well as jeopardized the future sustainability of Pharmaceutical sector. Methods: We analyzed health, pharmaceuticals and medicines public spending EU benchmark in real terms, GDP ratio and per capita. We also revise the System of Health Accounts methodology and conceptual framework of International Classification of Health Accounts. OECD definitions, sources and methods were also studied. Results: Despite the efforts to improve the comparability across countries through common and better information framework of greater relevance for policy concerns, we observed that such approach is not recommended. For a more comprehensive understanding of health spending in relation to GDP ratio, it should be considered together with health spending per capita. Portugal comparisons are an OECD example for this rational. Importantly, OECD measures of pharmaceutical spending exclude in-patient (hospital) spending. In addition, pharmaceuticals definition is broader than prescription medicines only. Conclusions: Important political decisions were taken assuming pharmaceuticals and prescription medicines as the same concept. Here, we demonstrate Objectives: To construct a comorbidity index according to iso-consumption of health resources and their associated costs in routine clinical practice. Methods: We made a multicentre study through review of computerized medical records. The study population consisted of patients assigned to seven primary care centres, two acute care hospitals and one social health centre, allmanaged by Badalona ServeisAssistencials SA. The following variables were analysed: age (0-100 years), sex, comorbidity (chronic diagnoses) and direct health care costs. We estimated the resource use and gross health care costs attributed to all patients requiring health care in 2013. Subsequently, an expert panel identified the most-prevalent chronic diseases (ICD-9). Subgroups were established according to age and number of chronic comorbidities. The main variables and comorbidities associated with the costs were identified using a multivariate model. This allowed a total score to be developed for each patient. Statistical analysis: Spearman’s correlation coefficient, multiple logistic models and linear regression were used. A value of p < 0.05 was identified as significant. Results: We recruited 103,764 patients (86.5% of the total). Mean age was 43.1 (23.9) years and 52.4% were female. The mean unit cost was € 836.9 (2032). The gross cost was associated with age (r = 44.8) and comorbidity (r = 0.512). The final score was obtained from the logistic model. The score obtained showed a good correlation with age (r = 81.8), comorbidity (r = 0.939) and the cost of health care (r = 0.696). The index obtained explained 44.7% of the gross cost. The index was better adapted to the adjusted costs. We describe the disaggregated results and the results according to levels of comorbidity (healthy, low, medium, high). Conclusions: The comorbidity index obtained was shown to be a simple potential predictor of the cost of care and may be applied in routine clinical practice. . . . A421 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PHP100 Correlation Between Hospitalized Morbidity and Cost of Treatment of Selected Chronic Diseases Ofierska-Sujkowska G 1, Jagodzinska-Kalinowska K 2, Matusewicz W 2 1Agency for Health Technology Assessment in Poland, Warsaw, Poland, 2Agency for Health Technology Assessment in Poland (AOTM), Warsaw, Poland . . . Objectives: In 2011 the budget of the public payer in Poland, National Health Fund (NHF) amounted to 58.2 bln PLN. 47.3% of these funds was allocated to fund hospital treatment which is based on diagnosis-related groups (DRGs) but also includes innovative, expensive drugs (i. a. chemotherapy, orphan). The aim of the paper was to identify the existing correlation between hospital morbidity and the average unit cost of hospital treatment in various chronic diseases. Comparison of individual hospital costs of treatment different chronic diseases with varying degrees of severity, different course and level of incidence (from common to ultrarare) open the possibility of presenting the correlation between morbidity and unitary medical costs incurred by the NHF for the treatment of various indications. Methods: The analysis was prepared for 23 therapeutic areas, which were divided into 4 categories (diseases: common, infectious, pediatric and rare /ultrarare), diversified both in terms of severity and morbidity. This choice gave a possibility of comparing a wide range of chronic diseases, which have a significant impact on quality of life and mortality. The analyzed material included hospital procedures and services indicated by the NHF in 2011. Results: Using the Pearson’s linear Correlation test, dependence between variables (the average cost of hospitalization and the number of patients) for 23 therapeutic areas was assessed. The results showed that the lower prevalence (fewer patients with the diagnosis) is connected with the higher average unit cost of treatment. Conclusions: In Poland, among all studied diseases the most expensive was the individual treatment of patients diagnosed with ultrarare diseases (i. a. Lysosomal storage diseases). Chronic ischemic heart disease, hypertension and diabetes mellitus were associated with the largest number of annual hospitalizations and globally were the most cost-intensive for NHF. At the same time hypertension is associated with the lowest average unit cost of hospitalization among all studied diseases. PHP101 Patient Reported Outcomes and Their Relevance in Reimbursement Decisions Stoddart S 1, Lis Y 2, Malmenas M 3 1PAREXEL International, London, UK, 2PAREXEL International, Uxbridge, UK, 3PAREXEL International, Stockholm, Sweden . . . Objectives: Payers all need evidence of clinical effectiveness, reasonable cost, and safety. Although, patient-reported outcomes (PROs) are an important component of this evidence it is not always clear which PROs to use and reliance on an incorrect PRO can have negative consequences. The objective was to describe the factors determining which PROs are most likely to meet payer evidence needs. Methods: A structured discussion of factors influencing PRO choice was held between experts in late phase studies, evidence review, and economic modelling. These were then validated against a sample of published reimbursement decisions across multiple disease areas. Results: Focus of most questions is on whether to use a generic, disease-specific or novel measure. Disease specific PROs are very frequently of value in demonstrating clinical effectiveness and may be the standard clinical outcome. For example, migraine studies include pain assessment and headache frequency. To facilitate Payers’ decision making generic PROs such as EQ-5D and SF-36 are commonly used for assessing quality of life (QoL). However these instruments may not be reliable in disease areas characterised by focal impacts on patient wellbeing. For example, pain is only a small component of the SF-36 consequently changes in pain levels cannot be measured with great precision and the extreme effects of pain may be missed when generating estimates of utility. Conclusions: When choosing a PRO, it is important to consider the disease, treatment and payer reimbursement decision context. Disease specific QoL measures may be more sensitive than a generic tool but may also need mapping to a general QoL measure. Acceptance will be contingent on the existence of robust evidence either from published literature or additionally collected data. Reviews of previous HTA submissions and payer decisions help identify requirements / standard practices in key markets and identify common criticisms or problems. PHP102 Interventions and Policy Measures in Health Care And Pharmaceutical Sector To Increase Effeciency And Recovery of Greek Health Care System Malovecká I 1, Minariková D 1, Foltan V 2 University, Bratislava, Slovak Republic, 2Faculty of Pharmacy, Comenius University, Bratislava, Slovak Republic . . . 1Comenius Objectives: Greece is under a very tough and strict memorandum enforced by the European Union that has designed and applied austerity measures in order to save Greek economy and minimizes its public debt. Also the Greek Heath care system should be improved, upgraded or redesigned from scratch in order to be more flexible and more efficient. The legislative intervention made by the European Union and the memorandum in Greek Heath Care system and the Pharmaceutical market constituted legislative framework for the Greek Heath Care and the Pharmaceutical sector. Methods: Data on total pharmaceutical expenditures from 2009-2011, some of them until 2012 were obtained from Hellenic Association of Pharmaceutical Companies. They were analyzed using basic statistical methods of observing. Results: In nearly all observed parameters was recorded their lowering. Total pharmaceutical expenditures per capita Δ x2009-2011= -20%, x2011= 5.073 mil € , pharmaceutical expenditure as% of health expenditure Δ x2009-2011= -6,5%, x2011= 24,1%, pharmaceutical expenditure as % of GDP Δ x2009-2011= -11,1%, x2011= 2,4%, private pharmaceutical expenditure Δ x2009-2011= -13%, x2011= 1.094 mil € , private pharmaceutical expenditure per capita Δ x2009-2011= -13,5%, x2011= 96 € , public pharmaceutical expenditure Δ x2009-2012= -36,8%, x2012= 3.215 € , public pharmaceu- tical expenditure per capita Δ x2009-2012= -37,6%, x2012= 282 € , rebate from pharmaceutical companies Δ x2009-2012= +364,1%, net public pharmaceutical expenditure Δ x2009-2012= -42,3%, x2012= 2.88 mil € , net public pharmaceutical expenditure per capita Δ x2009-2012= -43%, x2012 = 253 € , net public pharmaceutical expenditure as % of GDP Δ x2009-2012= -36,3% and in 2012 they constituted 1,4% GDP. Price changes of medicines Δ x2005-2011= -17,8%, Δ x2010-2011= -10,7%. Conclusions: Interventions implemented in 2010 had according presented parameters a great impact on effectiveness of Health care and Pharmaceutical sector and this improvement will continue for a few years more. The betterment of Health care and Pharmaceutical sector might help to lower the public debt and recover economic credibility of Greece. PHP103 Determinants of Hospitals’ Attractiveness For Patients: Application to Expensive Drugs Siorat V 1, Degrassat-Théas A 1, Paubel P 2, Parent de Curzon O 1, Sinègre M 1 1General Agency of Equipment and Health Products (AGEPS), PUBLIC WELFARE HOSPITAL OF PARIS (AP-HP), PARIS, France, 2Paris Descartes University, Paris, France . . . . . Objectives: Hopital’s attractiveness can impact the number of inpatients, and so health care expenditures. The objective is to identify parameters related to patients or drugs, wich influence the proportion of inpatients not residing in the hospital’s region, in order to better understand expensive drug expenditures. Methods: A database on expensive inpatient drugs listed on a national formulary was set up, with information related to drugs -from National and European Health authorities- and patients -from our university hospital centre’s database (UHC) of the Paris region. A censored regression model, Tobit model, was developed in which the dependent variable was, for each drug, the ratio of the number of inpatients not residing in the Paris region, on the total number of inpatients receiving this drug. Explanatory variables were related to drugs and to patients. All statistical tests were run in Stata/IC13®. Results: During 2012, 526 091 doses -of the 113 drugs list- are administered to 30 499 UHC inpatients. 13% of inpatients come from off the Paris region. According to the Tobit model, three variables positively impact the ratio (innovative status, percentage of women and Diagnosis Related Group coded as severe) and three negatively (orphan designation, percentage of patients over 65 years old, biological origin). Therapeutic classes and Major Diagnostic Categories also impact significantly the ratio according to the statistic class. For instance, innovative status, defined by a market authorization granted in the five last years and a high level of improvement in actual benefit, improves the ratio of 8.6 points. To benefit from the most innovative treatments, patients tend to be cared in UHC’s hospitals. Conclusions: This study represents the first step of modelling significant determinants of hospital’s attractiveness for patients. It focuses, quite uniquely, on patients and drugs factors, and could be extended to other variables. PHP104 Economic Impact of Clinical Researches to the Research Centers and Reimbursement Systems in Turkey Kockaya G 1, Daylan Kockaya P 2, Demir M 3, Uresin A Y 3 Economics and Policy Association, Ankra, Turkey, 2Zirve University, Gaziantep, Turkey, 3Istanbul Medical Faculty, Istanbul, Turkey . . . . . 1Health Objectives: Clinical researches are an important tool for the improvement of medicine and significant economic value. The aim of this study was to investigate the economic impact of clinical researches to the research centers and reimbursement system in Turkey. Methods: Budget of clinical trials were calculated from the raw data of the Report of Istanbul Medical Faculty Clinical Researches (ITFKAR) between years 2006-2010. In addition, the possible cost of the drugs that were used in the clinical researches for SGK was calculated for showing the cost of medicines acquired with clinical researches. It has been accepted that predicted budgets of reviewed files were spent for research. Similarly, it has been accepted that the number of patients was not changed during the studies. Thus, approved budgets could be accepted as drug investment of the approval year. For the calculation of drug costs, discounted reimbursement figures of Social Security Administration (SGK) for licenced products were used. If not licenced and imported via Turkish Chamber of Pharmacists, the prices of abroad drug list of SGK were used. Calculations were performed with the drug prices for the year 2013. Results: The average of drug cost savings per patient with participation to the clinical research and clinical research investment per patient were calculated as US$ 21.649 and US$ 4.879. It could be said that total budget of sponsored pharmaceutical researches was US$107 million and the government had a saving close to US$ 311.096.130 due to not reimbursing the cost of drugs of the patients who were included to the clinical researches in Turkey depending on the analysis. Conclusions: However study is based on assumptions, the findings are unique for the literature. The health policy makers can take account the study of the policy improvements about clinical trials in Turkey. PHP105 Pharmaceutical Developpement: Africa an Emergent Market Mahi L 1, Lemdasni M 2, Houari A 2 1Axelys Sante Affaires Medicales et Recherche Clinique, Paris, France, 2Stratance Health, Rabat, France . . . Objectives: With population of more than 1 Billion, Africa is the world’s second largest continent. The transition that African countries are undergoing is a real challenge for health care authorities. 24% of the global disease burden is accounted in Africa. The diseases that prevailed in the 60s and 70s have not completely disappeared, tuberculosis is still endemic as well as transmissible diseases like HIV/AIDS and Hepatitis still remained as a major problem while some others, such as cancers, cardiovascular and metabolic disorders are dramatically increasing, requiring the implementation of effective health care programs and leading to an increasing demand for drugs treatment. Methods: Inventory of the potential of Africa in clinical research. Results: In 2012, African Pharmaceutical market revenue was USD 18 Billion and expected to reach USD 30 Billion by 2016. Established countries in both Sub-Saharan and North African countries (South Africa, Nigeria, Cote d’Ivoire, A422 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Senegal, Egypt, Morocco, Algeria and Tunisia) are contributing # 80% of the pharma market in Africa. Despite maintaining regional offices within Africa, many major Pharma and device manufacturers frequently overlook the continent when sponsoring clinical studies. Cultural barriers, political upheaval and uneven infrastructure are certainly causes for the lack of interest. But Africa offers tremendous expertise and opportunity for drug and device companies looking for cost-effective study sites and appropriate patient drug market populations. Currently more than 45% of the whole continent’s clinical trials are being conducted in South Africa and hence the need for the next generation clinical trial destination for a drug and device manufacturers. These companies can also consider technology transfer by partnering with local drug manufacturers and research centers to diversify their business portfolio. Conclusions: Africa presents real opportunities that should encourage many pharmaceutical companies to really engage in innovative clinical research programs in a win-win approach. the treatment of asthma were not accepted due to comments about population and model input, lack of statistical calculations and size of the incremental effects. Ranibizumab’s results for age-related macular degeneration were not accepted because of a wrong comparator, lack of long term data, assumptions on risks and mortality, and uncertainties around health care costs and ICER sensitivity. Finally, ZiNL considered the appropriate use of alglucosidase alfa for Pompe’s disease to be unsubstantiated because of the population choice for the ICER calculation and the use of a higher dose than the registered dose which was not supported by scientific data. The outcome results of alglucosidase alfa for cost-effectiveness were accepted, however. Conclusions: Although at T= 0 ZiNL provided feedback regarding the design of the outcome research study, most of ZiNL’s comments were on the final methodology used. The outcome research studies were considered to be of mixed quality and the results could often not substantiate the claim of cost-effectiveness and appropriate use. PHP106 Market Analysis in Regard to Biologically Active Supplements And Medicines in Armenia PHP109 Opportunity Costs of Implementing Nice Decisions in NHS Wales Beglaryan M , Amirjanyan A Yerevan State Medical University, YEREVAN, Armenia . . Objectives: Although, whether biologically active supplements (BAS) are medicines or not is still debatable, BAS keep making their way to the customer basically through pharmacies. A worldwide tendency toward “greener” choices when purchasing health status modifiers (HSM) is well documented. Current endeavor studies the pharmaceutical market situation (PMS) in Armenia (2009 to 2013) in regard to growth trends both in US dollars turnover (USDT) and number of packs sold (NPS) of BAS versus medicines, stratified by five leading diseases (LD). Methods: Statistical data on morbidity and mortality from the MOH RA were used to identify the leading five disease groups in newly identified cases. Further, statistical data on pharmaceutical market from “Pharmexpert” Marketing Research Centre (analysis of wholesale sales volumes to the retail drug stores) were investigated to identify growth rate (GR) trends of interest. Results: A PMS analysis has shown 11.92% and 6.65% of GR (medicines and BAS combined) in USDT and NPS respectively. For medicines alone the results were: 11.56% and 6.23% GR in USDT and NPS respectively. As for BAS, USDT and NPS, the figures were 21.48% and 15.36% of GR respectively. A further stratification by five LM has shown the highest GR in medicines used for treatment of Uro-Genital diseases (13.24% and 10.01% for USDT and NPS respectively), whereas in BAS the highest GR was in the Cardio-Vascular group (63.84% and 92.82% for USDT and NPS respectively). Conclusions: The results of the study go in line with the worldwide trends in shifting whenever possible, from medicines to using HSM of natural origin, of which BAS are a major part. The study does not claim to identify the underlying compound factors influencing such a tendency. However, the reality at had compels for studying the levels of BAS administration and use literacy among both HSM prescribers and consumers. Karlsberg Schaffer S 1, Sussex J 1, Hughes D 2, Devlin N 1 1Office of Health Economics, London, UK, 2Bangor University, Bangor, UK . . . . Objectives: In the UK, when a technology is recommended by the National Institute for Health and Care Excellence (NICE), the NHS is mandated to provide the funding to accommodate it within three months. Explicit in NICE’s approach to health technology assessment (HTA) is the assumption that the approval of a new, cost-increasing technology will result in the displacement of an existing, less cost-effective health care programme from elsewhere in the NHS. The objective of this study is to identify the actual opportunity costs of specific NICE decisions by documenting how in practice local commissioners in Wales accommodated financial shocks arising from technology appraisals (TAs). Methods: Interviews were conducted with Finance and Medical Directors from all seven Local Health Boards (LHBs) in NHS Wales. These interviews covered prioritisation processes, as well as methods of financing NICE TAs and other financial “shocks” at each LHB. We then undertook a systematic identification of themes and topics from the information recorded. Results: The financial impact of NICE TAs is generally planned for in advance and the majority of LHBs have contingency funds available for this purpose. Efficiency savings (defined as reductions in costs with no assumed reductions in quality) were a major source of funds for all cost pressures. Service displacements were not linkable to particular TAs and there appears to be a general lack of explicit prioritisation activities. The Welsh Government has, on occasion, acted as the funder of last resort. Conclusions: The assumption that newly recommended technologies will displace existing NHS services does not appear to hold true in practice. As the additional cost pressures represented by new NICE TAs are likely to be accommodated by greater efficiency and increased expenditure, the true opportunity cost of HTA decisions is extremely difficult to quantify and may even lie outside the NHS. PHP107 Impact of 2011 German Health Care Reform on Prices PHP110 Funding Integrated Health Care Services Ackermann A C 1, Reinaud F 2, Ando G 3, Izmirlieva M 3 1IHS, Zurich, Switzerland, 2IHS, Paris, France, 3IHS, London, UK Belgaied W 1, Urbinati D 2, Toumi M 3 1Creativ-Ceutical, Tunis, Tunisia, 2Creativ-Ceutical, Luxembourg, Luxembourg, 3University Aix-Marseille, Marseille, France . . . . . Objectives: This study seeks to evaluate the incremental benefit scores granted to new medicines under the 2011 AMNOG reform in Germany, and if there is a correlation between that score and the extent of price reduction after negotiation with the statutory health insurance fund. Methods: Resolutions issued by the G-BA related to all drugs which achieved the early benefit assessment process between January 2011 and June 2014 were reviewed to determine whether the drugs were deemed to bring an added benefit. Under AMNOG, an added therapeutic benefit score is granted to medicines according to 6 categories (major, considerable, minor, non-quantifiable, no or less added benefit versus the comparator). As part of the study, IHS created an overall quantitative innovation score for each product, based on any and all qualitative ratings granted by the G-BA in each patient subgroup. The score ranged from 0 (no added benefit proven) to 4 (major added benefit) and was weighted against each patient population. An average price reduction per innovation score range was then calculated. Results: Out of the 76 drugs assessed in the study, a total of 44 were deemed to bring an added benefit over the appropriate comparator by the G-BA, of which 34 had achieved price negotiation as of end of May 2014. These ratings translated into an average innovation score of 1.77, and were subject to price cuts averaging 21.7%. Conclusions: Our analysis highlights that innovative medicines can be subject to significant price cuts in Germany. No correlation between the innovation score and subsequent price cuts could be derived, as many other factors, including the initial price difference between the new medicine and the comparator, enter into account as part of the price negotiations. PHP108 Real World Data (Rwd) At T= 4 in the Netherlands Lie X , Nijhuis T Quintiles Consulting, Hoofddorp, The Netherlands . . Objectives: If the cost-effectiveness and appropriate use of a drug in the Netherlands cannot yet be determined during the initial reimbursement assessment (T= 0), this will be done at the re-evaluation 4 years later (T= 4) using RWD. Outcome research is an essential part of the re-evaluation of conditionally reimbursed drugs. The objective was to assess the reasons for accepting or rejecting outcome research results to confirm cost-effectiveness and appropriate use of a drug at T= 4. Methods: The website of Zorginstituut Nederland (ZiNL) was searched for re-evaluations of drugs published between January 2006 and May 2014. ZiNL’s assessments of the outcome research were compared. Results: ZiNL published the outcome research results for four drugs. The outcome research results for agalsidase alfa & beta were accepted for demonstrating cost-effectiveness and appropriate use in Fabry’s disease. The T= 4 results for omalizumab for . . . Objectives: Western countries health care systems face growing challenges dealing with disability and death due to chronic diseases. Coordination of health care services has become unavoidable. Our objective was to identify the Integrated Health Services (IHS) in the major health care markets and understand the funding process. Methods: IHS provided in Europe, North America and Asia were identified through a literature review. Future perspectives were based on country policy and observed trends. Results: All studied countries developed IHS such as disease management, case management and telemonitoring services, with the United States (US) being the major market. However, levels of implementation, funding processes and stakeholders involved vary highly between different countries. Funding processes such as fee for services and capitation are widely used in all studied countries and Payment for performance (P4P), bundled payment and diagnosis-related group (DRG) for outpatient are in progress, used mainly in the US, and to a much lower extent in the United Kingdom (UK) and Germany. Multiple IHS exists in France, though inappropriate incentives hinder their development. In the US, under the Affordable Care Act, Accountable Care Organisations (ACOs) are testing a range of payment models (capitation, one-sided/two-sided shared saving fee-for-service, bundled/episode payments and P4P). Conclusions: IHS have become ubiquitous in all health organisations. All countries studied are expected to develop more IHS based on P4P schemes. The P4P of ACOs represents the ultimate gold mine for the development of paid health care services. Even if this concept is still in progress it will be leading this market. This will also change dramatically the way pharmaceutical companies will do business. Drugs will have to be integrated in a more complex selling process driven by medium to long term outcome impact. The management of confounding factors on outcomes is critical and represents the challenge for ACOs. PHP111 Acceptance Of Telemonitoring By Health Care Professionals in Germany: A Question Of Financial Conditions Leppert F , Dockweiler C , Eggers N , Webel K , Hornberg C , Greiner W Bielefeld University, Bielefeld, Germany . . . . . . Objectives: The comprehensive implementation of telemedical applications still lags behind expectations in Germany. One of the main barriers to innovation is a lack of both a willingness to adapt and user’s acceptance. Processes of adoption and acceptance are characterized by a network of different factors which influence attitude and behavior which differ in severity depending on each user group. One key factor for accepting and adopting an innovation is the economic framework. We therefore examined the influence of economic factors influencing the A423 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 attitudes of physicians in Germany towards telemonitoring. Methods: In the first step fundamental determinants of adoption and acceptance were identified using a systematic literature review and transferred to a theoretical effect model. This model was used to create a quantitative questionnaire which was then used to interview online 201 outpatient and inpatient physicians from different fields of medicine. The participants (84.3% male, mean age 53 years) were asked to assess the empirically based economic attributes of telemonitoring regarding its implementation. Results: The results show a lack of information regarding the financial risks of using telemedical technology, as only 14% of those interviewed said they felt sufficiently informed about the subject. Barriers to using telemedicine technology include missing arrangements for reimbursement, uncertain financial advantages and missing business models. In addition, the cost of implementation are indistinct for a broad majority of interviewees. On the other hand, in most queried dimension the user see a potential financial benefit due to telemedicine. The positive effects expected from telemonitoring were rated much better rated by those who already used telemonitoring in their professional life. Conclusions: In addition to a more precise legal framework, information security and quality-based guidelines, more emphasis must be placed on economic issues. It is also imperative that physicians be better informed about all aspects of telemedicine. PHP112 External Reference Pricing Impact of the Integration of the Amnog Discount In The List Price Thivolet M 1, Cetinsoy L 1, Vataire A L 1, Rémuzat C 1, Brunet J 2, Kornfeld A 1, Toumi M 3 1Creativ-Ceutical, Paris, France, 2Assistance Publique des Hôpitaux de Marseille, Marseille, France, 3University Aix-Marseille, Marseille, France . . . . . . . . Objectives: In Germany, the AMNOG law replaced free pricing by the early benefit assessment (EBA) since 2011. Manufacturers are free to set new drugs’ prices for up to one year after which the price is negotiated between manufacturers and insurers based on the EBA. The negotiated price, that is consistently lower than the one set by the manufacturers, is considered as the official list price since April 1st2014. The objective of this study is to evaluate the potential impact of this new law on the external reference pricing (ERP) in Europe. Methods: A simulation model, developed for the European Commission, was used to simulate ERP’s impact on Boceprevir and Telaprevir prices after five years, following the discount’s inclusion on the official list price in Germany. ERP impact on price was evaluated in Belgium, Czech Republic, France, Germany, Luxembourg, Norway, Slovakia, Slovenia, Sweden, Switzerland, The Netherlands and UK for Boceprevir and in Belgium, Finland, France, Germany, Luxembourg, Norway, Poland, Slovakia, Slovenia, Sweden, Switzerland, The Netherlands and UK for Telaprevir. National policy inputs were obtained from a literature review and consultation of international organisations’ representatives. Prices used at the start of the simulation were obtained from IMS. Results: After five years, the relative price variation of Boceprevir between the scenario with AMNOG discount integrated in the list price and with no discount integrated in the list price was null in Belgium, Luxembourg, Sweden and UK, of -8.1% in the Netherlands, -9.2% in Norway, nearly -10% in Czech Republic, France, Slovakia and Switzerland, and -14% in Slovenia. For Telaprevir, the price variation was null in Belgium, Finland, Luxembourg, Slovenia and the UK, of -0.8% in Slovakia, -2.5% in the Netherlands, -2.9% in Norway, -6.9% in France and -8.6% in Switzerland. Conclusions: Integrating AMNOG discount in the list price impacts significantly the price in European countries due to ERP. PHP114 A Comparison of Cobb-Douglas, Translog and Additive Models of the Production Functions of Inpatient Services In Public Hospitals Reyes-Santias F 1, Antelo M 2, Rodriguez-Alvarez M X 1, Vivas-Consuelo D 3, Cadarso-Suarez C 2, Caballer-Tarazona V 3 1Universidad de Vigo, Vigo, Spain, 2Universidad de Santiago de Compostela, Santiago de Compostela, Spain, 3Universidad Politecnica de Valencia, Valencia, Spain . . . . . PHP115 Gender-Related Behaviors in Drug Consumption In Italy Orlando V 1, Bruzzese D 2, Putignano D 1, Guerriero F 1, Tettamanti A 3, Menditto E 1 1CIRFF- Center of Pharmacoeconomics, Naples, Italy, 2University of Naples, Federico II, Napoli, Italy, 3IMS Health, Milano, Italy . . . PHP116 Value of Life And Cost of Pre-Mature Deaths With The Perspective Of Productivity As Net Tax Revenue: A Comparison In Usa, Canada, Japan and Australia Yenilmez F B 1, Tuna E 1, Atikeler K 1, Kockaya G 2, Tatar M 1 University, Ankara, Turkey, 2Health Economics and Policy Association, Ankra, Turkey . . . . . . 1Hacettepe Objectives: The Human Capital Theory emphasizes investments to the health care sector as an important element in achieving and sustaining economic development. Investments to health care sector improves macro and micro economic outcomes for the whole society. The aim of this study is to calculate the possible produced value for a life-time term (VLT) and cost of pre-mature deaths (CPD) from the productivity for USA, Canada, Japan and Australia. Methods: Net present value (NPV) of the taxes and spending for each year were calculated. For calculating NPV in the government perspectives, two modelling approaches were combined, human capital modelling based on lives saved and lost productivity, and generational accounting, which accounts for a range of other government fiscal transfers to citizens. The possible produced value for a life-time term for each country were assumed as calculating the total NPV for each country depending on the countries life expectancy. CPD for each countries were assumed as the difference between NPV on the year of life expectancy and each decades as life years 60, 50, 40, 30, 20, 10. The economic values for the model of each country derived from World Bank, OECD, UNESCO or WHO. Results: Possible produced value for a life-time term for each country were calculated as US $ 1.415.530, US$ 774.663, US$ 238.236 and US$ 2.917.835 for USA, Canada, Japan and Australia, respectively. CPD per person for USA were calculated as US$ -1.526.126, US$ -1.661.257, US$ -1.300.923, US$ -796.547, US$ - 351.827 and US$ -40.507 for the life years 10, 20, 30, 40, 50 and 60 respectively. The trend was same for Canada, Japan and Australia. Conclusions: However the study was based on a hypothetical model that calculated the NPV with the taxes and spending in a life-time term, the results of each country were parallel. . . Objectives: To investigate the adequacy of the widely used Cobb-Douglas and translog models of the production functions of hospital in-patient services. Methods: To investigate the adequacy of the widely used Cobb-Douglas and translog models of the production functions of hospital in-patient services, we fitted these and additive models (AMs) to 2002-2007 data for the gynaecology and obstetrics, general and digestive surgery, internal medicine, and traumatology and orthopaedic surgery services of 10 public hospitals in Galicia (NW Spain). Production, measured as admissions weighted in accordance with their diagnosisrelated groups (DRGs), was treated as a function of physician full-time equivalents as surrogate labour factor and number of beds as surrogate capital factor. Results: For the General Surgery specialty the findings for the CD model indicate a better fit than those for the Translog and the AM, as it is shown by AIC value while R2 (CD= 96.32, Translog= 96.30, AM= 98.30) prefers the flexible AM. This is a good example of using AM as a tool for checking the behaviour of existing parametric models. In this case we can be confident with Cobb-Douglas estimations. Findings for the Internal Medicine specialty indicate responses for the CD (AIC= -17.789) seems to be more “robust” than those based on the AM (AIC=-13.113) and Translog (AIC=-15.939) models, R2 (CD= 95.88, Translog= 95.80, AM= 97.90) shows better fit for the AM regression model. Conclusions: Our results suggest that while the Cobb-Douglas and translog models suffice to represent the production functions of services with low average DRG weight, the greater flexibility of models such as AMs is required for services with higher average DRG weight. . entire populations. The aim of this study was to describe difference in prevalence of drugs use in the entire population in Italy between men and women. Methods: We performed a cross-sectional study using 2012 data from the IMS LifeLink Treatment DynamicsTM LRx Database, an italian-based administrative database that includes all prescribed drugs that are reimbursed by the Italian National Healthcare System. Pharmacological groups accounting for > 90% of the total volume in Defined Daily Doses were considered. Crude and age adjusted differences in prevalence were calculated as risk ratios of women/men. Results: 31 therapeutic categories were analyzed and there are significant differences for 30 of them. The largest sex difference in prevalence was found for thyroid preparations that were more common in women (59.3/1000 women and 10.9/1000 men, respectively). This was followed by antiinflammatory and antirheumatic products (114.0/1000 women and 67.4/1000 men) and antidepressants (62.1/1000 women and 26.8/1000 men). The pharmacological groups with the largest relative differences of dispensed drugs were drugs affecting bone structure and mineralization (RR 12.4), calcium (RR 7.0) and thyroid therapy (RR 4.9), which were dispensed to women to a higher degree. Antigout-agents (RR 0.4), vasodilators used in cardiac diseases (RR 0.7) and ACE inhibitors (RR 0.7) were dispensed to men to a larger proportion. Conclusions: This is the first Italian study that shows substantial differences between men and women. Our findings are congruent with those reported previously in the literature. Some of differences may be explained by variations in disease prevalence and severity, pathophysiology, or by other biological differences. However, it is also evident that other differences lack a rational medical explanation. . . Objectives: Sex differences in drug use have been demonstrated in several therapeutic area. However, there is a lack of overviews on sex differences of drug use in PHP117 Home Pharmacies: An Insight In Self-Medication Practice Paut Kusturica M 1, Tomic Z 1, Bukumiric Z 2, Trivic B 1, Sabo A 1 1Faculty of Medicine, Novi Sad, Serbia and Montenegro, 2Faculty of Medicine, Belgrade, Serbia and Montenegro . . . . . Objectives: Based on the content of drugs in households, it is possible to examine the inclination of patients towards self-medication and groups of drugs that are commonly used for this purpose. Taking into consideration the above, the present study aimed to investigate the self-medication practice, with an emphasis on self-medication with prescription only medications. Methods: The study was performed in 8 months period and involved 383 randomly selected households. The study consisted of a personal insight into the inventory of all drugs. The interviewer recorded the total number of drugs in households and asked respondents for each the drug was obtained on prescription or bought in the pharmacy for self-medication. After the data were collected, drugs were classified according the Anatomical Therapeutic Chemical (ATC) Classification System. Drugs were also classified according to ALIMS (Medicines and Medical Devices Agency of Serbia) into two groups: prescription only medication (POM) and OTC (Over the Counter) drugs and then analyzed. Results: The total number of drug items present in the 383 households was 4384 with an average of 11.4±5.8 per household. More than a half of drugs in households were prescription only medications (58,5%). Approximately one third of these drugs were obtained without prescription. The most common prescription only medications obtained without prescription were anti-inflammatory and antirheumatic products (41,8%) and antibacterials for systemic use (12,4%). Ibuprofen (61,0%), diclofenac (27,8%) and nimesulide (8,2%) were the most common self-medicated drugs in the group of antiinflamatory drugs, while the most frequently self-medicated antibiotics were cefalexin (25,7%), doxyciclin (18,6%), cotrimoxazole (17,7%). Conclusions: In conclusion, our survey indicated that self-mediation with prescription drugs appeared to be a rather common practice, which is far away from the concept of “responsible self-medication”, especially regarding antibiotics. ACKNOWLEDGEMENT: This work was supported by the Ministry of Science and Technological development, Republic of Serbia, project No. 41012. A424 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PHP118 Patterns and Predictors of Hospital Readmission in Taiwan Cheng J S 1, Ku H P 1, Chang C J 2 Gung University, Tao-Yuan, Taiwan, 2Chang Gung University, Kwei Shan, Tao Yuan, Taiwan . . . . . . 1Chang Objectives: Hospital readmissions have been an important issue, as they reflect suboptimal quality of medical care and incur high health care expenditures. However, limited information is available on the patterns of hospital readmission in the entire population to support a thorough planning to prevent hospital readmissions. Therefore, this study aimed to examine the patterns and economic burden of hospital readmission in Taiwan, and identify predictors of hospital readmissions. Methods: This study used the National Health Insurance Research Database of enrollees randomly selected from those enrolled in the National Health Insurance program in 2005. Individuals who were admitted to acute hospitals in 2005 were selected and their readmission patterns one-year after discharge were examined. Cox proportional hazards regression model was adopted to identify predictors of hospital readmission. Results: The 30-day, 6-month and one-year readmission rates were 11%, 25%, and 34%, respectively. During the one-year follow-up, 52% of total health care expenditures were due to hospital readmissions. Of those who were readmitted to hospitals, 56% were readmitted once and took up 29% of the cost of rehospitalization. However, those readmitted for more than three times (5%) accounted for 30% of the cost. The major disease category of the highest 30-day and one-year readmission rates was neoplasms. The disease of the highest 30-day and one-year readmission rates were cancer of bronchus and lung (36%) and cancer of liver and intrahepatic bile duct (74%), respectively, and the most frequent reason for readmission was the disease itself. Age, gender, place of residence, previous hospitalization history, comorbidities, and length of stay of the index hospitalization were risk factors of hospital readmissions. Conclusions: This study identified diseases of higher short-term and long-term readmission rates, causes of shortterm and long-term hospital readmissions, and predictors of hospital readmission. The information is of importance for planning interventions to reduce hospital readmission rate. PHP119 Secure System for Iv Administrations: Health Economic Impact Of A “Smart” Infusion Safety System Schmidt A , Bénard S st[è]ve consultants, Oullins, France . . Objectives: More than half of medication errors are reported during administration. “Smart” Infusion Safety Systems (SISS) have been developed to avoid administration errors and provide data for continuous quality improvement (CQI). Few data produced by SISS have been analysed. Given the lack of clinical trials, a model was developed in order to assess the economic benefits of SISS from the perspective of a French public hospital. Methods: Comprehensive data from 6 intensive care units (ICU) were analysed. Two types of alerts were defined: hard (absolute) and soft limits. An avoided error was defined as the detection of a scheduled infusion over a hard limit or a scheduled infusion over a soft limit that was later overridden by staff. The severity of consequences was estimated on the HARM INDEX score, which is based on the pharmacologic risk, overdose detectability, overdosing range and the type of hospital unit. According to this score, errors were categorised as minor, significant, and serious. The economic value was estimated based on the hypothetical resulting length of stay had the error not been avoided, which increases according to error severity, and the financial investment for SISS. Results: Overall, 207,025 infusions were analysed from the database of 6 ICUs and 8,503 of them were associated with a safety alert (4.1%), including 987 errors (0.48%). Applied to one public ICU equipped with 5 SISS, the model estimates that SISS would prevent 78 errors per year (54 minor, 12 significant, 12 serious) and avoid 173 days of hospitalisation per year. This corresponds to an annual savings of € 139,491 for the hospital (39% and 61% from significant and serious errors, respectively). Conclusions: This model demonstrates the high economic burden of medication errors for French public hospitals and the need for CQI. To date, no equivalent evaluation has been conducted in France. PHP120 Value of Life and Cost of Pre-Mature Deaths With The Perspective of Productivity As Net Tax Revenue: A Comparison in France, Germany, Italy, Spain, United Kingdom Tuna E 1, Yenilmez F B 1, Atikeler K 1, Kockaya G 2, Tatar M 1 University, Ankara, Turkey, 2Health Economics and Policy Association, Ankra, Turkey . . . . . . 1Hacettepe Objectives: The Human Capital Theory emphasizes investments to the health care sector as an important element in achieving and sustaining economic development. Investments to health care sector improves macro and micro economic outcomes for the whole society. The aim of this study is to calculate the possible produced value for a life-time term (VLT) and cost of pre-mature deaths (CPD) from the productivity for France, Germany, Italy, Spain, UK (UK). Methods: Net present value (NPV) of the taxes and spending for each year were calculated. For calculating NPV in the government perspectives, two modelling approaches were combined, human capital modelling based on lives saved and lost productivity, and generational accounting, which accounts for a range of other government fiscal transfers to citizens. The possible produced value for a life-time term for each country were assumed as calculating the total NPV for each country depending on the countries life expectancy. CPD for each countries were assumed as the difference between NPV on the year of life expectancy and each decades as life years 60, 50, 40, 30, 20, 10. The economic values for the model of each country derived from World Bank, OECD, UNESCO or WHO. Results: Possible produced VLT term for each country were calculated as US$ 993.347, US$ 629.814, US$ 245.885, US$ 48.818 and US$ 1.628.957 for France, Germany, Italy, Spain, UK, respectively. CPD per person for France were calculated as US$ -1.060.571, US$ -1.148.852, US$ -887.348, US$ -551.873, US$ - 224.070 and US$ -10.972 for the life years 10, 20, 30, 40, 50 and 60 respectively. Although numbers were different, the trend was same for Germany, Italy, Spain, UK. Conclusions: However the study was based on a hypothetical model that calculated the NPV with the taxes and spending in a life-time term, the results of each country were parallel. PHP121 Switching Patients With Primary Antibody Deficiencies to HomeBased Subcutaneous Immunoglobulin: Economic Evaluation of An Interprofessional Drug Therapy Management Program Perraudin C , Bourdin A , Berger J , Bugnon O School of pharmaceutical sciences, University of Geneva, University of Lausanne, Lausanne, Switzerland . . . . Objectives: Lifelong immunoglobulin G (IgG) replacement is the standard therapy for patients with primary antibody deficiencies. It can be administered either intravenously (IVIg) by health care providers in hospital or subcutaneously (SCIg) by patients at home. However, self-administration requires patients’ education and support over long term to ensure proper adherence and optimal efficacy and safety. Every patient who switches to SCIg is proposed by the Policlinique Médicale Universitaire (Lausanne, Switzerland) a drug therapy management program with a nurse and a community pharmacist including training, coaching and follow-up. The aim of the study was to evaluate if switching to SCIg at home including the management program was cost-effective compared to IVIg at hospital. Methods: Assuming that both therapies provide similar efficacy, a 3-years cost-minimization analysis based on a simulation model was performed from a societal perspective. Health care costs (IgG, time of professionals, infusion pumps and disposables) were derived from administrative data. Transport and losses of productivity were estimated. One-way sensitivity analyses were performed. Results: Under base case assumptions, SCIg at home was estimated at 34960 CHF per patient the first year and 30 030 CHF in subsequent years against 34 170 CHF per year for IVIg. The total savings for a switch to SCIg at home with the program was 7490 CHF per patient over 3 years. Results were relatively sensible to the assumptions. Conclusions: Homebased SCIg therapy including an interprofessional therapy management program may be an effective and efficient alternative to hospital for patients with primary antibody deficiencies. Additional costs from purchase of equipment and management program in the first year were offset by hospital costs avoided in short term. Additional studies are ongoing to analyse the retention in the therapy in medium term and the impact on quality of life. PHP122 Gender Medicine in Germany: What is so Difficult about its Implementation? - An Empirical Study in Germany – Chase D P 1, Mitar I 2, Oertelt-Prigione S 3, Hess N 4, Amelung V E 5 for Applied Health Services Research, Berlin, Germany, 2Pfizer Pharma GmbH, Berlin, Germany, 3Charité Berlin, Berlin, Germany, 4Cardiology Practice, Berlin, Germany, 5Hannover Medical School, Hannover, Germany . . . . . . . 1Institute Objectives: Personalized medicine is currently a popular topic in health care debates. Yet, the basic differentiation between females and males is hardly found in care delivery/health management programs. This study aimed at elucidating the opinion of German Statutory Health Insurance (SHI) managers and internal specialists regarding gender-specific care in order to understand their perceptions on responsibilities and possible implementation opportunities. Methods: Between April and June 2013, a questionnaire on the implementation of gender medicine in the current health care landscape was developed. Based on literature review and expert consultations, it included open- and closed-ended questions on expectations/prioritization, need for action, and implementation of gender medicine. Forty-eight insurance managers of the largest German SHIs, covering over 95% of the market, and approximately 16,000 physicians of the German Society for Internal Medicine (DGIM) were contacted to complete a web-based survey. Descriptive analyses, Chi-square tests, and Pearson correlation coefficient were used to investigate the research objective. Results: According to both, insurance managers (76%) and physicians (60%), gender-specific care is not sufficiently incorporated into standard medical care. Respondents claim the responsibility lies with the ministry of health, physicians and medical staff, as well as their associations. Specifically, more evidence is needed to incorporate gender aspects in treatment guidelines, an idea which is well-supported by insurances (65%) and physicians (70%). A top-down approach for implementation is preferred by 65% of insurance managers and 50% of physicians, whereas fewer participants encourage bottomup mechanisms. Conclusions: German SHIs expect a significant governmental influence and/or support of self-governing bodies to achieve an incorporation of gender medicine into daily practice. Primary responsibility for the integration of gender-specific approaches is perceived to lie with physicians. As soon as critical hurdles in the medical field will be removed, the positive perception of both participating parties can be integrated in the implementation process of gender-medicine. PHP123 Value of Life And Cost of Pre-Mature Deaths With the Perspective of Productivity As Net Tax Revenue for Turkey Kockaya G 1, Tuna E 2, Yenilmez F B 2, Atikeler K 2, Tatar M 2 Economics and Policy Association, Ankra, Turkey, 2Hacettepe University, Ankara, Turkey . . . . . . 1Health Objectives: The Human Capital Theory emphasizes investments to the health care sector as an important element in achieving and sustaining economic development. Investments to health care sector improves macro and micro economic outcomes for the whole society. The aim of this study is to calculate the possible produced value for a life-time term (VLT) and cost of pre-mature deaths (CPD) from the productivity for Turkey where the life expectancies was noted 75 years. Methods: Net present value (NPV) of the taxes and spending for each year were calculated. For calculating NPV in the government perspectives, two modeling approaches were combined, human capital modeling based on lives saved and lost A425 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 productivity, and generational accounting, which accounts for a range of other government fiscal transfers to citizens. The possible produced value for a life-time term was assumed as calculating the total NPV depending on the life expectancy. CPD was assumed as the difference between NPV on the year of life expectancy and each decades as life years 50, 40, 30, 20, 10 and new born. The economic values for the model were derived from World Bank, OECD, UNESCO or WHO. Results: Possible produced value for a life-time term for Turkey was calculated as US$ 483.298. Cost of pre-mature death per person was calculated as US$ - 102.064, US$ - 271.716, US$ -441.679, US$ -583.726, US$ - 518.753,14 and US$ - 483.986 for the life years new born, 10, 20, 30, 40 and 50 respectively. Conclusions: However the study was based on a hypothetical model that calculated the NPV with the taxes and spending in a life-time term, cost of premature death was calculated as the highest in early ages and was decreasing up to the retirement age. The results may be reference for the decision makers. Health policy makers may improve the access to the treatments in the early life years for the possible increased cost of premature deaths in Turkey. PHP124 Exposure To Potential Drug-Drug Interactions in Teaching Hospital Of South Punjab, Pakistan Khan H M S 1, Butt H 1, Shah N H 2 1Islamia University Bahawalpur, The Islamia University of Bahawalpur, Pakistan, 2Bahauddin Zakarya University, Multan, Pakistan . . . . . . Objectives: Drug-drug interaction is a common error in poly-medication and contributes a major part in adverse drug reactions. The aim of the present study was to evaluate the percentage of potential drug-drug interactions in the prescription in the region of South Punjab Pakistan prescribed by the medical practitioner Methods: A total of 100 prescriptions were included in this study. Institution based retrospective study was performed in Nishter Hospital Multan, Pakistan, the 3rd oldest medical institution of Pakistan, which has a capacity of 1800 beds and having a best facility to hold a large number of emergencies at a time. Prescriptions were collected to see the Drug - Drug interactions and compared with the standard reference of important DDI’s. The potential drug-drug interactions were categorized according to their severity, effect and mechanism. Results: The study showed that among 100 prescriptions (543 medicines), 41% of prescriptions have potential DDI’s. Mostly prescribed drugs were the antibiotics (38%), and the drugs belonging to class analgesic were found to contribute mostly in drug-drug interactions (26.50%). The survey showed the total of 71 interactions and their severity level accounting as major (20.10%), moderate (63.4%) and minor (15.90). The mechanism by which the drugs interact with one another showed that there were pharmacokinetic (60.5%), pharmacodynamic (38.6%) and few of them interact by the mechanism which was not specified in the available literature (0.9%). Conclusions: Drug-drug interactions occur in poly-medication and need to be evaluated and monitored for the positive impact on the medication use system and improvement of quality of patient care. A clinical pharmacist with its accurate knowledge of drug, their effect on human organs and their interaction with other can monitor and manage these drug interactions. PHP125 Value of Life And Cost of Pre-Mature Deaths With The Perspective Of Productivity As Net Tax Revenue: A Comparison In Argentina, Brazil And Mexico Atikeler K 1, Tuna E 1, Yenilmez F B 1, Tatar M 1, Kockaya G 2 1Hacettepe University, Ankara, Turkey, 2Health Economics and Policy Association, Ankra, Turkey . . . . . . Objectives: The Human Capital Theory emphasizes investments to the health care sector as an important element in achieving and sustaining economic development. Investments to health care sector improves macro and micro economic outcomes for the whole society. The aim of this study is to calculate the possible produced value for a life-time term (VLT) and cost of pre-mature deaths (CPD) from the productivity for Argentina, Brazil and Mexico. Methods: Net present value (NPV) of the taxes and spending for each year were calculated. For calculating NPV in the government perspectives, two modelling approaches were combined, human capital modelling based on lives saved and lost productivity, and generational accounting, which accounts for a range of other government fiscal transfers to citizens. The possible produced value for a life-time term for each country were assumed as calculating the total NPV for each country depending on the countries life expectancy. CPD for each countries were assumed as the difference between NPV on the year of life expectancy and each decades as life years 60, 50, 40, 30, 20, 10. The economic values for the model of each country derived from World Bank, OECD, UNESCO or WHO. Discount rate and inflation for wages were taken as 3% per year for all countries. Results: Possible produced VLT for each country were calculated as US $ 716.945, US$ 663.129 and US$ 653.598 for Argentina, Brazil and Mexico, respectively. CPD per person for Argentina were calculated as US$ -666.234, US$ -657.861, US$ -512.554, US$ -338.675, US$ - 164.722 and US$ -24.615 for the life years 10, 20, 30, 40, 50 and 60 respectively. The trend was same for Brazil and Mexico. Conclusions: However the study was based on a hypothetical model that calculated the NPV with the taxes and spending in a life-time term, the results of each country were parallel. PHP126 Community Pharmacist Perceptions of Generic Medicines: A Survey in Tehran Mehralian G 1, Noee F 2, Yousefi N 1, Peiravian F 1 1Shahid Beheshti University of Medical Sciences, School of Pharmacy, Tehran, Iran, 2Azad University, Tehran, Iran . . . . Pharmaceutical expenditure as has been increasingly grown during the last decades so that policy makers have paid more and more attention to develop strategies to deal with this challenging issue. Pharmacists play an important role in health continuum to implement such strategies through dispensing and delivering generic medicines to consumers. So, the main objective of this study was to explore the perception of Iranian pharmacists regarding generic substitution, and also to explore how Iranian pharmacists think about patients’ role, physicians’ role and government role in generic substitution. A cross sectional descriptive study involving the entire population of Tehran community pharmacies (n = 2000) was performed using a self-administrated anonymous questionnaire. A total of 1205 questionnaires were returned indicating a response rate of 60%. Regarding to the first section, 62% of participants agreed that pharmacists should be given right to generic substitution, and 45% viewed that the generic medicine are bioequivalent with brand ones. Majority of the respondents (73.6%) stated that, they do substitution once generic medicine is available. However, they think that patients, physicians and government have important role in generic substitution. More than half (75.5%) of pharmacists believed physicians’ prescription behavior highly influenced by the marketing and promotion activities of foreign companies. 93% of respondents pointed to full coverage of generic medicines by providers as a strategy to generic substitution. Summary, evidences indicated that Iranian pharmacists have good insights regarding implementation and promotion of generic substitution strategy. PHP127 Building A “High-Alert/High Risk Medications Reminder System” To Improve Patient Safety Kao L C Mackay Memorial Hospital, Taipei, Taiwan . . Objectives: This paper was aimed to introduce how we developed a “high-alert/ high risk medications reminder system” operated in the computer system to improve medication safety in our hospital. Methods: This reminder system was designed in a 2,400-bed health institute with 2,000 nurses, consisting of 12 nursing divisions and 75 units. The new module was added to the current BCMA system with VB. net. The functions of this computer system include seven steps. Results: A satisfaction and cognition survey on the application of the new, computerized “highalert/ high risk medications reminder system” was done to MICU nurses (N= 36). The survey response rate was 100%. The results showed high rate of positive support on the new system due to easy to access and easy to learn. The satisfactory rate comparing the old paper form system and the new computer system was 62.8% vs. 96.1%. More importantly, the “high-alert/high risk medications reminder system” appeared to reduce the administration error rate from 5 cases in 2011 to 0 cases in 2012. Conclusions: Our early experience showed that an efficient and userfriendly “High-Alert/High Risk Medications Reminder System” could be helpful for medical staffs to improve medication safety although a longer follow-up time to evaluate the efficacy of the computerized system is still needed before a final conclusion can be established. In addition, we are also keen to investigate whether this system could reduce the waste of time, expenditure, and manpower in a medical center in our future study. PHP128 Cost-Effectiveness of Telemedicine: Lessons To Learn From An International Review Bongiovanni-Delarozière I 1, Le Goff-Pronost M 2, Rapp T 3 1French National Authority of Health (HAS), Saint Denis La Plaine, France, 2Telecom Bretagne, Brest, Brittany, France, 3University of Paris Descartes, Paris, France . . . Objectives: At the global level the large deployment of telemedicine raises needs for cost-effectiveness evaluations. The objective of this literature review is to explore to what extend telemedicine innovations that were implemented in many countries were cost-effective. Specifically, we explore whether the model used to evaluate the cost-effectiveness can be adapted to the evaluation of telemedicine technologies. So, what answers can a review of the international literature relating to the medico-economic evaluation of telemedicine provide? Methods: Following analytical reading of 286 articles published between 2000 and 2013, 74 studies that implemented economic evaluation of telemedicine are analysed. Three axis of analysis are considered: the act concerned by the telemedicine intervention, the medical speciality, the economic evaluation method implemented. Results: The descriptive analysis showed significant heterogeneity in studies characteristics: economic analysis method, telemedicine applications, medical specialities, and organisational practices. The qualitative analysis underlines that most studies face methodological issues and provide reduced evidence of the economic impact of the telemedicine interventions. The telemedicine technologies are too individualised (by the context and the organisation) to be evaluated using the standards of cost-effectiveness analysis. Conclusions: This literature review did not allow proposing a classification for telemedicine practices identified as efficient, depending on the strategies compared, field of application or speciality, types of telemedicine or an organisation of care model. Despite all this, one focus can be the management of chronic diseases that remains a central topic at the international level. The increase in the number of medical specialities or fields of application concerned with telemedicine and the increased volume of activity necessitate the dissemination of methodological recommendations to promote the coherent development of economic evaluations. Our literature review shows that there is need to develop innovative methods to assess the cost-effectiveness of telemedicine technologies. PHP129 Assessment of Validity of An Indicator Module In Identification of adverse Drug Events in Patients of Medicine Department Mallayasamy S , Shravya J , Mehta A , Vilakkathala R Manipal University, Manipal, India . . . . Objectives: An indicator is a clue that helps a health care organization to identify adverse drug events and assess the overall harm that occurs from medical care within that organization. The main aim of the study was to investigate use of an indicator list for identification of adverse events in the health care setting studied. Methods: The study was a prospective observational study in a tertiary care teaching hospital. The study mainly involves the review of medical records of patients in general medicine department who were admitted due to drug related problems with the help of trigger tool. When the presence of indicator is identified, those cases were thoroughly scrutinized to identify adverse drug events and confirm A426 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 such events in the light of available evidence and discussion with the treating physicians. Results: A total of 275 patients were included in the study for case review as per the study criteria. Out of 275 patients, 150 patients had at least one indicator (55%) and detection of adverse events was about 19.2%. Ratio of actual adverse event detected to the presence of indicators was calculated as true positives. The average true positive value of indictors was 33%. Average number of indicators present per patient was determined to be 3.82. The harm categorization was carried out for the observed adverse events as per National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP). Conclusions: The present study showed that indicator tool could be used to review the cases prospectively and identify adverse events. The identified indicators showed the pattern and frequency of adverse events. Findings of our study supported the idea of making the indicator tool as a practical aid for identification of adverse events. PHP130 Barriers of Pharmaceutical Care in Community Pharmacies: Evidences From Emerging Counteries invited; 22 [8HQ/14LAF] tentatively agreed to participate and 15 [7HQ/8LAF] were finally interviewed. Compared to LAF, HQ perceive the GVD more useful (HQ: 100%; LAF:63%) and capable of positively influencing the final outcome of submissions (HQ:86%; LAF:50%). Compared to HQ, LAF consider the GVD is delivered too late (LAF:88%; HQ:13%) and that LAF should be involved in GVD development (LAF:100%; HQ:71% but only for review). HQ consider that LAF involvement should be restricted to the GVD review while LAF think they should participate in building the value story (63%) or the economic model (37%). LAF prefer a modular GVD with each component delivered separately (LAF: 88%; HQ: 66%), request national data to be included (LAF:63%; HQ:14%), and cannot see the utility of new technologies for the GVD (LAF:75%; HQ:33%). HQ and LAF expect the GVD to be updated continuously (HQ:6%; LAF:75%) and be supplemented with phase IIIb data for potential resubmissions (HQ:100%; LAF:88%). Conclusions: HQ and LAF expectations differ particularly on how soon the GVD should be made available, the involvement of affiliates during its development and the applicability of the GVD contents to local submissions. How to address these discrepancies to ensure alignment will be discussed. Mehralian G 1, Javadi arjmand A 2, Ranghchian M 1, Peiravian F 1 1Shahid Beheshti University of Medical Sciences, School of Pharmacy, Tehran, Iran, 2Azad University, Tehran, Iran PHP133 Management of Crises and Risks in German Hospitals - Factors Influencing Medication Errors The pharmaceutical care concept has been popular in the world during past decades, but it did not reach well to its maturity in some countries, so the investigation on barriers to the implementation of the pharmaceutical care in these countries would be great of interest. Objectives: The aim of the study was to identify and prioritize barriers to the provision of pharmaceutical care in Iranian community pharmacies based on Tehran community pharmacists’ perceptions. Setting The present study was conducted with participation of pharmacists working in community pharmacies settled in Tehran, the capital city of Iran. Methods: A cross-sectional descriptive study was performed using an anonymous questionnaire between August and November 2013. 505 community pharmacists expressed their perceptions on items by a 5-point Likert-type scale. Besides descriptive analysis, data was also analyzed through structural equation modeling. Main outcome measured was the importance perceived by community pharmacists in the case of 5 area including resources, attitude and vision, education and training, skills, regulatory and environmental issues. Results: Five major dimensions included in the survey instrument were confirmed by confirmatory factor analysis. According to the model developed based on pharmacists’ opinions, lack of pharmacists’ skills and lack of appropriate regulation and environment are the two most important barriers of the provision of pharmaceutical care, and the least important is lack of resources. Conclusions: Findings of the present study showed that importance of various barriers vary from country to country based on the pharmacists’ perceptions, and corrective actions should be made accordingly. Draheim M 1, Flessa S 2, Glasberg R 1, Hartmann M 1, Schermuly C 1, Stanchev V 1, Tamm G 1, Hessel F P 1 1SRH University Berlin, Berlin, Germany, 2University Medicine Greifswald, Greifswald, Germany . . . . PHP131 Costs and Effectiveness of the Mediguide Technology in the Conditions of the Czech Republic Jurickova I , Jarosova V Czech Technical University in Prague, Kladno, Czech Republic . . Objectives: MediGuide technology is designed to monitor the position of diagnostic and therapeutic catheters equipped with a sensor MediGuide and navigation in vascular or cardiac intervention in the environment of electro-physiological laboratory. MediGuide compared to conventional technology, offers the 3D display on the live fluoroscopy or recorded background, radiation burden reduction, fluoroscopy time and total time performance. The aim of this thesis is to analyze the clinical and cost effectiveness of new technologies MediGuide in the Czech Republic. Results of the analysis are being compared to the studies of the German Cardiac where technology is positively evaluated. Methods: The evaluation method is cost-effectiveness analysis, which compares technology system MediGuide to a comparator angiography Siemens Artis Zee. Value of the effect was determined by TOPSIS in which they were entered weight values obtained by Saaty method of pair comparison by group of experts. The sum of direct costs includes costs associated with the operation of both technologies. Calculated values of costs and effects were achieved in incremental ICER assessment ratio, fixing the amount of costs per unit of effect. Results: More appreciated criteria in Saaty method were: safety and effectiveness of the technology. The fundamental values of the following criteria in this method were used: radiation burden, which in the Czech Republic reached values using conventional technology 575±511μ Gy. m2 and in Germany 14,453±7,403μ Gy. m2, fluoroscopy time values for Czech Republic 7.5±4.5min; Germany 34±9min and total time performance of Czech Republic 148±91 min; Germany 157±51min. The cost of power at the treatment of 100 patients was in the first year to CZK 181,711; MediGuide under the same conditions CZK 276,765. Conclusions: Manufacturer of MediGuide values technology as an effective, safe and value-added reduce the radiation burden of the patient, fluoroscopy time and shorten the overall time performance. PHP132 Can A Global Value Dossier Meet Headquarters and Affiliates’s Expectations? Kooreman P J 1, Casamayor M 2, Schuchardt M 1, Van Engen A 1 Consulting, Hoofddorp, The Netherlands, 2Quintiles Consulting, Barcelona, Spain . . . . . . . . . . . . . . Objectives: In German hospitals there is a need for optimized crisis- and risk management. About one third of all hospital-acquired patient damages are due to medication errors. The aim of our interdisciplinary project “Risk Management in Hospitals” is to analyze the behavior of the various key players in the fields of medical care, human resources, supply and IT-systems in the context of various simulated risk scenarios as well as to develop adequate risk management tools. In this abstract, the first results on medication errors are presented. Methods: Through a network of experts, systematic literature searches and expert workshops, the five most relevant crises from different disciplines (medical, HR, supply and IT) were identified and specified. In a standardized survey among German hospital managers further data were collected on these crises, their management and influencing factors. The results on medication errors presented here are based on the survey data using a linear regression model. Results: The analysis was based on fully completed questionnaires by 97 German hospitals. 70% of the hospitals reported that crises due to medical errors occurred in the last 5 years, which could not be handled with established risk management tools. Although the crisis-appearance was lower in private hospitals (28%) compared to publicly funded (31%) and nonprofit (41%) hospitals, this difference was not significant. We recognized that a high frequency in the occurrence of medical errors was significantly associated with a lower evaluation of risk management issues (e.g. the presence of a staff member responsible for risk management, scenario analysis and crises trainings). Conclusions: Our study confirms that German hospitals have to expand their targeted risk management activities in order to prevent the occurrence of hospital crises especially due to medical and medication errors. All survey results will be incorporated in a decision-making and benchmarking-tool for hospital managers to improve crisis management in German hospitals. PHP134 Comparison of Expected Versus Actual Cost Consequence Of Reimbursed Drugs in the Netherlands Between 2009 and 2013 Chatterjee A 1, Tempelaar S 1, Verheggen B 2 International, York, UK, 2Pharmerit International, Rotterdam, The Netherlands . . . 1Pharmerit Objectives: A Budget Impact Analysis (BIA) analyses the financial consequence associated with the uptake of a new treatment option in the market; which can impact planning and forecasting and influence changes to pharmaceutical and health care budgets. Therefore, BIA forms a key consideration of the Dutch Health Care Insurance Board (ZiNL) when making a reimbursement decision regarding a new treatment intervention. The current analysis investigates consistent trends observed between the predicted budget versus the actual budget spent on reimbursed drugs in the Netherlands. Methods: Forecasted data specific to new drugs reimbursed under list 1B between 2009 and 2013 were extracted from drug reimbursement reports available from the ZiNL website. Actual data of the selected drugs were extracted from the Drug and medical devices Information Project database (GIP database). Per year and cumulative (between 1 and 5 years) data on total drug cost, total number of users and total cost per patient were compared between the forecasted values and the actual values. Results: In total 20 drugs were included in the analysis, of which 12 presented data for 3 years or more. Compared to the actual data, the expected total drug cost was overestimated for 14 drugs, total number of users was overestimated for 12 drugs and for 10 drugs the total cost per patient was overestimated. Total number of users was most accurately estimated, presenting a pooled cumulative overestimation of 5 times the actual number of users. Total drug cost was least accurately estimated (pooled cumulative overestimation of 13 times the actual total cost), since the total drug cost includes the uncertainty reflected in both the total number of users and total cost per patient. Conclusions: The expected budget of 1B drugs in reimbursement submission reports generally overestimated the actual budget presented in the GIP database between 2009 and 2013. 1Quintiles Objectives: Gaining reimbursement is a complex process that involves generating the right evidence and communicating it successfully to stakeholders. To ensure alignment of the value story across countries and help affiliates in their local submissions, headquarters (HQ) develop the global value dossier (GVD). We have investigated whether the current GVD meets HQ and local affiliates (LAF) expectations. Methods: Pharma HEOR and market access professionals from HQ and LAF were invited to participate in telephone interviews to assess their views on current use and utility of the GVD. Results were analyzed to identify differences and alignments between HQ and LAF. Results: 43 [16HQ/27LAF] professionals were PHP135 Predicted Versus Actual Budget Impact of High-Cost Drugs In Ireland Fogarty E 1, Barry M 2 1National Centre for Pharmacoeconomics, Dublin, Ireland, 2HSE Medicines Management Programme, Dublin, Ireland . . Objectives: Budget impact analyses are an essential component in the economic evaluation of new drugs. These analyses allow the health care payer to assess the likely impact of the drug on the payer’s budget, and to plan for short- and long- term A427 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 resource allocation. The objective of this study is to compare the company-predicted budget impact with the actual budget impact of high-cost drugs reimbursed in Ireland. Methods: All drugs submitted to the health service executive (HSE, health care payer in Ireland) for reimbursement under the high-tech drug scheme (a scheme used to administer high cost drugs) from 2009 to 2012 were included in the review. Company estimates of the likely budget impact of the drug in 2013 were extracted from submissions and compared with actual expenditure in 2013 from the health service executive-primary care reimbursement service (HSE-PCRS). Only drugs for which budget impact estimates were available and which were reimbursed by the HSE in 2013 were included in the analysis. Results: Ten drugs were included in the analysis, including six cancer drugs, two immunomodulators for multiple sclerosis and rheumatoid arthritis, and two orphan drugs for cystic fibrosis and idiopathic thrombocytopenic purpura. The cumulative expenditure on these drugs in 2013 was € 55.8 million compared with a predicted gross budget impact of € 53.4 million, representing a € 2.4 million underestimate in company submissions. The most significant underestimate related to the drug for multiple sclerosis (€3.4 million) while the biggest overestimate related to the orphan drug for cystic fibrosis (€2.9 million). Conclusions: Company submissions have been shown to both under- and over-estimate budget impact predictions. It is important that budget impact estimates are as realistic as possible in order to effectively inform decisions on resource allocation or reimbursement. PHP136 Redel Study: Differences in Reimbursement Delays in Cee Countries Komáromi T , Bagi P , Salfer B Healthware Consulting Ltd., Budapest, Hungary . . . Objectives: The dynamic of the reimbursement politics shows a very different pattern in different countries. The REDEL study examined the elapsed time from marketing authorization to the starting date of reimbursement of the original medicines in Central and Eastern European Countries (Austria, Bulgaria, the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia and Slovenia). Methods: The basis of comparison were 216 products and their ATC codes selected from the database of the European Medicines Agency which were granted a marketing authorization between 1st January 2007 and 1st July 2013. In the case of these products the research studied the dates, when countries adopted them into their reimbursement system. The adoption was the subject of the study between 1st January 2010 and 1st July 2013. The following three different indicators were calculated in the study: REDEL - the delay between marketing authorization date and reimbursement date; INNREIMB - the number of reimbursed INNs according to a specific country or MAH; SR - Success Rate as the ratio of reimbursed INNs to examined INNs. Results: While an average of 403 days elapsed between the authorization and the starting date of reimbursement in Slovenia (mean of 76 products), the same period was 1295 days in Poland (mean of 21 adoptions). The average is 632 days. The top three in the ranking of REDEL of active substances (ATC 1st level – anatomical main group) were dermatologicals (D), respiratory system (R) and Muscolo-skeletal system (M). Conclusions: The results show that even threefold differences exist among the studied countries with regards to the reimbursement delay. An average of almost two years elapse until a producer can have the given product adopted into the reimbursement system in a country (the REDEL steadily increasing in the studied period, while the number of reimbursed products decreasing). PHP137 Black Box Amnog Rebates: What Is Driving the Price in the Negotiation With The Gkv-Spitzenverband? Freiberg M 1, Pruefert A 2, Hepp S 3, Schalber T 3 1Quintiles Consulting, Neu-Isenburg, Germany, 2Quintiles Consulting, Hoofddorp, The Netherlands, 3Quintiles Commercial, Mannheim, Germany . . . . Objectives: Price negotiations of a pharmaceutical company with the German GKV-Spitzenverband (Association of Statutory Health Insurance) after early benefit assessments can be considered a black box. This analysis aims at providing insights on the parameters that drive the final rebate to predict future pricing decisions and to enhance the negotiation strategy and therefore secure optimal pricing. Methods: Published benefit assessments from the G-BA (Federal Joint Committee) website and the products’ prices as listed in the German pharmacy selling system were used as a basis for research. The latter allows a comparison of launch prices (manufacturer selling prices) and prices after the negotiation with the GKV-Spitzenverband (reimbursement prices). The following parameters were analyzed: Rebate size, rebate by added benefit rating, and rebate by therapeutic area. Results: By June 2014, 36 products had been through price negotiation, with the rebate of the launch price ranging from 5-71% (average: 25%). The rebate of products with considerable benefit rating ranged from 10-35% (average: 21%). Products with minor added benefit reached rebates between 5-48% averaging at 23%. Products with no quantifiable benefit yielded rebates ranging from 11-44% (average: 24%). Products with no additional benefit had a rebate between 5-71% averaging at 27%. Products in oncology yielded an average rebate of 27%., followed by endocrinology (23%), central nervous system (22%), cardiovascular (21%), and infectious diseases (16%). Conclusions: The better the added benefit rating of a product, the lower is its negotiated rebate. However, only marginal average differences were observed. The rebate per therapeutic area did not reveal obvious patterns: Assessments for oncology products resulted in above average rebates, while rebates for products for infectious diseases were far below the average. PHP139 Comparison of Post-Authorisation Measures From Regulatory Authorities With Additional Evidence Requirements From HTA Bodies in Germany Ruof J 1, Staab T R 1, Slawik L 2, Orben T 3, Bosch S 4 1Roche Pharma, Grenzach-Wyhlen, Germany, 2Bristol-Myers Squibb GmbH und Co. KGaA, Munich, Germany, 3vfa, Berlin, Germany, 4Bayer Business Service GmbH, Leverkusen, Germany . . . . . . Objectives: Regulatory authorities such as the European Medicines Agency (EMA) can make marketing authorisation contingent upon post-authorisation measures (PAMs) so as to fill in information gaps in efficacy and safety. PAMs are generally formulated in agreement with manufacturers, and evaluate clinical hypotheses in an ethical and practical way. In Germany, novel medicines must also undergo an early benefit assessment (EBA) by the Federal Joint Committee (G-BA) following marketing authorisation. G-BA may demand additional evidence in order to formulate an opinion on added therapeutic value, which then leads to determination of reimbursement. We compared selected PAMs with the corresponding G-BA demands to see if they were similar. Methods: Medicines that received a restricted EBA from G-BA before 15 June 2014 were evaluated and compared with their marketing authorisations by EMA. PAMs from EMA, and EBA restrictions from G-BA, were assessed in terms of their required additional evidence. Results: Twenty-eight percent of all 79 medicines assessed by G-BA received a restricted EBA. Only nine of those had obligations for PAMs. Four of these were conditional approvals or approval under exceptional circumstances, while five received unconditional marketing authorisation. G-BA justified restricted EBAs for the four conditional approvals based upon agreement with the EMA opinion. For the five unconditional approvals, G-BA required considerably more information than EMA. The additional evidence requested by the two bodies rarely corresponded to one another. EBA restrictions were more influenced by transferability to the German health care context, choice of subgroups and appropriate comparator, than were the corresponding EMA PAMs. Conclusions: G-BA often demands more evidence than specified in EMA PAMs from medicines granted unconditional approval. Although PAMs are discussed and agreed between EMA and manufacturers, G-BA demands and restrictions are not. The possibility for such discussions with G-BA would be an improvement for the future. PHP140 Standard Costs for Health Economic Evaluations: An International Comparison Mostardt S , Sandmann F G , Seidl A , Zhou M , Gerber-Grote A U Institute for Quality and Efficiency in Health Care (IQWiG), Cologne, Germany . . . . . . . Objectives: Country-specific lists of standard costs can reduce the variability of results in health economic evaluations that are attributed to differences in employed data sources and approaches for defining the resource use and unit prices. Moreover they potentially speed up the conduct of health economic evaluations. We aim to investigate which HTA agency officially recognizes and applies a standard cost list and, where such list exists, explore pre-specified procedural and methodological aspects. Methods: Reviewing all national pharmacoeconomic guidelines published on the ISPOR Website http://www.ispor.org/PEguidelines/index.asp in English (i.e., 30 out of 37). Standard cost lists mentioned in the guidelines were, inter alia, compared on the following aspects: i) objective, ii) authorship, iii) release interval, iv) data sources, v) costing perspective, vi) cost categories, and vii) health-state costing. Results: Out of the 30 pharmacoeconomic guidelines available in English, a standard cost list was officially recognized and applied by 4 HTA agencies (Canada, England, Australia and the Netherlands). All 4 lists aim to reduce heterogeneity between health economic evaluations in order to increase the comparability. Compiling the standard cost lists was commissioned to external scientific institutions in all 4 countries. Updates of the lists have been published periodically, spanning from anually (e.g. England) to when required for methodological reasons or to ensure currency (e.g. the Netherlands). Data collection was primarily based on claims data and/or official statistics; in the Netherlands, published research and expert opinions can be employed (for estimating resource use). Costs were derived from a payer perspective. Both direct and indirect costs were stated, except for Australia (only direct costs). No country presented costs in relation to health states. Conclusions: Standard cost lists are mentioned in 13% of the pharmacoeconomic guidelines available in English. The 4 lists conicide on many procedural and methodological aspects. Heterogeneity arises mainly from country-specific costs. PHP141 Regulatory Approval To Patient Access, An Evaluation of Eu5 And Us National Timing Differences: An Update Mycka J 1, Dellamano R 2, Lobb W 1, Dalal N 1, Pereira E 1, Dellamano L 2, Sagaydachnaya O 1 1Medical Marketing Economics, LLC, Montclair, NJ, USA, 2ValueVector, Milan, Italy . . . . . . . Objectives: Examine the time between regulatory approval and launch/pricing and reimbursement (P&R) approval in the US and EU5 countries. Methods: Examined new molecular entities, formulations and combinations with EMA approval between Jan 2009 and May 2014. Additional analysis of products launched between April 2013 and May 2014. Time comparison for general medicines vs. orphan/oncology indications was made including shifts over time. Data gathered from: USA: FDA approval date, wholesale; acquisition cost effective date; UK/Germany: Product availability/ introduction; France: P&R decision (Agrément collectivités/date published in Journal Officiel); Italy: First Official Gazette P&R Decree publication; and Spain: Date of commercialization. Results: Overall, for launches between Jan 2009 and May 2014, the average time from FDA approval to US launch was 6 weeks (oncology 4 weeks; orphan drugs 2 weeks). Across the EU5, Germany remains the fastest to market. Analysis of new products launched between April 2013 and May 2014 shows time to access in Spain has increased vs. the previous 5 years (75 vs. 54 weeks). Limited numbers of orphan drugs have completed full Spanish P&R process with 106 weeks to launch far exceeding all drugs (75 weeks). Italian average time to complete P&R is 69 weeks, while average time to be listed in class C-nn, without national reimbursement, is only 18 weeks. UK average time to oncology launch appears short (16 vs. 20 weeks), however HTA assessments often mean significant access delays. French orphan drugs assessment is faster vs. all drugs (46 vs. 50 weeks). Conclusions: The time to P&R post-regulatory approval increased ~3 weeks in Italy and ~21 weeks in Spain for products launched in the latest year vs. all drugs with EMA approvals between Jan 2009 and Dec 2013. A428 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PHP142 Impact of Chronic Diseases on Health Care Expenditures: A Multivariate Linear Model From Pharmaceutical Reimbursement Data Karakaya G , Van Tielen R , Vanrillaer V , Umbach I MLOZ, Brussels, Belgium . . . . Objectives: The aim of the study is to analyze the burden of chronic diseases on public health care expenditures using pharmaceutical data regarding about 2 million individuals. Methods: Data come from the administrative database of the third Belgian health insurance funds. Without explicit diagnoses of diseases in the database, chronic diseases are mainly estimated using drugs prescription (reimbursed medications only) for a treatment of at least 90 days/year. A multivariate linear model based on the OLS method is used to analyse the impact of 23 chronic diseases on health care expenditures, while controlling for age, sex, marital and social status, share of hospital expenditures, residential areas and year of death. Results: Monthly average health care costs for people with one or more chronic conditions is 6 times greater than the ones without any chronic conditions (€ 423 vs. € 71). All chronic diseases (except psoriasis) have a significant impact on health care (at 1% level). The last months of life, developing or living with cancer, chronic renal failure, rare disease and mental disorder are the factors having the greatest impact on monthly reimbursed health care expenditures. All things being equal, a person at the end of life costs more than € 2,236 per month to the health insurance compared to a person not at the end of life. Respectively this amounts to € 3,557, € 3,008, € 2,042 and € 1,151 for people living with a rare disease, chronic renal failure, mental disorder and cancer. Conclusions: Results found in this study are similar to those observed in other countries. Our findings show in particular that chronic diseases significantly drive health care expenditures. Using secondary data allows to classify chronic diseases according to the financial weights while controlling for characteristics of the analysed population. Results highlight the high financial burden of chronic diseases for public health care expenditures and allow decision-makers to take appropriate public health measures. PHP143 The Effect of China’s Basic Medical Insurance Schemes on Health Service Utilization Lee C 1, Sun H 2, Guan Q 3, Wasserman M 3 1Double Helix Consulting, London, UK, 2Nanyang Technological University, Singapore, 3Double Helix Consulting, Singapore . . . . Objectives: China’s medical insurance coverage surpassed 95% nationwide by 2011 under three basic medical insurance schemes UEBMI, URBMI and NCMS, partly due to the government’s heavy investment since 2009. It is still inconclusive whether the increased coverage rate has improved access to care. Past studies using the China Health and Nutrition Survey (CHNS) focused on data before 2009. This study investigates effects of the three different insurance schemes on health services utilization after the 2009 new health reform. Methods: An analysis was conducted on 2009 and 2011 data from the CHNS (sampling totally 23202 people from 9 provinces). To control for confounding factors, propensity score matching models were developed controlling for relevant factors such as age, gender, income, educational level and health status. The level difference of health service utilization was compared between each insured and the uninsured group. Attempts were also made to compare average treatment costs per episode; however the data set is incomplete for statistical analysis. Results: In the matched samples, the UEBMI group on average used 2.7% less outpatient services (p< 0.05) but 1.2% more inpatient services than the uninsured group (p< 0.05). By contrast, the URBMI group was more likely to pay both outpatient and inpatient visits than the uninsured, although both improvements were not statistically significant (p> 0.05). People in the NCMS group were 5.6% less likely to use outpatient services (p< 0.05); they also tended to use inpatient services less (p> 0.05). Conclusions: Access to care across the varying schemes is not equal. Differences between the UE/URBMI and the uninsured may be resulted from different health statuses, but also they could reflect the tendency among the employed to delay care seeking. The relative underutilization of care by the rural population points again to high co-payment requirement. More research is necessary to understand the interplay of care infrastructure and individual care utilization in China. PHP144 Price Negotiation For Pharmaceuticals in Germany: High Influence Of Eu Price Weighting Method Neubauer A S , Voss P, Minartz C , Gmeiner A , Neubauer G IfG Institute for Health Economics (www.ifg-muenchen.com), München, Germany . . . . . Objectives: To investigate quantitatively, which influence the chosen EU price weighting method has in a theoretical framework model of pharmaceuticals price negotiations. Methods: Three components theoretically determine the reimbursed price of a pharmaceutical under AMNOG conditions in Germany: 1) the level and certainty of added benefit, 2) prices of comparable pharmaceuticals in Germany and 3) the weighted EU price level (post rebates). In a theoretical model the influence of various weighting methods of EU price levels on resulting reimbursement was investigated. Results: Several weighting models for EU prices are theoretically feasible. When defining the 15 EU basket countries the German arbitration board had considered a weighting based on volume, approximated by country population size and an adjustment by purchasing power parity. Operationalisation of each of those factors showed a significant influence on results in our simulation: in several example cases a population and PPP health basket (Eurostat) was set as base case. Using a weighting based on GDP (in PPP) instead, resulted in a ~10-15% increase and based on Eurostat general goods basket (in PPP) in a ~20-25% increase in the calculated average EU price level. Actual observed EU drug volumes appeared to rather support such cases than the weighting based on the Eurostat PPP health basket. Which currency exchange rate was used, e.g. yearly vs. daily, Eurostat vs. other sources, had only a minor influence in most constellations. The launched country basket showed an overall high impact, with a frequently high importance of the U. K. as an EU country being launched relatively early. Conclusions: The chosen weighting method of EU prices has a high impact on results of EU average price, which is an important component of German price negotiations. No undisputed method exists at the time of market entry. PHP145 A Payers Perspective to Pharmaceutical Market Access: Defining Market Access Sendyona S Creativ-Ceutical, London, UK . Objectives: The payer’s perspective is considered after a drug has been granted market access and launched. This is often too late as payers may not want to reimburse expensive drugs, especially if the current gold standard is cost effective compared to alternatives. This is mostly encountered if the drug was, during the clinical trials, compared to sub-standard comparators. This research aims to establish whether there are interventions to consider during a drug development cycle, and how, by using Key Intervention Points, a drug can be successful in the market. The first in a series of questions aims to define what exactly market access is. Methods: A literature review was conducted on Ovid MEDLINE to establish a whether a clear and internationally validated definition of market access has been proposed. Included were articles and/or reviews concerning market access of pharmaceuticals for human use published from January 1999 to present day. Results: Market access can be thought of as either gaining regulatory approval from bodies such as Food and Drug Administration or European Medicines Agency to make a drug available to patients, or as developing a drug that achieves blockbuster status through successful reimbursement, either through a high number of sales, or a higher-than-competitor premium price. Regulatory approval and reimbursement are often thought of as being inherently different, yet both share the central principle of balancing the benefits and harms in deciding availability of drugs. The main difference between regulatory approval and reimbursement is the scope of benefits and harm, and the population they consider. Conclusions: Market access is difficult to define, with different opinions and perspectives. There currently lacks a clear, internationally validated method of defining exactly what it market access entails, and what it means to have successfully achieved it. There is a need to definitively define this important concept. PHP146 Market Access Pathway for Medical Nutrition in Europe and The Us Chevrou Séverac H 1, Droeschel D 2, Walzer S 2 1Takeda Pharmaceuticals International, Zurich, Switzerland, 2MArS Market Access & Pricing Strategy GmbH, Weil am Rhein, Germany . . . Objectives: This analysis aims at presenting different market access pathways of Medical Nutrition (MN) products in the US, France, Germany and the UK. Methods: Systematic review of submission processes for MN and food for special medical purpose (FSMP), combined with experience of the authors from previous research on health economics, Market Access and reimbursement. Results: When considering MN delivered in the ambulatory care setting, only in France, innovative MN presenting with therapeutic value faces the medical device reimbursement process. In the UK, the process is handled by the ACBS and focus mainly on clinical outcome and safety; additionally this process sets a reimbursed price. In the US and Germany, there are reimbursed categories for MN linked to composition of the product and dedicated to patients with inability to have their nutritional needs covered by normal food intake (set by CMS in the US and G-BA in Germany). Creating new reimbursed categories linked to new MN either bringing innovative therapeutic value or targeting new disease area is highly difficult in all countries. For MN delivered in hospital settings, products delivered enterally or orally are mainly covered by hospital budget. The budget can be either from the hospital’s kitchen for thickened and thickening products, under the diagnosis-related group funding scheme related to each countries, by the nutritionist budget (mainly UK and US) or by the hospital’s pharmacy budget for specialties. For standard products, access is obtained based on tenders. In long-term care and nursing home, coverage and funding are more heterogeneous: they vary from highly regulated reimbursed scheme based on composition of the products and disease area to per diem fee per patient covering both food and MN. Conclusions: The market access pathways for granting reimbursement or coverage of the medical nutrition category are very heterogeneous between the analyzed countries. PHP147 The Reimbursement of Expensive Drugs In Hospitals in Western European Countries Falk K , Brown A , Murray G , Martensson M Abacus International, Manchester, UK . . . . Objectives: Over the past decade increasing numbers of expensive drugs have entered the market, placing a financial burden on hospitals in particular. Many high-cost therapies require use or at least initiation in the hospital setting. Many European countries reimburse hospitals via diagnosis-related group (DRG) systems. However, DRG systems are often insufficient to cover the costs of expensive drugs. Here we assess how expensive drugs are funded in hospitals across markets, what the requirements are, and the process of additional funds being granted. Methods: Publicly available documents, governmental guidelines and regulations were assessed to understand the different processes and requirements expensive drugs need to meet to receive additional funding. Countries included were the UK, France, Germany, Italy, Spain, the Netherlands, Sweden, Denmark, Switzerland, Austria and Portugal. Results: In all included countries, hospital treatments are reimbursed via DRG systems, and most provide additional funding for expensive drugs. There A429 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 is overlap in the criteria to qualify for funding. Many countries consider drugs for additional funding when a drug can be used in more than one indication and cannot be grouped to a specific DRG. Other criteria identified relate to, for example, drug prices and indications. Some countries grant additional reimbursement for drugs prior to assessment by a national reimbursement process, while others only grant additional reimbursement after the drug has been available for a certain period of time and funding decisions are based on historical data. In most countries, additional reimbursement is considered annually. Hospitals and expert groups can suggest additional reimbursement for expensive drugs to the responsible authority. Conclusions: Many countries have adapted to the need for additional funding for expensive drugs, and have established systems to grant this funding to hospitals. There are differences in criteria to qualify for, and timelines for receiving, additional funding after drugs are launched. PHP148 Pay-For-Performance: Balancing Cost and Care Ziskind M A 1, Pierce C A 2 1Johnson & Johnson Healthcare Systems Inc., Horsham, PA, USA, 2The Resource Group, Richfield, OH, USA . . . . Objectives: Initiatives aimed at improving both the quality and efficiency of United States health care are commonly grouped under the broad category of “pay-forperformance” (P4P) programs. Typically these programs award bonuses to providers that attain pre-determined quality and cost goals, but may also impose financial penalties on those that fail to meet those goals. Fueled by the Affordable Care Act, P4P programs have recently expanded significantly within the public sector and are expected to grow. This project was designed to review Medicare P4P cost measures, discuss the implications for provider prescribing patterns, and recommend possible alternatives. Methods: Two P4P programs, both well known under the health reform law and having potential to impact a large portion of the Medicare population, were evaluated: 1) Accountable Care Organizations (ACO), and 2) the Physician Value Based Payment Modifier (VBPM). Each program’s cost measure components and calculation methodologies were isolated, described and evaluated for the potential to influence prescribing patterns. Results: Cost measures for the ACO and VBPM programs are based on payments made under Medicare Part A (Hospital Insurance) and Medicare Part B (Supplemental Insurance) but do not include Medicare Part D (Prescription Drug Benefit) costs. Whether performance is measured against the provider’s historic costs, or compared to national benchmarks, only Medicare Parts A and B costs are included. Conclusions: Medicare Part D costs are not included in the cost measure calculation, thereby eliminating prescription drug expense from the performance rating. This methodology may encourage providers to shift Part B drug costs to Part D, thus limiting patient access to therapies that may only be covered under Medicare Part B. PHP149 Review of Pricing And Reimbursement Systems in South-Eastern Europe Zibelnik N 1, Odeyemi I A 2 1Astellas Pharma South-East Europe, Ljubjana, Slovenia, 2Astellas Pharma Europe Ltd, Chertsey, UK . . . Objectives: To provide an up-to-date description and comparative analysis of pricing and reimbursement policies in South-Eastern Europe (SEE), and to identify factors influencing reimbursement decisions. Methods: Payers and decision makers in Slovenia, Croatia, Romania and Bulgaria were interviewed by questionnaire. An additional literature survey covered country-specific legislation and publications (PubMed 2009–2014), and relevant documents from web sources including national hospital insurance funds, drug agencies, ministries of health, Eurostat, pharmacoeconomic and outcomes research conference proceedings, the Organisation for Economic Co-operation and Development, and Business Monitor International. Results: The four countries spent 5.1%–8.8% of gross domestic product on health in 2012. Price controls are commonly used, applied via negotiation with marketing authorisation holders or indirectly through the application of copayments. Key policies are based on international and internal reference pricing. Bulgaria, Romania and Slovenia base pricing on the lowest manufacturing or retail price, while Croatia uses average pricing. Reimbursement requires demonstration of clinical and economical benefits over current standard therapy. When the importance of benefits are recognised and recommendations from reference health technology assessment authorities in Europe are published, budget impact analyses focusing on key drivers such as target population and price are applied. Cost containment mechanisms are also applied, including net price negotiation, rationing, decision postponement, payback, clawback and risk sharing, with the aim of protecting overall budgets. Legislative changes to pricing and reimbursement systems are very common. Out-of-pocket expenses in Romania and Bulgaria are among the highest in Europe, while the reimbursed: retail price difference is largely covered by supplementary insurance in Slovenia and Croatia. Conclusions: The middle/lower income SEE countries use reference pricing, and have some of the lowest prices in the European Union. Reimbursement of innovative drugs is restricted, there are downward trends in pricing, and risk-sharing agreements based on outcomes are finance-driven. PHP150 Pricing and Reimbursement Environment For A Biologic Obtaining A Licence In A Second Indication in Key European Countries Ling C S 1, Balp M M 2, Bjoerk B 2, Quijano M 2, Heyes A E 1 1RTI Health Solutions, Manchester, UK, 2Novartis Pharma AG, Basel, Switzerland . . . . . . . . Objectives: To gain a better understanding of the pricing and reimbursement processes and evidence requirements at national, regional, and local levels with regards to a biologic obtaining a license in a second indication in the UK, France, Italy, and Spain. Methods: In countries with a largely national system (UK, France), five telephone interviews were conducted; as Spain and Italy also have local and regional systems, 17 and 18 interviews, respectively, were conducted. Stakeholders included payer-advising clinicians, hospital administrators and pharmacists, regional payers, and local payers. Results: In the UK and France, pricing and reimbursement is agreed at a national level, with few restrictions at regional and local levels. In the UK, NHS England is likely to be responsible for funding of new biologics and relies on guidance from NICE before adopting a product in a new indication; therefore, demonstrating cost-effectiveness is key. In France, the ASMR issued by the transparency commission is important; funding usually is through the groupe homogéne de séjour. In addition to demonstrating clinical benefits, pharmacoeconomic studies may be required for high-cost drugs. In contrast, in Italy, although AIFA needs to approve a new product on a national level, subsequent requirements vary by region and sometimes specific location. Similarly, in Spain, once approved nationally, regions develop their own recommendations, and local decisions are made by hospital formularies. Evidence requirements in Italy and Spain vary at national, regional, and local levels. Conclusions: A biologic obtaining a license in a new indication must undergo the same procedure as a new product. The process and restrictions for biologics may be stricter than for other medications due to the perceived high cost. The level of national, regional, and local requirements and restrictions varies; it is important that appropriate evidence is submitted to decision makers at each level. PHP151 Cost and Quantity Characteristics of Medical Devices In Slovakia Malovecká I 1, Minariková D 1, Lehocká L 1, Foltan V 2 University, Bratislava, Slovak Republic, 2Faculty of Pharmacy, Comenius University, Bratislava, Slovak Republic . . . . 1Comenius Objectives: Medical devices, together with pharmacotherapy are supportive treatment of many acute and chronic diseases. The place of their dispensing and direct sale is the dispenser of medical devices. Many medical devices are reimbursed from public health insurance funds entirely, for others, particularly advanced functional types of medical devices, there must the patient participate on the price or he can buy them according own decision. Methods: The target of the work was to analyse the data from paid databases of Slovak authority National Center for Health Information that collects the outputs of provided health care services. The most recent data were from 1.1. - 30.9.2014. Results: Referring to the Center for Health Information until 30.9.2013 there were 226 registered establishments that sell medical devices. Their specialization were dispenser of medical devices (n= 163), dispenser of orthopedic devices (n=48) and dispenser of audioprotetic devices (n=15). Expenditure of reimbursed medical devices from public health insurance funds amounted n-monthly packaging= 8,8mil and n-monthly value= 12,7 mil €. The highest shares had the group medical devices for incontinence and urinary retention (n-packages= 7,7 mil, n-packages%= 86.9, n-values= 3,9 mil € , n-value%= 30.9), the group plasters and bandaging materials (n-packages=0,5 mil, n-packages%=6.1, n-values=2,6mil €, n-value%=10.2) and medical devices for ostomates (n-packages=0,4 mil, n-packages%=4.6, n-values=1,6 mil €, n-value%=8.1). Direct sale of medical devices to the patients reached n-monthly packaging= 0, 9 mil and n-monthly value=1,1mil €. The highest shares had the group medical devices for incontinence and urinary retention (n-packages=5,7 mil, n-packages%=50.3, n-values= 2, 6 mil €, n-value%=26.2), the group plasters and bandaging materials (n-packages=3,5 mil, n-packages%= 30.9, n-values= 2,0 mil € , n-value%= 20.4) and medical devices for diabetics (n-packages= 0,5 mil, n-packages%= 4.5, n-values= 0,7 mil € , n-value%= 7.3). Conclusions: Medical devices are reimbursed frompublic health insurance funds or paid by patient and their proportion constitutes 7: 1 in packages and 11: 1 in Eur. PHP152 The Greek Health Care Reform After Troika’s Involvement: The Potential Impact on Global Pricing and Access Strategy Anastasaki E 1, Bradshaw S 1, Shah S 2 Solutions LLC, London, UK, 2Market Access Solutions LLC, Raritan, NJ, USA . . . 1Market Access Objectives: Troika’s measures to support Greece’s financial recovery have targeted all sectors of economy including health care. Since 2012, policy reforms have changed the way health care is funded, managed and delivered, and how pharmaceuticals are priced, accessed and reimbursed. This study examines the changes to the Greek system and tries to understand the wider possible impact on global pricing and access strategies. Methods: To better understand the recent reforms we conducted a literature review of public domain sources, including the Greek Government Gazette, PubMed and other websites. Searches were conducted in English and Greek-language, and materials were translated into English. From our findings a road map diagram was developed, and this was validated by interviews with health policy experts. Results: Part of troika’s campaign to reduce public spending has seen the Greek government focus on pharmaceutical markets and introduce policies to contain costs. The drug budget for 2014 has been cut to 2 billion euros, a billion lower than 2013. Considerable price cuts have been agreed on both novel and generic agents on top of clawbacks and rebates for high cost drugs. Prescribing is controlled through electronic prescription and physician budget caps. Introduction of price-volume agreements and risk sharing schemes are being considered, however the infrastructure to support implementation is still under development. Demonstrating value by health economics and outcomes research can still help manufacturers to achieve premiums. Conclusions: With a small population and an ever-decreasing expenditure on health care, it is tempting to overlook Greece when developing a product launch strategy. However, with Greek drug prices being referenced by several EU and non-EU countries, ignoring Greece may no longer be an option, especially when considering the indirect effect on the big EU5 prices. Understanding the reforms and assessing the impact on launch sequencing will be key in developing optimal pricing strategies. PHP153 Swisshta Recommendation as an Optimal approach for Pragmatic HTA Evaluations? An International Comparison Walzer S , Droeschel D MArS Market Access & Pricing Strategy GmbH, Weil am Rhein, Germany . . A430 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Switzerland’s regulation of prices for reimbursed drugs is based on referencing across countries and within the therapeutic class for products with comparators. The SwissHTA initiative involving all key stakeholders in the health care systems (sickness funds, industry, physicians, academia, Kantons) has published consensus papers for new benefit criteria and measurements. Methods: A comparison was executed comparing the new proposed criteria against benefit assessments in HTA systems in Germany and the UK. Results: In terms of clinical benefit assessment the suggestion by SwissHTA follows accepted evidence-based methods. In comparison to Germany the Swiss approach suggests a pragmatic application by applying disease specific standards. This disease focus allows also accepting different levels of evidence given the characteristics of the disease. This pragmatic approach allows Swiss decision-makers accepting lower evidence levels at the time of launch (e.g. in case of comparison with non-Swiss standard-of-care) coupled with a post-reimbursement commitment. The Swiss method looks similar to the medical benefit application by NICE. In terms of health economic (HE) evaluations SwissHTA suggests focusing on technical efficiency instead of QALY comparisons across the whole system as in the UK. Such an approach avoids the application of arbitrarily defined cost-effectiveness thresholds. In Germany the HE focus is solely based on cost comparisons. In terms of decision-making in Germany the focus is based on an assessment of the available evidence against a theoretical maximum standard of evidence. In the UK coverage decisions are based on cost-effectiveness assessments allowing for context-specific adjustments. In the SwissHTA recommendation a multi-criteria decision-making should be applied with an equal focus on all key aspects (e.g. clinical benefit, public relevance, social preferences, etc.). Conclusions: In comparison to HTA systems in Germany and UK the SwissHTA recommendations seems to be more pragmatic and would follow a broader multi-criteria decision making approach. PHP154 Product Quality Aspect in Reimbursement of Medical Devices: Comparison of Turkey VersUS Europe Seyhun O 1, Erdogan E 1, Can H 1, Erdol S 1, Guler I 2, Bayazit A 2 1Medtronic, Inc., Istanbul, Turkey, 2Turkish association of social security experts, Ankara, Turkey . . . . . . Objectives: FDA has long recognized that dramatic increase in adverse event reports due to medical devices and recalls may reflect quality flaws. While some of this increase can be explicated by FDA’s greater outreach emphasizing reporting requirements, failures in product design and manufacturing process cause more than half of all product recalls. Therefore, FDA’s concern regarding low quality products remains. In the EU, medical device pre-market quality is assured by CE mark authorization. This regulation is the prerequisite for market registration also for Turkey. However, due to heterogeneity and complexity of devices, manufacturers, imported devices and multiple use environments, there is strong need for postmarket quality assurance. Methods: This study investigates whether post-market quality assurance (measured by less adverse events/better health outcomes) can be applied through local reimbursement policies. First, it is investigated whether there are reimbursement rules in Europe acting as post-market quality assurance. Then, a comparison is made with Turkey’s existing reimbursement scheme. Results: Our comparative analysis reveals only Belgium and France implement quality or brand based reimbursement rules. In Turkey, there is no quality based reimbursement scheme; however current reimbursement application guideline requirements may act as a gate keeper for lower quality products. Our Results show in addition to pre-market regulations, post-market quality can be assured by local reimbursement authorities. Conclusions: There are several opportunities to improve quality assurance and reduce risk across medical device industry; i.e. enhancing visibility of comparative quality to harness market forces and increasing the collaboration between stakeholders. From health policy perspective, implementation of new value based reimbursement models require providers to prove that they’re meeting quality standards and benefitting patients while cutting costs. Therefore, while value based payment contracts are still in their infancy in Europe and Turkey, they will have a direct impact on the assurance of continued medical device quality. PHP155 A Comparison of Factors Influencing Reimbursement and Coverage Decisions in Scotland (Smc), The Netherlands (Nzi) and Germany (G-Ba) Charokopou M 1, Alleman C J M 1, Verleger K 2, Spoorendonk J A 1, Schmidt R 2, Schoeman O 2, Heeg B 1 1Pharmerit International, Rotterdam, The Netherlands, 2Pharmerit International, Berlin, Germany . . . . . . . . . . Objectives: In Germany, Scotland and the Netherlands, the manufacturer’s submission is assessed by the HTA bodies; G-BA, SMC and NZi. In Germany, the submitted evidence is used to assess the drug’s additional benefit, followed by price-rebate negotiations with the GKV-Spitzenverband. In Scotland and the Netherlands, the submitted evidence is evaluated for reimbursement decision. This study aims to compare factors that influence the reimbursement recommendation by SMC and NZi, the additional benefit by G-BA and the rebate by GKV-Spitzenverband. Methods: Three databases were created consisting of 463 SMC applications, 262 NZi evaluations and 68 G-BA decisions. Logistic regression analyses were conducted to assess the impact of the submitted evidence on the recommendation by SMC and NZi and the effect of variables on the additional therapeutic benefit by G-BA. The impact of variables on the rebate was examined through linear regression analysis. Results: In Scotland, 57% of the applications received positive recommendation and the NZi recommended 83% of the submissions. In Germany, 60.3% of the products demonstrated an additional benefit. In Scotland, the multivariate analyses showed that the performance of a cost-minimization analysis and beneficial cost-effectiveness outcomes were the strongest positive predictors of the recommendation. In the Netherlands, univariate analyses showed that the decision was significantly affected by whether the product under assessment was a life-saving intervention and the inclusion of (positive) economic evidence. In Germany, univariate analyses demonstrated that the therapeutic indication and the overall survival benefit, along with improved morbidity and adverse events meaningfully influenced the benefit assessment. Analysis showed that the rebate was significantly reduced by 13% for products that demonstrated additional benefit. Conclusions: Even though reimbursement submission requirements of Scotland and the Netherlands look similar, SMC weights the cost-effectiveness outcomes more, while NZi focuses on the variables related to additional clinical benefit; variables that also significantly influence G-BA’s decision. PHP156 A Comparison of Additional Benefit Scores in Germany (Gba) and France (Has) Soussi I 1, Thivolet M 2, Kornfeld A 2, Brunet J 3, Toumi M 4 1Creativ-Ceutical, Tunis, Tunisia, 2Creativ-Ceutical, Paris, France, 3Assistance Publique des Hôpitaux de Marseille, Marseille, France, 4University Aix-Marseille, Marseille, France . . . . . Objectives: The Pharmaceutical Market Restructuring Act (AMNOG) has brought a sustainable change to the reimbursement of new drugs in Germany. The G-BA assesses the additional benefit of the drug, compared to an appropriate therapy. AMNOG law is perceived to be one of the toughest drug evaluation process in Europe. In France the high authority for health (HAS) assesses the level of improvement of actual benefit (IAB). The objective of this study was to compare the additional benefit score issued under AMNOG law to IAB scores granted by the HAS. Methods: All G-BA’s additional benefit scores until June 1st 2014 and HAS IAB score were compared. Results: In Germany, a total of 76 completed early benefit assessments. From the best available score perspective, the G-BA assessed the additional benefit as considerable in 20% of drugs assessed (score 2), as minor in 30% of drugs assessed (score 3), as unquantifiable in 8% of drugs assessed (Score 4) and as none in 38% of drugs assessed (Score 5). No drug has been granted a major additional benefit (score 1) and 4% of drugs were directly allocated to a reference price group. In France, the transparency committee granted a major improvement in 0.2% of cases (IAB I), an important improvement in 1.3% of cases (IAB II), a moderate improvement 2.5% of cases (IAB III), a minor improvement in 9,2% of cases (IAB 4) and no clinical improvement in 86.8% of cases (IAB V). Conclusions: This study shows that the G-BA assigned an additional benefit (scores from 1 to 4) to more than half of drugs whereas the HAS granted an additional benefice rating to less than 14% of case. This study suggests that there is a more favourable benefit rating in Germany than in France. PHP157 Hta Status of Biosimlars Across the UK and Ireland Mildred M , Davies K Boehringer Ingelheim Ltd, Bracknell, UK . . Objectives: Biosimilars have the potential to revolutionise the health care landscape by realising cost savings over originator biologics and thus increasing access to innovative medicines. The biosimilars marketplace in the UK and Ireland is relatively new, however the landscape is rapidly developing. The objective of this analysis was to map the HTA status of biosimilars in the UK and Ireland to provide insight for stakeholders involved in the assessment of new biosimilars. Methods: The HTA status of all EMA authorised biosimilars was identified by searching the websites of all four HTA agencies in the UK and Ireland, namely, NICE, the SMC, the AWMSG, and the NCPE. All previously assessed medicines and on-going technology appraisals were screened for the inclusion of biosimilars using the non-proprietary (common name) and proprietary (brand) names. Results: Sixteen (84%) of the nineteen biosimilars submitted to the EMA have been authorised, eleven of which (69%) have been considered by HTA agencies. The SMC has approved 100% of the biosimilars it has considered (n=7); the largest positive reimbursement rate amongst all HTA agencies considered. The AWMSG has considered the largest number of biosimilars (n= 11), of which five, (45%) received a positive reimbursement status. Both NICE and the NCPE have approved one biosimilar, however three additional biosimilars are currently being considered by NICE. Conclusions: The reimbursement status of biosimilars in the UK and Ireland is not consistent across HTA agencies. The timing of HTA submissions to different HTA agencies may play an important factor in the reimbursement status of biosimilars given that this landscape is relatively new and assessment processes vary. Marketing authorisation holders for biosimilars may want to consider the strategic importance of submitting evidence to each of the HTA agencies in the UK and Ireland, and the impact timing may have on the uptake of their biosimilar. PHP158 Does Not Reaching an Agreement on the Final Nice Scope Have Any Impact on The Final Apraisal Outcome? Casamayor M 1, Heemstra L 2, Van Engen A 2 1Quintiles Consulting, Barcelona, Spain, 2Quintiles Consulting, Hoofddorp, The Netherlands . . . Objectives: Identifying the right patient population, comparator and endpoints is key to increase the likelihood of reimbursement. Manufacturers do not always agree with payers’ views on these items. Disagreement may lead to funding rejection. We assessed the rate of mismatches between manufacturers and NICE and their impact on the final appraisal outcome. Methods: All manufacturer submissions (MS) from January 2011 until June 2014 were reviewed. For these submissions, the initial proposed scope, the manufacturer’s comments, and the final scope and appraisal outcome were analysed. All changes to the initial scope suggested by the manufacturer were recorded and their impact on final outcome investigated. Results: In the time period reviewed there were 101 MS of which 7 were suspended and not included in our analysis, while comments were not available for another 18. Manufacturer comments are published for 76 MS. The manufacturer disagreed on ≥ 1 section of initial scope in 93% (71/76) of MS. The areas where manufacturers and NICE disagreed most commonly are the comparator(s) (43/71; 61%) and population (40/71; 56%) to be assessed. The final scope implemented all and some of the manufacturer’s comments in 56% (40/71) and 28% (29/71) of submissions, respectively. A431 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Rejection was more common for manufacturer’s comments on outcomes (6/8; 75%) and comparators (8/13; 61.5%). Rate of final recommendation by NICE was higher for those MS where all (29/40; 74%) or certain changes (14/20; 74%) requested by the manufacturer were implemented in the final scope than for those where NICE rejected all manufacturer requests (7/11; 64%), and similar to overall recommendation rate (66/91; 73%). Conclusions: These data highlight that the initial scope frequently does not meet manufacturer’s expectations. However, manufacturer’s suggestions are often incorporated in the final scope. NICE not implementing manufacturer’s suggestions to the final scope does not decrease the likelihood of being granted funding. PHP159 An Examination of the Regulatory and Reimbursement Processes for Biobetters and Comparison With Biosimilars de Silva S U , Dimova M , Bending M W Mapi, London, UK . . . . . Objectives: Biosimilars and biobetters are subsequent versions of licensed innovator biotherapeutics. Whereas biosimilars are comparable to the originator product in terms of quality, safety and efficacy, biobetters incorporate intentional modifications to the originator molecular profile with the aim of producing a superior product. This distinction between biosimilars and biobetters has important implications from a regulatory perspective, with biosimilars following class-specific guidance whereas biobetters are considered innovator drugs. This study sought to examine and compare the regulatory and reimbursement approaches to the appraisal of biobetters and biosimilars. Methods: Biobetters and biosimilars of the same product class were identified, and qualitative analyses of the recommendations by indication, evidence considered, and key decision drivers were undertaken using available regulatory and HTA reimbursement decision documentation from six European countries. Results: Findings for filgrastim are presented as an example; 7 biosimilars, and the pegylated filgrastims (pegfilgrastim and lipegfilgrastim) considered biobetters, were identified. Biosimilar filgrastims were granted European marketing authorisation based on demonstration of clinical comparability to the originator filgrastim in one indication and extrapolation of the results to all 5 approved indications. Pegfilgrastim demonstrated clinical non-inferiority to filgrastim in one indication and was approved solely for this indication; the subsequently developed lipegfilgrastim was approved for the same indication but used pegfilgrastim as the comparator. Similar to biosimilar filgrastims, economic evidence in the form of cost-minimisation analyses was considered in HTA recommendations of both pegylated filgrastims. This differs from the approach for certain other biobetters that have demonstrated clinical superiority and cost-effectiveness versus their originator. Conclusions: Biosimilars and biobetters are subject to distinct regulatory processes and the decision driving factors for reimbursement also differ among currently licensed biobetters. With the development of these products gaining momentum, it will be interesting to observe how the appraisal processes evolve to address the scope and variety of emerging biobetters. PHP160 Time Limits Restriction in Germany Thivolet M 1, Soussi I 2, Kornfeld A 1, Toumi M 3 . . . . 1Creativ-Ceutical, Paris, France, 2Creativ-Ceutical, Tunis, Tunisia, 3University Aix-Marseille, Marseille, France Objectives: In Germany, with the introduction of the Pharmaceutical Market Restructuring Act (AMNOG) in January 1st2011, pricing and reimbursement decisions for new drugs have been driven by the early benefit assessment (EBA). G-BA can decide to set or not a time limitation to the decision. The objectives of this study were, first, to review the number of time-limited decisions over time and second, to identify drivers of these decisions. Methods: G-BA’s decisions, from the introduction of AMNOG Law to June 1st2014, were reviewed. Exempted and/or cancelled procedures were excluded. Results: As of June 1st 2014, 76 EBAs were concluded and time limits, from 1 to 5 years, were imposed on 28% (21/76) of these decisions. Short-term restrictions (≤ 2 years) accounted for 52% (11/21) of the timelimited decisions and long-term (> 2 years) for 48% (10/21). Time-limited decisions concerned largely oncology drugs (62%; 13/21), followed by endocrine/metabolic drugs (19%; 4/21) and neurology drugs (10%; 2/21). The number of time limited decisions increased over the studied period, from none (0/2) of the decisions in 2011 to 16% (3/19) in H1 2012, 38% (3/8) in H2 2012, 20% (3/15) in H1 2013, 35% (7/20) in H2 2013 and reaching 42% (5/12) from January 1st to June 1st 2014 decisions. Time-limited decisions were triggered by one or several factors, with safety concerns being the major driver (38%; 8/21). Other drivers were uncertainties of outcomes (33%; 7/21), ongoing studies (33%; 7/21), lack of data (24% (5/21), European Medicine Agency’s (EMA) conditional approval (19%; 4/21), design uncertainty (10%; 2/21), inappropriate comparator (10%; 2/21), quality of life concerns (10%; 2/21), and EMA requirements for post-authorisation studies or risk management plan (10%; 2/21). Conclusions: An increasing trend for time-limited decisions was observed. Time restricted decisions have become a major uncertainty management tool in Germany. PHP161 Reimbursement Trends and Evidence Requirements for UltraOrphan Therapies Across Europe: Optimising Market Access in Increasingly Challenging Markets Morawski J 1, Paul A 2, Ransom J F 3, Spinner D S 4, Doyle J J 5, Faulkner E C 6 1Quintiles, Cambridge, MA, USA, 2Quintiles Consulting, Durham, NC, USA, 3Quintiles Global Consulting, Hawthorne, NY, USA, 4Quintiles, Durham, NC, USA, 5Quintiles, Hawthorne, NY, USA, 6Institute for Pharmacogenomics and Individualized Therapy, Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, NC, USA . . . . . . . . . . Objectives: Ultra-orphan diseases are extremely rare conditions many of which are severe, chronic, and progressive with high mortality rates. There is a growing number of therapies for ultra-rare diseases currently on the market. Reimbursement decisions for these therapies have been characterized by reduced evidence require- ments with unmet need weighing heavily into health technology assessment (HTA) and reimbursement decision-making; as well as a generally wide pricing latitude. To gain insight into evolving market access requirements, we conducted a review of pan-European ultra-orphan therapy HTA requirements and reimbursement decisions. Methods: Applying the National Institute for Health and Care Excellence (NICE) definition for ultra-orphan diseases (prevalence of ≤ 1/50,000), full European HTA reports on ultra-orphan therapies published through May 2014 were identified and reviewed to compare evidence requirements and reimbursement decisions across countries for health economic, clinical, and value based criteria. Results: Over sixty published ultra-orphan HTAs were identified across nine markets. A small portion of these submissions were rejected for reimbursement largely due to lack of evidence on clinical benefit. For therapies recommended with access restrictions, payers often requested additional follow-on studies or ongoing monitoring of patients by manufacturers. With respect to economic evidence evaluation, reimbursement decisions predominately hinged on therapy cost per patient per year, rather than cost-effectiveness. More recent assessments also evaluated quality of life evidence and input from patient groups. Conclusions: As health care budgets become more strained, ultra-orphan therapies priced at a premium have come under increased scrutiny from HTA agencies and payers to demonstrate value for money. In order to achieve optimal market access, manufacturers must consider continually evolving stakeholder evidence requirements and develop clinical and health economic value plans that demonstrate how their ultra-orphan therapies provide health gain instead of disease stabilization. PHP162 Global HTA Assessments of Ultra-Orphan Products: A Case Study Of Eculizumab (Soliris) and Iduronate-2-Sulfatase (Elaprase) Paul A 1, Morawski J 2, Spinner D S 3, Doyle J J 4, Faulkner E C 5, Ransom J F 6 1Quintiles Consulting, Durham, NC, USA, 2Quintiles, Cambridge, MA, USA, 3Quintiles, Durham, NC, USA, 4Quintiles, Hawthorne, NY, USA, 5Institute for Pharmacogenomics and Individualized Therapy, Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, NC, USA, 6Quintiles Global Consulting, Hawthorne, NY, USA . . . . . . . . . . Objectives: Ultra-orphan diseases affect a very small patient population, defined by the National Institute for Health and Care Excellence (NICE) as those diseases with a prevalence of ≤ 1: 50,000. Medicines for these indications are difficult to develop in part due to challenges associated with recruiting for clinical trials from a small patient population. Within this context, global payer bodies have assessed these therapies with modified evidence requirements and opportunity for very high prices. We performed a health technology assessment (HTA) review of two ultra-orphan products – eculizumab/Soliris and iduronate-2-sulfatase (IDS)/ Elaprase – to gain insight into the evolving HTA evidence requirements for ultraorphan medicines and comparatively evaluate key decision drivers across geographies. Methods: We scanned global HTAs published before end of May 2014 to identify the two most widely assessed ultra-orphan therapies that have variable reimbursement decision outcomes (eculizumab/Soliris and IDS/Elaprase). To evaluate pivotal decision drivers, we analyzed HTAs across several criteria, including clinical efficacy, unmet need, strength of evidence, cost-effectiveness and burden of illness. Results: We identified HTAs in seven countries. For both products, reimbursement decisions varied across agencies. Key decision drivers included cost-effectiveness, clinical efficacy, risk-sharing schemes, and lowered evidence requirements/ special criteria for ultra-orphan medicines. Assessments rejecting Soliris and Elaprase (e.g., Australia, Canada, UK) did so based on cost-effectiveness and lack of long-term survival data. Notably, the NICE Highly Specialized Technology Committee requested unprecedented justification of Soliris pricing. Some agencies (e.g, Scottish Medicines Consortium [SMC]) preemptively rejected the products due to manufacturer non-submission of required data. In Australia, Soliris gained recommendation alongside a risk-sharing scheme while Elaprase gained recommendation under Life Saving Drugs Program criteria. Conclusions: Eculizumab and IDS are among a select list of therapies commanding very high prices globally. This study demonstrates variability in decision criteria and approaches across HTA agencies for such high-priced ultra-orphan products. PHP163 Evidence-Based Market Access Value Resource: Navigating The Hurdles for A Biologic Obtaining A License In A Second Indication in Key European Countries Hogue S 1, Bjoerk B 2, Walker A 3, Balp M M 2, Fernandez M 4, Quijano M 2, Ling C S 5, Heyes A E 5 1RTI-Health Solutions, Research Triangle Park, NC, USA, 2Novartis Pharma AG, Basel, Switzerland, 3University of Glasgow, Glasgow, UK, 4RTI Health Solutions, RTP, NC, USA, 5RTI Health Solutions, Manchester, UK . . . . . . . . . . . Objectives: Market access for an innovative technology, such as a biologic obtaining a license in a second indication, can be complex and time consuming. Reimbursement is critical to rapid adoption of and optimal patient access to a new technology. This study aimed to determine the best approach for communicating value and providing field-based staff with value resources to facilitate dialogue with stakeholders in various scenarios. Methods: We conducted desktop research of published literature, health technology assessment reports, clinical trials data, and third-party websites to identify the critical path and data most valuable to reimbursement decision making in order to prepare a communication resource. We conducted a country-affiliate workshop and qualitative one-on-one interviews with payer decision makers in several key markets to understand funding flow and the most appropriate means of communicating value to external decision makers. Results: The process and restrictions for biologics may be stricter than for other medications because of perceived high cost. There are multiple appropriate access pathways for various settings of care, all with varying requirements and value drivers. It is critical to understand the needs of external decision makers and provide field-based staff with a consistent yet customizable means of communicating the value of new technologies. All evidence and insights were synthesized into an evidence-based market access value resource for key stakeholder A432 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 engagement. Conclusions: The evidence-based market access value resource approach provides a clear, concise, and globally integrated value story that will assist in market access and form the basis of consistent communication regarding value at the national, regional, and local level across external stakeholders (e.g., payer decision makers, physicians, patient advocates). Access for a biologics will be complex; regardless of pathway, decisions regarding reimbursement and adoption of a new technology are diverse and dispersed across and within countries, with varying levels of required evidence. PHP164 Medicare Provider Utilization and Payment Data: The Book to Bill Gap Casazza Liberatore C 1, Connelly N 2, Culp J L 2, Doyle J 3 1Quintiles Consulting, Hawthorne, NY, USA, 2Quintiles Consulting, Durham, NC, USA, 3Quintiles Global Consulting, Hawthorne, NY, USA . . . . . The US significantly outspends all other top 10 developed nations, with no increase in life expectancy. Increasing transparency in health care spending could help address this cost to outcomes gap. CMS’ recent release of Medicare Part B Utilization and Payment data, covering 880,000 providers and $77 billion in Medicare payments, is a significant step toward this goal. Objectives: Understand differences in billed versus paid amounts for provider and procedure types reported in Medicare data. Methods: Descriptive and inferential statistics were run on provider specialties representing greater than 2% of claims to describe the differences between maximum allowed Medicare payment amount, amount billed by providers, and the amount reimbursed. Geographical variation was also explored. Results: Amount billed is at least double the amount paid and double the maximum allowable amount for all of the specialties explored; amount billed versus paid varies significantly by specialty, with some specialties billing as much as six times what they are paid. Largest discrepancies were in anesthesiology (on average billing $838, versus $139 allowed and $110 paid). Other specialties with significant disparity include cardiology, diagnostic radiology, emergency medicine, ophthalmology, and orthopedic surgery; within certain specialties, specific procedures showed ranges of billing and payment ratios; across all specialties, average amount paid was about 90% of maximum allowable; and each state is represented in line with census data for the Medicare population; state differences in amount paid versus billed and allowed is not significant. Conclusions: In analyzing payer and provider data to increase transparency on health care spend, we recognize there is heterogeneity between what is paid versus billed across specialty. There is an opportunity to focus attention on narrowing this gap for high-value procedures through evidence and education of patients, payers, and providers to ensure patients receive appropriate treatment, and providers are appropriately reimbursed. PHP165 Claims Reimbursement Analysis of the National Health Insurance Scheme in Ghana Nsiah-Boateng E National Health Insurance Scheme, Accra, Ghana . Objectives: To assess the value and service quality of the National Health Insurance Scheme (NHIS) benefits. Methods: A review method was employed to analysis medical claims for the 2011 to 2013 period. The medical claims were retrieved from the database of the Ashiedu Keteke District NHIS Office. The incurred claims ratio, promptness of claims settlement, and claims rejection ratio indicators of benefit value and service quality were analyzed. Results: A total of 421,574 medical claims with a cost of GHS7.3 million (USD2.6 million) were analyzed. These claims came from thirteen accredited health care providers-three public health facilities, four private clinics and six community pharmacies. The incurred claims ratio increased significantly from 4.3 to 7.2 over the period, 2011-2013. The proportion of claims settled beyond 90 days increased consistently from 26% to 90% over the same period. Although, the proportion of claims rejected increased from 0.9% to 3.6% over the period under review, overall, it was low. The reasons for rejection included provision of benefits to inactive subscribers and breach of sublimit on certain expense category. Conclusions: There is increased awareness and utilization of health services; however, there are considerable delays in claims settlements. It would be necessary for management of the NHIS to settle claims in time to ensure that health care providers are financially resourced to render service to subscribers. PHP166 National Health Insurance Fund Drug Expenditure in Bulgaria, 2007-2012: Reference Based Pricing Alone or in Combination With Other Approaches to Pricing Djambazov S N 1, Vekov T Y 2, Petrov D 3 1Cancer clinics Doc Dr Valentina Tsekova, Sofia, Bulgaria, 2Medical University Pleven, Pleven, Bulgaria, 3Bulgarian Medical Union, Sofia, Bulgaria . . . . . Objectives: Our team wanted to compare the economic effect of restricted market access and reference based pricing (RBP) vs. RBP alone in two consecutive periods, 2007—2009 and 2010—2012. Methods: We used the officially published cash flow statements and income statements from the financial reports of the National Health Insurance Fund (NHIF) for each of the years from 2007 to 2012. Then we compared the data about the expenditure for drugs, as well as the data about the budget control. Results: While restricted market access and RBP has been applied between 2007 and 2009, the NHIF drug expenditure increased with 15% (from EUR 144 mln to EUR 166 mln). For that period, the drug expenditure was generally 100% within the budget. From 2010 to 2012, while only RBP has been applied, the expenditure increased with 62% (from EUR 187 to EUR 303 mln). For the period, the drug expenditure exceeded the NHIF budget with 5% in 2010, 34% in 2011, 20% in 2012. Conclusions: RBP alone cannot control the drug expenditure in a long-term. Additional measures are needed together with RBP. Performancebased pricing, differential pricing, comparative pricing, profit control and price- volume agreements may be considered as additional to RBP measures for pricing and budget control. PHP167 Evaluation of Social Welfare Part of Hungarian Drug Provision System Andriska P , Komáromi T , Gyáni G Healthware Consulting Ltd., Budapest, Hungary . . . Objectives: The part of Hungarian drug provision system, which is available on social welfare list, changed several times and in different extents in recent years. The introduction of drug budget in 2006 and simultaneously the abolition of social welfare drug list implied the most significant change. The chief aim of our analysis to give a comprehensive overview about the main trends on this field examining the range of available products on social welfare list based on different aspects, also concerning the trends of demand, product structure, expenditures and patient burdens. Methods: Changes of key points of relevant acts concerned in drug provision, the aggregated (not patient level) public turnover data of the Hungarian Health Fund and published number of patients involved in this reimbursement category were considered as the key sources and indicators of our evaluation. Results: After regulation changes, demand (DOT) and number of patients within social welfare list category significantly decreased after 2006 until 2009, but since then both of them stagnate. Significant growth of reimbursement outflow within this segment can be observed until 2011, but after slight decrease. Part share of higher level reimbursement categories within reimbursement outflow increased. Demand moved to more expensive products, while patients were able to access to more innovative active ingredients in higher level reimbursement categories. Strong correlation can be observed between changes of acts and breaks in turnover trends. Conclusions: Results of the analysis may support objective judgement of the present social welfare provision system, as well as may contain considerable consequences regarding to potential ways of future structural changes, considering both interests of entitled patients (right to access innovative therapies) and the Health Fund (increase savings, improve efficiency). Further analysis based on real world (patient level) data may result more complex investigation opportunities of this patient segment and reimbursement category. PHP168 Analysis of New Model of Therapeutic Positioning Reports As A P&R Decision-Making Tool in Spain Rodrigues T , Izmirlieva M , Ando G IHS, London, UK . . . Objectives: This study aims to determine if the recently proposed model for therapeutic positioning reports (IPTs) in Spain is actually being used as a supportive tool for pharmaceutical pricing and reimbursement (P&R) decisions whilst delivering greater transparency and regional market access harmonisation. Methods: Primary research was conducted with regional payers on the proposed model of IPTs, which contain a comparative evaluation on effectiveness and safety, as well as criteria of use, for newly authorized medicines and older medicines which have a high potential health or economic impact. This was supplemented by secondary research. Results: The publication of IPTs has brought quicker uptake and greater homogenization to the market access of certain medicines. Notably, according to primary research, protease inhibitors were made available in all autonomous communities (CCAAs) two months after the IPT publication. However, the deadline of 3 months for the publication of IPTs is generally not being met. As delineated in the model and shown by primary research, IPTs are starting to be used as a basis for P&R decision-making in Spain and play an important role in the identification of the most suitable target patient population. IPTs are also delivering on promises of generating consensus amongst the Spanish CCAAs. Conclusions: IPTs are being used to support the Inter-ministerial Commission of Medicine Prices and the Directorate General of Pharmacy and Basic Services role in P&R. If accepted by the majority of the CCAAs, IPTs will also serve as a tool to evaluate and limit discrepancies in innovative medicines access throughout Spain. However, there is lack of consensus of whether an IPT should include an economic evaluation from the start. Contrary to manufacturers, regional payers generally believe its inclusion would help with the selection of more cost-effective medicines whilst aiding with the adoption of IPTs in general. HEALTH CARE USE & POLICY STUDIES – Health Care Research & Education PHP169 Determination of Referral Utilization Rate: Linking Households To Health Care System; Kenyan Rural Setting Experience Mogere D M 1, Loum C S 1, Kaseje D 2 1Great Lakes University of Kisumu, KISUMU, Kenya, 2Great Lakes University of Kisumu, Kisumu, Kenya . . . . . Objectives: Referral utilization is defined as the number of patients referred and seen by physicians. The objective was to establish referral utilization rate among sick persons identified at the households, counseled and referred to link hospitals by community health workers. Sick persons identified were issued with referral and counter referral slips to take to the hospital. Methods: This was a quasi -experimental study carried out in two sub- locations in rural Kenya. One hundred community health workers were trained on community based referral and counter referral model and issued with referral tools. Each was assigned 25 households, instructed to regularly visit them in order to identify sick persons, counsel and refer them to link hospitals for care. One hundred villages comprising 2209 households with a population of 11,000 people were covered and referral model implemented for 12 months. Results: A total of 322 patients were identified, advised and referred, with a referral response rate of 93% (298/322.) Seventy percent were five years and above. The study showed that 82% (263/322) of the patients arrived in hospital with referral slip the same day A433 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 they were referred, 5% (15/322) arrived the second day, 6% (20/322) third day and over and 7% (24/322) did not arrive in the hospital. Conclusions: The study provides evidence that community health workers if properly trained, equipped and supported would identify sick persons in the households, counsel and refer them to hospitals for specialized care. Patient referral arrival rate on same day is reasonably high because households were regularly visited and sick persons encouraged to seek care. Similar studies carried out in the US and Great Britain suggest that in a population of 1000 adults, 750 will experience an episode of illness, of these 250 will consult a physician of whom 6 will require to be referred to the physician. PHP170 Provincial Entry of New Drugs In China Wan W Y H , Conti C GfK NOP, London, UK . . . . Objectives: To improve health care provision whilst containing cost at the same time, Chinese authorities allow the entry of new drugs at provincial level. Whilst different provinces have different economic situation and needs, it is essential for pharmaceutical companies to strategize provincial drug entry. The purpose of this research is to explore province(s) that may provide the best chance for new drug entry. Methods: To capture all the drugs that were newly included in the 2012 National Drug List (NDL) but not the 2009 NDL, multiple databases including the two NDLs, Adis Insight and the Chinese database Yao-zhi Shu-ju were used. Provincial reimbursement status of these drugs was identified using the Chinese database. Provincial population and economic data (including gross domestic product (GDP), health expenditure and consumer price index (CPI)) in 2011 were captured on the National Statistical Database. The correlation between the numbers of provincially reimbursed drugs (with/without restrictions) and GDP per capita, and between the numbers of provincially reimbursed drugs (with/ without restrictions) and regional health expenditure, were investigated. All databases were accessed in September 2013. Results: 245 new formulations were identified, among which 35 of them are alternative formulations of 26 new drugs. Among the 23 provinces and 4 municipalities provinces, Anhui (48%) followed by Jiangsu (20%) approved most new drugs without restrictions. Guangxi (21%) was the autonomous region approved most new drugs without restrictions. Among the provincially reimbursed drugs, Jilin (62%) was the province approved most drugs associated with restrictions. The number of approved drugs was not correlated to regional GDP per capita (p=0.18, r2=0.06) nor with regional health expenditure (p=0.78, r2=0.003). Conclusions: There is a substantial variation in the number of approved new drugs between provinces in China. The economic status of the provinces is not likely to be the sole factor causing such differentiation. PHP171 Using The Ispor 2013 European Congress as a Big Data Case Study Furnback W , Wang B C M Alliance Life Sciences, Somerset, NJ, USA . . . . Objectives: The availability of concentrated big data sets can help researchers analyze current trends in an industry and drive future growth. Over 1,600 poster presentations related to pharmacoeconomics and outcomes research were presented at the 2013 ISPOR European meeting. We analyzed the titles of the posters to find trends in research. Methods: We analyzed the titles and associated sections of all the poster presentations accepted for the ISPOR 2013 European meeting held in Dublin, Ireland. The distribution by sector (research on methods, health care use & policy studies, cancer, cardiovascular, etc) was calculated in addition to the type of study (cost, PRO & patient preference, clinical outcomes, etc.). We also searched the titles of posters for key words to divide studies into the following four categories: cost-effectiveness, budget-impact, burden of illness and literature review. Results: In total, 1,679 poster title were analyzed. For the sector distribution, the top three studies were as follows: research on methods (238; 14.18%), health care use & policy studies (217; 12.92%) and cancer (210; 12.51%). The most popular types of studies were: cost studies (617; 36.75%), patient reported outcome & patient preference studies (212; 12.63%), and clinical outcomes studies (200; 11.91%). Of all the posters, cost-effectiveness studies made up 12.27%, while budget impact studies, literature reviews and burden of illness studies made up 2.5%, 1.67% and. 6%, respectively. The top three sectors for cost-effectiveness studies were cancer (31; 15.05%), cardiovascular disorders (29; 14.08%), and infections (28; 13.59%). Conclusions: Analyzing trends at conferences may help researchers and key stakeholders understand the current issues in the field. The most common type of study for the 2013 ISPOR European were cost studies. The cost-effectiveness studies were concentrated in cancer, cardiovascular disorders and infections. PHP172 Common Illnesses Identified by Community Health Workers in the Households and Referrred to Primary Health Facilities For Care Mogere D M 1, Loum C S 1, Kaseje D 2 1Great Lakes University of Kisumu, KISUMU, Kenya, 2Great Lakes University of Kisumu, Kisumu, Kenya . . . . . Objectives: To determine common illnesses identified and referred to the health facility for care by community health workers during routine household visitations. Methods: Quasi-experimental study was carried out in two sub-locations in rural Kenya where one hundred community health workers were trained on community based referral and counter referral model and issued with referral tools. Each was assigned 25 households, instructed to regularly visit them in order to identify sick persons counsel and refer them to link hospitals. One hundred villages comprising 2209 households with a population of 11,000 people were covered where the counter referral model was implemented. Results: In total the community health workers identified, counseled and referred 322 sick persons to health facilities for health care. Those identified for referrals were categorized as either below or above five years. Under fives referrals accounted for 30% (97/322) where the top five reasons for referrals included; general illnesses for under fives, 5.2 (18%); postnatal care for infant, 30 (9.3%); immunization defaulters (3.1%); malaria, 3 (. 9%) and diarrhea, 3 (. 9%). Top five reasons for above five referrals included; general illnesses for above fives, 89 (27.6%); ANC clinic 34, (12.7%); health check up, 24 (7.5%); Chronic cough, 19 (5.9%) and PNC for mother 6 (1.9%.) Conclusions: Community Health Workers were able to identify, counsel and refer 322 persons with 38 various types of ailments. The study concludes that if CHWs are trained well, supplied with referral tools and provided with regular support supervision, they have the capacity to identify common ailments at household level and provide advice on the appropriate health action required to be taken. It is the view of the authors that community health workers referral model be scaled up. PHP173 An Evaluation of Patient Satisfaction In Pafos, Cyprus With The “Europep” Instrument Panayides N 1, Latsou D 2, Geitona M 2 1Neapolis University, Pafos, Cyprus, 2University of Peloponnese, Corinth, Greece . . . Objectives: To evaluate patients’ satisfaction with the general practice care provided at the Pafos General Hospital outpatient units. Methods: The EUROPEP questionnaire, standardized and validated into Greek, was distributed to outpatients, with the method of random sampling and filled out with personal interviews, from February to April 2014. EUROPEP includes 23 items and 5 dimensions regarding doctor – patient relationship, medical care, information - support, organization of care and accessibility. The sample size was based on 50% of annual outpatients’ visits. For the assessment of the questionnaire internal consistency, the coefficient a Cronbach was used. Student’s t-test and analysis of variance have been performed in order to determine the significant differences between the dimensions and sociodemographic characteristics. Results: In total 345 out of 383 outpatients filled out the questionnaire (RR= 90%). The majority of the sample (61.2%) was female and belonged to the 61-80 years old group. 43.5% of sample answered moderate health condition and 43.8% bad. The Cronbach’s alpha of questionnaire met the criterion of 0.858. The doctor – patient relationship, medical care, information – support and organization of care scored 4.09±1, 3.56±0.8, 3.97±0.9 and 4.01±1.2 respectively, which mean ‘satisfied’. However, the accessibility regarding the getting through to the practice and waiting time was scored much lower (2.3 ±0.8), meaning ‘dissatisfied’. Statistically significant differences were found between gender and doctor – patient relationship (p= 0.019), information - support (p≤ 0.001) and organization of care (p= 0.003), while age group was related to information – support (p≤ 0.001) and organization of care (p≤ 0.001). Patients’ health condition was found significant with the doctor – patient relationship (p= 0.004), medical care (p= 0.007) and accessibility (p= 0.001). Conclusions: Patients were satisfied with the provision of general practice and sociodemographic characteristics appear to be significant predictors of satisfaction. Also, problems of accessibility can be solved through the forthcoming reform of primary health care in Cyprus. PHP174 Clinical Trial Activity in Greece From 2010 To 2012: Still Missing the Opportunities? Athanasakis K 1, Boubouchairopoulou N 1, Alexiou V 1, Baroutsou B 2, Kyriopoulos J 1 1National School of Public Health, Athens, Greece, 2Sanofi - aventis AEBE, Athens, Greece . . . . . Objectives: Clinical trials (CTs) lead to innovative medical treatments, and present positive externalities both for clinical practice and the economy. The study aimed at portraying the CT activity in Greece for 2012, built upon a previous survey conducted in 2010, and at highlighting any discrepancies. Methods: The survey was conducted among the members of the Hellenic Association of Pharmaceutical Companies (SFEE). Each company was requested to return a structured questionnaire for each interventional CT approved by the National Ethics Committee (NEC) during 2012. Questionnaire items focused on the main characteristics of each CT: duration, phase, budget, number of recruited patients, affiliation of recruiting sites and therapeutic area of the investigational agent. Results: All SFEE members participated returning completed questionnaires (response rate: 100%). Data on 70 interventional CTs were received. The majority was phase-III trials (64.3%) as in 2010 (68.3%) with a mean duration of 36.3 months (regardless of phase). Most CT sites were affiliated to a university or NHS hospital (45.7% and 42.3% respectively). A contractionary phase in the CT activity was observed compared to 2010, as demonstrated by the basic characteristics of approved CTs, i.e. average number of patients/trial: 35 vs 98, average number of patient per participating center: 8.68 vs 32.5, average total budget/trial 218,556€ vs 296,598€ for 2012 vs 2010 respectively. Conclusions: In 2012, the health care sector in Greece entered a recession phase due to the severe economic crisis and the subsequent fiscal adjustment program. These measures, accompanied by non-efficient processes in the administrative setting for CT approvals in Greece, appear to have an impact on the country’s CT activity. Given that CTs represent vital investments for the human capital and the economy, joint action from all stakeholders is of paramount importance in order to surpass existing hurdles and to promote investment in research. PHP175 Predictors of Working Motivation in Job Satisfaction Among Nurses in Cyprus Polyviou C 1, Latsou D 2, Geitona M 2 1Neapolis University, Pafos, Cyprus, 2University of Peloponnese, Corinth, Greece . . . Objectives: To investigate factors affecting nurses’ working motivation on job satisfaction in public and private hospitals in Pafos. Methods: The study included all hospitals in Pafos, with 410 nurses out of which a random sample was collected from February to March 2013. Two generic, self-administered, instruments have been used. The first questionnaire refers to working motivation in terms of recognition, responsibility, personal development, job interest, working relationships etc, developed by Everard κ α ι Morris (1999). The second instrument refers to job satisfaction by Spector (1985). Both instruments score 1: “agree/satisfied very A434 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 much” to 4 “disagree/dissatisfied very much”. Correlation analysis was performed to identify the relationship between working motivation and job satisfaction. Also, multivariate regression analysis with stepwise method was conducted between overall job satisfaction and various working motivation dimensions. Results: 141 out of 150 nurses filled out the questionnaire (RR 94%). The majority of the sample (68.8%) was female and belonged to the 21-30 years old group. Working motivation scoring was ranged from 2.54 to 2.04, meaning positive, except from working relationships (1.8 ±0.7). Also, the mean of overall job satisfaction was 2.3±0.6, meaning satisfied. Positive correlation was found between working motivation and job satisfaction (p≤ 0.01), with the exception of salary and working relationships. Personal development (β = 0.227), job interest (β = 0.254), responsibility (β = 0.149) and recognition (β= 0.154), were significant predictors of overall job satisfaction (p≤0.05). Reporting high motivation from work was positively and significantly associated with higher job satisfaction. Conclusions: Working motivation of health professionals contributes to the enhancement of their job satisfaction and consequently to the improvement of health services provision. PHP176 Factors Influencing Job Satisfaction Among Nurses of Pafos General Hospital in Cyprus M 1, Latsou D 2, Geitona M2 drugs were found using the online Drug Future Chinese Marketed Drugs Database (accessed December 2013 to January 2014). Results: 260 drugs for neurological diseases, 106 drugs for oncology and 76 drugs for respiratory diseases were identified; 92, 42 and 26 drugs respectively had the marketing years of both import brands and local brands were available. 32 (~35%) neurological drugs, 13 (31%) oncology drugs and 11 (42%) respiratory drugs had local ‘branded’/generic versions available before the imported (originator) version. Most of these drugs in all three therapeutic areas were marketed between 2002 and 2003, whilst their imported (originator) versions were marketed 1-7 years later. Since 2009, no originator drug has been found marketed later that their local ‘branded’/generic versions. Conclusions: The low number of local ‘branded’/generic drugs marketed before the imported drugs in recent years suggests that the IP issue in China might have improved since the 2009 health reform. Current Results also suggest that respiratory medicine was most popular for inappropriate local manufacturing. It is yet to find out whether this observation was related to the costs of the originator drugs, the market demand, difficulties in copying new molecules due to their complexity, or high cost of manufacturing. These results may act as the basis to further explore future IP issues in China. PHP179 The Ecology of Medical Care in Japan Revisited Nikolaou 1Neapolis University, Pafos, Cyprus, 2University of Peloponnese, Corinth, Greece Takahashi O , Ohde S St. Luke’s Life Science Institute, Tokyo, Japan Objectives: To assess the job satisfaction of the nursing staff of the Pafos General Hospital in Cyprus. Methods: A cross-sectional survey was conducted in April 2014. All 150 nurses (RR= 100%) working at the General Hospital of Pafos filled out a standardized, self-administered questionnaire with a 5-point Likert scale, regarding the assessment of job satisfaction, working conditions and personal well being. The questionnaire scoring from 1 represented “very dissatisfied” to 5 “very satisfied”. In order to investigate the association among job satisfaction as dependent variable and sociodemographic characteristics of the sample, satisfaction of working conditions and personal well being as independent variables, multiple linear regression analysis was performed with backward method. Also, correlation analysis, using Spearman rank coefficient, was conducted to quantify the strength of association among job satisfaction and independent variables. Results: The majority of the sample (64.7%) was female and permanent employees (60.7%). The mean age was 33.7 (S.D. 9.7) years. The personal well being score was 4.1 (S.D. 0.6), which means satisfied. The working conditions and job satisfaction scores were 2.2 (S.D. 0.7) and 2.9 (S.D. 0.9), showing low and moderate satisfaction respectively. Age (β = -0.321, 95% C.I. -0.650, -0.008), permanency of job (β = 0.346, 95% C.I. -0.005, 0.697), satisfaction of working conditions (β = 0.532, 95% C.I. 0.331, 0.732) and personal well being (β = 0.444, 95% C.I. 0.188, 0.700) were significant predictors of overall satisfaction (p≤ 0.05). Positive correlation was found between job satisfaction and working conditions (rs = 0.349, p ≤ 0.001) and personal well being (rs = 0.185, p ≤ 0.024) respectively. Conclusions: The assessment of nurses’ working conditions appears to contribute to the improvement of job satisfaction, the enhancement of productivity and nursing care outcomes. Objectives: Studies on the ecology of medical care have provided a framework for health care systems, medical education, and clinical research. Ten years ago, Fukui et al. reported the ecology of medical care in Japan. However, medicine and health care organization has changed since then. In the current study, we conducted an updated evaluation and compared it to results observed 10 years ago to understand health care seeking behaviors of people in Japan. Methods: A population weighted random sample from a nationally representative panel was used to estimate the number of health-related symptoms, self-care, and health care utilization per 1,000 individuals based on a prospective health diary recorded for one month. Variations in terms of age and gender were also examined. Results: Based on 4,548 persons (3,787 adults and 797 children), on average per 1,000 persons, 794 (95% Confidence Interval (CI): 782 - 805) had at least one symptom, 447 (95% CI: 433 - 462) used overthe-counter drugs (OTC), and 265 (95% CI: 252 – 278) visited a physician’s office, 206 (95% CI: 194 – 218) a primary care physician, 4 (95% CI: 2 - 5) a hospital emergency department, and 117 (95% CI: 107 - 126) a professional provider of complementary or alternative medical (CAM) care. Children had more physician visits, and adult used more OTC and CAM. Females were more likely than males to have symptoms, to visit their physician and to use OTC and CAM. Conclusions: Compared with the data from 2003, health care seeking behaviors of people with symptoms in Japan are remarkably similar, with the exception of an observed increase in the use of CAM. Results of this study would be useful for further delineation of health care seeking behaviors of people in the context of a health care system unique to Japan. . . . PHP177 Legal and Ethical Implications of Using Data From Social Media Websites Khankhel Z , Abogunrin S , Martin A Evidera, London, UK . . . Objectives: Most social media (SM) sites have common terms and conditions dictating how data from the sites may be used, but how these relate to health care research is usually unclear. We analyzed studies included in two separate systematic literature reviews to assess whether researchers have evaluated the legal and ethical implications of using SM as a clinical tool or as a data source for health care research. Methods: In the first review, MEDLINE and Embase were searched (2008-2013) for articles evaluating the use of SM as a tool or application in health care. In the second review, MEDLINE and Embase were searched for English language studies on the use of SM in the context of adult vaccination. We assessed whether any of the individual studies considered the legal and ethical implications of using SM sources. Results: The first review identified 3,232 unique abstracts; 36 reported on interactive, internet-delivered programs, Facebook, and mobile apps, for improving health outcomes of patients with cancer, or inflammatory, mental health, musculoskeletal, neurologic, ophthalmologic, or sexual health-related disorders. The second review identified 1,264 publications, of which 32 used SM to communicate with patients, analyze content, or recruit patients for studies related to perceptions of adult vaccines. Among these 68 publications, no study reported whether the legal or ethical implications of using SM content were considered prior to conducting research. Conclusions: The evidence suggests that legal and ethical implications of utilizing data from SM websites are not reported, and may not be considered, in current SM research. It is unclear whether this reflects the lack of clarity about what these restrictions mean for health care research, or whether researchers are not aware that such restrictions may exist. An informed debate to raise awareness of these issues and come to an understanding of the best way forward is urgently needed. PHP178 Phenomenon of Inappropriate Drug Manufacturing in China - Past and Present Wan W Y H GfK NOP, London, UK . . . Objectives: It was an unspoken fact that Chinese manufacturers copying patentprotected drugs inappropriately (e.g. aripiprazole (neurology), pemetrexed (oncology), tiotropium bromide (respiratory)). The purpose of this study is to explore this phenomenon among the drugs currently included in the National Drug Lists (NDLs) and trends in the recent years. Methods: Reimbursed drugs from three therapeutic areas (neurology, oncology and respiratory) were identified from the 2012 NDLs. The marketing years of the import (originator) drugs and local ‘branded’/generic . . PHP180 What Is The Proportion of Patients Who Return Counter Referral Slips to the Original Referral Service For Confirmation of Arrival to the Hospital? Mogere D M 1, Loum C S 1, Kaseje D 2 1Great Lakes University of Kisumu, KISUMU, Kenya, 2Great Lakes University of Kisumu, Kisumu, Kenya . . . . . Objectives: Counter referral success rate is defined as number of patients received back at the original referring service with counter referral slips divided by the total number of patients referred and received at the referral hospital. Study objective was to determine the number of patients who returned counter referral slips to the original referral service for confirmation of arrival to the hospital. Methods: Quasi-experimental study was carried out in two sub-locations in Kenya where hundred community health workers were trained on community based referral and counter referral model and issued with referral tools. Each was assigned 25 households, instructed to regularly visit them in order to identify sick persons counsel and refer them to link hospitals. One hundred villages comprising 2209 households with a population of 11,000 people were covered where the counter referral model was implemented. Results: During the 12 months of implementation, a total of 322 patients were identified, counseled and referred to link hospitals of whom 93% (298/322) arrived the hospitals with referral and counter referral slips. The study showed that forty six percent (138/298) of counter referral slips had the name, signature and date of referring health worker, only 12% (35/298) indicated CHW’s telephone contact. Forty one percent (121/298) had the name and signature of attending physician of which 96% (116/121) had hospital stamp. Further, forty seven percent (140/298) of the patients returned counter referral slips to the original referral service for confirmation of arrival at the referral hospital. Fifty one percent (117/228) were satisfied with referral and counter referral process. Conclusions: It is concluded that referral and counter referral model is both feasible and acceptable. However, referral protocols and other resources are necessary for the process to be successful. Further, both ends should be willing and prepared to receive those being referred there. PHP181 What Is The Patients’ Median Delay From The Referring Community Health Worker To Arrival in the Link Health Facility? Mogere D M 1, Loum C S 1, Kaseje D 2 1Great Lakes University of Kisumu, KISUMU, Kenya, 2Great Lakes University of Kisumu, Kisumu, Kenya . . . . . Objectives: Median delay could be defined as time taken in days from referring health facility to arrival at receiving link health facility. The purpose of the study was to determine the amount of time (in days) patients take to arrive at the link health facility for health care after being referred from the household by community health workers. Methods: Quasi-experimental study was carried out in two A435 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 sub- locations in Kenya where hundred community health workers were trained on community based referral and counter referral model and issued with referral tools. Each was assigned 25 households, instructed to regularly visit them in order to identify sick persons counsel and refer them to link hospitals. One hundred villages comprising 2209 households with a population of 11,000 people were covered where the counter referral model was implemented. Results: Three hundred and twenty two sick persons were identified, counseled and referred to the two link health facilities by one hundred community health workers who covered one hundred villages. It was further shown that 82% (263/322) arrived in the referral health facilities with referral slips on the same day, 5% (15/322) second day, 6% (20/322) third day and over while 7% (24/322) did not arrive in the hospitals. Conclusions: It was observed that community health workers regularly visited households. This observation is consistent with findings of previous similar studies. Only about less than a third of all the referrals did not arrive the link referral hospital the first day. It is therefore concluded that community health workers are a critical link between the households and primary health care facilities. They are likely to identify common illnesses during routine household visitation, counsel and refer sick persons to health facilities for care. PHP182 Choosing Important Health Outcomes For Comparative Effectiveness Research: A Systematic Review Gargon E 1, Gurung B 1, Medley N 1, Altman D 2, Blazeby J 3, Clarke M 4, Williamson P 1 1University of Liverpool, Liverpool, UK, 2University of Oxford, Oxford, UK, 3University of Bristol, Bristol, UK, 4Queen’s University, Belfast, Ireland . . . . . . . Objectives: A core outcome set (COS) is a standardised set of outcomes which should be measured and reported, as a minimum, in all effectiveness trials for a specific health area. This will allow results to be compared, contrasted and combined as appropriate, as well as ensuring that all trials contribute usable information. The aim of this review was to identify studies which sought to determine which outcomes or domains to measure in all clinical trials in a specific condition, and to describe the methodological techniques used in these studies. Methods: We developed a multi-faceted search strategy for electronic databases (MEDLINE, SCOPUS, and Cochrane Methodology Register). We included studies which sought to determine which outcomes/domains to measure in all clinical trials in a specific condition. Results: 198 studies (250 reports) were included in the review. Studies covered various areas of health, most commonly cancer, rheumatology, neurology, heart and circulation, and dentistry and oral health. A variety of methods have been used to develop COS, including semi-structured discussion, unstructured group discussion, the Delphi Technique, Consensus Development Conference, surveys and Nominal Group Technique. The most common groups involved were clinical experts and non-clinical research experts. Thirty-one (16%) studies reported that the public had been involved in the process. The geographic locations of participants were predominantly North America (n= 164; 83%) and Europe (n= 150; 76%). Conclusions: This review has brought together the existing research in a single place. COS will increase the efficiency and value of the research process and make it easier for people to make well-informed decisions about health care. Design of new trials will be simplified, risk of measuring inappropriate outcomes reduced, and selective reporting of outcomes less likely. We have highlighted future areas of research, including the need for methodological guidance for COS development and better reporting of COS studies. PHP183 The Role of Patients in Clinical Research and Evidence Based Decision Making as Reported Via A Survey of Patient Advocates Holtorf A P 1, Palacios D 2, Brixner D 3 1Health Outcomes Strategies, Basel, Switzerland, 2Novartis Pharma AG, Basel, Switzerland, 3University of Utah, Salt Lake City, UT, USA . . . . Objectives: Increasingly, patients are becoming more active contributors to research. The purpose of this survey was to understand the experience and expectations of patient organizations (POs) with clinical research and patient reported outcomes. Methods: An online survey was conducted in English language during May 2014 among 40 participants of a global cross disease patient forum. The information was used to provide background regarding patients as active contributors in clinical research. The participants represent a broad range of disease specific and general patient organizations from various countries including USA, European countries, Asia, Latina America, Middle East and Australia. Results: The most important deficiencies of current clinical research practice were described as a lack of patient friendly reporting, incomplete reporting and a lack of patient involvement in the study design. Over the half of respondents stated that clinical trials are conducted in a way that does not yield information relevant to the patients. Active involvement in clinical trials was specified through 79 different expectations which fell into 7 categories: (1) Accessibility of new medicines (2) More representative patient selection for studies (3) Transparent conduct of the study (4) Better information on safety and risks to participants (5) Active versus passive role of patients (6) Contribution to innovation (7) Improved result reporting; especially to patients. One third of the respondents have already been actively involved in clinical research and felt that this was a mostly positive experience. Globally, patients and patient organizations are becoming increasingly involved in clinical research through multiple formal or informal routes of engagement. Conclusions: Patients should be better represented in clinical research and consulted for patient relevant study aspects in order to provide information of greater relevance to patients in their own therapeutic decisions. PHP184 Drug Repurposing as An Efficient Strategy In Drug Development – Example Of Cns Area Toumi M 1, Murteira S 2, Caban A 3, Kornfeld A 4 1University Aix-Marseille, Marseille, France, 2Lundbeck Japan KK, Tokoyo, Japan, 3CreativeCeutical, Paris, France, 4Creativ-Ceutical, Paris, France . . . . Objectives: Traditionally, drug repurposing has been considered as a cost-effective and a reduced-risk strategy for developing new drugs. Currently little is known and documented regarding efficiency of repositioning strategies in drug development. The objective of this article is to assess the meaning of this process with focus on CNS area. We aim to identify the repositioning strategies that conducted to the discovery and development of new CNS products as well as identify the source and target indication and development status. Methods: In order to identify repurposing cases that target CNS, an extensive research was performed. It included a literature review as well as search of websites of organizations engaged in promoting and development of drug repurposing. Cases were extracted from the source material. For each molecule, its initial and target indication as well as development status was recorded. Each case was classified by type of repurposing strategy (repositioning, reformulation or both). Results: 109 source products were identified. They were repositioned 183 times. About one third of the source products have been repositioned more than one time. Half of new indications was approved. The majority of cases were repositioned while only 10 were reformulated and 16 were reformulated and repositioned at the same time. The highest number of repurposed molecules was originating from CNS therapeutic area (67 molecules targeted 127 indications). Among new therapeutic indications alcohol / opioid / drug dependence was targeted most often (20 repositions). Regarding repositions within CNS area, Schizophrenia, Epilepsy and Depression were the richest sources of repositioned drugs with 10 and more products each. Conclusions: Drug repurposing in CNS is an efficient and currently very active drug development strategy, exemplified by the considerable amount of new indications that have been found via this strategy, with approximately half of the target indications being currently under development. PHP185 Components of Sustainable Health Systems: What Is Known About The Cost-Effectiveness of Clinical Care? Tordrup D , Stephan L , Bertollini R World Health Organization, Bruxelles, Belgium . . . Objectives: Evidence to ensure cost-effective purchasing of services are an important pre-requisite for ensuring financial sustainability of health systems. The Research Agenda for Health Economic Evaluation project aims to map out the economic evidence available for the treatment of the 10 highest burden conditions in the European Union and identify evidence gaps. Methods: The 10 highest burden diseases are selected by Disability Adjusted Life Years (DALYs) based on the Global Burden of Disease study for Europe (Ischemic Heart Disease, Low Back Pain, Stroke, Major Depressive Disorder, Lung Cancer, Falls, COPD, Diabetes, Other musculoskeletal, Neck pain). Clinical pathways are generated for these conditions by merging published clinical guidelines from public and professional bodies. Economic (cost-benefit and cost-effectiveness) evidence indexed by MEDLINE is stratified by treatment pathways to identify evidence gaps. Existing reviews are analysed to assess coverage of primary evidence. Results: Results are presented for diabetes: 1,208 articles were retrieved from MEDLINE, of which 315 mapped to the treatment pathway. Most evidence was published after 2000. 74 reviews were identified, of which 24 were published between 2009-13. Volume of economic studies for each treatment ranged from 0 to 39 studies. The most intensively studied were diabetic retinopathy (n= 39), foot ulcers (n= 34) and renal disease (n= 31). Areas with no economic evidence included non-pharmacological blood pressure and dyslipidemia control. In some cases economic evidence was available, but no recent reviews of the evidence were identified (eg. islet or pancreas transplantation, diabetic neuropathy, interventions for weight loss). Conclusions: Economic evidence was available for the majority of treatment modalities for diabetes, however for several treatments no studies were identified. Importantly, reviews were not identified for several treatments where economic evidence was available. Lack of economic evidence prevents cost-effective commissioning, and lack of reviews of evidence may be a further barrier to translation. PHP186 Health Care Reform in China And The United States: A Tale of Two Nations Seeking To Improve Access To Affordable Coverage Moore R Decision Resources Group, Nashville, TN, USA . Objectives: China and the United States have approved massive health care reforms seeking to increase access to affordable health care services. Both countries employed a composite of government programs and private enterprises designed to improve efficiencies in their health care markets. This study compares the results of these health care reforms so far, explores the longer-term prognoses and considers how lessons learnt can inform health care reform elsewhere. Methods: In the U.S., 80 MCOs and 240 physicians were surveyed about their participation in reform programs. Quarterly data on the uninsured and health care cost trend data were also assessed. In China, 670 patients were surveyed regarding their access to a health care provider. Results: Early returns show mixed results that stem from policy decisions inherent in the original programs as well as from national political considerations. Overall, both the U.S. and China health care reforms are enjoying success but have shortcomings. In the U.S, partisanship, technology issues, and implementation uncertainties have undermined the reform legislation. However, the nation’s uninsured rate declined from 17.1% in Q4 2013 to 13.4% in Q2 2014, and the health care cost trend has fallen. Moreover, the predicted spike in emergency room utilization has not yet appeared. In China, the percentage of citizens with coverage increased from 30% in 2003 to 95% in 2011, while patients’ share of out-of-pocket costs fell. However, improvement is felt less in rural regions and demand for services is increasing. Consequently, Chinese citizens still have difficulty accessing health care providers, which, combined with inadequate reimbursement rates, means health care in China remains suboptimal. Conclusions: Together with early coverage statistics, responses from both the U. S. and China to challenges in their respective health care reform efforts indicate that long-term success is achievable. Their experiences underscore the A436 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 need for political unity, substantial financial investments, and adaptability when faced with obstacles and suboptimal results. PHP187 The Future of Education In HTA and Health Economics Carroll C , Beecroft C , Miller L The University of Sheffield, Sheffield, UK . . . Objectives: An increasing range of online education is available to those working in the field of pharmaco-economics and Health Technology Assessment (HTA). The aim of this study was to explore the market for free online education in HTA and to assess the practicalities of delivering a free, interactive short-course. Methods: We delivered a 5-week online course at the end of 2013 following the principles of the Massive Open Online Course (MOOC) approach: the HTA MOOC. The programme covered the basics of HTA: What it is; how new technologies are identified; how they are evaluated; the principal means of assessing clinical and cost-effectiveness; and how HTA is used within different health systems to inform reimbursement decisions. The course was provided through the Coursesites, a free, open-access, education platform. Materials were adapted from an existing online MSc programme. Descriptive statistics of participants were recorded and a survey conducted of participants’ experiences. Results: 2039 individuals registered for the HTA MOOC, of which 1508 completed a voluntary survey giving basic sociodemographic data about themselves. Participants were from Europe (55%), Asia (16%) and North America (13%) and accessed the MOOC to gain knowledge about the subject (78%), and specifically for career development (56%). 531/2039 (26%) of those registered in Week 1 completed the MOOC (i.e. submitted all required weekly assessments by the end of Week 5), compared to an average of 5-6% for MOOCs generally [THES 2013]. Students reported the principal strength of the programme to be the quality and usefulness of the learning materials, and the principal issue to be time required (workload was considered by some to exceed the 4-5 hours/week intended). Conclusions: The MOOC approach offers scope for delivering flexible, effective and accessible education to small or large international cohorts of professionals, patients and providers working within HTA and pharmaco-economics. PHP190 Pharmacoeconomic Education in Brazilian Schools of Pharmacy Freitas G 1, Balbinotto G 2 . . 1UFRGS, Porto Alegre, Brazil, 2Universidade Federal Do Rio Grande Do Sul, PORTO ALEGRE, Brazil Objectives: The objective of this study was to survey the pharmacy schools in Brazil to determine the extent of education in pharmacoeconomics offered during the school year 2012-2013. Methods: A questionnaire based on previous studies was developed. This was emailed to 55 pharmacy schools in Brazil during October and December 2013. The schools were selected from the Ministry of Education website. Public and private University schools (only those that have high concepts in the National Examination Performance of Students) were included. In addition, a search was made in the database directories of research groups from National Council for Scientific and Technological Development (CNPq). Results: Of the 55 questionnaires sent, 16 were returned (29%). Only two schools do not address the education of pharmacoeconomics in any moment. Most of schools address some concepts in different subjects (8 hours). Five schools have formal courses that teach only pharmacoeconomics and health technology assessment (over 30 hours). All agree that the education of pharmacoeconomics is important and ten schools believe that very few hours are devoted to the teaching pharmacoeconomics at his university. In search of directories of research groups 23 groups that develop research in the area of pharmacoeconomics in Brazil were found. Conclusions: There is a large deficit in the availability of courses pharmacoeconomics in Brazil at both the undergraduate and graduate. There is a great opportunity for experienced individuals to fill this gap. Provide an education in pharmacoeconomics for pharmacy students is especially important in the context of evidence-based decisions and when health issues and allocation of scarce resources is a priority for Brazil. PHP191 Application of A ‘Nice Post-Hoc B/S Analysis’ To the Nice Appraisal Process Macaulay R HERON Commercialization, London, UK . Objectives: Post-hoc subgroup analyses are still used in clinical trials of medical technologies to identify patient populations in whom greatest benefits can be achieved, despite this being derided as an analytical approach. Indeed, using such an approach, aspirin has been shown to be ineffective versus placebo in acute myocardial infarction patients born under the star signs of Libra and Gemini (ISIS-2, 1988, Lancet) and endarterectomy is only efficacious in treating symptomatic stenosis patients born on a Monday, Wednesday, or Friday (ECST group, 1998, Lancet). This research aimed to determine what effect the name of a drug has on the National Institute of Health and Care Excellence (NICE) appraisal process by applying a post-hoc analysis that compares the rates of acceptance by the first letter of the drug or technology name using a Chi-squared test. Methods: All final appraisal determinations resulting from Single Technology Appraisal (STA) or Multiple Technology Appraisal (MTA) processes were identified up to April 2014 from which the first letter of the generic name and the decision were extracted. Results: 481 appraisals were identified, 371 (77%) of which were approved (defined as ‘recommended’ or ‘optimised’), spanning all letters of the alphabet except J, K, W, X, and Y. The lowest approval rates by letter were for B (64%, 14/22) and S (69%, 18/26). Drugs beginning with a B or S were significantly less likely to be approved than drugs beginning with any other letter (p= 0.0072). Conclusions: Based on this ‘NICE post-hoc B/S analysis,’ manufacturers should consider lobbying the World Health Organization to give their new pharmaceuticals International Nonproprietary Names beginning with ‘B’ or ‘S’ to optimise success rates in the NICE appraisal process. Alternatively, regulators and HTA bodies should continue to view efficacy claims on the basis of post-hoc sub-group analyses with great scepticism. PHP192 Impact of Story Books on Promoting Knowledge and Behavior of 4th Stage Elementary Students About Rational Use of Medicines in Kermanshah Provine of Iran 2008 Mohammadhosseini N National Committee on Rational Drug use, Tehran, Iran . Objectives: Evaluation of the impact of children stories on promotion of the knowledge and behavior of 4 th stage elementary students about rational use of medicines in the Kermanshah elementary schools. Methods: 64 elementary schools were selected in Kermanshah province of Iran. The schools were categorized into 3 groups (rich, medium and poor level) based on their students’ economic level. Both boys and girls were involved. Two checklists were designed in order to evaluate students’ knowledge and behavior before and after the intervention. A story book which was published by National Committee of Rational Drug Use (RUD) conveyed to the students as the intervention. Results: The correlation between pre and post intervention and the schools economic levels and the gender of students were obtained. The knowledge change rates were 21.83%, 22.28% and 17.40% for rich, medium and poor schools respectively. The change rate for girls was greater in comparison to that of boys. Conclusions: It is concluded that children stories as an educational intervention causes valuable positive changes on students’ knowledge and behavior regarding hygiene and medicine. Implementing such interventions among children, result in promoting rational use of medicines in future of the society. PHP193 Clinical Trials in France: An Underexploited Opportunity Mahi L , Mahi I Axelys Sante Affaires Medicales et Recherche Clinique, Paris, France . . Objectives: Despite high-performance infrastructures and recognized expertise, Clinical trials (CT) are declining in France. Health Professionals deplore the burdens of the French administration that leads to a real concern regarding the international scientific competition. In addition patients are reluctant to participate in CT especially after the benfluorex scandale. Methods: one of the witnesses of the competitiveness of France in CT is the activities among Clinical Investigation Centers (CIC). These are plateforms fully dedicated to clinical and scientific research acting as an interface between INSERM units (Public National Institut of Scientific Research) and CHU (University Hospital). Results: There are 54 CIC in France. They have been created between 1992 and 2009. CICs are located in university hospitals (CHU) dessiminated in the most important regions of the country. The specific needs of CHU and researchers have led to implement several types of CICs: multi-topics (24 CIC-P), Clinical Epidemiology (9 CIC-EC), integrated in biotherapy (11 CIC-BT) and technological innovations (8 CIC-IT). During the last quadrennium, CIC-P supervised about 1000 protocols, of which 2/3 were therapeutic and 1/3 on physiopathology. 25% of protocols conducted in CIC-P were translational in collaboration with INSERM units, 20% were for rare diseases, 35% were funded by industry and 36% by the PHRC (Academic funding). The AP-HP (Academic Hospitals of Paris Area) is the major sponsor in France and one of the first in Europe, with 500 clinical trials enrolling more than 17,000 patients in 2010. Despite this development, the number of CT submissions to ANSM (French Regulatory Agency) decreased from 1,000 in 2008 to 895 in 2012. ANSM approuved 705 protocols in 2012 versus 790 in 2008. Conclusions: Despite appropriate structures and willingness to encourage scientific production, clinical trails remain underexploited due to different causes which need to be deeply evaluated. HEALTH CARE USE & POLICY STUDIES – Health Technology Assessment Programs PHP194 Will Value Based Assessment (VBA) Revolutionise The Nice Assessment? Heemstra L 1, Purchase J L 2, Van Engen A 1 1Quintiles Consulting, Hoofddorp, The Netherlands, 2Quintiles Consulting, Reading, UK . . . . Objectives: In March, the National Institute for Health and Care Excellence (NICE) revealed their plans to implement Value Based Assessment (VBA) in their Technology Appraisal process: VBA will replace NICE’s end-of-life criteria. Instead wider societal impact (WSI) and burden of illness (BOI) will be systematically included in the assessment through absolute and proportional Quality Adjusted Life Year (QALY) shortfall. However, it is unclear whether these elements will impact NICE’s recommendations. Methods: The BOI and WSI for the 26 examples provided by NICE was compared with an average displaced treatment in the NHS. Assuming that both elements were weighted equally, an average QALY shortfall rating was calculated. This was then transformed into a hypothetical willingness to pay (WTP) threshold ranging from £20,000 for the drug with the lowest rating to £50,000 per QALY for the drug with the highest rating. The hypothetical WTP was compared with the most plausible ICER and recommendation from NICE. Results: Multiple Sclerosis, rheumatoid and psoriatic arthritis, as well as oncology indications were associated with a high BOI/WSI. Six of the 26 examples had higher most plausible ICERs than the estimated threshold. Two of these six were still recommended using end-oflife criteria (metastatic melanoma and mCRPC), whereas the remainder were not recommended by NICE. Three of the 26 examples had ICERs below the estimated WTP threshold but higher than £30,000, and were still recommended. All of these recommendations mentioned additional value elements such as innovation, end-of- A437 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 life criteria or even QoL impact on family. Conclusions: This analysis presented a possible scenario for the implementation of VBA. None of the drugs with a negative recommendation would get a positive recommendation under this scenario. The products at most risk are those currently accepted using end-of-life criteria. However, how the new value elements will be weighted has yet to be determined. PHP195 The Evolution of International Reference Pricing: An Analysis of 39 Countries Lockwood C , Marinoni G , Ando G IHS, London, UK . . . Objectives: To characterise country-level changes to international reference pricing (IRP) policy frameworks across 39 markets, and understand how countries amend this tool. Methods: Qualitative interviews were conducted with 50 stakeholders across 39 markets, representing 37 payers or payer influencers and 13 industry stakeholders. These interviews focused on country-level IRP methodology, both past and present. Extensive secondary research of government websites and existing literature was also conducted. A qualitative and semi-quantitative analysis of these findings was undertaken to identify key trends in how IRP has been modified as a policy tool since implementation. Results: Of the markets considered, IRP has remained comparatively stable – in terms of both the countries comprising the reference basket and the underlying formula for determining the reference price remaining unchanged – in just under one-third of the 39 markets. In contrast, just over one-third of these markets have modified their IRP baskets to include one or more additional countries with the objective of lowering prices. A smaller number have made substantive changes to the IRP formula itself, notably with four markets transitioning from taking the average price of their basket as a reference to either taking the average of the three lowest or taking the lowest. Conclusions: IRP serves as a dynamic policy tool, changing to reflect individual country circumstances and broader policy reform over time. While a number of markets have maintained a stable IRP regime since inception, a larger number have made some changes, the most common being to add one or more markets to the reference basket. Although changes to IRP are mostly concentrated in Eastern European markets, as well as Western European markets most impacted by the recent economic crisis, the present study also suggests that there is considerable variation across geographies and time in how countries adapt their use of IRP. PHP196 Making Sense of Nice’s ‘New’ MTA and STA Process Guide: A Narrative Synthesis Urbich M , Mildred M Boehringer Ingelheim Ltd, Bracknell, UK . . Objectives: NICE’s guides to the process of Multiple Technology Appraisals (MTA) and Single Technology Appraisals (STA) provide a valuable source of information to enable stakeholders to engage in Health Technology Appraisals (HTAs). NICE recently conducted a review of the process guides which has implications for all stakeholders involved with MTAs and STAs. The objective of this analysis was to identify the number and type of amendments within the guides in order to highlight the most important changes for consultees and commentators. Methods: Narrative synthesis was used to systematically identify, classify and explore the impact of the proposed amendments to the MTA and STA process guides. The hypothesis was that the amalgamation of the MTA and STA guides would simplify both TA processes whilst increasing rigour and transparency. Sources of reference were the draft of the new process guide, the NICE Senior Management Team Board Cover Paper, and the 2009 MTA and STA process guides. Results: Amendments were classified as relating to the process itself (26%), data and confidential information (44%), terminology (7%), and others (22%). Of the 27 amendments to the MTA and STA process guides, 4 (15%) were identified as major amendments which warrant specific appreciation. Major amendments included: (1) the STA decision problem meeting moving to after the Department of Health referral; (2) stricter rules around the marking of confidential information; (3) full publication of MTAs; and (4) MTAs now have the opportunity to go straight to Final Appraisal Determination (FAD) following the first committee meeting. Conclusions: The draft process guide suggests that the new MTA and STA processes will foster greater engagement between stakeholders early on, increase transparency, and enable patients to have quicker access to innovative medicines which are able to go straight to FAD. PHP197 Methodological Requirements Regarding Quality of Life Measurement in the Early Assessment of Benefit In Germany Blome C , Augustin M , Lohrberg D University Medical Center Hamburg-Eppendorf, Hamburg, Germany . . sis, and interpretation emerged. Dominant topics included: the appropriate level of disease-specificity of QoL instruments; required evidence on an instrument’s validity and on the validity of a minimal important difference; appropriate duration of QoL assessment; consequences of potential bias due to unblinded study design or missing data; interpretation of results that differed between subscales of an instrument; non-acceptance of surrogate endpoints for QoL. Conclusions: Evidence on QoL can have high impact on the additional benefit determined by the G-BA. Therefore, QoL assessment and analysis in clinical studies that shall enter benefit dossiers should confirm with a range of methodological requirements. . Objectives: In Germany, an early assessment of benefit (EAB) is required for new medicines since January 2011. The pharmaceutical manufacturer submits a dossier on additional benefit over comparative treatment which is subsequently evaluated by the Institute for Quality and Efficiency in Health Care (IQWiG). Stakeholders can comment on the evaluation in a formal comments procedure. The final decision on additional benefit is made by the Federal Joint Committee (G-BA); it provides the basis for price negotiations between manufacturer and statutory health insurance funds. Quality of life (QoL) is one of four criteria for benefit evaluation. This qualitative study aimed to determine methodological requirements for QoL measurement in the German EAB. Methods: A qualitative content analysis according to Mayring was conducted. Documents of all EABs completed until December 2013 (including dossier, IQWiG evaluation, protocol of the oral hearing, and G-BA decision) were searched for the term QoL or synonyms. Relevant passages were extracted and reduced to key content by two researchers independently. On the basis of subsequent consensus building, recurring themes of the term’s usage in the EAB process were identified. Results: In the 66 early assessments of benefit included in the analysis, a range of methodological requirements regarding QoL assessment, analy- PHP198 Exploring The Flaws in Cost-Effectiveness Models That Lead to Rejection of Nice Submissions Griffiths E A , Hendrich J PAREXEL, London, UK . . . Objectives: New health technologies are required to demonstrate both clinical and cost-effectiveness before recommendation by the National Institute for Health and Care Excellence (NICE) for reimbursement in England; however, a large proportion of submissions are rejected due to non-robust economic analysis. Published NICE guidance includes a comprehensive critique of submitted economic evidence so, to help inform future submissions, we assessed the flaws in cost-effectiveness models leading to rejection by NICE. Methods: All NICE single technology appraisals from January 2006 to May 2014 were included in the analysis. Multiple technology appraisals, resubmissions, vaccination programmes, requests for advice, and submissions where an incremental cost-effectiveness ratio (ICER) could not be determined were excluded. Recommendations and reasoning across decisions were extracted, with a focus on the critique of the economic evidence. Results: 121 NICE submissions met the inclusion criteria, 28 (19.8%) of which were rejected. Non-robust economic analysis was one of the listed reasons for rejection in 75.0% (21) of cases, and in all cases where the submitted ICER was below the £30,000 cost-effectiveness threshold. Within these submissions, the key drivers behind rejection due to non-robust cost-effectiveness modelling were: a high level of uncertainty in inputs (leading to a sensitive or unreliable ICER) (in 90.5% of cases); mishandling of data (e.g., overstated treatment effect or failure to account for adverse events) (85.7%); misalignment to the reference case (e.g., inappropriate comparator or weak methodology) (38.1%); and unrealistic assumptions (38.1%). Conclusions: Non-robust economic analysis is one of the main reasons behind rejection of NICE submissions, largely due to high uncertainty, or selective use of data to favor the health technology being appraised. Early modelling may allow manufacturers to identify and address sources of uncertainty or weakness early in the clinical development process, in order to construct a convincing and robust economic argument ahead of reimbursement submission. PHP199 Is it Possible to Predict The Market Access of a New Pharmaceutical in Germany? A Systematic Evaluation of Federal Joint Committee Decisions on Early Benefit Assessments According to The German Law For Reforming The Market of Pharmaceuticals Schwander B 1, Banz K 2, Kaier K 3, Walzer S 4 GmbH, Loerrach, Germany, 2Outcomes International, Basel, Switzerland, 3University of Freiburg, Freiburg, Germany, 4MArS Market Access & Pricing Strategy GmbH, Weil am Rhein, Germany . . . . 1AHEAD Objectives: As of 1st January 2011 the German drug market is regulated by the act of the reorganization of the pharmaceutical market (AMNOG). Since then the normal procedure for reimbursement of a new pharmaceutical is an early benefit assessment by the joint federal committee (G-BA) which determines one of six additional benefit levels. According to AMNOG any specification of the reimbursement price shall be based on the outcomes of the early benefit assessment. Hence this assessment takes a key role for market access of a new drug in Germany which poses the question whether it is possible to predict the level of additional benefit that will be established by the G-BA. Methods: In order to evaluate a possible predictor of G-BA decisions, the ‘evaluation of pharmaceutical innovations (EVITA)’ score was calculated and retrospectively compared with 40 published G-BA decisions. The EVITA algorithm evaluates a new compound for a given indication and in relation to a relevant comparator on the basis of randomized controlled trial (RCT) evidence. EVITA translates the RCT outcomes on the therapeutic benefit and risk profile into rating points, which are expressed as a total EVITA score. Results: Univariate ordinary least squares and ordered logit regression analyses show statistically significant correlations between EVITA scores and the G-BA additional benefit levels. Moreover, for the prediction of an additional benefit level of at least ‘minor’, an EVITA score cutpoint of ≥ 3 is associated with a sensitivity of 100% and a specificity of 80%. For the prediction of an additional benefit level of at least ‘considerable’, an EVITA score cutpoint of ≥ 7.5 is associated with a sensitivity of 100% and a specificity of 93.1%. Conclusions: The present investigation indicates that the EVITA score may have the potential for the prediction of G-BA decisions related to AMNOG early benefit assessments. PHP200 Selection of Topics For Nice Technology Appraisal 2005-2011: What Matters Most? Ward D J 1, Shiyka A 2, Fellows R 1 1NIHR Horizon Scanning Centre, Birmingham, UK, 2University of Birmingham, Birmingham, UK . . . . Objectives: The National Institute for Health and Care Excellence (NICE) undertakes appraisals of selected health technologies, and those judged to be costeffective must be funded for use on the National Health Service in England and Wales. Obtaining a positive appraisal decision is therefore important to commercial developers seeking market access, who require consistent application of topic selection criteria (which were amended in 2009). We sought to establish which characteristics of drugs or their indications were most important in the decision to undertake an appraisal. Methods: All marketing authorisations (MAs) granted A438 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 2005-2011 for new drugs and new indications for existing drugs were identified from the European Medicines Agency (EMA) website. The decision to undertake an appraisal was obtained from the NICE website and NIHR Horizon Scanning Centre records, and the associations between this and characteristics of the drug and intended patient population were then determined. Results: For 2005-2011, we identified 134 MAs granted by the EMA (116 new drugs and 18 new indications) of which 72 (54%) were selected for appraisal. The decision to undertake an appraisal was significantly associated with an MA granted 2009-2011 (OR= 2.3, p< 0.01), the drug being a biological agent (OR= 3.9, p< 0.01), administered on a long-term basis (OR= 1.8, p< 0.05), indicated for a patient population < 1 in 1,000 (OR= 2.1, p< 0.05), or for malignant disease (OR= 5.1, p< 0.01). It was not associated with an indication for more severe disease (OR= 2.0, p= 0.06), an MA issued for a new indication (OR= 1.4, p= 0.50), or whether a drug was first-of-kind (OR= 1.8, p= 0.10). Conclusions: We identified several characteristics associated with the decision to undertake an appraisal relating to both the drug and intended patient population that do not completely match published topic selection criteria (e.g. severity). Further analyses are required to determine which are the most relevant factors in this decision. recommendations, comparison of patient pathways and specific type of patient input desired. Two reviewers extracted methodological details, study designs, and outcomes into summary tables. Results: We identified 21 articles out of a total of 18,829 studies. Articles covered multiple subject areas. Process improvements were most common (4 studies) followed by current perceptions (3 studies), comparison of patient pathways (2 studies) and specific type of patient input desired (1 study). Research methodologies and stakeholders varied widely including telephone, web audit, interview /questionnaire and literature review. Stakeholders varied between national & international HTA agencies, experts and patient groups. Three studies involved patient groups and one involved patients. These studies informed the role, process and nature of input but did not address the impact on HTA decision making. Conclusions: Compared to other HTA areas there is a lack of published material on PAG involvement. There have been many attempts to provide a framework for patient involvement but so far none has been used in HTA decision-making. Existing data does not help to quantify role of the patient in HTA decision making. Additional research is needed to understand and quantify patient group input in HTA decisions. PHP201 Determination of Cost-Effectiveness Threshold For Malaysia PHP204 Trends and Key Decision Drivers For Rejecting An Orphan Drug Submission Across Five Different HTA Agencies Lim Y W , Shafie A A , Chua G N , Hassali M A A Universiti Sains Malaysia, Penang, Malaysia . . . . . . . . . PHP202 Systematic Review of Economic Evaluation of Health Technologies Developed In Brazil From 1980-2013 Decimoni T C 1, Leandro R 1, Soarez P 1, Craig D 2 1Sao Paulo University, São Paulo, Brazil, 2University of York, York, Brazil . . . . . Objectives: The aim of this study is to review published economic evaluation of health technologies conducted in Brazil. Methods: Systematic review of economic evaluations studies published in MEDLINE, EMBASE, LILACS, SciELO, NHS EED, HTA Database, Web of Science, SCOPUS, BVS ECOS and SISREBRATS from 1980 to 2013. Full (Cost consequence analysis - CCA, cost minimization analysis - CMA, costeffectiveness analysis - CEA, cost-utility analysis - CUA, and cost-benefit analysis - CBA) and partial (cost description - CD and cost analysis - CA) economic evaluation studies were eligible for inclusion if at least one of the authors was Brazilian and was affiliated to a Brazilian institution. Two independent reviewers screened articles for relevance and carried out data extraction. Disagreements were resolved through discussion or through consultation with a third reviewer. We performed a qualitative narrative synthesis. Results: We identified 11946 records and 557 met inclusion criteria. One hundred and ninety (34.1%) were full (of these, 56.6% CEA, 20.3% CCA, 12.7% CUA, 5.6% CMA, and 4.7% CBA), and 367 were partial economic evaluation (of these, 64.7% CD and 32.3% CA). The main health problem studied were Infectious and Parasitic diseases (17.1%), Diseases of the Circulatory System (12.3%) and Neoplams (10.3%). The majority (72.9%) was conducted by authors from the southeast region, and south region (12.6%), mainly linked to academia (69.5%), and 54.2% were published in medical and 18.9% in public health journals. Seventy-two (14.7%) studies reported to be funded by industry and 16% was considered to have conflict of interest. Conclusions: There was a considerable growth in the conduct and publication of economic evaluation studies in Brazil. A qualitative evaluation of the methodology used in those studies is important to legitimize their use in the process of local decision-making. PHP203 A Literature Review of Patient Advocacy Group (Pag) Involvement in HTA Hicks N 1, Hawken N A 2, Arvin-berod C 1, Toumi M 3 1Commutateur, Paris, France, 2Creativ-Ceutical, Luxembourg, Luxembourg, 3University AixMarseille, Marseille, France . . . . Mardiguian S , Stefanidou M , Sheppard F PAREXEL, London, UK . Objectives: Decision on the cost-effectiveness (CE) of health care technologies usually creates an argument especially when alternatives are more expensive but more effective. In this situation, external criterion in the form of CE threshold or willingness-to-pay for a quality-adjusted life-year (WTP/QALY) needs to be applied to decide on its CE. Nevertheless, the lack of empirical and well-accepted CE threshold in Malaysia is recognized as one of the most important barriers in using health technology assessment for decision making. This study was mainly done to determine the CE threshold value across Malaysian population, estimated in terms of societal WTP for a QALY. Methods: A cross-sectional, contingent valuation study was conducted using stratified multistage cluster random sampling technique in the states of Penang, Kedah, Selangor and Kuala Lumpur Federal Territory. Respondents were asked for the socioeconomic background, quality of life and their WTP for a hypothetical EQ-5D health state scenario (treatment, extended life in terminal illness and life saving situations with three health severities – mild, moderate and severe, and two QALY gained levels – 0.2 QALY and 0.4 QALY) using pre-designed questionnaires. Interval model analysis was applied to determine the CE threshold. Results: One thousand thirteen respondents aged between 20–60 years old who can understand either English or Malay language were interviewed face-toface. The mean value of CE threshold was determined at the range of MYR 19,929 to MYR 28,469 (~ USD 6,200 to USD 8,900). Conclusions: By comparing our results to Malaysian GDP per capita in the year 2013; ~ MYR 33,754 (~ USD 10,548), we noted that the mean WTP/QALY is ranged between 0.59–0.84 times of GDP per capita. . Objectives: Patient input is an important part of the assessment process, yet sometimes seen as having a low evidence base. Previous work by the authors shows more research is needed on identifying how the patient group contribution is impacting decision making. Our objective was to review and critically appraise existing publications on PAG involvement in HTA. Methods: A search in Pubmed, Cochrane and ISPOR databases since 2009 was undertaken to identify studies on patients or PAG involvement in the HTA decision. Studies were evaluated for relevancy. We extracted information on perceptions of patient input, process improvement . . Objectives: Access to orphan drugs is often inconsistent, and is hindered by difficulties in demonstrating value in HTA appraisals due to the small patient populations and insufficient data. To inform future submissions, we examined the trends and key decision drivers that resulted in a submission being rejected across five HTA agencies. Methods: The Orphanet database was searched for orphan drugs with a marketing authorisation between 2002 and 2014. To assume a certain level of competition, awareness and commercial potential, rare diseases for which two or more orphan drugs were available were selected. Decisions from five HTA agencies were considered: AWMSG (Wales), CADTH (Canada), NICE (England), PBAC (Australia), and SMC (Scotland). Assessments that resulted in a rejection were examined for key decision drivers, and for trends and variation by disease type. Results: A total of 28 licensed orphan drugs were available for the treatment of eight rare diseases. The number of orphan drugs assessed, and rejection rates, varied by HTA agency; PBAC and SMC had the lowest rejection rates (4/18; 22% and 6/22; 27%, respectively), while NICE had the highest rejection rate with 40% (4/10). Uncertainties regarding clinical efficacy, and concerns over the robustness of economic evidence were the key decision drivers that led to a rejection. Examination of data by disease type indicated a trend towards higher rejection rates for diseases with a higher prevalence rate. Conclusions: The proportion of rejected submissions varied by HTA agency, particularly within the HTA bodies in the UK, highlighting inconsistencies in decision-making. An association between prevalence rate and the proportion of rejected submissions was found, with lower rates of disease prevalence correlating with higher acceptance rates. This is most likely due to the lower budget impact incurred in smaller patient populations. PHP205 Predictors of German Amnog Decisions and GKV Rebate Negotiations: A Database Analysis Verleger K 1, Schoeman O 1, Schmidt R 1, Wilke T 2, Heeg B 1 1Pharmerit International, Berlin, Germany, 2IPAM - Institute for Pharmacoeconomics and Medication Logistics, Wismar, Germany . . . . . Objectives: G-BA, IQWiG and GKV are the main governmental stakeholders in the German AMNOG process. Based on manufacturer-submitted dossiers, the G-BA assesses the drugs’ additional benefit per pre-defined subgroup. Subsequently, the GKV negotiates rebates by drug. This research aims to describe factors influencing GB-A decisions and assess the association between additional benefit and rebate. Methods: All G-BA decisions up to March 2014 were analyzed. Univariate logistic regression was used to investigate the relationship of G-BA decisions (dependent variable: additional benefit (y/n) per subgroup) with study characteristics. Study characteristics included were disease area (ATC-code), superiority/ non-inferiority study design, comparators used in the submitted trials (in/direct; in/adequate comparator according to GBA [“ZVT”]), main area of claimed benefit e.g. overall survival (OS). Linear regression was used to assess the impact of added benefit (in at least one subgroup) on rebate. Results: Sixty-eight G-BA decisions, with in total 137 G-BA subgroups, were included and analyzed. In total, 60.3% of assessments resulted in an additional benefit. Most commonly, dossiers were submitted to the G-BA for ATC-codes L and A (39.7%; 19.1%). Out of 40 ATC-code L subgroups (27 drugs), 70.0% resulted in a positive assessment, with 50% demonstrating a benefit in OS. Univariate logistic regression showed a significant relationship between added benefit and: ATC-codes A/J/L; improvements in morbidity; adverse events; direct comparators; and the ZVT (ORs: 0.1; 11.2; 6.0; 55.2; 24.3; 20.9; 15.2; all p< 0.05). All drugs showing an OS advantage received a positive benefit assessment. Added benefit reduced the rebate significantly by 13.1% (p< 0.05). Conclusions: Key factors for a positive G-BA benefit assessment are improved OS, morbidity, and adverse events, demonstrated through the use of direct “ZVT” comparators. ATCcodes J and L carry the highest chance of gaining a positive assessment. The rebate negotiated with the GKV decreases significantly if an added benefit is determined. PHP206 Nice Restrictiveness Compared To the Market Authorization Jaksa A, Westbrook L , Rubinstein E , Daniel K , Ho Y S Context Matters, Inc., New York, NY, USA . . . . . Objectives: To determine how often NICE recommendations are more restrictive than the market authorizations. Methods: 161 NICE Technology Appraisal decisions from 2007-2013 were evaluated. These reviews included 80 unique drugs from 37 disease conditions. For each generic drug included in a review, the corresponding A439 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 market authorization was retrieved from the EMA or MHRA. NICE positive decisions were compared to the market authorizations. Any decision that included language that restricted the population eligible for reimbursement for a given therapy was categorized as “recommend with restrictions.” NICE positive decisions that were not more restrictive than the market authorization were categorized as “recommend.” Negative decisions were categorized as “do not recommend.” Restrictions were also quantified and categorized. Results: NICE issued “do not recommend” decisions in 32% of the reviews from 2007-2013. The overall rate at which NICE issued “do not recommend” decisions increased after 2010, but this did not pass traditional levels of statistical significance (p= .21). NICE issued positive decisions in 68% of reviews, but the decision was more restrictive than the market authorization in 52% of the positive decisions. NICE’s restrictiveness has decreased since 2007, with the exception of 2013 where 60% of NICE’s positive decisions were “recommend with restrictions.” For the “recommend with restrictions” reviews, there are 1.7 restrictions on average (range 1-4, s.d. 11) added to the market authorization. The most prevalent type of restrictions were for contraindicated or intolerance.” Conclusions: In 2007-2013, NICE issued “recommend with restrictions” decisions in 36% of reviews and issued both “recommend” and “do not recommend” decisions in 32% of reviews. NICE was more restrictive than the market authorization in 52% of the positive decisions, though NICE’s restrictiveness seems to be declining over time. An independent analysis of NICE decisions in 2007-2013 found a statistically significant different distribution of decisions than reported in the NICE website (p= .01). PHP207 Multicriteria Decision Analysis (MCDA) In HTA – Pilot Study in the Czech Republic Hajek P 1, Pecen L 2, Bulejova L 3, Cook M 4, Dolezal T 5, Dolezel J 6, Duba J 7, Dukova I 8, Fuksa L 9, Heislerova M 10, Jaskova K 8, Karasek P 11, Klimes J 12, Kminek A 13, Kucera Z 14, Vesela Š 15, Vothova P 1, Svihovec J 10 1Pfizer s.r.o., Prague, Czech Republic, 2CEEOR s.r.o., Prague, Czech Republic, 3InterMune, Muttenz, Switzerland, 4BAYER s.r.o., Prague, Czech Republic, 5VALUE OUTCOMES, Prague, Czech Republic, 6GlaxoSmithKline, s.r.o., Prague, Czech Republic, 7OAKS Consulting s.r.o., Prague 9, Czech Republic, 8Merck, Prague, Czech Republic, 9General Health Insurance Company of the Czech Republic, Praha, Czech Republic, 10University Hospital in Motol, Prague, Czech Republic, 11Novartis s.r.o., Prague, Czech Republic, 12iHETA, Prague, Czech Republic, 13AstraZeneca, Prague, Czech Republic, 14Sanofi-Aventis, Prague, Czech Republic, 15Janssen, Prague, Czech Republic . . . . . . . . . . . . . . . . . . Objectives: Multicriteria Decision Analysis (MCDA) is an analytical quantitative instrument focused on supporting the decision-making process between alternative products based on multiple criteria. Methods: In the pilot study on MCDA application in HTA in the Czech Republic, the following criteria were chosen by the experts: efficacy/effectiveness, safety, budget impact, disease severity, cost effectiveness and unmet needs. The number of evaluators was 10. Each evaluator determined weights within the range from 1 to 10 (from the least to the most important). The resultant weights were displayed as an arithmetic mean of weights of the individual evaluators and as a trimmed mean with the minimum and maximum values discarded. The weights were also calculated by discarding the last evaluator, i.e. there were 4 sets of weights examined, each time normalized by 100%. Results: Each evaluator rated 5 chosen medicines with weights 0, 1, 2, 3 within the chosen categories. Afterwards, the mean scores and trimmed means with the lowest and the highest values discarded were determined for each of the 5 medicinal products chosen. All 8 estimates (4 weights times 2 mean scores) lead to the identical classification of medicinal products which proves the robustness of the approach. The biggest divergences between the evaluators‘ assessment of the same medicinal product was observed in case of its safety, whereas the slightest were considered the budget impact and cost-effectiveness. On the other hand, the differences in the cost-effectiveness assessment of the 5 medicinal products considered were followed by the greatest discrepancies as regards the budget impact. The MCDA s was compared with the classifications of the medicinal products based on the ICER only which revealed significant differences (e.g., 2nd place according to the ICER vs. 5th according to the MCDA). Conclusions: The MCDA brings new information with respect to the each criterion‘s separate application. PHP208 Out With The Old – In With The New: Would New Social Preference Weights For Eq-5d Inevitably Require A Reappraisal of Previous CostEffectiveness Determinations? Kind P , Meads D M University of Leeds, Leeds, UK . . . Objectives: Social preferences are widely used in economic evaluation required by regulatory agencies. In the UK, NICE requires the use of EQ-5D and its associated set of TTO preference weights for computing QALYs. The weights in question date back nearly two decades. It is reasonable to question whether they continue to represent contemporary social preferences. Were a revised set of EQ-5D weights to be produced then would this necessitate the revision of all past appraisal decisions? This paper presents the 1stphase of work designed to address that question. Methods: The ICER is defined by the ratio of marginal cost (Δ C) /marginal benefit (Δ B). For a given ΔC the ICER falls as ΔB increases. For a given threshold (λ ) and for a fixed incremental cost (Δ C), there is a minimum health benefit Δ Bmin (given by Δ C/λ ) which must be achieved to produce an ICER that comes below that threshold limit. TTOweighted scores were computed for all 243 health states defined by the 3-level version of EQ-5D. A difference matrix was created in which D(i,j) contains the numeric difference between the ith and jth state. The number of differences below a given Δ Bmin was computed for each column (health state). Threshold values were varied (£20,000-£50,000). Cost differences were varied (£500-£10,000). Results: Less than 10% of health state value differences failed to meet the minimum Δ Bmin of 0.0125 (Δ C = £500; λ = £20,000) indicating susceptability to changes in health state value, however this proportion rose to 57% for higher incremental costs (e.g. Δ C = £3,000). 81/243 health states account for 50% of the differences that exceed Δ Bminat all tested levels of Δ C and λ . Graphical representation of these Results can be used to assess the need for reappraisal. Conclusions: For higher cost interventions, relatively small differences in EQ-5D weights can generate ICERs with the propensity to reverse previous cost-effectiveness decisions. PHP209 Decision Drivers For Brazil: An Analysis of Conitec Recommendations Kreeftmeijer J 1, Skaltsa K 2, Palazzolo D 3 1Quintiles Consulting, Hoofddorp, The Netherlands, 2Quintiles Consulting, Barcelona, Spain, 3Quintiles Consulting, Reading, UK . . . Objectives: Since its establishment in December 2011, Brazil’s HTA body CONITEC (National Commission on the Incorporation of Technologies), has published more than 90 assessments. The objective of the present study was to perform an analysis on CONITEC’s positive and negative decisions, in order to understand the main decision drivers. Methods: All assessments published by CONITEC between December 2001 and April 2014 were included in our analysis. The rationale for the decisions was analyzed for both positive and negative recommendations. Reasons for recommending or rejecting a technology were summarized into categories. Results: In total, 101 publications were identified: 67 of those assessed drugs, 12 a procedure or intervention and 11 a medical device. The remaining 11 were clinical guidelines (not included in the analysis). Overall, 46 recommendations were positive and 44 negative. The main reasons for rejection were concerns about the economic evidence (23 reports) or lack to demonstrate significant additional clinical benefits (22). The main reasons for positive recommendations were demonstrated clinical efficacy benefit (21), low budget impact (19) and fulfillment of high unmet needs (15). In the majority of cases, the decision was based on multiple factors. Conclusions: Brazil has set the way for a more transparent process for technology assessment following a formal process including pharmacoeconomic guidelines. However, insufficient clinical benefits and methodological concerns about the economic evaluation as major rejection drivers reveals that manufacturers are not yet addressing CONITEC’s requirements. Full transparency on the evaluation of outcomes is still missing, providing additional complexity for manufacturers towards a positive recommendation. Our results demonstrated a positive relation between acceptance and demonstrated clinical efficacy, as well as a low budget impact. CONITEC seems to follow a specific pathway in their decision that should be leveraged by manufacturers in order to increase their likelihood of receiving a positive recommendation. PHP210 The Risky Business of Drug Development: The Final Say of National Hta Agencies on A Pharmaceutical’s Benefit During The Last Stretch of An Expensive, Long-Lasting And Arduous Development Journey – As Illustrated By The Decisions Of Germany’s Gemeinsamer Bundesausschuss (G-Ba) Maetzel A 1, Staab T R 2, Ruof J 2 LLC, Chapel Hill, NC, USA, 2Roche Pharma, Grenzach-Wyhlen, Germany . . . . 1Berlikon Objectives: Since 2011 Germany follows a formal process of evaluating new pharmaceuticals for their incremental benefit vs. an appropriate comparator to inform price negotiations with Insurers. This study summarizes the rationale underlying the German authorities’ (G-BA) final assessment of manufacturers’ submissions following successful approval by regulators. Methods: G-BA decisions (1/2011 to 2/2014) were evaluated for their alignment (full, partial or none) between manufacturer’s development programs and expectations concerning: (1) target population; (2) comparator; (3) clinical endpoints, including indirect comparisons. Also addressed was the role of safety and how the G-BA addressed the potential for bias. Results: Of 69 completed submissions, 3 were resubmissions and 7 lacked a dossier. 59 completed submissions were subjected to a detailed review. Ten (17%) were for orphan disease indications. Major disagreement existed for 37 (63%), of which 17 (46%) were considered fully inadequate, and 20 (54%) inadequate for significant subgroups. Main reasons for inadequacy were: wrong comparator (27 of 37 [73%], wrong endpoint 6 [16%] and use of historical controls (3 [11%]). For 34 (92%) the major disagreement also led to a lower benefit judgment. All 19 indirect treatment comparisons were considered flawed. Safety was a differentiator for 24 of the 59 submissions, either primary (2) or in addition to efficacy (22). G-BA disagreed with the manufacturer on safety for 12 (50%) of the 24 submissions. Conclusions: This analysis of the first 3 years of G-BA’s early benefit appraisal illustrates that a majority of the submissions fail to convince the German authorities despite having obtained licensing approval. A wrong comparator was the main reason for full or partial rejection. Indirect treatment comparisons were never accepted. Decisions taken early in the development program have important repercussions on reimbursement negotiations with authorities in Germany. PHP211 Hta Epidemiology Data in Different Geographical Regions: Investigation of Requirements For Oncology Drugs Chadda S , Chandler T PHMR Associates, Newcastle upon Tyne, UK . . Objectives: Harmonisation of health technology assessment (HTA) processes between countries is a logical and efficient solution to complex data gathering exercises required of pharmaceutical manufacturers when preparing submissions. However, harmonisation is a slow process and potentially substantial differences between countries exist. Further, some types of data will inevitably need to be country-specific to meet local HTA requirements. Epidemiological data can be considered one such source of information. Methods: We reviewed HTA requirements in Australia, England and Wales, Japan and Scotland for epidemiology data requirements in their submission. Specific data types were identified and compared across the geographical regions. Results: Clear requirements were available for Australia (Pharmaceutical Benefits Advisory Committee [PBAC]), England and Wales (National Institute for Health and Care Excellence [NICE]), and Scotland (Scottish Medicines Consortium [SMC]). As of April 2014, there is an ongoing development A440 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 of the HTA process in Japan, which is excluded from this analysis. NICE, PBAC and the SMC all require the number of patients with the indicated disease and a clear statement of patient numbers eligible for treatment per year for 5 years. Beyond this, the requirements of the Australian PBC and the SMC were similar, specifying prevalence, incidence and mortality data, whereas NICE requires a measure of disease burden (not clearly defined) and life expectancy among those with the disease. The epidemiology requirements did not differ by disease area. Conclusions: HTA bodies stipulate the inclusion of epidemiological data to estimate economic impact of interventions. Some requirements are common to all agencies, but there are also some important differences. Specific epidemiological data needs for individual agencies must be considered by drug developers when planning and gathering information for HTA submissions. PHP212 Inclusion and Consideration of Patient Preferences in Amnog Early Benefit Assessments Obradovic M , Rauland M GfK Market Access, Nuremberg, Germany . . Objectives: To explore the inclusion and evaluation of patient preference data in AMNOG early benefit dossiers. Methods: Patient perspective is becoming increasingly important in health care decision making. Health technology assessment (HTA) agencies include patient views in their appraisals in different ways and to various extents. Patient views and preferences can be captured either by engaging patient organisations, patients themselves, or informal carers in the HTA process (as is done for example in the UK or Canada). Such patient aspects can be also explored in qualitative and quantitative studies which are considered as part of evidence documentation by the assessment committees. We have reviewed value dossiers and corresponding benefit assessments in Germany from 1st January 2011 to 31th March 2014. Types of patient preference data included in the value dossiers and their consideration in the assessments were collected and summarised. Results: A total of 68 dossiers were analyzed. 18 dossiers (26%) included data on patient preferences. Those data related to: a) relevance of different treatment endpoints from the patient’s perspective in oncology, hepatitis C, diabetes, or HIV infection; b) patient preference for a specific drug administration route (e.g. oral vs. injections), administration system (e.g. different inhalation systems), or administration frequency; c) patient preference for therapy duration or type of therapy. In none of the assessment reports did the evaluating committee specifically address the evidence presented on patient preferences. A comment on patient preference data was stated in one assessment report only (for aclidiniumbromid). Conclusions: About a quarter of value dossiers referred to data on patient preference. Surprisingly, it appears that the evidence on patient preference has not been considered in the AMNOG benefit assessments despite the fact that benefit to the patient is the central criterion of the AMNOG early benefit assessment. PHP214 Rapid Relative Effectiveness Assessment of Pharmaceuticals: Transferability and Completeness of Information Derived From Global Value Dossiers To Complete A Eunethta Submission Neeser K 1, Lister J 1, Stanisic S 2, Stengel C 1, Mueller E 1 1LASER ANALYTICA, Loerrach, Germany, 2LASER ANALYTICA, Milano, Italy . . . . . Objectives: EUnetHTA is currently evaluating the applicability of the Core Model® for Rapid Relative Effectiveness Assessment of Pharmaceuticals of selected drugs for which a market authorization is intended between 2013 and 2015. A global value dossier (GVD) represents an important tool for pharmaceutical manufacturer (PM) to consolidate information on considered disease, its management and treatment, epidemiology, as well as health economic outcomes. The current study is intended to quantify the information that may be transferable from a well prepared, comprehensive GVD to EUnetHTA submission via gap analysis. The EUnetHTA submission comprises of 4 modules: 1. Description of the health problem; 2. Technical characteristics of technology and appropriate comparator(s); 3. Clinical effectiveness, and 4. Safety of the new drug. Methods: A GVD, developed for a pharmaceutical product, was used as data source to evaluate the feasibility to address all questions in the four EUnetHTA modules. Results: In modules 1 and 2 about 60% of data that are required to complete the EUnetHTA submission are missing from the GVD. Most of this information (e.g. implementation of the new drug, current use of the drug) may be easily obtained from other documents held by PM. In modules 3 and 4 about 70% of queries cannot be answered without additional assessments (e.g. complementary evidence synthesis, study quality assessments), systematic searches (e.g. on-going unpublished studies) and additional sources. Conclusions: A GVD can be a useful pool of knowledge for a new drug in a specific indication. Despite this large body of evidence, a considerable part of the information that is required for a EUnetHTA submission may be missing from GVDs and needs to be derived from existing clinical study reports, extensive systematic literature searches and additional evaluations. Applying of validated and systematic methods during the GVD development process may reduce additional work for assessment reports. PHP215 Encepp-HTA Working Group Survey on Capacity To Conduct Research in Support Of Health Technology Assessment Toussi M 1, Lis Y 2, Qizilbash N 3, Blake K V 4, Ehrenstein V 5, Kurz X 4, Moore N 6, Sinclair M 7 Health, Paris La Défense, France, 2PAREXEL International, Uxbridge, UK, 3OXON Epidemiology Limited and Department of Primary Care and Public Health, Imperial College, London, UK, 4European Medicines Agency, London, UK, 5Aarhus University Hospital, Aarhus N, Denmark, 6INSERM CIC-P 0005, Université de Bordeaux, CHU de Bordeaux, Bordeaux, France, 7University of Ulster, Newtownabbey, UK . . . . . . . . . 1IMS Objectives: In light of growing interest in Health Technology Assessment (HTA) related outcomes being incorporated into post-authorisation studies (PAS) of medicines, an HTA Working Group of the European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP) has been established. The Working Group’s current focus is on establishing baseline data to shape the network’s activities in meeting emerging demands for evidence generation in PAS. Therefore, an exploratory survey was undertaken to map current capabilities related to research that supports HTA and to identify relevant training needs. Methods: An online survey was developed and piloted within the HTA Working Group prior to distribution to over 200 researchers on the ENCePP mailing list. The items in the survey were both qualitative and quantitative. Data will be analysed using simple descriptive statistics and thematic analysis. Results: The survey is ongoing and will close in July 2014. Preliminary results indicate that (50%) of those sampled had conducted studies involving patient reported outcomes (PROs) including quality of life data. The most frequently used data collection tools were: paper (80%), handheld devices (50%), interactive voice (30%) and webbased (30%). Clinical outcomes were the most frequent primary study endpoints (100%), although at least some data on PROs and cost outcomes were included in 86% and 71% of studies, respectively. The most common design was primary data collection prospective cohort studies. ENCePP guidelines were the most frequently referenced followed by ISPOR and ISPE methodological good practice guidelines. Full data analyses will be available in September 2014. Conclusions: The findings of the ENCePP-HTA WG survey provide baseline data on ENCePP expertise and resources in research supporting HTA that are of interest to regulatory and HTA stakeholders across Europe. PHP216 What The English Could Learn From The Irish: Making The Nice Approval Process More Cost-Effective Macaulay R HERON Commercialization, London, UK . Objectives: The National Institute of Health and Care Excellence (NICE) is a body that has gathered worldwide respect for its fully comprehensive clinical and economic review of pharmaceutical agents. However, it has been criticized for taking too long to conduct assessments and, as a consequence of this, only being able to appraise a subset of all new agents coming to market. Other Health technology Asessment (HTA) bodies achieve a much higher output of treatment appraisals with a much shorter turnaround time (e.g. in 2013 NICE appraised 27 drugs compared to 57 by the National Centre for Pharmacoeconomics (NCPE), Ireland). Methods: A thematic analysis of the appraisal processes of the NCPE was conducted to determine if any key criteria could be extrapolated that could potentially increase the speed and output of the NICE appraisal process. Results: One particularly interesting aspect of the NCPE system was revealed – a 2 to 4 week rapid preliminary review that all drugs must undergo to determine the necessity of a full pharmaco-economic assessment (PEA). Full PEAs are reserved for products with high cost, significant budget impact, and/or questionable value for money. This enables products that are cheaper and at least as efficacious as the relevant comparator to gain rapid market access. It also enables the NCPE to focus their time and resources on the products that will have a substantial/questionable clinical and/or economic impact. Conclusions: Adoption of such a systematic rapid review procedure into the NICE appraisal process with reimbursement conditional on NICE approval could similarly enable a better focussing of NICE resources to where it is most impactful. This could also incentivise pharmaceutical companies to drop their prices to gain rapid market access rather than going through the expensive and time consuming procedure of a full NICE submission. PHP217 Going Beyond The Qaly In Assessing The Benefits of Medical Devices Wilkinson G 1, Drummond M 2 School of Economics and Political Science, London, UK, 2University of York, Heslington, York, UK .. . 1London Objectives: Many medical devices offer improvements over current care that may be difficult to assess using standard methods of benefit such as the qualityadjusted life-year (QALY). The objective of this research was to identify whether these benefits could be measured and valued by alternative approaches, such as contingent valuation (willingness-to-pay) or conjoint analysis (discrete choice experiments). Methods: We undertook a systematic review of the literature from 1996 to 2013 to identify empirical studies of the benefits of medical devices using the following methodologies: willingness-to-pay (WTP), discrete choice experiments (DCEs), multi-criteria decision analysis (MCDAs) and subjective well-being (SWB). We recorded the number and category of individuals surveyed, the attributes explored and the key findings of each study. Results: The search resulted in 2772 hits, of which 2016 were considered not relevant and 76 were duplicates. On further examination, 47 of the remaining 680 papers were found not to meet our inclusion criteria, 240 were methodological papers and 363 involved non-device technologies. This left 30 relevant empirical studies, of which 18 were WTP and 12 DCEs. The types of devices studied included hearing aids, hip and knee implants and colostomy bags. Comfort, convenience or ease of use was the most common attribute explored other than effectiveness and quality of the device. Where both were studied, patient preferences sometimes differed from those of physicians. Conclusions: This research demonstrates that it is feasible to measure and value the benefits of devices using alternative approaches to QALYs, but that the literature is quite small as compared with that for non-device technologies. This is surprising given the often-repeated claim that devices have benefits that are difficult to assess using standard methods. Whilst the studies demonstrate the relative importance of each attribute, thought is required on how to incorporate these estimates into comparative cost-effectiveness studies. PHP218 Reality in Market Access In Germany and France – Comparative Analysis of Added Benefit Decisions on Innovative Pharmaceutical Therapies Droeschel D 1, Chicoye A 2, Lepretre M 2 University - SRH FernHochschule Riedlingen, Riedlingen, Germany, 2ESSEC Executive Education, Paris-La Défense, France . 1Riedlingen . . A441 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: The German market access system for drugs have been changed significantly in the last years, by introducing a similar focus on benefit assessment as in the French system. The research question remains whether they produce consistent results in terms of additional benefit (AB) for pharmaceuticals which have passed the assessment in both systems. Methods: The G-BA and IQWiG as well as the Transparency Commission (TC) databases were searched systematically to identify those products, which have been processed in both systems between Jan 2011 and Dec 2013. For further comparison a data grid consisting of 26 items for evaluation has been developed including study comparator, primary clinical endpoints, health related quality of life inclusion. Results: Overall, 140 new therapies have been assessed in France by TC, and 80 in Germany by the G-BA. According to inclusion criteria, 44 products could be identified which have passed through both systems including 7 orphan drugs. Thirteen products (30%) had no AB granted by both Agencies, whereas 9 (20%) were in both cases granted with a minor AB, (assuming that “minor” values are equivalent between the two systems), amounting to 22/44 cases with a similar resolution. Five cases (11%) showed a discrepancy in added benefit, all times TC = no and G-BA = yes. However, varying magnitudes appeared to be the greatest difference (n = 17 (39%) remaining drugs), conditioned by lacking concordance of both scale grade systems. Conclusions: Decisions of the agencies in both countries show partial heterogeneity in driving criteria like benefit levels (ASMR and AB). Although the evidence package for initial assessment in both countries is largely similar, preliminary results suggest their contextualization and scales are different. Further analysis based on results of the grid is needed to better assess criteria leading to different benefit levels and their reimbursement impact. PHP219 Factors Influencing Dutch Drug Reimbursement Recommendations; A Database Analysis Alleman C J M , Spoorendonk J A , Charokopou M , Hensen M , Heeg B Pharmerit International, Rotterdam, The Netherlands . . . . . . . . Objectives: In the Netherlands, manufacturers need to apply for reimbursement of outpatient drugs on either list 1A (no added benefit) or 1B (added benefit). For expensive drugs, hospitals can receive additional reimbursement if the drug is included on the expensive drug list (EDL). Pharmacoeconomic evidence is only required for list 1B and EDL evaluations. The National Health Care Institute (NZi) evaluates submissions and makes (provisional) reimbursement recommendations to the Dutch government. The aim of this study was to identify explanatory variables for the recommendation by NZi. Methods: A database of published evaluations from February-2006 to March-2014 was created, consisting of the final reimbursement recommendation and a range of corresponding explanatory variables such as the therapeutic indication, clinical and economic characteristics. Univariate analyses were performed to assess the impact of the individual explanatory variables on the recommendation by means of odds ratios. Results: In total 262 applications were included; the number of positive recommendations by NZi were 121/122 (99%) for 1A, 77/107 (72%) for 1B and 19/28 (68%) for EDL. Pharmacoeconomic analysis was reported in 36/107 (34%) 1B evaluations, of which 27 (75%) were recommended. For the EDL category, pharmacoeconomic analysis was reported in 20/28 (71%) evaluations, out of which 17 (85%) received a positive recommendation. Univariate analyses for the 1B subgroup showed that NZi recommendations were significantly (α = 0.05) influenced by clinical trials with life-saving primary endpoint (positive), non-inferior trial outcomes compared to placebo (negative) and budget impact below € 2,500,000 (positive). Whereas, the univariate analyses on EDL evaluations demonstrated that ATC-code L (antineoplastic and immunomodulating agents), clinical trials with life-saving primary endpoint and reporting of economic analysis outcomes had a significant and positive impact on the final NZi recommendation. Conclusions: These univariate analyses demonstrated that for 1B and EDL evaluations indication, clinical and economic factors impact the NZi reimbursement recommendations. PHP220 Measuring Extent of Access For Nice Health Technology Assessment Decisions: Trends From 2008 to 2013 O’Neill P , Devlin N Office of Health Economics, London, UK . . Objectives: When assessing trends in NICE HTA decisions it would be useful to ascertain their implications on access for groups of technologies. A specific issue is to understand the degree of access associated with ‘optimised’ decisions, where usage has been restricted to a subgroup of patients relative to the scope of the appraisal. Using a previously developed method, we calculate the degree of recommended access for medicines and assess trends between 2008 and 2013 by therapeutic area and over time. Methods: In a previously published paper we developed a measure, M, to assess access associated with NICE technology optimised appraisal decisions. This was defined as M=(p/P)x100, where M is a measure of the level of patient access (0 equals no access, 100 full access), P is the set of patients considered in the guidance as potential candidates for treatment (given the scope of appraisal and license), and p is the number of patients for whom NICE did recommend. Applying measure M to NICE HTA decisions for medicines between January 2008 and December 2013 we assess trends by therapeutic area and over time. In this paper, to understand trends, we extend the analysis to include recommended and not recommended decisions. We assume a recommended decision scores 100 using measure M, a not recommended decision 0, and optimised decisions, where not possible to determine M, a score of 50. Results: For 201 decisions between 2008 and 2013, on average, M was equal to 52, ranging from 37 in 2008 to 57 in 2011. At therapy level, M scored between 38 for cancer medicines to 100 for Hepatitis C treatments. Conclusions: The results for this period suggest around half of patients have been recommended by NICE to receive treatment, relative to scope of appraisal and license. These considerations address access not implementation issues. PHP221 A Comparison of International Health Technology Assessment Systems – Does The Perfect System Exist? Kerr A 1, Todd C 1, Hebborn A 2, Ulyate K 1 1Roche Products Pty. Ltd., DEE WHY, Australia, 2F. Hoffmann-La Roche AG, Basel, Switzerland . . . . Objectives: There are a number of common elements considered good practice in Health Technology Assessment (HTA) that have been published by organizations representing the field. These components include: clear processes and decisionmaking, including scope for pragmatic approaches and appeal; transparency in methodology, value judgments and decisions; and a facility for stakeholder involvement. The objective of this study was to compare international HTA systems to rank their performance against the ideal components of HTA. Information was also collected on emerging topics such as combined regulatory-payer scientific advice, coverage with evidence, evaluation of drug-diagnostic pairs and disinvestment. Methods: A survey was designed to collect information on the HTA systems in the United Kingdom (UK), France, Germany, Italy, The Netherlands, Sweden, Central Eastern Europe, Canada, Australia, New Zealand (NZ), Korea and Taiwan. Questions were grouped under the topics: process, methods, data, societal input and transparency. The survey was completed by Roche affiliates with first-hand experience working with the HTA system in their country. Results: The majority of countries give consideration to rare diseases and low budget impact with leniency in decision making and/or process. Transparency in decision-making is lacking in many of the countries surveyed. Whilst consumer members sit on decision-making committees in several countries, only the UK involves a group of citizens in setting the decision making criteria applied by the committee. Combined regulatory-payer scientific advice is only available in European countries. Australia is the only country to evaluate drug-diagnostic pairings for both costs and outcomes. Only the UK and NZ have routine disinvestment reviews. Conclusions: Each country is performing well in some elements of their HTA system, but none met all the requirements of an ideal system. HTA systems can learn from the experiences in other countries when considering improvements to processes and efficiency. PHP223 Trends In Early Engagement Between Industry And HTA: Analysis of Scientific Advice Service Provided By Nice Since 2009 Maignen F M 1, Osipenko L 1, Gajraj E 1, Chivers R 2 Institute for Health and Care Excellence, London, UK, 2National Institute for Health and Clinical Excellence, Manchester, UK . . . . . 1National Objectives: Regulatory Scientific Advice (SA) provided by EMA, FDA, MHRA and other agencies is highly demanded by manufacturers but health technology assessment (HTA) scientific advice is still far from becoming a routine step in the product development cycle. NICE has been running an advisory service for 5.5 years. Methods: This work presents analysis of requests to the programme: types of advice projects, number and type of requests per company, clinical indication, stage of clinical development when the advice is sought, reason for seeking advice and current development and regulatory status of products. Results: Between 2009 and 2014 NICE conducted 109 advisory projects (107 medicinal products and two diagnostic tests). 23 of these projects were done in parallel with regulatory agencies and/or other HTA bodies. 78% of all requests were in the following four therapeutic areas: oncology, neurology, rheumatology and cardiology. Majority of products (61%) were in phase II of clinical development when advice was sought. At the time of this analysis, 71 products (66%) were still in development, 6 (5.5%) were subject of a review for a marketing authorisation (MA), 8 (7.5%) had received a MA, the authorisation was not granted to 2 products (2%) and the clinical development was discontinued in 20 cases (19%). Most products that received NICE scientific advice are yet to be referred to the technology appraisals programme. Conclusions: Over the last few years, requests for scientific advice diversified into personalised medicines, regenerative medicines and products for rare and very rare diseases. Most HTA scientific advice requests continue to come from top 20 Pharma companies, however we are starting to see an increasing number of inquiries and project bookings from small-medium size companies. PHP224 Exploring Uncertainty in Economic Evaluation Of Medicines: A Review of The First Manufacturers’ Submissions To the French National Authority For Health (Has) Ghabri S 1, Hamers F F 1, Josselin J M 2, Harousseau J L 1 1Haute Autorité de Santé (HAS), Saint-Denis La Plaine, France, 2Université de Rennes 1, Rennes, France . . . . . . . Objectives: Since October 2013, HAS is required to provide the inter-ministerial Economic Committee on Health Care Products (CEPS) with an economic evaluation on innovative medicines likely to have a significant budget impact on the national health insurance scheme. HAS economic evaluations are based on critical appraisals of cost-effectiveness analyses (CEA) submitted by manufacturers. Exploration of uncertainty around incremental cost-effectiveness ratio is critical to assess the robustness of CEA. Our objective was to assess how uncertainty exploration has been undertaken by manufacturers, using HAS guidelines on economic evaluation as an analytical framework. Methods: Manufacturers’ submissions assessed by end of May 2014 (n= 13) were reviewed. Three sources of uncertainty were considered: uncertainty around model input parameters, uncertainty around model structure and methodological uncertainty. Tools to explore uncertainty included deterministic sensitivity analysis (DSA), probabilistic sensitivity analysis (PSA), as well as overall compliance with HAS guidelines. Results: Model input parameters were the most frequently explored source of uncertainty. Both DSA and PSA were systematically used. However, reporting of DSA varied substantially across submissions, with frequent lack of justification of parameters ranges. Regarding PSA, the choice of distribution was not systematically justified and lacked consistency across similar parameters. Most submissions failed to consider parameters correlations. Exploration of uncertainty around model structure was rarely presented. Where applicable, alternative methods for extrapo- A442 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 lating health outcomes were reported in two-thirds of the submissions. However, worst-case scenarios were hardly presented. Compliance with HAS guidelines for exploring methodological uncertainty (e.g. perspective, discounting, time horizon) was fair. However, the choice of the comparator(s)–an essential component of a CEA–was considered problematic in nearly 40% of submissions. Conclusions: Overall, reporting of DSA and PSA is complying with HAS guidelines. More work is needed to explore uncertainty, in particular, to account for correlations between model input parameters and to enhance the analysis of structural uncertainty. PHP225 Conditional Resolutions in the Amnog Early Benefit Assessment Assmann G , Schmidt U , Drechsler M , Pfannkuche M Boehringer Ingelheim Pharma GmbH & Co. KG, Ingelheim, Germany . . . . Objectives: Since 2011 the early benefit assessment (EBA) is mandatory to obtain reimbursement for new drugs in Germany. By law EBA of the G-BA (Federal Joint Committee) can be conditional for a certain period of time. This study explores the number, duration and rationale of conditional G-BA resolutions. Furthermore the provided, later given (during statement process), accepted and additionally requested study data is analysed in this context. Methods: AMNOG dossiers and resolutions of the G-BA including justifications until June 2014 were reviewed for relevant information and analysed with special regard to the submitted and accepted study data. Drugs which were directly allocated to a fixed-reference price group were excluded. Results: 22 new drugs with conditional resolutions were identified (29.3% of all resolutions; 2012: 6, 2013: 10, 2014: 6). Duration varied between 1 and 5 years (1: 4, 2: 8, 3: 5, 4: 1, 5: 4). Conditional resolutions were most common in cancer (n= 13; 59.1%) drugs. 91.0% had at least a benefit (none: 2, minor: 12, considerable: 6, not quantifiable: 2) in one subgroup. G-BA limited resolutions due to (a) no sufficient data to assess the benefit (n= 9), (b) conditional approval and/or further data requests by EMA (European Medicines Agency) (n=6), (c) expected new study results (n= 6), and (d) due to a formal incorrect dossier (n= 1). To date, only vemurafenib was assessed again with same result as before. Conclusions: From 2011 on an increased number of resolutions and especially cancer drugs were conditional. Only in a few cases G-BA defined in the resolution which specific data has to be shown in the re-assessment. In these cases the requirements concerning further evidence were not part of the G-BA voting. As currently just one drug was re-assessed, no conclusions can be drawn how re-assessment changes the extent or certainty of additional benefit. PHP226 Pricing Additional Benefit in Germany Daniel K , Jaksa A , Ho Y S Context Matters, Inc., New York, NY, USA . . . . Objectives: Gemeinsame Bundesausschuss (G-BA) assesses the additional benefit of newly-approved drugs compared to the G-BA-determined appropriate comparator. As part of the submission, manufacturers provide annual drug cost estimates. This research tests whether manufacturers submit higher annual costs for drugs that provide greater additional benefit, that is, whether added benefit is priced. Methods: The influence of G-BA’s additional benefit assessment on the manufacturer’s estimated per patient annual drug cost was estimated via ordinary least squares based on 73 reviews. The model also included controls for the (log) size of the target population and the annual cost of the comparator. These variables were collected from the Federal Gazette publication or the “Beschluss” document. Results: Our model explains three-quarters of the variation in annual drug cost (R2= .76). Conditional on the target population and the cost of its comparator, a drug assessed as having unquantifiable or minor added benefit is estimated to have a per patient annual cost about 2 times greater than one with no added benefit (p= .02). A drug with the highest assessment, “Considerable additional benefit,” is estimated to have an annual per patient cost that is 9 times greater (p< .01). There is evidence of a “quantity discount”: doubling the size of the target population is estimated to reduce the per-patient annual cost by 18% (p< .01). Conclusions: If, as seems plausible, manufacturers’ assessments of their own drugs are aligned with G-BA’s assessments, our results are consistent with manufacturers setting higher prices for more beneficial drugs. The price a manufacturer uses to estimate annual cost represents in effect a proposed allocation of its drug’s benefits between itself and patients/payers. This research raises at least two important questions: How do manufacturers propose to share the gains and what sharing emerges from subsequent price negotiations? PHP227 Risk of Bias in Trial-Based Economic Evaluations Adarkwah C C , Evers S M , Hiligsmann M Maastricht University, Maastricht, The Netherlands . . . . . Objectives: The objective of our research is to give first an overview of the risks of bias in trial-based economic evaluation, and second to identify how key sources for bias can be revealed and overcome (bias-reducing strategies) in future trialbased economic evaluation in the fields of general practice and health psychology. Methods: A scoping review was performed using PubMed and the NHS Economic Evaluation data base. It was complemented with experiences of the authors. Sources of bias in trial-based economic evaluation as well as bias-reducing strategies are discussed. Results: The different forms of bias are presented, and assigned to a particular trial phase. A distinction is made between pre-trial biases, biases during the trial and biases that are relevant after the actual trial. All potential forms of bias are discussed in detail and strategies are shown to detect and overcome these biases. Conclusions: In order to avoid bias in trial-based economic evaluations, one has to be aware of all of the possible forms of bias and all stakeholders have to examine trial-based economic evaluations in a rigorous and stringent manner. Our research findings can be helpful in this examination as they provide an overview of the possible biases which researchers should take into account. PHP228 UK Value-Based Assessment: Will Scotland’s Smc Approach It In The Same Way? Hamilton M L Access Partnership, London, UK . . Objectives: In May 2010 the UK government set out its intentions to reform the current method of pricing branded medicines and introduced a new system of value-based pricing (VBP). This was to replace the Pharmaceutical Price Regulation Scheme (PPRS) which expired at the end of 2013. With discussion still ongoing, and the PPRS renewed up until end of 2018, it has been indicated that VBP will no longer relate to medicines’ pricing, but instead will be a reshaping of the Health Technology Assessment (HTA) model employed by the UK in its appraisal of new medicines, and renamed Value Based Assessment (VBA). As medicine assessment is devolved in the UK, performed by NICE in England and the SMC in Scotland, we seek to understand whether the SMC will adopt the same approach to VBA as NICE. Methods: The research was conducted through in-depth secondary research and interviews with stakeholders, including payers and KOLs, in UK. Results: Both NICE and the SMC have indicated they will continue to use QALY as a measurement of clinical and cost effectiveness while also incorporating issues such as burden of illness and wider societal impact in their assessment. However, their approach to conducting such assessments may differ with the SMC suggesting using a new system of patient and clinician engagement (PACE), currently in use for the appraisal of medicines for end of life or rare conditions (orphans), as a wider process to determine Scotland’s requirement for a value based approach to assess all new medicines. Conclusions: Manufacturers would be encouraged to closely follow the outcomes from the new PACE system, incorporated into the SMC assessment for end of life and orphan therapies, to ensure readiness for the introduction of VBA in Scotland. PHP229 Discrepancy Between National Drug Recommendations And Local Uptake in the Swedish Inpatient Sector McGee M A 1, Izmirlieva M 2, Ando G 2 Life Sciences, London, UK, 2IHS, London, UK . . . . 1IHS Objectives: The current study seeks to assess discrepancies between national inpatient drug recommendations issued through the cost-effectiveness pilot project (Klinikläkemedelsprojektet) in Sweden and local uptake reflected by Stockholm Country Council´s (SLL) procurement activity. Methods: Secondary research investigated the 15 inpatient drug recommendations issued by the national New Pharmaceutical Therapies group (Nya Lakemedels Terapier, NLT) in 2013 and compared these against the SLL’s 2014 procurement list. Results: In 2013, the NLT group issued a total of 15 recommendations for products in a variety of therapeutic areas (TA). Out of these, ten gained positive national decisions, 50% of which were for oncology drugs. Of those recommended, cancer drug pertuzumab (Perjeta; Roche, Switzerland) saw the highest incremental cost-effectiveness ratio (ICER), at SEK2,565,000 (USD383,730) per quality-adjusted life year (QALY) - greatly exceeding the informal cost-effectiveness threshold of SEK800,000 in Sweden. There was a significant discrepancy in the number of nationally recommended drugs (ten) in 2013, versus the SLL uptake (two) by 2014. The two recommended products subsequently procured were Roche´s (Switzerland) ophthalmology drug ranibizumab (Lucentis) and Takeda´s (Japan) oncology therapy brentuximab vedotin (Adcetris). Conclusions: The Swedish government is contemplating making the Klinikläkemedelsprojektet permanent, and as such the alignment between national and local level priorities is a critical component of its integrity. The results show that eight nationally recommended products had not been procured, indicating that there is still a discrepancy between those recommendations and uptake at the local level, based on the current data. One explanation for the discrepancy could be attributed to the funding system, where local payers are responsible for funding inpatient drugs. PHP230 HTA Approach in Italy. Structure, Methods, And Process Of Veneto Region’s Evaluation Of Pharmaceutical Effectiveness Unit (Unità Di Valutazione Dell’efficacia Del Farmaco, Uvef) Young K E 1, Urbinati D 2, Toumi M 3 1Creativ-Ceutical, Milan, Italy, 2Creativ-Ceutical, Luxembourg, Luxembourg, 3University Aix-Marseille, Marseille, France . . . . Objectives: Veneto leads the nation in terms of better health status and pharmaceutical expenditures within budget. This study aimed to assess the region’s Health Technology Assessment (HTA) approach (structure, methods, and process of conducting HTA) by observing Evaluation of Pharmaceutical Effectiveness Unit’s (UVEF) evaluations on drugs, the concomitant Regional Technical Commission (CTR) decisions on formulary inclusion, and the evidences considered. Methods: We reviewed all HTA reports in the UVEF website evaluating the HTA dimensions assessed, if they were in line with the HTA scoping document by the Regional Decrees (2008), and changes over time. Further, this research evaluated the presence of explicit explanations behind each CTR decision. Results: 223 HTA listings were retrieved in the UVEF website (2004-2011). 50.23% were published before the actual establishment of UVEF (2008) and were probably labelled HTA retrospectively. Fewer dimensions were assessed in the earlier documents and the most extensive changes were seen in 2008. The most assessed dimensions were Efficacy, Safety, and Costs and the least assessed were Budget Impact Analysis, Target Population, Place in therapy, and Assessment of Innovation. The Regional Decrees (2008) were impactful on the structure of the HTA reports, however these are still not fully compliant to the scoping document. A total of 72 CTR decisions were found in the website from 2009 to 2011. On average, 92% of all CTR documents have a documented explanation behind each decision. Conclusions: The HTA approach in Veneto is evolving towards the development of uniformity in the documents and a single transparent approach. The upward trend of attaching CTR decisions with explicit explanations A443 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 shows a positive development within the region’s HTA. The Regional Decrees were impactful however the HTA reports are not fully compliant to the scoping document and it is central to understand the reason behind the challenge. PHP231 Quality Of Life – A Rarely Acknowledged Key Category Within The Amnog Process in Germany Eheberg D 1, Shlaen R 2, Batscheider A 2, Gohlke A 2 1IMS HEALTH, Munich, Germany, 2IMS Health, Munich, Germany . . . . Increasingly, quality of life becomes a more important part of the HTA assessments of new products. Assessing quality of life is trickier than changes in clinical parameters as changes of quality of life tend to occur slower and with a high level of variance. Additionally, it is sometimes necessary to assess quality of life indirectly, especially in cognitive disorders. However, the available data are often driven by the requirements for marketing authorization and rarely fit to the requests of HTA agencies. Quality of life is, among mortality, morbidity and safety, one of the key patient relevant outcomes categories in the AMNOG (Arzneimittelmarktneuordnungsgesetz; Law on the Reorganization of the Pharmaceutical Market) process in Germany. Objectives: Our focus was to review the impact of quality of life data presented by manufacturers since the introduction of the AMNOG in 2011 on the level of additional benefit claimed by the manufacturer and the evaluation by the Federal Joint Committee (G-BA) and the Institute for Quality and Efficiency in Healthcare (IQWiG). Methods: We screened the IMS HTA database and the assessments published on the G-BA website for assessments including additional benefit claims based on quality of life. We compared these with the corresponding IQWiG reports and the final decision of the G-BA (if available). Results: Data on quality of life was part of 36 additional benefit claims. In most cases, IQWiG (n= 23) and G-BA (n= 28) considered quality of life data as well. However, in 6 cases IQWiG and G-BA assessed quality of life, even though the manufacturer didn’t include these data. There are only few additional benefit claims acknowledged by G-BA based on quality of life. Conclusions: Even though quality of life is seen as highly relevant factor for the HTA assessment of a new drug or technology, it is rarely taken into consideration. PHP232 Understanding The Role of Subgroup Analysis And Tests For Homogeneity or Interaction in the Amnog Dossier Hofmann-Xu L 1, Bonduelle D 2, Eheberg D 3 1IMS Health Germany, Munich, Germany, 2IMS HEALTH GmbH & Co. OHG, Munich, Germany, 3IMS HEALTH, Munich, Germany . . PHP233 The Inverse Correlation Between Internal And External Risk Under International Reference Pricing: An Analysis of Six European Countries . . PHP234 An Archetype For Classification and Comparison of Hta Activities in Latin America Skaltsa K 1, Allen N 2, Van Engen A 3, Blogg K 4 1Quintiles Consulting, Barcelona, Spain, 2Centre for Innovation in Regulatory Science, London, UK, 3Quintiles Consulting, Hoofddorp, The Netherlands, 4Quintiles Consulting, Reading, UK . . . . Objectives: HTA processes are being implemented in Latin America in an effort for a more efficient resource allocation. The objective of this study was to explore the HTA environment and classify its diversity based on previously developed taxonomies. This enables the comparison within the region, as well as with other regions in which similar maps have been developed, such as Europe. Methods: Two taxonomic groupings were used as described in Allen et al 2013 Health Policy 113; 305– 312. The first one is based on the position of a national HTA agency in relation to the position of the regulatory and the coverage body. The second set focuses on the key tasks performed by the HTA agency. The following countries were examined: Mexico, Cuba, Costa Rica, Colombia, Venezuela, Ecuador, Peru, Bolivia, Brazil, Uruguay, Argentina and Chile. No information on HTA activity could be identified for the rest of the countries on official websites or publications. Results: Seven different archetypes were identified by combining different values of the two taxonomic sets in these twelve countries. There were two main groups identified: one consisting of Brazil, Cuba and Chile where the HTA and coverage decisions are performed by the same agency and regulatory is independent, and the other one consisting of Costa Rica, Venezuela, Peru and Bolivia where no HTA process was identified and external evaluations or decisions from reference countries are used to inform decisions. Conclusions: Although the landscape is expected to change in the coming years, this first high-level overview shows that currently, there is a high degree of heterogeneity in HTA processes in Latin America despite the fact that these countries tend to share expertise in various levels, with few countries following similar processes and a number of countries with no HTA bodies in place at any decision level. . Objectives: Although subgroup analysis in clinical trials is often criticized, it is still considered an important part of the AMNOG Dossier when describing effect modification in different patient groups. An identification of effect modification in the subgroups will either support or weaken the total additional benefit (Zusatznutzen) of a drug. The purpose of this article is to give a detailed background with regard to the statistical inference in subgroup analysis and a brief review of the effect of tests for homogeneity/interactions on the final grading of the additional benefit, according to the decision from IQWiG. Methods: This article covers: Understanding tests for homogeneity, individual versus pooled data and influence of subgroup analysis on recent IQWiG benefit assessment. A research of the recently published AMNOG Dossiers was performed. A description was given to the number of drugs that had an additional benefit and the subgroups that were involved. Clustering analysis was performed to investigate the hidden structrue in the subgroups. Regression models were used to analyze the relationship of the subgroups and the additional benefit. Results: Some subgroups such as age, gender and weight play a major role in the AMNOG assessment. Other subgroubs are more specific for certain disease areas e.g. BMI or hypertension for diabetis. In a worst case the benefit of a drug may disappeare when a certain subgroup is taken into consideration. Conclusions: Subgroups exert a profound influence upon the overall effect of a drug. Despite of the weakness of the statistical inference, subgroup analysis plays an important role in the AMNOG Dossier. Some subgroups are frequently related to the overall effect of a drug and some might even change the whole story in the AMNOG Dossier. In conclusion, subgroup analysis should be understood properly and the results should be interpreted carefully. Marinoni G , Lockwood C , Ando G IHS, London, UK ments in the internal risk index. Conclusions: The inversion seen between the two risk indices reflects a key aspect of the current IRP landscape. Namely, mature markets exert considerable influence on prices internationally, while their use of IRP exerts limited downward pressure on their own prices. Conversely, while IRP policy in emerging markets is more likely designed to secure the lowest possible prices, these markets carry less influence on prices internationally. . Objectives: To index six countries on the basis of the risk they pose to pharmaceutical prices with regards to international reference pricing (IRP), from both an internal (how IRP is used by the country) and external (how other countries use this country for IRP) perspective. Methods: Details on IRP methodologies were obtained from primary and secondary research. Achievable drug price levels in Bulgaria, France, Germany, Portugal and Romania were derived from these markets’ IRP formula and the relative drug price levels in their reference markets (based on existing literature). Furthermore, based on the IRP formulas and relative drug price levels of each of the markets referencing these five and the United Kingdom, the markets most likely influenced by the six were identified. This number of markets, the fraction of them representing major pharmaceutical markets, and the relative price levels in each of the six countries were assigned weightings to rank them by external risk. Results: While Bulgaria, Portugal and Romania represent the greatest internal risk for pharmaceutical prices (i.e. greatest risk of obtaining a low price as a result of IRP), these markets fall into the low and moderate risk segments in the external risk index. Conversely, while France, Germany and the UK pose the greatest external risk (i.e. as a function of the number of significant markets referencing them), they fall into the low and moderate risk seg- PHP235 Combining Headroom and Return on Investment Analysis To Rank Potential Commercial Value of Six Medical Devices in Development Markiewicz K 1, van Til J A 1, Steuten L M G 2, IJzerman M J 3 1University of Twente, MIRA Institute for Biomedical Technology & Technical Medicine, Enschede, The Netherlands, 2University of Twente, Enschede, The Netherlands, 3MIRA Institute for Biomedical Technology & Technical Medicine and University of Twente, Enschede, The Netherlands . . . . . . . . Objectives: The development process of medical devices strongly depends on the financial resources available and the expected return on investment to manufacturers. The aim of this paper is to analyse the potential commercial viability of two disruptive and four incremental medical devices in different stages of development. Methods: The headroom method combined with the return on investment analysis was performed for one therapeutic and five diagnostic devices for different clinical target areas. Information regarding maximum additional benefit that could be obtained with new device, the estimated production price and expected sales volume was gathered from literature and expert opinions. A willingness-to-pay threshold for one additional QALY of € 30,000 was assumed for headroom analysis. Results: The devices were ranked according to their potential commercial viability. The analysis showed that two disruptive and two incremental devices had reasonably good balance between headroom and unit cost, and two incremental devices had no good balance. The device with the highest potential commercial viability was a disruptive therapeutic device for the cartilage repair treatment in the first clinical trial stage, with estimated headroom for the cost of the new treatment: € 74,600 and an expected production cost of the therapy: € 8,000 per unit. The market volume size was calculated based on the incidence of cartilage defects: 65% in routine knee arthroscopies. The disruptive diagnostic device for home brain monitoring of epilepsy patients in the prototype stage of development had the lowest potential commercial viability, with an estimated headroom of € 81,000 and an expected production costs per unit: €120,000 that resulted in the unfavorable return on investment. Conclusions: The headroom method combined with a return on investment analysis, offers insight in the potential commercial viability of medical devices under development. The research on the impact of that analysis on actual R&D decision making will still be determined. PHP236 Transferability of Economic Evaluations To Central and Eastern Euroepan and Former Soviet Countries Mandrik O 1, Knies S 2, Kaló Z 3, Severens J L 4 1Institute of Health Policy & Management, Erasmus University Rotterdam, Kiev, Ukraine, 2Institute of Health Policy & Management, Erasmus University Rotterdam, Rotterdam, The Netherlands, 3Eötvös Loránd University (ELTE), Budapest, Hungary, 4Erasmus University Rotterdam, Rotterdam, The Netherlands . . . . . Objectives: Data availability and decision makers methodology requirements are two important factors influencing transferability of economic studies. Applying qualitative assessment of experts’ opinion and systematic review of published economic studies, we aimed to analyze transferability of economic evaluations in Central and Eastern European (CEE) and former Soviet countries. Methods: Firstly, eleven reimbursement experts from eight countries were interviewed on their background and current practice of using economic evaluations, opinion regarding transferability of economic evaluations and importance of individual Welte’s transferability factors. Secondly, we analyzed peer-reviewed English-language economic evaluations A444 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 of sixteen countries from the region according to different criteria: characteristics, quality (using Drummond’s checklist), use of local data, addressed inputs limitation and results transferability. Results: Economic evaluations are used in CEE countries for informing decision making, while critically considering methodology, quality and study’s reliability. Experts acknowledged limited generalizability of study results both between and within geographic regions. Meanwhile, despite these constraints, facing limited health technology assessment (HTA) capacity experts were still using foreign evidence. At the same time, the usefulness of studies published in CEE and former Soviet countries to inform their decision making is limited because of insufficient transparency in reporting, unaddressed uncertainty, limited insight on inputs and transferability of results. Although local costs, baseline risk and resource use data are required, experts accept evidence originating from health care settings outside CEE and former Soviet countries regarding relative effect and utilities values. Conclusions: HTA priority setting and transferability assessment of economic evidence are important issue in health care decision making in CEE and former Soviet countries, since HTA research capacity is limited and local evidence is scarce. For this purpose, quality, transparency, and transferability should be addressed explicitly in published economic evaluations originating from CEE and former Soviet countries. PHP237 Ten Years Of Development Studies In Health Technology Assessment In Brazil: Profile Of Studies And Operational Indicators Koury C D N 1, Elias F T S 2 1FIPE -Fundação de Ensino e Pesquisas Econômicas, Brasilia, Brazil, 2Ministry of Health of Brazil, Brasilia, Brazil . . . . . . Background: The Department of Science and Technology (DECIT) at the Ministry of Health (MoH), since 2003, has financed studies to support demands from the MoH technical areas regarding the decision making process about health technologies. Objectives: To analyze DECIT performance in financed Health Technology Assessment (HTA) studies to improve the capacity building of HTA in Brazil. Methods: A retrospective descriptive study based on the analysis of documents and official records built in a single database in Excel containing the studies promoted by DECIT from 2003 to November 27, 2013. The variables pre-classified and collected from Brazilian Network for Health Technology Assessment’s database, SISREBRATS, were revised by two reviewers and conferred or supplemented with data from the Database of Health Research, Annual Reports Management by DECIT, and Final Reports and Cooperation Agreements by co-financier partners. Results: 284 HTA projects were financed entirety for R$ 25 million (± R$ 1,69 million) between 2003 and 2013. The average value financed was R$ 346 thousand and the average bias was ± RS 61 thousand. Of these, 57% (162/284) are pharmacotherapy; and 35% (99/248) are systematic reviews. The majority of the studies 58% (165/284) were financed by MoH calls in years 2005 to 2010 and contributed to 110 masters and doctors conclusions in HTA landfield. Conclusions: These numbers show despite the HTA culture in Brazil having been in its initial stage, DECIT is increasing its capacity to respond to demands with a variety of studies types, specily in pharmacotherapy. However, evaluation demands do not always correspond to health needs of promotion and prevention. PHP238 The Definition and Role of Quality of Life In Germany’s Early Assessment of Drug Benefit Lohrberg D , Augustin M , Blome C University Medical Center Hamburg-Eppendorf, Hamburg, Germany . . PHP241 Romanian Quick-HTA Development In 2013 Radu C P , Cernea R Roche Romania, Bucharest, Romania . . . Objectives: The Objectives are to present the characteristics of the quick-HTA started in June 2013 and to show the Results of the first quick-HTA process in Romania. Methods: The health care context and the Romanian legislation covering HTA were studied by considering: the reasons behind HTA introduction, the key stakeholders and the HTA process. A critical appraisal was done covering public HTA reports and the decisions taken by the Ministry of Health (MofH). Results: The introduction of quick-HTA began mid-June 2013 when MofH released the legislation and the HTA Unit started receiving dossiers for new drugs/indications. The HTA model was based on a “score-card” system with 6 criteria: HAS (France) opinion, NICE/SMC/AWMSG (UK) opinion, the number of EU countries with reimbursement, relative efficacy, relative safety and relative patient-reported-outcomes (PRO), but no role of budget impact. From June to December 2013, 167 HTA dossiers were evaluated by the HTA Unit with an acceptance rate of about 80%. Most of the drugs accepted for reimbursement were oncological (23%); other main therapeutic areas were diabetes with 16 drugs/indications receiving positive evaluations, rheumatology (14), oncohaematology (8) and neurology (7). The HTA included also biosimilars, all 4 of them receiving positive decisions. Unfortunately, early 2014, the new Government abrogated this HTA legislation and the already-published HTA reports, claiming that the process didn’t mentioned the criteria for de-listing reimbursed drugs nor the budget impact, within HTA reports. Moreover, the HTA process was moved into the responsibility of the National Agency for Drugs. Conclusions: The implementation of the quick-HTA in Romania took a good start, using a mixture of information, from benefits and costeffectiveness in other countries, to relative effectiveness, safety and PRO. However, the lack of consideration for the local context and the political disagreements led to a temporary suspension of this quick-HTA process in Romania. . Objectives: Since January 2011, pharmaceutical manufacturers are obliged to submit dossiers presenting benefits over comparative treatment when launching new products in the German pharmaceutical market. These dossiers are usually evaluated by the Institute for Quality and Efficiency in Health Care (IQWiG). Based on the dossier, the IQWiG’s evaluation and a subsequent hearing process, the Federal Joint Committee (G-BA) decides on additional therapeutic benefit related to mortality, morbidity, quality of life (QoL) and side effects. As QoL is among the more contested of these criteria and remains comparably unspecified, this qualitative study’s aim was to analyze definitions and the role of QoL in early assessments of benefit (EAB) in Germany. Methods: As most of the documents are freely available on the G-BA’s website, this study included all relevant documents of the first completed 66 assessments (11.2011-12.2013). We conducted a qualitative content analysis screening the dossiers, IQWiG evaluations, G-BA decisions and the protocols of the oral hearing for relevant links to QoL and synonyms. In a process of independent analysis and joint consensus building by two researchers, relevant text passages were extracted and reduced to key content on the term’s usage. During analysis, a system of codes was developed accounting for a wide variety of recurring QoL-related definitions and references to its importance. Results: Even though key players did not necessarily share the same QoL-definition, the concept’s relative importance was highlighted in numerous references. G-BA decisions criticize the lack of or the inadequate presentation of QoL data in the manufacturer’s dossiers. G-BA and IQWiG apply a narrow understanding of QoL, while manufacturers failed to establish wider notions of QoL linking factors such as patient satisfaction to the concept. Conclusions: QoL in a particular sense is of pivotal importance in Germany’s early assessment of benefit. The demand for reliable QoL data is growing. PHP240 An Update on Clinical And Economic Evidence Requirements for Advanced-Therapy Medicinal Products in Europe Anastasaki E 1, Walker A 2, Bradshaw S 1 1Market Access Solutions LLC, London, UK, 2University of Glasgow, Glasgow, UK . models that Payers have come to expect at the launch of pharmaceuticals. Thus, manufacturers face difficulties when negotiating the price for ATMPs and may not be able to realize their full potential. This paper aims to use case studies of ATMPs launched in EU5 to outline access pathways and review the clinical and economic evidence requirements. Methods: Secondary research identified ATMP approvals since 2008 and a framework was created to develop hypotheses on clinical and economic evidence requirements, considering alternative routes to market. Hypotheses were then validated during in-depth interviews with key stakeholders across EU5. Results: Payers are yet to be convinced about the full benefit of ATMPs and are reluctant to pay premium prices if they are not sure about long-term efficacy and safety. Additionally, in their cost-benefit analyses, Payers tend to consider only the direct costs associated with a condition, ignoring broader societal benefits and savings in the long run. Therefore, price negotiations are sometimes lengthy. Risk sharing and novel payment-by-result schemes are often agreed to mitigate risks. Conclusions: Payers are not yet familiar with the potential value of ATMPs, and, in most cases specific evaluation criteria don’t exist. Manufacturers need to invest in educating Payers on the huge differences between ATMPs and traditional therapies, particularly to show that manufacturing costs are substantial, and work together to identify relevant measures for clinical and economic evaluations of this new therapy class. . . Objectives: Advanced-therapy medicinal products (ATMPs), such as gene therapy, cell therapy and tissue engineering are a new class of medicines in the EU. ATMPs offer prospects in prevention and treatment of fatal and/or chronic debilitating diseases where no effective treatments exist. However, with complicated mechanisms of actions and benefits often being anticipated in the longer term, it is challenging to demonstrate hard clinical evidence and create robust cost-effectiveness PHP242 Hta In The Brazilian Health Care System and Potential Lessons Learned For Other Brics States Stuwe L T 1, Bellanger M M 2, Picon PD 3 University, Paris, France, 2EHESP French School of Public Health, Rennes, France, 3Federal University of Rio Grande do Sul, Porto Alegre - RS, Brazil . . . . . 1Pierre-and-Marie-Curie Objectives: The objective of the study was to provide an external assessment of recent HTA institutionalization in Brazil, and identify a set of lessons learned potentially applicable to BRICS States. Methods: This research is based on a quantitative and qualitative assessment. A literature survey between 2000 and 2014 was conducted in English, Spanish and Portuguese in PubMed/Medline, Science Direct, LILACS and SciELO. Epidemiological and socio-economic data was retrieved from national health accounts as well as WHO/PAHO, OECD and World Bank. For the Brazilian National Committee for Incorporation of Technologies (CONITEC), available reports on the incorporation of medicines into the National Unified System (SUS) for the first two years of operation (2012 and 2013) were analyzed. A matrix containing quantitative and qualitative criteria was elaborated to analyze reports by the outcome of decision, therapeutic class, author (s) of the request and public consultation. Results: A total of 92 available CONITEC reports for 2012 (n= 38, 33 for medicines) and 2013 (n= 54, 42 for medicines) were analyzed. 45% of reports on medicines recommended incorporation into the SUS. Most of the positive recommendations were clearly related to public health priorities as identified by the government, translating a strong commitment for improved access to medicines within the SUS i.e. anti-cancer drugs. Overall, the creation of the CONITEC represents a substantial step toward the institutionalization of HTA, with more transparency and accountability in decision-making processes, considering ethical, organizational, social, and legal aspects. Conclusions: Whereas lowest in Russia, India and South Africa, and at a transitional stage in China, Brazil has a comparable degree of institutionalization of HTA as countries with a long-lasting HTA experience. A best-practice assessment in the area of HTA within the BRICS has still to be elaborated. Transferability of lessons learned might be a strong tool for improving HTA development within the BRICS. PHP243 Is G-Ba Strategically Discounting The Benefit Assessment of Relatively High Cost Drugs? Jaksa A , Daniel K , Bergemann R , Ho Y S Context Matters, Inc., New York, NY, USA . . . . . A445 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Gemeinsame Bundesausschuss (G-BA) states that it assesses additional benefit strictly on clinical grounds, but it also requires that manufacturers submit drug and comparator costs. This raises the possibility that G-BA’s assessment might be influenced by price, possibly to provide leverage during subsequent price negotiations. This research tests the hypothesis that high cost drugs (relative to the comparator) are more likely to receive poor benefit assessments. Methods: The following variables were collected from the Federal Gazette publication or the “Beschluss” document: additional benefit assessment, annual cost per patient of drug and comparator, and estimated target population. The Scottish Medicines Consortium (SMC) clinical rationale for the same drugs and indications were collected to control for clinical efficacy. After excluding orphan drugs, reviews using best supportive care comparators, and reviews without SMC reviews, 58 reviews remained for analysis. G-BA’s additional benefit assessments were ranked from least benefit to most. The influence of drug cost relative to the comparator on the G-BA assessment was estimated via an ordered logit model. The model also included controls for the (log) size of the target population and clinical efficacy (SMC’s clinical assessment). Results: An increase in the cost difference between the drug and the comparator is estimated to result in a modest, statistically significant increase in the odds of receiving an additional benefit assessment greater than a “no additional benefit” assessment. Conclusions: Our results are inconsistent with the alternative hypothesis that G-BA is strategically discounting its assessment of relatively high cost drugs. The positive estimated relationship is consistent with manufacturers’ setting higher prices for more beneficial drugs (The data available provide no way to statistically account for this plausible source of endogeneity). Our results provide no support for rejecting the null hypothesis that G-BA assesses added benefit independently of drug cost. PHP244 Development of Hta in Turkey Tuna E 1, Tatar M 2, Ergin G 1, Senturk A 1, Atikeler K 2 1Polar Polar Health Economics & Policy, Ankara, Turkey, 2Hacettepe University, Ankara, Turkey . . . . . Objectives: The aim of this study was to assess the development of HTA in Turkey. Methods: In this regard, organization structures of the Ministry of Health (MoH) and Social Security Institution (SSI) and presentations of first HTA meeting held in April 2014 have been analyzed. Results: There are three main HTA agencies in Turkey. One is under the payer institution called SSI. The HTA committee of SSI assesses all the new health technologies to define whether they will be reimbursed or not. In other words, this committee is the major decisive HTA committee. Other two HTA committees are under the MoH. One of these is under the General Directorate of Pharmaceuticals and Medical Devices. This committee assesses certain drugs which are specifically asked to be evaluated by the SSI, MoH or other Ministries. One of the projects completed by this committee is the evaluation of top 100 selling drugs according to the effect of price, regulation, market and qualitative characteristics of drugs. The second committee of the MoH is under the General Directorate of Health Research. This committee assesses more general issues like obesity, KOAH etc., instead of certain health technologies and publishes national reports. One of the reports published by this committee was the importance of obesity surgery in the treatment of obesity. In addition to all these three committees, HTA studies also being carried out by a MoH hospital called Ankara Numune Training and Research Hospital (ANHTA). They have been working on hospital based HTA. Conclusions: Despite valuable studies being conducted as stated above, HTA is still in its infancy in Turkey and compared to other EU countries like Germany, UK etc. there is not an autonomous HTA agency. There are more than one committee, working on different aspects of health technology assessment under the supervision of government. PHP245 Disease Burden in Brazil and Health Technology Assessment: A Retrospective of Ten Years of Supporting Koury C D N 1, Elias F T S 2 1FIPE -Fundação de Ensino e Pesquisas Econômicas, Brasilia, Brazil, 2Ministry of Health of Brazil, Brasilia, Brazil . . . . . PHP247 Correlation Between End-of-Life Status of A Treatment And Likelihood of A Patient Access Scheme in The Setting of A Nice Review in The UK Izmirlieva M , Ando G IHS, London, UK . . Objectives: This study aims to assess the existence of a correlation between the applicability of end-of-life treatment criteria and the likelihood of NICE requiring a Patient Access Scheme (PAS) to recommend the treatment for funding. Methods: A review of all patient access schemes in existence as of March 2014 for NICE-recommended drugs was conducted to assess how many of those were for medicines which met the end-of-life treatment criteria and whether the supplementary criteria for end-of-life treatments had any bearing on the final NICE recommendation. Results: In total 42 PAS were identified. Of those, end-of-life treatment criteria were met and had bearing on the final NICE guidance in 7 cases (16.7%). End-of-life treatment criteria were considered but were not met in full in the case of 3 NICE reviews (in one of the three NICE considered that end-of-life criteria were not met in another review, even though the manufacturer had not applied for those criteria to be considered in the present review). End-of-life treatment criteria were also considered for one additional review where they were a focal point of the manufacturer appeal against the NICE guidance. In one additional case, end-of-life criteria were applied for but had no bearing on the final NICE guidance as the cost-effectiveness threshold was met without the application of special considerations. Conclusions: Given the high cost of drugs meeting end-of-life criteria (most of which are for oncology indications), as expected, many of them are subject to a PAS in the UK. However, the opposite correlation does not hold true - i.e., the requirement for a PAS in the UK is not restricted to end-of-life treatments. HEALTH CARE USE & POLICY STUDIES – Patient Registries & Post-Marketing Studies . Background: Defining health technology assessment priorities has been a challenge for the Department of Science and Technology who adopted a prioritization criteria strategy (epidemiologic relevance, services/policy relevance, state of the art, operational feasibility and social demand) for demands from MoH technical areas. However, evaluation demands do not always correspond to health needs. Objectives: To analyze the relationship between projects financed from 2003 to 2013 and disease burdens in Brazil. Methods: Systematization of the summaries from financed projects through searches in the information from www.saude.gov.br/ rebrats and http://pesquisasaude.saude.gov.br/bdgdecit/ and categorization according to the twenty sub-groups of diseases and injuries of disease burden (Dalys) research in Brazil (SHARMM ET al, 1998). Results: 284 HTA projects financed between 2003 and 2013. Of these, 24% (68/284) apply to the twenty main causes of loss of life years by premature death or incapacitation (Dalys, 1998). The first three largest are equal to 15% – diabetes, coronary ischemia, acute myocardial infarction, angina, cerebral infarction and stroke– corresponded to 13% of the (37/284) projects. Conclusions: The percentage found allows for the questioning of the prioritization starting point where the criteria are applied to subjects selected due to external influence and pressure from the market and not necessarily based on the needs impacting the population’s health. Uniting the two dimensions while also taking into account the strategic innovations in order to prioritize and finance assessment will be important for health systems sustainability. PHP246 First Experiences With The New Testing Examination and Treatment: Methods in Germany: Is This A New Amnog Clone? Bonduelle D 1, Eheberg D 2, Plantoer S 1 1IMS HEALTH GmbH & Co. OHG, Munich, Germany, 2IMS HEALTH, Munich, Germany . Objectives: In a new regulation the Federal Joint Committee (G-BA) can pass directives for testing examination and treatment methods with not yet sufficiently proven benefit, but which show potential as essential treatment alternatives (§137e SGB V). The objective of the present study was to compare the requirements for a successful application with the existing AMNOG (Law on the Reorganization of the Pharmaceutical Market) HTA requirements in Germany. The applicants must submit valid data on the potential of the method in question, among other requirements. In one of the first applications in Germany, done by the authors, we see a lot of parallels with the HTA process for drugs but also a lot of uncertainties. Methods: We compared the requirements of the IQWIG Methodology paper 4.2 and the G-BA rules of procedure for the early benefit assessment for medical drugs (§35a SGB V) with the new potential analysis for examination and treatment methods. This analysis was made for different criteria’s like study and endpoint design, certainty of results of the studies and others. We used our business case as template to extract the key-learning’s and identify the pitfalls in the new process. Results: The new legislation will have a strong impact on the study design and evidence to show the potential of new examination and treatment methods as essential treatment alternatives. A lot of evaluation criteria’s came from the drug assessments but are hardly applicable to proof the potential of new examination and treatment methods. Conclusions: The legislation uses parts of classic HTA assessment on medical drugs to evaluate the potential of new examination and treatment methods. In most cases this is not possible and will decrease the level of evidence of available clinical data for new examination and treatment methods due to grey zones and loopholes in the legislation. . . PHP248 The Costs and Effects of Post-Authorisation Safety Studies For New Active Substances Bouvy J C 1, De Bruin M L 1, Hoekman J 1, Stolk P 2 University, Utrecht, The Netherlands, 2University of Utrecht, Utrecht, The Netherlands . . . . . . 1Utrecht Objectives: At market entry, there usually is uncertainty regarding a new medicine’s benefit-risk profile. Therefore, regulatory authorities may request additional pharmacovigilance (PhV) activities. Regulatory Authorities can request a PostAuthorisation Safety Study (PASS) such as a registry, database study, survey, or clinical trial to reduce the uncertainty regarding certain safety risks. We aimed to assess the costs and effects of PASS for centrally approved new active substances (NAS) in Europe in 2007. Methods: We compared two scenarios for all NAS (n= 47): (1) Full regulation: routine PhV activities (spontaneous adverse drug reaction (ADR) reporting) with additional PASSs for some NAS; (2) Limited regulation: only routine PhV activities. For a follow-up period of six years after marketing we assessed the safety-related labeling changes for NAS and identified the source of these changes (PASS, spontaneous ADR reporting or other). Data on labeling changes was extracted from the European Medicines Agency’s website. A survey among pharmaceutical companies was used to estimate the costs of all requested PASSs. Results: For 23 of the 47 NAS, at least one PASS (33 PASS in total) was requested in 2007. After six years, on average 8.1 safety-related labeling changes were identified per NAS. Requested PASS were the source of ~4% of all cases of new safety information identified. The total estimated costs of the 33 requested PASS were between € 50 and € 150 million. Conclusions: For the 2007 cohort of NAS approved in Europe, the total costs of all requested PASS were substantial and yet these PASS contributed to the identification of only 4% of all new safety information identified post-marketing for NAS. However, PASS primarily aim to reduce uncertainty regarding safety risks and the (societal) value of this uncertainty reduction might not fully be captured by assessing health effects alone. A446 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PHP249 Supporting Interoperable Eu Patient Registries: Survey of Registry Holders’ Needs Pristas I 1, Doupi P 2, Meglic M 3, Karanikas H 4, Zaletel M 5, Brkic M 1, Plese B 1, Zuriaga Llorens O 6 1Croatian Institute of Public Health, Zagreb, Croatia, 2National Institute for Health and Welfare (THL), Helsinki, Finland, 3National Institute of Public Health, Slovenia, Ljubljana, Slovenia, 4National & Kapodistrian University of Athens, Athens, Greece, 5Nacionalni Institut za javno zdravje, Ljubljana, Slovenia, 6Conselleria de Sanitat. Generalitat C. Valenciana, Valencia, Spain . . . . . . . . Objectives: Due to the diversity in EU Member States’ specific needs and legislation, the complexity of the health domain and the variable progress in IT-implementation there is currently limited standardization across registries and other eHealth tools. PARENT Joint Action (www. patientregistries. eu) aims at supporting the development of comparable and interoperable patient registries, thus enabling secondary data usage for public health and research purposes in a crossborder setting. As part of mapping the EU patient registry landscape, we undertook a survey of registry holders with the purpose of gathering information on current practices, as well as future needs and expectations with regard to IT-supported data exchange. Methods: A survey of EU registry holders (n= 177) was conducted and data was collected via an online questionnaire. Gathered information defines additionally patient registry metadata and is part of the development process of the Registry of Registries (RoR) – an envisioned go-to source for up-to-date information about patient registry metadata across Europe. Results: We found a high degree of interest (82%) for participation in a single IT-enabled platform on the EU level, indicating IT tools, quality control systems and networking opportunities as the most potentially desirable services of such a system. The registries that indicated an interest to link their data to other data sources were more likely to have previously established standardized data exchange procedures and formats (Pearson χ2= 6.909, df= 1, p= 0.009). There were no statistically significant differences in presence of routine data exchange procedures among patient registries according to Registry type (condition-based, product-based or service-based), Geographical coverage and/ or Holding institution type. Conclusions: Achieving patient registry interoperability requires a complex service-oriented approach taking into account the whole registry context (political, legal, organizational, semantic and technical levels). Our survey findings will serve more advanced in-depth assessments of registry quality dimensions to be performed through the PARENT Framework. HEALTH CARE USE & POLICY STUDIES – Population Health PHP250 Determination of Major Chronic Diseases by Using Pharmaceutical Reimbursement Data From A Large Belgian Health Insurer Claims Database Van Tielen R , Karakaya G , Vanrillaer V , Umbach I MLOZ, Brussels, Belgium . . . PHP251 Use Of Factor Analysis To Obtain Independent Health Performance Indicators . . PHP252 Impact of Economic Crisis on the Greek Health Care System and on the Population Health Kaitelidou D 1, Kalogeropoulou M 2, Galanis P 1, Theodorou M 3, Charalambous G 4, Liaropoulos L 1 1National and Kapodistrian University of Athens, Athens, Greece, 2Center for Health Services Management and Evaluation, National and Kapodistrian University of Athens, Athens, Greece, 3Open University of Cyprus, Latsia, Nicosia, Cyprus, 4Hippokratio Hospital of Athens, Athens, Greece . . . . . . Objectives: The aim of the paper is to map the impact of economic crisis on the Greek health care system and on the population health. Methods: A systematic literature review was conducted in databases Pubmed, Embase and Scopus for the years 1980-2013, focusing mainly to the impact of the economic crisis in Latin America, Eastern Europe and Eastern Asia. In order to assess the impact of crisis in Greece, Hellenic Statistical Authority (EL. STAT.), Hellenic Center for Disease Control and Prevention (HCDCP), OECD, Eurostat and World Bank databases were used. Results: Greece entered a deep economic crisis in 2009 and in only four years lost more than 25% of its GDP. Unemployment rate exceeded 27% of the population in 2013, the number of uninsured population reached 2 million according to the Ministry of Labour (2014) whereas the rate of population at risk of poverty or social exclusion has increased from 27.6% in 2009 to 34.6% in 2012. Total current health expenditure decreased by € 5.4 billion (23.7%) while public current health expenditure fell by a greater proportion, 25.2% or € 4bn between 2009 and 2012. At the same time the use of public services (e.g. hospital admissions) and waiting times increased. Regarding the health outcomes, an increase in the incidence rate of HIV/AIDS per million population by 10% was recorded. Incidence of Hepatitis A was increased by 187.5% and meningitis by 87% between 2011-2013. Total mortality rate was increased by 3.2% while expressed unmet health care needs were increased by 55% between 2009-2011. Conclusions: The efforts to reform the Greek National Health System have been focusing mainly on short-term effects by reducing expenditure. However, health-promoting policies providing equitable access, especially regarding the vulnerable groups, and infectious diseases prevention should be considered as a more cost-effective alternative. . Objectives: The objective of this exploratory study is to define the (chronic) health status of about 2 million affiliates from the Independent Sickness funds (MLOZ, Belgium). Methods: Data were extracted from the administrative database of MLOZ. Information of specific reimbursed medications was used as proxy for diagnosis of 20 chronic diseases. We processed data of pharmaceutical consumption in the ambulatory sector (minimum 90 DDDs/year) to characterize the health status of patients. Results: The analysis included 1,962,526 affiliates. The prevalence of chronic diseases is 23.3% in the population. This population is spending 64.3% of the total health expenditures reimbursed for all affiliates. Hypertension is the most common with 16.2% of the population, followed by depression (5.1%), diabetes (3.8%) and a group called “cardiovascular event” with 2.1%. Patients suffering from respiratory disease represent 3.8% (Asthma (1.7%) and COPD (2.0%)). Regarding the average health care costs, the picture is totally different. Patients with a rare disease induce on average € 48,476 health care cost, followed by patients with renal failure (€ 40,429) or cystic fibrosis (€ 20,700). Patients with hypertension “only” cost € 4,938. Conclusions: Claims databases from health insurers are attractive for researchers because of their size and detailed computerized records of all reimbursed health care procedures on a recipient-specific basis. In terms of public health, it seems interesting to concentrate (public) financial efforts towards chronic pathologies concerning a large number of patients with important average health care costs. Using secondary data of insurer claims databases allows (with some caution) to characterize chronic health status of large populations. Results may help decision-makers in defining priorities in resource allocation. Van Wilder P , Bormans V , Leemans E SMART&BI, Zaventem, Belgium the variance. The 3 iHPIs loaded respectively on ‘mortality’, ‘prevention’ and ‘AIDS incidence’. The 7 European regions differed significantly for 7 of the 8 initial HPIs and for each of the 3 iHPIs (p always < 0.02). Cluster analyses based on the 3 iHPIs, avoiding multi-collinearity, generated regions with different country composition. Both Malta and Georgia became isolated; Finland and Portugal shifted from their geographical region. Conclusions: HPIs are important to assess and compare the impact of health care policies across regions and countries. After extracting 8 correlated HPIs linked to direct health care policies from a published set of HPIs, factor analysis allowed us to convert these into 3 independent components, which were subject to further multivariate analyses offering additional and different insights. . Objectives: Health Performance Indicators (HPIs) provide a quantitative tool to assess the performance of health care policies. Available HPIs may be strongly correlated, limiting further inferential use. In this study we converted published HPIs into a set of independent HPIs (iHPIs) using factor analysis. We subsequently used iHPIs to perform regional health care performance comparisons. Methods: We used the set of 27 HPIs on 43 European countries from 7 geographical regions of Mackenbach (Mackenbach JP, McKee M. European Journal of Public Health 23 (2), 195-201, 2013). We extracted only indicators related to direct health care services having limited missing data (< 20%) and applied factor analysis to obtain iHPIs. The performance of the 7 regions was analysed using iHPIs in cluster analysis and non-parametric ANOVA (significance level at 0.05). Results: 8 correlated indicators met our analysis criteria: systolic blood pressure, cervical cancer screening, teenage pregnancy, newborn and mother mortality, measles vaccination, mortality and AIDS incidence. The factor analysis reduced the set to 3 components or iHPIs and explained 77% of PHP253 Estimation of Stress Resistance of Medical Students in LVIV Blavatska O 1, Lototska L 1, Blavatskyj I 2 Halytskyj Lviv National Medical University, Lviv, Ukraine, 2Lviv Politechnic National University, Lviv, Ukraine . . . 1Danylo Objectives: High demands for speed and amount of students’ trainings in health care education is one of the stress causes. Presence of stress can negatively affect the state of progress and level of physical health of medical students. Methods: Testing of 275 students III-IV years of studying at medical faculty (145 girls, 130 boys) using methods of Friedman and Rosenman modifications «tendency to stress» and «self- rating of stress resistance». Results: Test results revealed: 34% of girls & 29% of boys showed below average level self-rating of stress resistance, including group tendency to stressful situations. Average level was characterized for 39% of girls & 27% of boys. This indicates the sufficient resistance to stress and rejection as failure stress. Higher than average level of stress resistance was observed for 15% of girls & 26% of boys. These students, despite the negative influence of different factors fight stress and try to prevent it. High level of stress resistance was exposed only by 12% of girls & 18% of boys. Conclusions: Gender differences in the aptitude to stress of medical students were established. High aptitude was recorded for 11% of girls & 7% of boys. Relatively high frequency of getting into stress was inherent for 25% of girls & 21% of boys. Presence of stress and attempts to avoid it were found for 29% of girls & 40% of boys. For 22% of girls & 26% of boys was found the ability to resist stress by communicative & harmony lifestyles. High level of self-regulation, the ability to achieve goals as the result of low aptitude to stress was inherent for 10% of girls & 7% of boys. Reducing adaptation to stress among medical students was registered. Formation of skills for stress resistance by implementation of healthsaving technologies in student lifestyle is necessary. PHP254 Breastfeeding and Its Effect On The Probability of Occurance of the Probability of Occurance of the Disease And Related Costs Lehocká L 1, Malovecká I 1, Minariková D 1, Foltan V 2 University, Bratislava, Slovak Republic, 2Faculty of Pharmacy, Comenius University, Bratislava, Slovak Republic . . . . 1Comenius Objectives: Breastfeeding and nutrition of a child by breast milk has distinct influence on the child’s health during childhood as well as in their adult age. Insufficient nutrition of infants and young children is one of the factors influencing morbidity in children and can cause increased susceptibility for certain chronic, non-infectious (so called civilization) diseases in adulthood, for example allergic diseases, obesity, diabetes mellitus that are closely associated with future costs of health care. Methods: Analysis of medical records of randomly selected children between the age of 1 to 11 (n= 100) from four pediatrician offices. The following data was analyzed: age, weight, breastfeeding, duration of breastfeeding and incidence of selected diseases (gastroenteritis, laryngitis, bronchitis, pneumonia, A447 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 allergic rhinitis, cow’s milk allergy, eczema and obesity). Results: Of 100 studied children 64% were breastfed (BF), while 36% had artificial nutrition (AN), xBF≤ 6months =48%, xBF> 6months=52%. Gastroenteritis: xBF =1,42 diseases/child/life (D/CH/L), xAN = 1,64 D/CH/L, xBF≤ 6months = 1,84 D/CH/L, xBF> 6months= 1,23D/CH/L. Laryngitis: xBF= 0,27D/CH/L, xAN = 0,64 D/CH/L, xBF≤ 6months = 0,74D/CH/L, xBF> 6months= 0,20D/ CH/L. Bronchitis: xBF= 3,73D/CH/L, xAN= 4,86 D/CH/L, xBF≤ 6months = 3,89D/CH/L, xBF> 6months= 3,34D/CH/L. Pneumonia: xBF= 0,11D/CH/L, xAN = 0,31 D/CH/L, xBF≤ 6months = 0,26D/CH/L, xBF> 6months= 0,03D/CH/L. Allergic rhinitis: xBF= 0,16D/CH/L, xAN = 0,28 D/CH/L, xBF≤ 6months =0,21D/CH/L, xBF>6months=0,11D/CH/L. Cow’s milk allergy: xBF= 0,11D/CH/L, xAN = 0,17 D/CH/L, xBF≤ 6months = 0,16D/CH/L, xBF> 6months= 0,03D/ CH/L. Eczema: xBF= 0,09D/CH/L, xAN = 0,42 D/CH/L, xBF≤ 6months = 0,21D/CH/L, xBF> 6months= 0,03D/CH/L. Obesity: xBF= 0,03D/CH/L, xAN = 0,17 D/CH/L, xBF≤ 6months = 0,05D/CH/L, xBF> 6months= 0,00D/CH/L. Conclusions: Results of the analysis indicated that breastfed children showed less frequent incidence of all of the studied diseases compared to children who had artificial nutrition. Breastfeeding and even more importantly the duration of breastfeeding has important positive impact on child’s health, lower the occurrence of the diseases, frequency of medication use and overall total expenditure on health care. HEALTH CARE USE & POLICY STUDIES – Prescribing Behavior & Treatment Guidelines PHP255 Evaluation of Medicine Prescription Pattern Using World Health Organization Prescribing Indicators in Iran: A Cross-Sectional Study Soleymani F 1, Taheri F 2 of Health, Tehran, Iran, 2Minitry of Health, tehran, Iran . . 1Ministry Objectives: objective was to quantify the specialists’ prescription pattern in Iran and to point out the prescribing behavioral differences among several specialties. Methods: A retrospective cross-sectional study was carried out on the claim data. National prescription data were obtained on the basis of the claims that the pharmacies submitted to the insurers during 1 year period of the study. More than 85 million prescriptions were analyzed using “Rx-Analyst” software that is designed and applied by National Committee of Rational Use of Medicines in Iran. Specified medical specialties were considered and the World Health Organization prescription indicators were used to evaluate the physicians’ prescribing behavior. Results: Average items per prescription were ranged from 3.68 in cardiologists’ to 2.06 in dermatologists’ prescriptions. The highest and the lowest mean price were belonged to neurologists’ and ophthalmologists’ prescriptions, respectively. In addition, 45% of patients received antibiotics, 41% of patients received injectable form of drugs, and 23% received corticosteroids. A high tendency toward prescribing corticosteroids and antibiotic as well as an injectable form of medicines was observed among general physicians. Conclusions: There is an inevitable need to improve prescription habits among different specialties, especially among general practitioners. This causes the policymakers to put more emphasis on priorities such as continuous education. and optimise the reporting of such evaluations. To our knowledge, there are no such published guidelines in Spain. Methods: This guidance was designed with the main aim of providing authors with recommendations, in the form of a checklist, to optimise design and reporting of EE and BIA to be submitted to the Catalan Health System (CatSalut) within its harmonization program for pharmaceutical innovation. Results: For a given evaluation, two separate guidelines with corresponding checklists assess its methodology, quality, and reporting based on several dimensions. Conclusions: The steps outlined in this first guidance in Spain, although not compulsory, will provide useful practical tips for how to go about designing and reporting of an EE or a BIA for pharmaceuticals in Catalonia (Spain). PHP258 Guidance for Risk Sharing Agreements / Pay Per: Results SCHEMES FOR PHARMACEUTICALS IN CATALONIA (SPAIN) Mora-Ripoll R 1, Gilabert-Perramon A 1, Espinosa-Tome C 1, Segú L 2, Gasol-Boncompte M 1 1CatSalut, Barcelona, Spain, 2Consorci Salut, Barcelona, Spain . Tsiantou V 1, Kyriopoulos J 1, Lionis C 2 School of Public Health, Athens, Greece, 2University of Crete, Faculty of Medicine, Clinic of Social and Family Medicine, Heraklion, Greece . . . 1National Objectives: Monitoring and measuring prescribing patterns can provide valuable information regarding the use of medicines and the overall prescribing behavior. In Greece, until recently there was no organized system to monitor and record prescribing patterns. The aim of this study is to describe prescribing patterns of GPs in primary health care. Methods: A prospective cross-sectional study was conducted (March–June 2011). Two geographical areas (rural and semirural) in Crete and one urban in Athens were randomly selected and all GPs working in these areas in public or private practices were invited to participate. Each GP answered a structured questionnaire regarding his/her prescribing patterns and collect data on 100 patients within a specified period. Prescribed drugs and diagnoses were then classified according to the ATC classification system and the ICPC-2 classification system respectively. Results: 13 GPs (7 from Crete) participated in the study giving data for 1202 patients. 84.5% of the patients received a prescription and the mean items prescribed were 2.72. At the 12.4% of the total encounters the patient was not present while at 9.9% of all encounters the reason was to ask a prescription for medicines already bought from the pharmacy. GPs prescribed at the 96.7% and 97.5% of these cases respectively. Diseases of the circulatory (24.9%), endocrine (23.5%) and musculoskeletal (10.4%) systems were the most commonly recorded diagnoses. Overall, 62% of all prescribed medicines were from the cardiovascular system, alimentary tract and metabolism and nervous system. 4.8% were antibiotics, 2.7% were injections, 19.9% were launched in the Greek market after 2001, 4% were OTCs, 3.7% were narcotics, 10.8% were generics. Conclusions: This preliminary descriptive analysis gives an overview of the prescribing patterns of GPs and highlights the areas that pharmaceutical policies should be focused. Further analysis is needed to identify the factors that determine this behavior. PHP257 Guidance For Economic Evaluation and Budget Impact Analysis For Pharmaceuticals in Catalonia (Spain) Mora-Ripoll R 1, Gilabert-Perramon A 1, Oliva-Moreno J 2, Puig-Junoy J 3 Castilla La Mancha, Toledo, Spain, 3Universitat Pompeu Fabra, Barcelona, Spain . . . . 1CatSalut, Barcelona, Spain, 2Universidad Objectives: Health economic evaluations (EE) and budget impact analysis (BIA) for pharmaceuticals are increasingly being used for decision making. Guidelines can be very useful to help authors with the use of proper methodology, assess the quality . . . HEALTH CARE USE & POLICY STUDIES – Quality of Care PHP259 Type of Multimorbidity and Patient-Centered Care Among Elderly Medicare Beneficiaries Garg R 1, Shen C 2, Sambamoorthi N 3, Ajmera M 1, Sambamoorthi U 1 University, Morgantown, WV, USA, 2The University of Texas MD Anderson Cancer Center, Houston, TX, USA, 3Northwestern University, Evanston, IL, USA . PHP256 Prescribing Patterns of General Practitioners (Gps) in Primary Health Care: Evidence From Greece . Objectives: In recent years, spending on prescription drugs contributes substantially to the continuous growth in health expenditure which is influenced by both a rise in the use of existing drugs and by the adoption of new and expensive drugs. However, their effectiveness outside of clinical trial settings is often uncertain at the time they gain market approval. This uncertainty may reflect a lack of real-world outcomes data, as opposed to clinical trials data, for a typical patient population. A risk-sharing agreement is a contract between a drug manufacturer and a health care provider/payer to help manage uncertainties regarding the cost and effectiveness of those drugs and serves the interests of both. Guidelines can be very useful to identify these uncertainties and select the pharmaceuticals which are most eligible for such an agreement to be implemented. To our knowledge, there are no such published guidelines in Spain. Methods: This guidance was designed with the main aim of identifying uncertainties, helping select the pharmaceuticals most eligible, and defining best type of risk sharing agreement (RSA) / pay per Results scheme (PRS) to be implemented between the Catalan Heath System (CatSalut) and health care providers/pharmaceutical companies in Catalonia. Results: For a given pharmaceutical, two questionnaires identify and score following four variables related to main uncertainties involved: (1) efficacy/effectiveness uncertainty; (2) budget uncertainty; (3) budget impact; and (4) incremental cost-effectiveness ratio (ICER) and willingness to pay. Final score guides the best type of agreement to be implemented accordingly. Conclusions: The steps outlined in this first guidance in Spain, although not compulsory, will provide useful practical tips for how to go about setting up an RSA or PRS in Catalonia (Spain). This guide is not a substitute for suitable regulatory or legal advice. . . . . 1West Virginia Objectives: Patient-centered care (PCC) is a critical component of health care management of elderly with multimorbidity, especially for those with co-existing chronic physical conditions with mental illnesses. However, little is known about the association between type of multimorbidity and PCC. The objective of this study was to analyze the association between type of multimorbidity and PCC. Methods: A retrospective cross-sectional study design was used. Data were derived from 2011 Access to Care module of the Medicare Current Beneficiary Survey (MCBS). The study sample (N = 10,158) consisted of community-dwelling elderly Medicare beneficiaries aged 65 years or older with at least one physical chronic condition. Types of multimorbidity (MM) defined as: 1) single physical condition (No MM); 2) two or more physical conditions without mental illnesses (MM-PI); and 3) both mental and physical conditions (MM-MI&PI). PCC was defined using patient-physician communication and patients’ confidence in physician. Chi-square tests and logistic regressions were used to test the unadjusted and adjusted associations. Results: A lower percentage of elderly Medicare beneficiaries with MM-MI&PI had optimal patient-physician communication (73.5%) and high confidence in their physicians (86.1%) as compared to those with No MM (79.1% and 90.2%). Without adjustments for health status, elderly with MM-MI&PI were less likely to have optimal patient-physician communication (OR = 0.74, 95% CI [0.61, 0.88]) and confidence in their physicians (OR = 0.67, 95% CI [0.53, 0.86]) as compared to elderly with no MM. The relationship between multimorbidity and PCC was not statistically significant after adjusting for general health and functional status. However, those with MM-MI&PI were as likely as those with MM-PI to have optimal patient-physician communication and confidence in their physicians. Conclusions: Those with multiple conditions faced poor communication and low confidence in doctor as compared to those with single physical condition. However, type of multimorbidity was not associated with PCC. PHP260 Impacts Of Bar-Code Medication Administration (Bcma) On Patients’ Safety in Taiwan Lee M Wan Fang Hospital, Taipei city, Taiwan . Objectives: A BCMA system was introduced to the hospital in June 2012. The purpose of this study is to evaluate the impact of the BCMA system on reducing medication errors and nurse and pharmacist satisfaction with the system. Methods: The nurse satisfaction questionnaire and the pharmacist satisfaction questionnaire for the overall system were designed by 8 experts using content validity index (CVI). Medication error rates before and after the implementation of the A448 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 BCMA system was compared. We also collected the numbers of phone calls for tracking stat drugs from Oct 2nd to Nov 15th in 2012. Results: After the implementation of the BCMA system, the medication error rate was significantly reduced from 0.18% to 0.12% (p < 0.05). Pharmacists agreed that the BCMA system provides assistance in tracking prescriptions (90%), identifying medications (60%), and reducing medication errors (53%). However, pharmacists also think that the system has increased their routine workload (57%). Nurses agreed that the BCMA system can improve patient identification (73%), administration of right medication (57%), medication administration time (47%), and access to correct prescription/medication information (42%). Both pharmacists and nurses were dissatisfied with the stability of the system (53%, 69%) and poor barcode sensitivity (47%, 64%). From Oct 2nd to Nov 15th in 2012, we received 469 phone calls. The average of tracing call for drugs is 1.67 times every hour. Conclusions: BCMA system significantly decreases medication error rate in the hospital. It is recognized to be able to improve medication safety by both pharmacists and nurses. Improving the quality of associated computer equipment is the next important step. HEALTH CARE USE & POLICY STUDIES – Regulation of Health Care Sector PHP261 Pharmacoeconomic Evaluation For Reimbursement Purposes in Bulgaria: Recent Updates Stoimenova A 1, Savova A 1, Benisheva-Dimitrova T 2, Kamusheva M 1, Petrova G 3 University, Faculty of Pharmacy, Sofia, Bulgaria, 2NATIONAL COUNCIL ON PRICES AND REIMBURSEMENT OF MEDICINAL PRODUCTS, Sofia, Bulgaria, 3Medical University Sofia, Faculty of Pharmacy, Sofia, Bulgaria . . . . . 1Medical Objectives: Our study evaluates the preferred methods which the pharmacoeconomic evaluations submitted for reimbursement purposes in Bulgaria are using. Methods: This is retrospective, descriptive study. We have examined all applications submitted to National Council on Pricing and Reimbursement in 2013 on which positive opinion was issued. Quantitatively and qualitatively descriptive analyses towards the new molecules were carried out based on the submitted pharmacoeconomic evaluations. Results: 22 new INNs were submitted and approved for reimbursement during the studied period (average time for approval: 194.55 days). Only 18.18% (n= 4) were orphan drugs. The most commonly used types of pharmacoeconomic evaluation were cost-effectiveness analysis (n=10), followed by cost-minimization analysis (n=4). Only one cost-utility analysis was provided. In all other cases only budget impact analyses (BIA) was submitted that correspond with the national regulatory requirements. Only one analysis included hospitalization costs calculation. The chosen perspective was the payer’s perspective for all analyses. The time horizon varies between 10 days and 5 years. 68.18% (n= 15) of the pharmacoeconomic evaluations were adjusted to 1 year calculations. 72.73% of the INNs were approved for 100% reimbursement rate, 3 INNs – for 75% and two-for 50% reimbursement rate. 45.45% of the submitted evaluations contained local epidemiological data. Only one analysis did not include the relevant comparator and only two haven’t discussed any additional benefits of the treatment. Pharmacoeconomic evaluations of new medicinal products was often inadequately validated by incorrect type of pharmacoeconomic analysis chosen, lack of Bulgarian epidemiological data, costs and effects were not correctly analyzed which led to prolongation of the time need for approval. Conclusions: The need for more detailed guidelines is obvious in the light of increasing importance of pharmacoeconomic evaluations for reimbursement purposes. PHP263 The Effect of Procurement Design on Entry and Success of Generic Drug Firms 1Comenius University, Bratislava, Slovak Republic, 2Faculty of Pharmacy, Comenius University, Bratislava, Slovak Republic Objectives: The economical crisis and the requirements for structural changes in all aspects of public health brought a new era of reforms in the country as well as the health care system itself. The Pharmaceutical Legislation, both European and national, were implemented to protect public and state interests. Application of new legislation influenced provider of health care services in different areas. Therefore it is important to evaluate the impact and performance on costs and revenues of health care services provider. Methods: Monitoring, calculation and assessment of costs and revenues with the help of financial analysis indicators for years 2003-2012, using financial statements was conducted, with respect to profitability, debt, liquidity, working capital and efficiency ratios. These ratios constituted significant information of implementing new health care legislation and price policies. Results: In case of profitability, parameter gross profit ranged from x2003–2011= 16.1–22.8% (xaverage= 19.2%, Xmean= 19.8%, σ = 2.4), but in 2012 decreased on 14.3%. Net profit ranged x2003–2011= 12.8–18.3% (xaverage=14.6%, Xmean=16.6%, σ=4.9%), while in 2012 reached only 2.3%. Debt parameters varied from x2003–2012= 2.33-4.8 (xaverage= 3.44, Xmean= 3.06, σ = 0.82), liquidity parameters current ratio x2003–2012= 1.11.7 (Xaverage= 1.43, Xmean= 1.45, σ = 0.15) and quick ratio x2003–2012= 0.7-1.3 (xaverage= 1.16, xmean= 1.09, σ = 0.15), working capital ratio x2003–2012= 2,7-12,9, (Xaverage= 9.6, Xmean= 10.1, σ = 3.1) and efficiency ratios were measured either. Conclusions: Implementation of new health care legislation and price policy that were intended to apply restrictive measures to increase system efficiency and cost savings had a significant impact on health care services providers by worsening profitability and liquidity parameters as key indicators of provider stability. Despite the improvement in debt ratio, working capital ratio and efficiency ratios, stability may be threatened and may affect the amount of health care services providers. PHP265 The Effect of Decentralised Public Health Care Provision on Accessibility To Medicines in Bosnia And Herzegovina Melck B , Ando G , Izmirlieva M IHS, London, UK . . . Objectives: This study aims to assess the extent of the restriction on access to medicines in Bosnia and Herzegovina in relation to the country’s decentralised provision of health care, with three separate state entities and 10 cantons within the federation of Bosnia and Herzegovina all having a role in the provision of public health care. Methods: A comparison was made between the contents of the federal essential drug list and the drug lists of the various entities and cantons, in order to ascertain which medicines were available, at what price, and with what level of reimbursement. Results: The federal essential list contains the basic INNs which must be provided by each canton, but there were some important differences between the lists of the various entities and cantons. For example, the range of patented and originator medicines available in the Republic of Srpska was greater than in the cantons of the federation of Bosnia and Herzegovina in general. Also, within the federation, there was some variation in the price of medicines between cantons, and an even greater variation between the reimbursement levels paid for them. For example, in the Tuzla canton, a package of generic losartan produced by Krka was priced at 9.30 convertible marks and reimbursed at 20% by the canton’s health insurance fund, while the same package at the same price was reimbursed 50% in the West Herzegovina canton. Conclusions: The lack of centralised control over drug funding means that citizens face variability in access to medicines across the spectrum, depending on which entity or canton they live in. Recent regulatory changes aimed at bringing greater control into drug pricing have so far failed to have any notable impact. Laitenberger U 1, Hunold M 2 for European Economic Research, Mannheim, Germany, 2European Commission, SaintJosse-ten-Noode, Belgium HEALTH CARE USE & POLICY STUDIES – Risk Sharing/Performance-Based Agreements Objectives: Competitive tendering of active ingredient-specific contracts for generic drugs has emerged in Germany due to regulatory changes in 2009. The central goal of these changes is to accomplish that small and medium sized firms (SME) have better chances to succeed. Another goal is more participation of manufacturers, including outright market entry. We study how different designs of the competitive tendering affect participation and the outcome of generic drug manufacturers. Methods: We use a newly collected data set on almost all of the around 4000 public competitive awardings of procurement contracts by purchase groups in Germany from the first one in 2008 to the end of 2012. An important source of cross-sectional variation is caused by the nine purchasing groups, which differ in how they design their awarding procedures. Major differences include the lot size, the number of possible contracting firms and whether a drugs of a specific active pharmaceutical ingredient is auctioned in one or different lots. We use OLS regressions with fixed effects for unobserved active pharmaceutical ingredient (API) market characteristics, as well as count data and discrete choice models. Results: Our econometric analysis reveals that there are more bidders in tenders where contracts with more than one supplier are possible, when the drugs for one API are auctioned in a single lot and when the lot size is smaller. Also small firms are more likely to win when multiple winners are possible and the lot size is smaller. Conclusions: We find that the design of awarding procedures influences participation in tenders and, furthermore, the chance of new suppliers to win the contract. This is relevant for both purchasing groups and policy makers as low bids and thus drug expenses of the statutory health insurances can only be achieved with a competitive industry structure. PHP266 The Comparison of the Risk-Sharing Schemes Proposed in Reimbursement Applications Received By Ahtapol In 2012 and 2013 . . 1Centre PHP264 10 Years Evaluation of Costs And Revenues Of Heath Care Services Provider Under The New Health Care Legislation and Price Policy Malovecká I 1, Minariková D 1, Foltan V 2 . . . Zawodnik S 1, Iwanczuk T 1, Hermanowski T R 2, Matusewicz W 1 1Agency for Health Technology Assessment in Poland (AOTM), Warsaw, Poland, 2Medical University of Warsaw, Warsaw, Poland . . . . . Objectives: To compare the Risk-Sharing Schemes (RSSs) proposed in reimbursement applications received by Agency for Health Technology Assessment in Poland (AHTAPol) in 2012 and 2013. Methods: Comparative analysis of RSSs proposed in reimbursement applications received by AHTAPol in 2012 and 2013 was conducted. RSSs were quantitatively and qualitatively analyzed and classified on the basis of both Carlson’s approach and the Polish Act on the reimbursement of medicinal products. Results: In years 2012 and 2013, 56 reimbursement applications with 26 proposed RSSs and 80 reimbursement applications with 52 RSSs proposed received by AHTAPol, respectively. They were classified into 5 categories according to the Act on the reimbursement. The most common category in 2012 was making the official sales price dependent on the applicant providing supplies at a reduced price (34.62%). On the other hand, in 2013 the most common category was making the official sales price dependent on a pay-back of a part of the reimbursement obtained to the entity which is obliged to finance benefits with public funds (48.08%). Further categories were also analyzed. Among all RSSs proposed in each year, only 17 of 52 in 2013 and 9 of 26 (1 proposition included more than one category) in 2012 could be classified according to the Carlson’s approach. As a Results, most common categories were Price Volume Agreements (4 in 2012 and 10 in 2013) and Manufacturer funded treatment initiation (4 in 2012 and 6 in 2013). Both in 2012 and 2013, there were only one RSSs dependent on the health effects achieved. Conclusions: Most of the propositions both in 2012 and 2013 are not considered to be RSS according to the Carlson’s approach. In 2013 compared to 2012 there were less price reduction schemes and more pay-back based schemes. There is a strong need for further research. A449 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PHP267 Coverage With Evidence Development Activities Around The World: An Environment Scan dialogue between health care authorities and the pharmaceutical industry to create true risk-sharing agreements. Li C M , Risebrough N A , Hux M ICON plc (formerly Oxford Outcomes), Toronto, ON, Canada PHP271 Risk-Sharing Schemes In Poland - Analysis And Classification of Rss Proposed In Reimbursement Application Received by Ahtapol In 2013 . . . . . Objectives: A growing number of health technologies are coming to market with limited, yet promising, clinical data. Coverage with evidence development (CED), a conditional interim reimbursement scheme linked to research to reduce uncertainty, has been increasingly explored by payers worldwide. The objective was to summarize CED programs worldwide, and report final reimbursement decisions. Methods: A search was conducted using published literature, websites and grey literature to identify CED programs worldwide from 1998 to 2012. Results: Seventy-four CED schemes were identified in Canada (n= 23), Sweden (n=16), United States (n=14), UK (n=11), Australia (n=4) and Europe (n=6). CED schemes were found in oncology (n= 21), heart disease (n= 12), diabetes (n= 10), Neurological disorders (n= 7), multiple sclerosis (n= 2), mental health (N= 3), rheumatoid arthritis (N= 1), and other (N= 18). Drugs, imaging techniques, surgical procedures, and devices were most commonly evaluated. Most CED programs aimed to address more than one type of uncertainty. The most commonly encountered uncertainty was clinical benefit, followed by value for money, adoption and diffusion, and affordability. Study designs included interventional, clinical, observational, and economic studies. CED programs were generally managed by independent, government-funded nonprofit research organizations, university-based academic centers, or professional societies. In only 39% (N= 28) of the identified CED schemes, study outcomes and funding decisions were reported. In 74% (N= 17) of reported cases, the technology evaluated was successfully funded. One technology (lung volume reduction surgery) was only funded in a subgroup, and one technology was funded with price reduction. On many accounts CED has proven challenging to implement. Conclusions: Although a large number of CED activities were identified, detailed information, especially the study outcomes and final reimbursement decision, only limited information of final decisions was publically available. CED is a promising mechanism to reduce uncertainty and aid timely patient access, but with emerging implementation challenges. Iwanczuk T 1, Zawodnik S 1, Hermanowski T R 2, Matusewicz W 1 1Agency for Health Technology Assessment in Poland (AOTM), Warsaw, Poland, 2Medical University of Warsaw, Warsaw, Poland . . . . . Objectives: To analyze and classify the Risk-Sharing Schemes (RSSs) proposed in reimbursement applications received by Agency for Health Technology Assessment in Poland (AHTAPol) in 2013. Methods: Risk-Sharing Schemes proposed in reimbursement applications received by AHTAPol in 2013 were quantitatively and qualitatively analyzed. The classification of the RSSs was conducted based on both Carlson’s approach and the Polish Act on the reimbursement of medicinal products. Results: In the studied period, 80 reimbursement applications for medicines, special purpose dietary supplements or medical devices were received by AHTAPol. Among them, there were 52 RSSs for 51, medical technologies. They were classified into 5 categories according to the Act on the reimbursement. The most common category was making the official sales price dependent on a pay-back of a part of the reimbursement obtained to the entity which is obliged to finance benefits with public funds (48.08%). Further categories were: making the official sales price dependent on the applicant providing supplies at a reduced price as specified in the negotiations on the price of the medicine (15.38%), making the official sales price dependent on the level of turnover of the medicine (11.54%) and making the level of the applicant’s revenues dependent on the health effects achieved (1.92%). Other, nonclassified RSSs constituted 23.08% of all. Among 52 proposed RSSs only 17 of them could be classified according to the Carlson’s approach. As a results, 10 Price Volume Agreements, 6 Manufacturer funded treatment initiation and 1 Conditional treatment continuation were identified. Conclusions: Most of the propositions are not considered to be RSS according to the Carlson’s approach. The most common propositions were related to pay-back of a part of the reimbursement obtained for each reimbursed package and did not included any risk sharing. There is a strong need for further research. PHP268 Managed Entry Agreements in UK, Italy And Spain Tolley C , Palazzolo D Quintiles Consulting, Reading, UK HEALTH CARE USE & POLICY STUDIES – Conceptual Papers Objectives: To compare Managed Entry Agreements (MEAs) in the UK, Italy, and Spain, and analyse the type of MEAs, number of agreements, and therapy areas in which they exist. Finally, to determine MEA impact on market access delay in these countries. Methods: HTA databases were searched for types and quantities of MEAs (publicly available). The data were analysed by indication, and country specific knowledge applied, to quantify the average delay to market access. Results: All types of MEA have been granted previously in the UK, the majority were nonoutcomes based (76%). In Italy, and Spain, all MEA’s were outcomes based, with Italy focused purely on risk-sharing agreements. UK has 42 MEA’s since 2000, Italy has 44 MEAs since 2006 and Spain has 9 MEAs since 2010. Of 95 MEAs, 56% were for oncology drugs, 12% musculoskeletal, 10% ophthalmology, 7% CNS, 5% respiratory and 10% other therapy areas. NICE average time to HTA decision is 21 months, and is delayed up to 10 months depending on the type of MEA. AIFA average time to HTA decision is 8 months; however MEAs are part of the pricing negotiation, so it may be a way to gain market access faster. In Spain, the average time to HTA decision is 8 months with decisions for drugs with double pricing arrangements taking on average 14 months. For high cost hospital and orphan drugs, the delay can be up to a year. Conclusions: Negative reimbursement decisions can have a significant impact on achieving market access, and revenue generation. MEAs represent an avenue for overcoming these negative decisions and market access for high cost medicines. Companies aiming for MEAs in UK, Italy, and Spain, should be aware of the potential market access delay and the precedence of MEAs that the payers in these countries are amenable to. PHP272 Whatever Happened To Nice Value-Based Pricing? Welcome, ValueBased Assessment . . PHP270 The Possibility of Initiating True Risk-Sharing Agreements in the Current Economic Situation Germanenko A 1, Traulsen J M 2 of Copenhagen, Kobenhavn N, Denmark, 2University of Copenhagen, Kobenhavn, Denmark . . . 1University Objectives: The aim of this project was to evaluate risk-sharing agreements that are currently being negotiated in Europe between health care authorities and the pharmaceutical industry. Methods: A literature review of the grey literature (reports of the Organization for Economic Co-operation and Development and WHO Collaborating Centre for Pharmaceutical Pricing and Reimbursement Policies) as well as peer-reviewed literature (PUBMED) was done to explore the background of this topic. Semi-structured interviews were conducted with representatives of the pharmaceutical industry, consultant agencies and health care authorities to get their opinions about the current state of risk-sharing agreements. Results: The study identified multiple problems in risk-sharing agreements such as insufficient terminology and methodology, the absence of adequate infrastructure for implementation and a lack of trust and dialogue between the establishing parties: health care authorities and pharmaceutical industry. These issues question the crucial concept of sharing the risks equally and disgrace the agreements, even though it would be of great value to overcome current challenges. Such agreements could help the health care authorities to keep within their budget while still providing innovative pharmaceutical products/treatments to the patients. In addition, the pharmaceutical industry would have the opportunity to bring its products to the market and patients would benefit from adequate pricing and reimbursement. Conclusions: In spite of the current problems, the popularity of risk-sharing agreements in Europe is growing fast and the future is bright. For this reason there is a strong need for Langham S 1, Gemmell E 2, Kerrigan M 1, Wright A J 1, Chetty M 1 1PHMR Associates, London, UK, 2PHMR Associates, Newcastle upon Tyne, UK . . . . . . For several years, there has been much discussion about the central role the National Institute for Health and Care Excellence (NICE) would play in the move to valuebased pricing (VBP) of drugs in England and Wales as part of the new Pharmaceutical Price Regulation Scheme (PPRS) scheduled to start in 2014. To date, there is no VBP model, but NICE and the PPRS have been disentangled, and NICE is currently consulting on a proposal to include new terms of reference for value based assessment (VBA) in its Technology Appraisal Methods Guide. This would change the way in which it makes recommendations to the NHS and could become policy within a short time frame. Therefore, manufacturers need to be aware the implication of the VBA proposal on data requirements for NICE submissions. Currently, NICE reimbursement decisions are based on an incremental cost-effectiveness ratio threshold of £20,000/QALY, with factors such as innovation and National Health Service objectives acting to increase the acceptable threshold up to £30,000/QALY, and end-of-life benefits potentially increasing the acceptable threshold further to £50,000/QALY. A key feature of the proposed VBA is the use of modifiers to numerically adjust the ICER (up to a maximum of 2.5 times the base ICER of £20,000/QALY, i.e £50,000/ QALY). The burden of the illness for which the drug is approved as well as its societal impact are two such modifiers. Precise guidance on applying burden of illness and societal impact modifiers to ICERs is currently lacking, although information may become available during summer 2014. However, there is some draft information on quantifying these modifiers. We will review all available information, including any post-consultation guidance, to provide an overview of data requirements and possible changes to cost-effectiveness models that manufacturers may need to consider for future VBAs. PHP273 National Immunisation Technical Advisory Groups - A Framework For Assessment and Insights From Research Kleintjens J 1, King S 2, Hernández-Pastor L J 2, Abels M 1, Saka Ö 1, Bichon J 1 1Deloitte, Diegem, Belgium, 2GlaxoSmithKline, Wavre, Belgium . . . . . . . Objectives: A National immunization Technical Advisory Groups (NITAG) is a body of national experts that empowers and provides guidance to national health policy makers to enable them to make evidence-based decisions on immunisation. The aim of this study was 1) To develop a framework for the assessment of NITAGs that will allow to understand how closely their operations are aligned with best practice; 2) to investigate whether differences with best practice are influenced by economic or geographic differences between countries; and 3) to understand the impact of NITAG alignment on the adoption of vaccines in the country’s national immunisation programme. Methods: Building on previous initiatives mostly undertaken by the World Health Organization, we conducted interviews and in-depth research on 35 NITAGs worldwide and built an assessment framework with 48 indicators to evaluate NITAG Alignment with international guidance. Results: The assessment revealed that there is a high variability in the degree of NITAG Alignment between countries which could not be explained by differences in GDP per capita, health expenditure per capita, or geographic location. Countries with a reasonably well-aligned NITAG have a higher proportion of the WHO-recommended and additional vaccines in their A450 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 national immunization programme. Some lower and middle income countries with fairly strong alignment scores for recently established NITAGs have basic immunisation programmes. This study also found that NITAGs are usually developed following a stepwise maturing process. Conclusions: Our detailed analysis of data from 35 countries suggests that, with the right support, all countries – regardless of their GDP/ capita, health expenditures and geographical location – have the potential to establish highly performing NITAGs that are well-aligned with international recommendations. Well-aligned NITAGs are generally instrumental for having strong immunisation programmes. Through awareness of its position in this maturation process, a NITAG can focus on the appropriate next step for development and strengthening. PHP274 Discontinuities Between Health Technology Assessment (HTA) and Health Care Service Objectives of the NHS Brazier P J 1, Durand A 1, Tierney R 1, Kelly S 2 Knowledge, London, UK, 2Pfizer Limited, Surrey, UK . . . . . 1Matrix Objective: Advances in early cancer diagnosis and treatment are enabling patients to live longer with more fulfilling lives. The value assessment in such cases is compelling. Many novel treatments for late-stage cancer also extend life, though prognosis may still be poor. The valuation of such life-extending treatments can be significantly reduced by health care costs associated with managing patients for longer. This study highlights discontinuities between HTA for treatments that extend patients’ lives and the NHS’s objective to improve cancer patients’ survival rates. Methods and Results: Economic comparison of two treatments with an equivalent QALY gain, one that extends life while the other enhances the quality of life, indicates that to achieve a common cost per QALY outcome the life-extending treatment must be valued lower than the life-enhancing therapy. This anomaly arises primarily because the value assessment for life-extending treatments includes NHS costs of patient management during their extended life in addition to the new treatment costs. For long-term chronic conditions these additional costs may be easily offset; however, for severe, debilitating, or terminal diseases the impact can be significant. Furthermore, for new treatments added in combination to standard care, the greater the cost of existing care the lower the value that may be placed on the new life-extending treatment, to the point that new therapies may be deemed uneconomical even if available at no cost to the NHS. These findings challenge the equitable use of ICERs for HTA including the accounting for health services costs during the extended lifetime of a patient achieved with a new treatment. Conclusion: Value-based metrics used to appraise new treatments can inadvertently discriminate against life-extending therapies. Use of the ICER in HTA can result in inconsistency with health service objectives e.g. the UK Government’s goal to improve 1-year and 5-year survival rates for cancer patients. PHP275 The Economic Value of Vaccination: Why Prevention Is Wealth Remy V 1, Largeron N 1, Quilici S 1, Carroll S 2 1Sanofi Pasteur MSD, Lyon, France, 2Sanofi-Pasteur MSD, Maidenhead, UK . . . . Context: Theoretical and empirical evidence has demonstrated that health care is a major driver of economic growth. The European economic crisis has resulted in health care budget cuts conferring consequences for health systems. Preventative programmes, and particularly vaccination, are most vulnerable to short-term cuts because their benefits are not always immediately identifiable. Although its huge public health benefits are recognised, only a minor fraction of the health care budget is allocated to vaccination. It has been suggested that cost-effectiveness analyses, as used in HTA as part of recommendations and reimbursement decisions, may render a too narrow perspective of the overall economic benefits of vaccination. Objectives: The aim of this project is to demonstrate that, in addition to contributing to health care system sustainability, vaccines have importance for wider economic planning. A seven-chapter report was developed to highlight the full economic value of vaccination from different perspectives: macro-economic, health care system, society… Each chapter is illustrated with existing evidence retrieved from peer-reviewed publications. The objective of this report is threefold: 1) to demonstrate the full economic value of vaccination with real life examples; 2) to inform policy-makers on how immunisation contributes to health care systems sustainability and efficiency; 3) to launch a call for action for the consideration of the full economic value of vaccination. Discussion: Immunisation programmes require adequate value recognition to ensure quick population access and wide acceptability. Policy-makers should acknowledge that prevention through vaccination involves low levels of investment relative to the substantial incremental benefits it procures. As with other preventative interventions, it is difficult to evaluate the true economic value of vaccines given a number of benefits are intangible and thereby difficult to quantify in pure monetary terms. Taking into account the full economic benefits of vaccination will allow understanding why prevention is the one of the best ways to find efficiency gains. PHP277 The Cost-Effectiveness Threshold For Orphan Designations in Poland Based on Reimbursement Decisions Grzywacz K , Pelczarska A , Witkowski M A , Ofierska-Sujkowska G , Zawada A Agency for Health Technology Assessment in Poland, Warsaw, Poland . . dations is cost-effectiveness information. Currently in Poland there is no specific formal threshold for orphan designations, there is only a general cost-effectiveness threshold that equals 3 x GDP per capita for ICUR/QALY (for CUA) or ICER/LYG (for CEA), which in 2014 is approximately € 26 800. We extracted data on orphan drugs from our database of medical technologies assessments from 2009 to March 2014. Data on the cost-effectiveness (QALY / LYG) were put together with the decision of reimbursement. On the basis of these data the threshold of cost effectiveness in Poland for orphan designations was determined and summarized with the costeffectiveness thresholds current for a given time interval. Determination of the threshold of cost effectiveness for orphan designations, that would be different (higher) than the generally accepted cost-effectiveness threshold (due to high price of orphan drugs to provide value for money is unlikely), is particularly important from an ethical point of view, because of substantial therapeutic meaning of these drugs and/or absence of other treatment options of proven benefit for the disease. . . . PHP278 Conflict Of Interest in HTA Recommendations and Case Law In France Frybourg S 1, Kornfeld A 1, Brunet J 2, Toumi M 3 1Creativ-Ceutical, Paris, France, 2Assistance Publique des Hôpitaux de Marseille, Marseille, France, 3University of Marseille, Marseille, France . . . . Objectives: The slow reaction of French authorities to the “Mediator saga” in France led to investigations that questioned the way conflicts of interest are reported. This policy research reviewed the Loi Bertrand, known as French Sunshine Act, and reported case law from the French Council of State (COS) related to conflict of interest in HTA recommendations. Methods: Literature review and analysis of recent laws and decrees were conducted to understand French policy in the field of conflict of interest. A review of COS’ decisions related to conflicts of interest among members of HAS commissions was performed. Results: France implemented the Loi Bertrand in May 2013 with the aim of specifying the scope of disclosure obligations. It affects most of the agreements concluded between health care professionals and companies and covers a vast range of health products. Six cases examined by the COS were analyzed, most of them related to removal of products from refundable list. Four cases led to suspension or invalidation of decisions based on HAS recommendations due to conflicts of interest. In the two other cases the HAS provided the declarations of interest when required by the COS and the COS considered the conflicts of interest as irrelevant for the decision. It appears that the COS based its decisions on two main criteria: the acknowledgement of negative conflict of interest (link with competitors) and the unavailability of declarations of conflict of interest, which have to be provided by latest when required by legal authorities. Conclusions: The strengthening of the regulation on declarations of interest might lead to more transparency but also more cases ruled by the COS. PHP279 Measuring and Observing Positive And Negative Externalities Caused By Vaccines: Do We Have The Right Assessment Approach Available? Lefebvre C 1, Terlinden A 2, Standaert B 3 1Independent Researcher, Kiowa, CO, USA, 2Navigha, Tervuren, Belgium, 3GlaxoSmithKline Vaccines, Wavre, Belgium . . . Objectives: Vaccination not only results in direct protection to those being vaccinated, it also has the potential of inducing indirect protection among unvaccinated individuals (=herd protection or positive externality). However, negative unintended consequences or externalities may also result from vaccination programmes (e.g. rebound effects). It is our purpose to present how and when these positive and negative externalities can be observed and measured. Methods: We first identify under what conditions herd protection is most likely to occur. We then explore how negative rebound effects can also be manifested. Detailed illustrations of both positive and negative effects are presented for different infections in relation to mass vaccination programmes. Lastly, we discuss methods for observing and measuring these externalities. Results: Optimal herd protection is likely to be observed when the vaccine has a high quality-induced immunity, substantial effect on the force of infection, and appropriate vaccine coverage and distribution. Example: HPV vaccination of 12-16 year old girls resulted in a 50% decrease of anogenital warts in 15-19 year aged adolescents in Denmark observed over a 4-year period. Rebound effects may potentially occur due to vaccine-related age shifting, decreased natural immunity, serotype replacement, low-medium coverage and non-homogeneous vaccine distribution. Example: increased herpes zoster incidence in elderly postvaricella vaccine introduction. Those externalities can be captured through observational studies using real-life data, or may be estimated using dynamic transmission modelling techniques. Conclusions: Limitations are inherent in those studies and involve substantial ambiguity in the process of observing and quantifying the indirect effects, making accurate evaluation troublesome. However the nature of these outcomes could be critical for achieving good economic value when decision- makers are evaluating a novel vaccine for introduction into a particular region or people group. More investigation is needed to identify and develop successful assessment methodologies for precisely analysing these outcomes. . The aim of this study was to identify the cost-effectiveness threshold for an orphan designations in Poland. According to criteria specified by the European Medicines Agency (EMA) a medicine must meet a strict criteria to qualify for orphan designation, such as: treatment, prevention or diagnosis of a disease that is life-threatening or chronically debilitating; disease prevalence level in the European Union (EU) of no more than 5 cases in 10,000 patients is necessary; no satisfactory method of disease diagnosis, prevention, treatment or if such method exists, the drug must deliver significant benefits to patients. In Poland, orphan drugs undergo full pharmacoeconomic evaluations and coverage decision processes similar to any other innovative medicines. One of the important element of reimbursement recommen- PHP280 Market Access and Reimbursement: The Increasing Role of RealWorld Evidence Pietri G , Masoura P PAREXEL Consulting, London, UK . . Randomised controlled trials (RCTs) have historically been considered the primary source of evidence to support market access and reimbursement. However, real-world data (RWD) are increasingly being considered by industry and payers. The objectives of this research were to review the current perception of RWD across Europe and to assess how RWD can support market access and reimbursement. A review of the literature, guidelines from European health technology assessment (HTA) agencies A451 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 and governmental authorities, case studies and the latest publications in value-based assessment (VBA) was performed to summarise the current perception of RWD, and to identify the advantages and challenges of using RWD to support market access and reimbursement. Only 10 guidelines were found from 73 European HTA agencies or governmental authorities which cited RWD as a source for evidence. NICE acknowledges the difficulties of generalising RCT results to clinical practice, and supports the capture and analysis of observational data. In addition, recent developments in VBA anticipate a greater scrutiny of attempts to model natural history in economic evaluations, which may be addressed by using longitudinal observational data. Case studies have shown economic evaluations based on RCT data may lack external validity, and may consequently produce inaccurate estimates of economic endpoints. There is a consensus that RWD are valuable in providing clinical practice evidence on treatment pathways, resource use, long-term natural history and true effectiveness. However, there are methodological challenges (such as lack of randomisation) to be addressed before RWD are widely accepted as a complement to RCTs to support decision-making. RWD are increasingly recognised as a valuable source of evidence for market access and reimbursement, and as a complement to clinical trial evidence. Nevertheless, there are challenges that need to be addressed to ensure real world data provide valid evidence to the decision process. PHP281 Reimbursement Hurdles For High-Cost Brand-On-Brand Combinations and Impact on Patient Access Leoni G , Papadopoulou K ICON Plc, London, UK . . Combinations of high-cost branded drugs are becoming a reality. The synergistic value of combining two potent drugs is expected to considerably bolster the benefit to the patient in terms of efficacy and, in some cases, even safety. However, the synergistic cost of using branded combinations increases exponentially due to longer treatment durations, thus making the total treatment cost unaffordable to European health systems. This poster aims to explore the pricing and reimbursement issues that health systems will encounter during the evaluation of branded combination therapies and potential solutions to make these combinations affordable by the health care systems and ensure patient access to innovative drugs. To meet these objectives, an in-depth review of published sources was conducted, including a thorough analogue assessment. Moreover, targeted interviews with twenty payers involved at different levels of pricing and market access decision-making in the EU5 were also conducted to support analysis. Research revealed that synergistic costs of already expensive monotherapies, further exacerbated by longer duration of treatment, exceed payers’ cost thresholds. Therefore, on one hand payers will struggle to award a value-based price for the individual drugs as well as for the combination and will look to discount and/or restrict access. On the other hand, as this approach will not reflect the combination’s synergistic value and could threaten the life-cycle indications of each compound, manufacturers may not launch in some markets, thus, limiting patient access. Consequently, it is important to find a balance in setting a value-based price for individual indications and for the combinations to ensure broad patient access. Aligning patient, payer and manufacturer needs is paramount to find a win-win-win solution. In the context of brand-on-brand combinations, traditional pricing models are not the solution and alternative approaches need to be adopted. PHP282 The Case For Early Payer Engagement Leigh C 1, Faulkner E 2, Horowicz-Mehler N 1 Global Consulting, New York, NY, USA, 2University of North Carolina, Durham, NC, USA . . . 1Quintiles Background: Knowledge of payer evidence requirements is vital to manufacturers who are facing increasing development costs for uncertain market access outcomes. Failure to engage payers early in asset development could result in delay of approval and/or coverage. Objectives: Build the case for early payer engagement as a means of reconciling the needs of payers and manufacturers Methods: A literature search was performed and primary research with key opinion leaders in the US and EU was conducted to characterize 5 early engagement strategies (informal consultation, formal consultation, outcomes-based risk sharing, financialbased risk sharing, and formal partnerships). 7 major markets (Canada, France, Italy, Germany, Spain, UK, and US) were also assessed for their historic use of early engagement models. Results: Payers want more manufacturer involvement in evidence development, including input into clinical trial design and RWE development in phases II and III through formal and informal consultations. Articulation of an asset’s value story in the peri-launch phase and negotiations with regional and local payers through direct consultations allows manufacturers to position the asset for optimal pricing and reimbursement. When agreement cannot be reached on price or reimbursement terms, risk-sharing agreements allow broader access in exchange for the manufacturer bearing incrementally greater financial risk. Manufacturers have also built partnerships to uncover the real-world value of therapies and gain insight into usage and adherence patterns. Each market has its own challenges for promoting collaboration, requiring manufacturers to tailor their approach to the various national and local payers Conclusion: Early planning is imperative in value-focused health care. When early payer engagement succeeds, it provides manufacturers time to design informed strategies to meet payer valuation needs. Evidence development that is closely aligned with payer requirements results in therapies that are more cost effective and gain quicker market access, benefitting manufacturers, payers, and patients alike. PHP283 Early Nice Decision Problem Meetings: Implications For CrossFunctional Industry Teams Floyd D , Langham S , Chetty M PHMR Associates, London, UK . . . The National Institute for Health and Care Excellence (NICE) in England and Wales is currently piloting a process whereby decision problem meetings are held several months before starting a technology appraisal, rather than approximately 10 weeks after formal invitation for the manufacturer to submit evidence, per current protocol. In general, the purpose and outcomes of the meeting, involving the NICE team and representatives from the evidence research group (ERG), do not change other than happening earlier. However, the meeting does allow manufacturers and sponsors to signal potential regulatory developments during the appraisal, ahead of the submission, to indicate potential inclusion and handling of patient access scheme proposals. For the meeting, an outline is required to demonstrate how the manufacturer/sponsor intends to approach the decision problem. This outline is to include, but is not limited to: evidence sources to be used; evidence likely to become available during the appraisal and how this might be managed; the planned approach to disease and economic modelling; potential challenges in interpreting the evidence; proposed approach to handling of uncertainty. If adopted, there are several implications of this new process for manufacturers/sponsors: market access strategy will need to be considered earlier than currently, with implications for data availability and analyses, value story development, positioning and indications, etc; cost-effectiveness models and their base cases will need earlier definition and completion; intentions regarding patient access schemes must be made before submission; ERGs may be reviewing limited published evidence in fast-moving therapy areas;, manufacturer market access groups will require more information from clinical, regulatory, medical affairs, modellers, epidemiologists much sooner that they currently do. Therefore, this seemingly simple change of meeting date relative to time of submission has important implications for manufacturers beyond their market access teams that require careful consideration in terms of planning and communication. PHP284 The Irish Cost-Effectiveness Threshold: Does It Support Rational Rationing or Might It Lead To Systematic Damage Of Ireland’s Health System? O’Mahony J 1, Coughlan D 2 College Dublin, Dublin, Ireland, 2Newcastle University, Newcastle upon Tyne, UK . . 1Trinity Irish legislation recognises the need to consider the cost-effectiveness of health services, both for new interventions and their opportunity cost. Ireland did not have an explicit cost-effectiveness threshold until a 2012 agreement between the pharmaceutical industry and government established a € 45,000/QALY threshold. It was agreed as part of a deal that provided cost savings on existing medications and only applies to pharmaceuticals: there is no official threshold for non-drug interventions. Drugs with cost-effectiveness ratios within the threshold are guaranteed reimbursement, whereas those exceeding the threshold may be approved following further negotiation. A number of drugs far exceeding the threshold have been reimbursed in Ireland in recent years. There are four reasons for concern regarding Ireland’s threshold. Firstly, as a price floor not a ceiling it offers only a weak constraint on the introduction of cost-ineffective interventions, which leaves little scope for positive net health benefit. Secondly, that the threshold only applies to drugs creates potential for inconsistencies whereby relatively cost-effective non-drug interventions may not necessarily be approved, leading to sub-optimal resource allocation. Thirdly, the current threshold has no apparent empirical basis. Finally, recent efforts to determine the appropriate cost-effectiveness threshold in the UK have estimated a threshold of approximately £13,000/QALY. Assuming Ireland’s threshold should be broadly comparable, the current Irish threshold is most probably too high. Consequently, reimbursing new interventions at and above the € 45,000/QALY threshold is likely to result in net harm, as new drugs produce less health than the interventions they displace. The failure of Ireland’s threshold to be empirically determined by the cost-effectiveness of services foregone means the requirements of current legislation are not being met and reimbursement decisions cannot be considered fully evidenced-based. It is likely the current threshold is excessive and will lead to systematic damage of the health system. PHP285 An Ethic System Overview: Brazilian Perspectives For Observational Studies Minowa E 1, Bueno C C 1, Piedade A 1, Clark L G O 1, Santinho C S 1, de Castro Monteiro D C 2, Feijo L F 1, Ueda K 3, Matos G M 4, Hashimoto D A K 4 1Evidências Credibilidade Científica, São Paulo, Brazil, 2Roche Product New Zealand, Auckland, New Zealand, 3Harvard Medical School, Tokyo, Japan, 4Universidade de São Paulo, São Paulo, Brazil . . . . . . . . . . . . . . . . . . . Background: Observational studies have been one of the hallmarks for the development of public health and health economics fields. It includes epidemiologic studies, evaluation of patterns of care, use of resources, cost of illness, analysis of safety and effectiveness of interventions from real world. However, there are different patterns of requirements for ethics reviews concerning observational studies, including vastly available models of ethics systems among different countries. Therefore, the objective of this study is to evaluate the ethics system, regulations and guidelines concerning observational studies in the selected countries. Methodology: Guidelines and regulations from Brazil, Argentina, Japan, New Zealand, Australia, USA and UK were reviewed to evaluate the ethics system and available guiding principle for observational studies. Additionally, a literature review was performed in the database Medline and SciELO mesh using the terms “ethics”, “observational study” and “multicenter study” among other similar terms. Results: In Brazil, same ethics regulation is applied for both interventional and observational projects, plus there is unsatisfactory ethics review timelines and duplicity of ethics review when considering multicenter studies. Specific pathways for multicenter studies are available only in New Zealand, Australia, USA and UK. For the exception of Brazil, other evaluated countries have specific guidelines, recommendations or regulations for observational studies. Conclusions: Brazil and Argentina still have a lot of A452 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 challenges to overcome regarding the overall ethics system. Applying same ethics regulations or guidelines from interventional studies may not be the most adequate choice for observational studies. PHP289 Sustainable Health Care Systems: The Role of Therapeutic Value and Value Based Pricing Hulshof J A M Simon-Kucher & Partners, Bonn, Germany . PHP286 State of the Art Research In Austria: Dexhelpp - Decision Support For Health Policy and Planning: Methods, Models and Technologies Based On Existing Health Care Data Zauner G 1, Popper N 1, Breitenecker F 2 GmbH, Vienna, Austria, 2Vienna University of Technology, Vienna, Austria . . . 1dwh The Austrian health care system incurs costs of 30 billion/year, the bulk of the costs (77%) are publicly financed. Health policy and decision planning based on research evidence helps to tackle increasing costs. The urgent need for the evaluation of new health technologies, services, infrastructure, as well as for the development of improved technologies for the analysis, planning and control of health systems is met by DEXHELPP. Methods: Today, decision support in health care is usually based on evidence from studies of limited size, but not yet on the analysis of large volumes of routinely collected health care data. DEXHELPP is dedicated to filling the gap by combining academic excellence by research partners with professional implementation including knowledge of commercial partner institutions. By doing so special methods for statistical analysis, simulation and visualisation will be implemented as well as new methods for documentation and providing of k-anonymity for used individual data. For this task, existing cooperation schemes provide a firm substantial and conceptual knowledge base. All relevant fundamental technological competencies from academic and applied research are provided by the consortium, coming from universities, competence centres, and R&D SMEs. Results: Developed methods will help in (1) analysing the status quo, (2) making reliable prognoses, and (3) evaluating the consequences of interventions. A scientific research server with routine data in order to test developed methods will be run. The project covers all relevant areas within this complex process, from data management via analysis and modelling through to user friendly presentation of results and quality assurance. Some of the most important actual decision-makers in Austria complete the consortium with application-oriented expertise. Conclusions: DEXHELPP focusses on the development on high innovative technological methods. Future focus will lay on building up a network with more stakeholders to integrate those methods in national and international processes. . . It is a common perception that the cost of pharmaceutical care, driven by the price of medicines, is a major contributor to increasing public health care expenditure, the culprit of unsustainability of health care systems. This presentation shows how the net effect of value-based pricing of new, innovative treatments and competitive pricing of older, often generic, products defeats that perception. The growth of pharmaceutical expenditure is, in fact, leveling off in most European countries, a trend that started even before the economic crisis of 2008. Value based pricing is an approach by which the pricing strategy is determined by therapeutic value, economic value and cost-effectiveness. Value based pricing requires a substantial body of sophisticated evidence, generated throughout product development. Competitive pricing is an approach by which the pricing strategy of older products is defined relative to the price of direct competitors in order to maximize market share. Competitive pricing occurs in crowded markets with multiple equal or undifferentiated treatment options. In recent years, competitive pricing has brought the price of many widely prescribed drugs significantly down. These two approaches to strategic pricing take place against the backdrop of tightening public financing rules that aim at ensuring financial sustainability of publicly financed health care systems. Consequence thereof is a two-tiered market access decision-making routine: for high-value/high-priced innovative treatments centralized decision-making based on value assessment with evidence, restricting them to a tightly characterized severely ill patient sub-population with low volume and/or a sales ceiling; versus local decision-making based on lowest (net) price for high-volume multisource products or equivalent single-source products, leading to significantly lower (net) prices than the originator at time of launch. Sustainable pharmaceutical care with openness towards innovation is therefore within reach. PHP290 Economic Evaluation In Portugal – Establishment of The National Health Technology Assessment System (Sinats) Cortegaça P , Silva F , Viana D , Ramos R , Martins J INFARMED, I.P., Lisbon, Portugal . . . . . The rationale for patient and public involvement (PPI) in the Health Technology Assessment (HTA) process has been widely documented. Engagement of patients or public in the process facilitates a broadening of scope and delineates a role for the patient as an ‘expert witness’ who has unique insight to living with an illness and the potential benefits/disadvantages (including side effects) that medical technology may offer. Our objective was to explore the concept of practical and meaningful public participation in the HTA process and present a research protocol to propose a PPI framework applicable to the Irish context. A qualitative systematic literature review and concept analysis was used to identify key attributes relating to patient and public involvement. It is proposed that this knowledge will be supplemented through semi-structured interviews of key informants and a review of a purposive sample of HTA agency websites. Researchers, decision makers and policy makers will contribute to development and refinement a framework through a process of deliberation. Capturing the diverging perceptions of key informants will lead to the enhancement of the proposed framework. The role of the publics will be clarified, various levels and methods of PPI further defined. This research will explore the concept of PPI and suggest a study protocol for the development of a framework for PPI in the context of the Irish HTA process. Economics is a social science that has a focus on allocating limited resources efficiently. In health care is essential to invest properly, assuring access to health technologies with the best cost-effective profile, ensuring the sustainability of National Health Systems. In Portugal, the INFARMED – National Authority of Medicines and Health Products I.P., has been making economic evaluations of medicines for 15 years, and is currently developing the National Health Technology Assessment System – SiNATS, of the utmost importance to the National Health System. Nowadays the assessment of the technologies (relative effectiveness assessment and cost-effectiveness) is focused on medicines, within the reimbursement process and the preliminary assessment to its acquisition by the National Health System’s Hospitals. Consequently, it is always done before the reimbursement decision, as a supporting tool to the decision itself. The purpose of this model is to guarantee a global system, and at the same time, extending it to new technologies besides medicines, e.g. medical devices. In these cases, the cost-effectiveness evaluation will be done through the whole life-cycle of that technology, affecting its price and use considering its real performance; instead of only its market entry. The INFARMED, by developing SiNATS, intends to contribute to minimize expenditures in health and the citizens’ life quality, in order to assure the National Health System’s sustainability and efficient usage of public resources in health. Moreover, this system also aims to observe the technologies’ effectiveness and its usage, with the purpose of reducing wastes as well as to promote and award innovation and equal access to health technologies. This research intends to verify the applicability of the universal key principles of Health Technology Assessment Systems in SiNATS, in order to guarantee that the model achieves its proposing objectives. The study will be focused on the structure of system. PHP288 Patient Access To Life-Saving Medication; Preventing Stock-Outs Due To Parallel Trade PHP291 Market Access and Reimbursement Options For Orphan Durg Hospital Only Medicines In Europe – One Size Fits All? Ratcliffe M 1, Mbanya Z 2, Gielen V 1, Sparrowhawk K 1 1PHMR Associates, London, UK, 2PHMR Associates, Newcastle upon Tyne, UK Vosgerau S 1, Paulus G 1, Plantoer S 2 1IMS Health, Munich, Germany, 2IMS HEALTH GmbH & Co. OHG, Munich, Germany Manufacturers and suppliers of potentially life-saving drugs must ensure adequate supplies of in-date drug formulations for all markets across Europe. Failure to do so risks patient lives, but stock-outs have occurred even when a manufacturer has “over-supplied” a market with many times the amount of drug expected to meet local demand. These stock-outs have occurred as a result of parallel trade. In these cases, as little as a 10% variance on pack cost is enough to trigger parallel trade, which can result from price variations at launch across the EU and currency fluctuations following launch. These situations form a useful series of case studies to determine supply chain issues leading to stock-outs. Based on this case study analysis, a model is being developed to estimate the degree of exposure to risk of parallel trade and its associated potential cost to the manufacturer/supplier, and to patient access. The model identifies and quantifies trade flow patterns and key elements. Consequently, supply planning is possible based on essential metrics identified in the trade flow patterns. An allocation scheme to meet local needs is the key output of the model. The allocation scheme is applied to limit stock to the country to be in line with demand, plus a smaller variance (depending on local conditions) for that market compared to that previously applied. The model is intended for use in key EU markets (France, Germany, Italy, The Netherlands, Spain, Italy, UK) as well as countries in Central and Eastern Europe. The model should help to optimise supply chain planning and operations, thereby minimising the risk of stock-outs. Consequently, all patients requiring specific, potentially life-saving drugs should have access to these treatments. Orphan drugs are medicines used to the treat life-threatening or chronic diseases affecting very rare diseases. From April 2000 to October 2010, 720 drugs received orphan drug designation from the European Medicines Agency (EMA). Hereof, 63 were granted marketing a marketing authorization. Since then - with an annual orphan drug sales volume of more than $6 billion - there has been a steady increase in applications for orphan designation with the Committee for Orphan Medicinal Products (COMP), averaging ten positive recommendations per month. Although central approval of orphan drugs covers 30 European countries, it does not necessarily provide for national availability as each national authority has to agree to market access and reimbursement. Further, the hospital sector is known for its high diversity in terms of national access mechanisms and degree of centralization, exemplified by the Nordic Countries: In Denmark and Norway a highly centralized hospital sector prevails with one main hospital procurement agency (AMGROS in Denmark and LIS in Norway), while in Sweden and Finland several county councils are authorized to make decisions independently from each other. The objective of the present study is to identify and evaluate possible short cut routes to market access and reimbursement for orphan drug hospital only medicines (OD-HOM) in EU and further investigate country specific requirements for identified short cut routes to market access and reimbursement for OD HOM in EU as strategic options. A literature review and further desk research was performed. The following countries are included in the analysis: Nordic Countries, Germany, France, The Netherlands and Spain. The results of the ongoing OD-HOM research will be displayed as a summary PHP287 Proposed Framework for Patient And Public Involvement in the Hta Process In Ireland Walsh C Trinity College Dublin, Montenotte, Ireland . . . . . . . . A453 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 of options and requirements by country. The evaluation of market access options for OD-HOM can ease market access and reimbursement and influence international launch sequences of innovative orphan drugs. patient’s involvement and better health outcomes. Improving service development and access to health advice with care experience are other adjuvant interventions. So patient’s participation is a key better care plan. PHP293 Cost and Duration Of Regulatory Process in an Observational Study in Europe and USA PHP297 Turkish Public Procurement System for Medical Devices:A Guide for Reimbursement Policy? Ori A , Fiori G G , Fernandez S , Longo D , Simoni L MEDIDATA SRL, MODENA, Italy Seyhun O , Can H , Erdol S , Erdogan Ciftci E Medtronic, Inc., Istanbul, Turkey . . . . . . . Objectives: Despite the growing interest in observational studies, there is any defined regulatory process unlike clinical trials, which have a defined one. Currently, international observational studies have to deal with each county regulatory process that means to face times and costs very heterogeneous and not regulated yet by the EU. Here we report the experience of MediData in international observational study conduct. Methods: The start-up process in 7 countries was analyzed as for regulatory process in terms of time and costs. Results: In Europe the faster regulatory process takes 12 weeks (Turkey) and the slowest one takes almost twice as long (Italy 24 weeks), while in the USA they spend almost 16 weeks to conclude the whole process. Usually times and costs have an inverse relationship, but the observational studies analyzed shows that the second cheaper country is Turkey, that comes after France where the process is free. Another example of this variety is the case of Spain and France in the same duration of the process have a difference of spending of 12000 € . Conclusions: These duration and spending really differentiated have a huge impact on the observational study conduction. Especially at the international level, we might run into sponsored studies which could be conducted in countries with cheaper or faster regulation processes, with the real risk of a selection bias of the sites and, therefore, a very partial collection of data. Inevitably, the result will be unrepresentative epidemiological data. It’s very important to have a pan-European regulation that allows the observational study conduction in an homogenous and aligned way and also to have complete and representative epidemiological data. PHP295 Access To Orphan Drugs in France: The Case Of Siklos For The Treatment of Sickle Cell Syndrome Before The Council of State . . . In 2008, Public Procurement Law (PPL) was amended to implement and regulate e-Procurement in Turkey. Non-private, public purchaser hospitals are obliged to enter tender results of their medical device purchases into e-procurement system (EPS) to be eligible for a reimbursement by the Social Security Institution (SSI) in Turkey. Objectives of this research are to examine the impact of the EPS that is currently being used for medical device purchases in Turkey since 2011; and to investigate how the system is being used to define ceiling prices for reimbursement, concurrently assessing the quality and quantity of data uploaded by hospitals. EPS data is downloaded on a GMDN basis from Turkish National Database for Medical Devices, (TITUBB) which is an e-catalogue system that was launched to provide barcode level product registration, search and the e-procurement results. In accordance with the objectives, assessment is conducted on more than ten GMDNs and results are utilized to see the effect of EPS on pricing, comparing the prices on officially published SSI positive lists. Our study shows that a lowest price detected on EPS could be set as the reimbursement price, as was the case for the product; aortic stent graft, contralateral limb, where a defined SSI positive list price was reduced after determination of a lower price on EPS, on account of a mispriced tender record. An extrapolation while searching out the reimbursement prices is needed instead of SSI’s calculations based on merely a retrospective and detection of lowest price practice. Effectiveness of this policy depends on resolving the weaknesses of EPS data in terms of quality and quantity; a misdated tender or a mispriced product could be a ground for an erroneous price setting and tenders that are not recorded at all or deferred might lead to incomplete EPS data to define reimbursement prices. Murteira S 1, El Hammi E 2, Kornfeld A 3, Toumi M 4 1Lundbeck Japan KK, Tokoyo, Japan, 2Evidenz, Tunis, Tunisia, 3Creativ-Ceutical, Paris, France, 4University of Marseille, Marseille, France Disease-Specific Studies Background: Hydroxycarbamid is an antineoplastic molecule commercialized since 1968 in France under the name Hydrea®. In the absence of an alternative licensed drug for the treatment of Sickle Cell Syndrome (SCS), Hydrea® has been routinely used off-label for over 15 years in this indication. In 2007, Siklos® became the first licensed orphan drug indicated in the symptomatic treatment of SCS. However, during the pricing and reimbursement procedures, the French Transparency Committee decided to compare Siklos® to Hydrea® and as such Siklos® was evaluated as bringing a weak additional benefit (ASMR IV) which had a major impact on the price given by the French Economic Committee on Health Care Products (CEPS). The manufacturer of Siklos® entered in litigation with the CEPS and after a long legal procedure, the French Council of State issued two orders by way of interlocutory procedures suspending the decision of the drug price setting and determining the minimal threshold under which Health Authorities could not set the price of this drug. Discussion: The negative assessment of the French HTA could have set a precedent for future evaluations of orphan or pediatric medicines issued from drug repurposing. These decisions are unique in the history of the Council of State’s jurisprudence on several aspects: it was an unprecedented case where a CEPS decision is overturned by the Council of State and the first time that a judicial decision provides guidance on the price level for a pharmaceutical drug in France. This case law has also the consequence to enlarge the criteria taken into account by the CEPS in new drug price setting, especially for orphan drugs. Conclusion: Siklos® case law provided a new legal ground that is in contradiction with available regulation and practice and opens the way for future legal challenges to the CEPS decisions. Mental Health – Clinical Outcomes Studies . . . . PHP296 Patient’s Participation in Improvement of Health Related Outcomes: The Better Care Plan PMH1 The Risk of Metabolic Disorders in Patients Treated with Asenapine or Olanzapine: A Real World Data Study Conducted in Italy and Spain Maina G 1, Ripellino C 2, Pegoraro V 2, Caresano C 3 1University of Turin, Turin, Italy, 2CSD Medical Research S.r.l., Milan, Italy, 3Lundbeck Italia S.p.A., Milan, Italy . . . . Objectives: Second-generation antipsychotic drugs, knows as Atypical Antipsychotics, have a better tolerability than conventional antipsychotics but it has reported that its usage lead to a substantial weight gain, an increase risk of dyslipidemia and type 2 diabetes mellitus. In this article authors assessed the risk of metabolic adverse events associated with Asenapine in comparison with those associated with Olanzapine by studying real world data. Methods: The study was a retrospective analysis based on data extracted from Italian and Spanish Cegedim Stategic Data Longitudinal Patient Data databases. Patients were divided in two cohorts (Asenapine and Olanzapine) according to the inclusion criteria and data from these patients were analyzed starting from 2009 up to 2013. Diabetes and dyslipidemia registrations have been searched in order to assess the risk of developing metabolic adverse events. Results: The retrospective analysis showed a lower risk of developing type 2 diabetes and dyslipidemia associated with Asenapine treatment in comparison with Olanzapine. Scenario analyses supported the robustness of the results. Conclusions: Asenapine is associated with a lower risk of metabolic adverse events than Olanzapine, demonstrating a better safety profile with regard to metabolic effects. Singh A 1, Khera K 1, Rana R 2, Chauhan S 3 1Manipal College of Pharmaceutical Sciences, Manipal, India, 2R.M. Medical College & Hospital, Chennai, India, 3KLE’s College of Pharmacy, Belgaum, India PMH2 Efficacy of the Phosphorylated Tau P181 for the Alzheimer’s Disease Dementia - a Systematic Review and Meta Analysis Objective: Patient’s involvement in their health care is important to improve quality, efficacy and better patient care. It is a strategy to achieve improved health outcomes and lower costs. A main objective of the discussion is focused on “how to inform, encourage and educate the patient to participate in their health care decisions?” The purpose of the discussion also centered on the various patients focused interventions required to improve the health care outcomes by patient’s participation. Background: All patients want the information about the disease and complications they have, but not all want to involve in decision making. Most of the patients preferred to leave the final decision on doctors/clinicians. Engagement of patients in health care and decision making would improve the health outcomes and lowering the cost of therapy or treatment. Discussion: Health care is a complicated system to patients; they struggle to understand their medical conditions and services. All patients must have to know about basic health information and treatment options. This would be possible by encouraging patient’s to play an active role in their health care that could improve health outcomes, efficiency and quality. Patient’s participation and engagement are based on patient activation which refers to understanding, ability, health literacy and willingness of patient to manage his or her health and care. Patients can participate at different levels as direct care or their preference in treatment plan and/or in policy making for future recommendations. Patient focused interventions are discussed as companion to patient’s participation or engagement. Improvement in health literacy, patient safety, improvement in self care and clinical decision making are four primary interventions which promote the Mo J National Evidence-based Health Care Collaborating Agency & Inha University, SEOUL, South Korea . . . . . Objectives: The purpose of this study was to apply a systematic review of the literature to evaluate the diagnostic effectiveness of the Phosphorylated tau p181 for the Alzheimer’s disease dementia. Methods: A systematic literature review was used to evaluate the effectiveness of of the Phosphorylated tau p181 for the Alzheimer’s disease dementia. The Scottish Intercollegiate Guidelines Network (SIGN) tool was used by two evaluators to independently evaluate the quality of the ten studies. The literature review covered from October 27, 1946 to October 22, 2013, and eight domestic databases including KoreaMed and foreign databases including Ovid-MEDLINE, EMBASE, and Cochrane Library were used. Results: A total of 9studies (9 diagnostic evaluation studies) were identified to evaluate Phosphorylated tau. Theeffectiveness of this test was evaluated based on diagnostic accuracy. The diagnostic accuracy for identifying AD of ELISA was high (pooled sensitivity, 0.843 (95% CI 0.818-0.867); pooled specificity, 0.799 (95% CI 0.768-0.828); summary receiver operating characteristic area under the curve 0.9082±0.0236. Primary criteria for inclusion were valid studies on (i) patients with mild cognitive impairment with confirmed or suspected AD and non-AD dementia, and (ii) assessment of tau levels using appropriate comparative tests. Conclusions: Evaluated CSF Phosphorylated tau levels are of potential utility in the differential diagnosis of AD versus non-AD dementias and healthy controls. A454 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PMH3 Adaptive Video Games can Assess and Enhance Cognitive Health Sparrowhawk K 1, Kumar R 1, Harrison J 2 1MyCognition, London, UK, 2Metis Cognition, London, UK . . . Objectives: MyCognition has already demonstrated that its adaptive video game programmes are able to enhance cognitive health in healthy volunteers (FENS, July 2014). It is now using the same technology to assess and train cognitive health in psychiatric and neurological patient populations that have cognitive impairment. Methods: In both the psychiatric and neurological patient population it is planned that all with have their cognitive health assessed prior to commencement of the study to provide a baseline score. This will be for the 5 key cognitive domains (attention, psychomotor speed, episodic memory, working memory and executive functioning). All will then be randomised to either active or control group. The active group comprises adaptive video training of at least one hour per week over a 12-week period on top of standard care. The control group receives standard care alone. All the subjects will have their cognitive health assessed at 0, 4, 8 and 12 weeks. They will also have their condition assessed via a specific scale, e. g., PANS for Schizophrenia, UPDRS for Parkinson’s. The results will be calculated as change from baseline for both groups at 4, 8 and 12 weeks. Results: In previous tests in healthy volunteers a significant improvement in cognitive health was seen after 4 weeks of training. By 8 weeks it was possible to see a dose response to the duration of training, with longer training showing a greater effect. Conclusions: Detrimental cognitive health is seen in many neuropsychiatric conditions. To date, drug therapies have had a poor impact on cognitive disorders. This adaptive video training programme presents a potential therapeutic intervention that is safe and effective. The online assessment can be used to profile at risk subjects and the training could be used to prevent as well as treat cognitive decline. PMH4 Selecting Patients with Severe Personality Disorders using Concept Mapping Goorden M Institute for Medical Technology Assessment, Rotterdam, Zuid Holland, The Netherlands . Objectives: The costs of mental health care for patients with severe and complex personality disorders are high. It is likely that currently, these patients are not treated effectively due to limited early recognition. The aim of this study is to develop a set of criteria to match patients with severe and complex personality disorders to effective treatment in highly specialized treatment programs. Methods: The subsequent steps consisted of conducting a literature search, applying concept mapping, arranging an expert-meeting and validating the set of criteria. After the literature search, concept mapping, using cluster analysis and multidimensional scaling, was conducted. Goodness of fit and reliability were tested. A consensus meeting was arranged to determine cut-off points and operationalize the clusters. In this way, the definite set of criteria for matching patients with severe and complex personality disorder was developed. A pilot study on the checklist was conducted, including 20 therapists evaluating 45 patients and subsequently a validation study at 8 mental health institutes was performed. Additionally, a methodology to obtain sets of criteria for other psychiatric diagnoses was developed. Results: After the literature search and concept mapping, 6 different categories of criteria were defined. The concept mapping model provided a good fit (stress value= 0.30) and reliability (ρ = 0.49) was reasonably high. The bridging values were on general low, indicating homogeneity. The pilot study indicated that clinical judgment and the outcome of the checklist correlated high (0.8). The validation study has been accomplished at 8 mental health institutes in the Netherlands. The analysis concerning the validation study is now being conducted and the first results will be available in August 2014. Conclusions: The method is systematic, structured and repeatable and the results of the pilot study are promising. The protocol will provide an instrument to develop sets of criteria for other disorders to increase matching of patients in mental health care to (cost) effective interventions. PMH5 Systematic Literature Review and Mixed Treatment Comparison of Gxr Versus other Treatments in Children and Adolescents with Attention Deficit Hyperactivity Disorder (ADHD) Joseph A 1, Ayyagari R 2, Bischof M 3, Cai S 4, Xie M 2, Zhanabekova Z 2, Sikirica V 5 Group Inc., Boston, MA, USA, 3Shire, Eysins, Switzerland, 4Analysis Group Inc., New York, NY, USA, 5Shire Development, LLC, Wayne, PA, USA . . . . . . . 1Shire, Zug, Switzerland, 2Analysis Objectives: This study compared the clinical efficacy of ADHD treatments in children and adolescents. Methods: A systematic literature review was conducted, according to National Institute for Health and Care Excellence guidelines, to identify randomized controlled trials (RCTs) of guanfacine (GXR), atomoxetine (ATX), lisdexamfetamine (LDX), and methylphenidate (MPH) extended release (ER) and immediate release (IR) in children and adolescents with ADHD. A Bayesian mixed treatment comparison was conducted to compare baseline-to-endpoint change in ADHDRS-IV score, response (defined as a clinician global impressions – improvement [CGI-I] score ≤ 2), with meta-regression adjustments permitted by data availability (age and percent female). 95% credible intervals (CrIs) for treatment effects and the posterior probability that GXR was more efficacious than each treatment were estimated. Results: Of 5,619 records retrieved, 29 RCTs met the inclusion criteria. Five trials included GXR, 4 included LDX, 16 included ATX, 7 included MPH-ER, and 5 included MPH-IR. Per-arm patient sample size ranged from 29 to 222. The mean ADHD-RS-IV score change from baseline and 95% CrI (active minus placebo) were -8.68 (-10.63, -6.72) for GXR, -14.98 (-17.14, -12.80) for LDX, -6.88 (-8.22, -5.49) for ATX, and -9.33 (-11.63, -7.04) for MPH-ER. The relative risk and 95% CrI for CGI-I response (drug vs placebo) were 2.13 (1.68, 2.59) for GXR, 2.93 (2.47, 3.40) for LDX, 2.30 (1.79, 2.81) for MPH-ER, 1.97 (1.43, 2.58) for ATX, and 1.66 (1.02, 2.32) for MPH-IR. Among non-stimulants, GXR was more effective than ATX when comparing ADHD-RS-IV change (with a posterior probability of 93.91%) and CGI-I response (posterior probability 71.01%). Conclusions: This review found that LDX had greater efficacy compared with GXR, ATX, and MPH in the treatment of children and adolescents with ADHD with no overlap in CrIs. Among non-stimulants, although GXR had a higher probability of being more efficacious than ATX, their CrIs overlapped. PMH6 An Evaluation of the Comparative Effectiveness of Clomethiazole Against Diazepam in the Treatment of Alcohol Withdrawal Syndrome in Routine Clinical Practice Kamudoni P 1, Gründer G 2, Sychla H 2, Juha B 3 University, Cardiff, UK, 2RWTH Aachen University, Aachen, Germany, 3Cheplapharm Arzneimittel GmbH, Greifswald - Insel Riems, Germany . . . . 1Cardiff Objectives: The objective of this study was to assess the comparative effectiveness of clomethiazole against diazepam in the treatment of alcohol withdrawal syndrome (AWS) in the real-world. Methods: Following a retrospective design, this study is based on case report notes of patients consecutively treated for AWS at a University Clinic in Germany (Aachen), from 2008 to 2013. The primary outcomes were: scores of German version of the revised clinical institute withdrawal assessment for alcohol scale (CIWA-Ar) in the first four days of therapy; duration of therapy; duration of hospital stay and rates of complications. Patients in the two groups (diazepam and clomethiazole) were matched for demographic factors and severity of withdrawal using propensity scores. Results: Seventy nine patients (Diazepam = 42; Clomethiazole = 37) were included. Mean age of patients (M = 69%, F = 30.4%) was 45±15 years. Duration of excessive alcohol use was 16 ± 12 years. On admission mean systolic and diastolic blood pressure was 139±20 and 88±13, respectively. Pulse rate was 100 ±18 per minute. CIWA-Ar scores at the end of each of the first four days of therapy were not statistically significantly different (ANOVA, p < 0.5) for the two therapies. Average duration of therapy was 3.81±2.9 days vs. 5.8 ±3.2 days (ANOVA - F = 8.6, p = 0.004), for clomethiazole and diazepam groups, respectively. Duration of hospital stay was 14.92±10.1 vs. 15±13 (ANOVA – F, p > 0.2), respectively. There was no difference in complication rates, except for seizures (higher in Diazepam group, Chi-square test, p = 0.023). Conclusions: Clomethiazole and Diazepam seem to offer comparable effectiveness in resolving symptoms of AWS. Otherwise the notably shorter therapy duration and lower rates of complications for clomethiazole hints at some advantages such as faster recovery and lower costs of treatment, although this would have to be confirmed in future. PMH7 Predictors Pf Remission in the Treatment of Depression in the Middle East: Real-World Evidence From A 6-Month Prospective Observational Study Novick D 1, Jihyung Hong J 1, Montgomery W 2, Dueñas H 3, Elfatarany G 4, Haro J M 5 1Eli Lilly Holdings Limited, Windlesham, UK, 2Eli Lilly Australia, Sydney, Australia, 3Eli Lilly de Mexico, Mexico City, Mexico, 4Eli Lilly & Company, Saudi, Riyadh, Saudi Arabia, 5Parc Sanitari Sant Joan de Déu, CIBERSAM, Universitat de Barcelona, Barcelona, Spain . . . . . . . Objectives: To understand the potential predictors of remission among patients treated for major depressive disorder (MDD) in a naturalistic clinical setting in the Middle East. Methods: Data for this post-hoc analysis were taken from a 6-month prospective, non-interventional, observational study that involved 1,549 MDD patients without sexual dysfunction at baseline in twelve countries worldwide (n= 314 in the Middle East). Depression severity was measured using Clinical Global Impression of Severity (CGI-S) and 16-item Quick Inventory of Depressive Symptomatology Self-Report (QIDS-SR16). Depression-related pain was measured using the pain-related-items of Somatic Symptom Inventory (SSI). Remission was defined as a QIDS-SR16 score ≤ 5. Generalised estimating equation (GEE) regression model was used to examine baseline factors associated with remission at each follow-up visit. The model included age, sex, region, CGI-S score, QIDS-SR 16 total score, SSI pain total score, and treatment (duloxetine vs. a selective serotonin reuptake inhibitor [SSRI] ) at baseline. Other baseline factors were also included if they appeared to be significant at p< 0.1 in simple GEE models. The model analysed the patient observations up to the point where their initial medications were maintained. Results: Of the 240 patients analysed, 133 (55.4%) initiated duloxetine and 107 (44.6%) initiated an SSRI at baseline. Of these, 199 patients achieved remission at any visit during follow-up (91.0% in the duloxetine group and 72.9% in the SSRI group, p< 0.001). The GEE results showed that higher QIDS-SR16 scores at baseline (odds ratio [OR]= 0.88, p= 0.002) and more MDD episodes in the past 24 months (OR= 0.77, p< 0.001) were negatively associated with remission, whereas treatment with duloxetine (vs. SSRIs) was positively associated with remission (OR= 2.78, p< 0.001). Conclusions: Treatment with duloxetine (vs. SSRIs), a lower level of depression severity and fewer previous MDD episodes appeared to be strong predictors of achieving remission in the treatment of MDD in the Middle East. PMH8 A Network Meta-Analysis of the Relative Efficacy of Pharmacological and Psychological Interventions in Adults with Obsessive Compulsive Disorder Bryden P A 1, Caldwell D M 1, Welton N 1, Churchill R 1, Baxter H 1, Lewis G 2, Skapinakis P 2 1University of Bristol, Bristol, UK, 2University College London, London, UK . . . . . . . . . Objectives: Obsessive compulsive disorder (OCD) is the fourth most common mental disorder in the UK with a prevalence of 1.5%. Both pharmacological and psychological interventions are used in the treatment with the treatment course usually chronic. This study estimates the relative efficacy of pharmacological and psychological treatments or combinations of both. Methods: A systematic review was conducted to identify RCTs of clomipramine, SSRIs, venlafaxine, and psychological interventions with a behavioural, cognitive or cognitive behavioural component. A Bayesian random effects network meta-analyses (NMA) was used to obtain coherent estimates of relative efficacy of every pair of the different interventions, even when direct evidence was not available. The primary endpoint was mean change in Yale-Brown obsessive compulsive scale (YBOCS). We assumed that treatment effects for the different SSRIs were similar in the sense that they came A455 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 from a common Normal distribution of treatment effects with an overall SSRI class effect mean, and between treatment within class heterogeneity. Results: There were 55 eligible studies identified in the systematic review. The intervention with the greatest decrease in YBOCS was behavioural therapy (“exposure and response prevention”) showing a decrease of 13.86 (CrI 9.34 to 18.31). The second and third greatest decrease was cognitive therapy (12.80 CrI 7.39 to 18.18) and behavioural therapy plus clomipramine (12.47 CrI 5.80 to 19.08) respectively. The SSRI class effect showed a relative decrease in mean YBOCS of 2.89 (CrI 1.05 to 4.71) compared to pharmacological placebo. The results of the individual SSRIs ranged from a decrease of 2.49 (sertraline) to 3.10 (fluvoxamine). Conclusions: This analysis showed a combination of behavioural therapy plus clomipramine has the greatest decrease in YBOCS. There is little evidence to show a difference between SSRIs. PMH9 Systematic Review and Mixed Treatment Comparison of Lithium or an Atypical Anti-Psychotic (AAP) used to Augment a Selective Serotonin Reuptake Inhibitor (SSRI) in Treatment Resistant Depression (TRD) Edwards S J , Wakefield V , Nherera L , Trevor N BMJ, London, UK . . . . . Objectives: To estimate the clinical effectiveness of augmentation of SSRI antidepressant therapy with either lithium or an AAP in TRD, defined as failure to respond to two or more antidepressants in the current episode of depression. Methods: Systematic review of CENTRAL, EMBASE, MEDLINE, and PsycINFO was completed in August 2011. Additional data were obtained from manufacturers. Studies were assessed for quality using the Cochrane Risk of Bias Tool. Pairwise meta-analysis and mixed treatment comparison (MTC) were undertaken based on intention-totreat analyses. Results: Of the 3,721 papers found in the literature search, 12 randomised controlled trials (RCTs) were identified; 10 (SSRI + AAP vs SSRI + placebo/no treatment); 1 (SSRI + AAP vs SSRI + lithium); 1 (SSRI + lithium vs SSRI + placebo). The RCTs included in the primary analyses used fluoxetine as the SSRI and olanzapine as the AAP. Results of the MTC showed a non-significant trend in favour of lithium augmentation for response [lithium odds ratio (OR) 1.29; 95% credible interval (95% CrI): 0.11 to 5.32], mean change in Montgomery-Åsberg Depression Rating Scale (MADRS) score from baseline (mean difference -1.47; 95% CrI: -9.10 to 6.41) and all-cause withdrawals (OR 0.74; 95% CrI: 0.10 to 2.66). Conclusions: In patients with TRD, there is a lack of direct evidence comparing the clinical effectiveness of augmenting an SSRI with an AAP compared with augmenting with lithium. Augmentation of SSRIs with lithium or AAP is likely to be beneficial in people with TRD. The limited evidence indicates no statistically significant difference between the two augmentation strategies. PMH10 Relationship of Insight with Medication Adherence and the Impact on Outcomes in Patients with Schizophrenia and Bipolar Disorder: Results From A 1-Year European Outpatient Observational Study Novick D 1, Montgomery W 2, Treuer T 3, Aguado J 4, Kraemer S 5, Haro J M 6 1Eli Lilly Holdings Limited, Windlesham, UK, 2Eli Lilly Australia, Sydney, Australia, 3Eli Lilly & Company, Budapest, Hungary, 4Parc Sanitari Sant Joan de Deu, CIBERSAM, Sant Boi de Llobregat, Spain, 5Eli Lilly and Company Ltd, Bad Homburg, Germany, 6Parc Sanitari Sant Joan de Déu, CIBERSAM, Universitat de Barcelona, Barcelona, Spain . . . . . . . Objectives: Many patients with schizophrenia and bipolar disorder have impaired insight and low medication adherence. The aim of this post-hoc analysis is to explore the relationship between insight and medication adherence and their impact on the outcomes of patients with schizophrenia or bipolar disorder. Methods: We included 903 patients with schizophrenia or bipolar disorder who participated in an observational study conducted in Europe on the outcomes of patients treated with two oral formulations of olanzapine over a 1-year period. Evaluations included Clinical Global Impression (CGI), Global Assessment of Functioning (GAF), insight (Scale to Assess Unawareness of Mental Disorder, SUMD), non-adherence (Medication Adherence Rating Scale, MARS), and therapeutic alliance (Working Alliance Inventory, WAI). Correlations between variables were assessed by Spearman Correlation Coefficient (SCC). A path analysis was used to understand the relationship between insight, adherence, therapeutic alliance and outcomes. Results: 67.8% of patients had schizophrenia. GAF score was higher in bipolar vs schizophrenia patients (mean (SD) 58.4 (15.6) vs 51.9 (15.7), p< 0.001). Medication adherence was also higher in bipolar patients (mean MARS score (SD) 6.5 (2.8) vs 5.8 (2.7); p< 0.001). Patients with schizophrenia had lower insight (i.e. SUMD item 1, unawareness of mental disorder, mean (SD) of 2.5 (1.3) in schizophrenia vs 1.9 (1.2) in bipolar, p < 0.001). Better insight was associated with higher adherence (SCC, ranging from 0.39 to 0.49 for the three SUMD general items, p< 0.0001 in all cases) Higher insight was related to a stronger therapeutic alliance (SCC ranging from 0.38 to 0.48, p< 0.0001). The path analysis revealed a positive impact of insight on adherence and alliance and that stronger alliance was related to lower clinical severity (lower CGI score). Conclusions: Insight and adherence were found to be closely related. Insight impacts on the therapeutic alliance with mental health professionals. These factors are associated to treatment outcomes. PMH11 Social Contacts Reduce Negative Symptoms, Especially Emotional Withdrawal in Patients with Schizophrenia Millier A 1, Siegrist K 2, Amri I 3, Toumi M 4, Aballéa S 1 University Düsseldorf, Düsseldorf, Germany, 3Creativ-Ceutical, Tunis, Tunisia, 4University of Marseille, Marseille, France . . . . . 1Creativ-Ceutical, Paris, France, 2Heinrich-Heine Objectives: In schizophrenia, negative symptoms - especially emotional withdrawal (EW) - represent an important dimension, and are associated to a significant burden. Social contacts are likely to reduce negative symptoms and ameliorate quality of life (QoL) over time. Our objective was to test whether this hypothesis was verified in a large cohort of European patients with schizophrenia. Methods: We used data from the EuroSC study, a longitudinal cohort of 1208 patients with schizophrenia followed for 2 years. Every 6 months, the collected information included QoL-Interview, from which the Global Satisfaction Score (GLS) and the frequency of social contacts score were derived, and the Positive And Negative Symptoms Scale (PANSS), from which EW score was derived. After bivariate and correlation analyses, we tested whether few social contacts at baseline would predict greater EW and lower GLS after 2 years when adjusted on baseline level. Finally, randomeffects regression analyses were performed to test the longitudinal effect of social contact, adjusting on potential confounding factors. Results: Bivariate and correlation analyses established a link between frequency of social contact and both EW score (-0.24, p< 0.001) and negative factor scale (-0.30, p< 0.0001) at each time point. Few social contacts at baseline were associated with greater EW (p= 0.013) and worse negative factor score (p= 0.009), when compared to baseline. A trend for prediction of better QoL was also found, although not reaching significance. Random effects regressions confirmed the significant impact of social contacts over time on EW (p< 0.0001), negative factor score (p< 0.0001) and QoL (p< 0.001). Conclusions: Given consistent effects of social contacts on reduction of negative symptoms and improvement of QoL in schizophrenic patients, social contacts should be used as a therapeutic tool. A higher frequency of social contacts could be obtained by regular therapeutic groups offered to these patients. PMH12 Outpatient Treatment of Adolescents in Japan with Drugs for Attention Deficit Disorders Inagaki A 1, Nishimura Y 2, Otsuka H 2, Hirakawa H 2, Hatou K 2, Kubota Y 3, Watanabe Y 3, Miki K 3, Endoh Y 4 1Aoyama Gakuin University, Tokyo, Japan, 2Japanese Association of Mental Health Services, Tokyo, Japan, 3Japanese Association of Neuro-Psychiatric Clinics, Tokyo, Japan, 4Institute of Neuropsychiatry, Tokyo, Japan . . . . . . . . . Objectives: To examine prescription patterns of drugs for the treatment of attention deficit disorders in Japanese children and adolescents. Methods: We conducted a cross-sectional survey during October 2013 on outpatients aged 19 years or less in 34 private mental clinics. Patients who were prescribed at least one drug for the treatment of attention deficit disorders were analyzed in this report. Data were extracted on gender, age, principal psychiatric diagnosis (based on ICD-10), and types and doses of psychotropic drugs. Results: The samples consisted of 286 males and 51 females. The average age (standard deviation) was 11.6 years (3.1). The mean length of psychiatric treatment was 21.3 months (24.0). The most frequent principal diagnostic category was “behavioral and emotional disorders with onset usually occurring in childhood and adolescence” (F9; n= 237), followed by “disorders of psychological development” (F8; n= 99), and “mental retardation” (F7; n= 1). Of 337 samples, 247 (73.2%) were prescribedOROS methylphenidate (OROS-MPH), a psycho-stimulant, while 141 (41.8%) received atomoxetine (ATMX), a selective noradrenalin reuptake inhibitor. OROS-MPH/ATMX combination therapy was administered to 51 (15.1%) of 337 patients. Antipsychotics were concurrently prescribed in 80 (23.7%) patients. Mood stabilizers were co-prescribed in 20 (5.9%) cases. Antidepressants were co-prescribed in 19 (5.6%) patients. Anxiolytics/hypnotics were concurrently prescribed in 13 (3.9%) patients. Conclusions: In Japan, nearly one-sixth of the outpatients with attention deficit disorders received OROS-MPH/ATMX combination therapy. PMH13 The Quality of Prescribing for Psychiatric Patients Soerensen A L 1, Nielsen L P 2, Poulsen B K 2, Lisby M 2, Mainz J 3 University, Aalborg, Denmark, 2Aarhus University Hospital, Aarhus, Denmark, 3Aalborg Psychiatric University Hospital, Aalborg, Denmark . . . . . . . . 1Aalborg Objectives: Prescribing for adult psychiatric patients is often highly complex due to the nature of psychiatric conditions, but also due to somatic comorbidity. Therefore, the aim of this study was to identify prevalence and types of potential inappropriate prescribing (PIP), asses the severity of potential clinical consequences and identify possible predictive factors of PIP. Methods: The study was designed as a prospective study of PIP using medication reviews. Patients who were admitted during a 4 month period (August 2013 - November 2013) to a psychiatric university hospital were included (n= 219). The medication reviews, including an assessment of potential severity, were carried out by clinical pharmacologists after admission and after the attending physician had seen the patient. Frequencies and categories of PIP were analyzed in absolute numbers and as percentages. Severity of PIP was assessed using four categories. Logistic regression analysis was used to identify possible predictive factors of PIP. Results: The proportion of patients with one or more PIPs was 123/219 (56%). “Interaction between drugs” was the most common category for potentially serious and potentially fatal PIPs with 49/123 (40%) and 32/45 (71%), respectively. Of 32 identified potentially fatal drug-drug interactions, 15/32 (47%) involved two or more antipsychotic drugs and 12/32 (37%) involved antipsychotic drugs in combination with antidepressants. The remaining 5/32 (16%) potentially fatal drug-drug interactions involved propranolol, erythromycin, simvastatin and promethazine. After adjusting for age, gender, alcohol/substance abuse, number of prescriptions, number of somatic diagnoses and level of kidney function, only polypharmacy (> 5 prescriptions) increased the odds for a PIP significantly; OR= 4,82 (95%CI: 2.33-9.98), p< 0.0001. Conclusions: PIP is frequent and might have serious or fatal consequences. Special attention should be given to drug-drug interactions involving antipsychotics and antidepressants but also somatic medications and polypharmacy threatens medication safety. There is a pressing need to improve the quality in prescribing for psychiatric patients. PMH14 The Prevalence and Disease Burden of Treatment-Resistant Depression - a Systematic Review of the Literature Kubitz N 1, Vossen C 2, Papadimitropoulou K 2, Karabis A 2 1Janssen-Cilag GmbH, Neuss, Germany, 2Mapi, Houten, The Netherlands . . . . A456 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1Creativ-Ceutical, Paris, France, 2ZRx Objectives: Major depressive disorder affects approximately 10-15% of the population and is associated with significant morbidity and mortality. It is one of the leading causes of disability in young adults. A large proportion of the burden can be attributed to treatment-resistant depression (TRD). To understand the prevalence and disease burden of TRD in Western European countries, the US and Canada, a systematic literature search was performed. Methods: OVID, the Cochrane Library and the CRD database were used to retrieve TRD publications in English language from January 2003-October 2013. In total, 6306 abstracts were identified. Predefined selection criteria regarding study design, patient population (age ≥ 12 years; US, Canada, Germany, Italy, France, Spain or UK; TRD defined as one treatment failure and high symptom severity e. g. MADRS ≥ 31, or an inadequate response to ≥ two antidepressants) and outcomes of interest were applied. Results: Only seven studies included prevalence and/or disease burden data. Five studies provided prevalence estimates which adhered to the strict TRD definition used for this review. Study design and definition of the patient population were critical in determining the prevalence rates, with the lowest rates found in US employer claims databases (11-15%), higher rates in commercial health insurance databases (29-31%) and the highest rates in a European multicenter study (51-56%). The database studies mainly included employed patients thereby likely underestimating the prevalence, whereas the European study likely overestimated the prevalence due to a less stringent TRD definition. Inconsistent data were reported regarding treatment outcomes, comorbidities, hospitalization and work productivity. There was no information on other outcomes such as health-related quality of life or functioning. Conclusions: No consistent data were found in the literature from January 2003-October 2013 regarding the epidemiology and disease burden of TRD. To determine the prevalence and disease burden for TRD, further studies are needed. Objectives: Buprenorphine/naloxone (BUP/NAL) combination is a well known treatment for opioid dependence. As a chronic relapsing disorder, some patients alternate between periods of on treatment and off treatment. The aim of this study was to compare health care resource utilization and costs between these patients and patients treated continuously. Methods: Statistical analyses were conducted on a Medicaid insurance claims database (TruvenHealth MarketScan® Medicaid) from January 2007 to June 2012. Patients with at least two treatment episodes in the first year after the initial filled prescription were identified. The end of a treatment episode was defined as a period of 60 days with no filled BUP/ NAL prescriptions following the theoretical end of the last filled prescription. An ordered logistic regression model was used to analyze the impact of initial treatment episode duration on the number of new episodes in the year following the end of the first episode. Health care resource utilization and related costs during the first year after initiation were compared between the two groups. Results: 2,223 patients were included in the analysis. During the first year, 86% of patients had only one treatment episode, 13% had two and 1% had three. Compared to patients who remained in treatment continuously over 12 months, the multiple treatment episode group had lower medication costs (-$2,877) but higher psychiatric inpatient costs (+$720), non-psychiatric inpatient costs (+$2001) and emergency room costs (+430) over 12 months. Total health care costs over 12 months were higher among multiple treatment episode patients ($16,583 vs. $15.123, p= 0.0004). Conclusions: Despite lower medication costs, total health care costs over 12 months were higher among patients with multiple treatment episodes compared to patients treated continuously. PMH15 Prevalence of Metabolic Syndrome in Patients with Schizophrenia According to the Presence or Absence of Negative Symptoms PMH18 Treatment Cost Comparison: Paliperidone Palmitate Versus Risperidone Long Acting in Brazil Sicras-Mainar A 1, Ruiz-Beato E 2, Mauriño J 2, Navarro-Artieda R 3 1Badalona Serveis Assistencials, Badalona. Barcelona, Spain, 2Roche Farma, S.A, Madrid, Spain, 3Hospital Universitari Germans Trias i Pujol, Badalona, Spain Pititto L , Guarniero F , Antonio M Janssen Cilag, São Paulo, Brazil . . . . Objectives: The aim of this study was to estimate the prevalence of metabolic syndrome (MS) in patients with schizophrenia according to the presence or absence of negative symptoms. Methods: A retrospective, cohort study was conducted using electronic medical records from the health provider BSA (Badalona, Spain). All adult outpatients with a diagnosis of schizophrenia were followed for 12 months. Two study groups were defined by the presence or absence of negative symptoms based on the PANSS Marder Negative Symptoms Factor (N1-N4, N6, G7 and G16). MS prevalence was estimated using the NCEP ATP III criteria. Descriptive statistics and logistic regression models were applied. Results: We studied 1,120 patients (mean age: 46.8 ± 13.8 years; male: 58.4%). One or more negative symptoms were present in 52.5% of patients (95%CI: 49.6-55.4%). Dyslipidemia (48.7%), hypertension (38.2%), and diabetes mellitus (19.3%) were the most frequent comorbid conditions. Quetiapine, risperidone and olanzapine were the most common antipsychotic drugs administered. Patients with negative symptoms showed a greater mean number of comorbid conditions than patients without this symptomatology (8.5 and 7.0, respectively; p<0.001). Prevalence of MS was 38.6% (CI: 35.7-41.5%), higher among patients with one or more negative symptoms (43.9% vs. 34.9%, respectively; p= 0.002). MS was associated with the presence of negative symptoms, age, and comorbidity (OR= 1.6, 1.2, and 1.2, respectively; p<0.05). Conclusions: Further studies are necessary to elucidate the association between the presence of negative symptoms and MS among patients with schizophrenia as well as the underlying mechanisms involved. PMH16 The Potential Benefits of Long-Acting Atypical Antipsycothic Therapy in Preventing Relapse in Brazil . . . . . . 1Deloitte Access Objectives: To quantify the economic burden of schizophrenia relapse in Brazil, and to estimate the impact of atypical Long Acting Injectables (LAIs) on relapse. Methods: Administrative health service data from a Brazilian public system database (DATASUS) were used to estimate the number of relapse patients and related resource utilisation. Corresponding data for private system patients were estimated based on published literature and by extrapolating DATASUS data. A prevalence-based costing with a mixed bottom-up and top-down approach was used to quantify direct and indirect costs, disability adjusted life years (DALYS) and their associated monetary value. A decision-analytic model was constructed to evaluate the cost effectiveness of potentially transferring non-compliant patients from oral antipsychotics to atypical LAIs. All costs are presented in 2013 Brazilian real. Results: In 2011-12, 88,721 patients with schizophrenia in Brazil experienced 263,037 episodes of relapse that resulted in hospital or outpatient care. The potential avoidable health care cost of relapse was R$722.6 million. The estimated additional health care cost per DALY avoided was R$5,049 if non-compliant patients could be transferred to atypical LAIs to achieve 5% overall utilisation. Reducing relapses would give Brazil potential avoidance of 1,335 DALYs, which corresponds to a saving of R$482.8 million in the stock of health capital. Conclusions: The economic burden of schizophrenia relapse in Brazil is significant. Brazilian policymakers should provide greater access to LAIs. PMH17 Analysis of ‘Revolving Door’ Patients in Opioid Dependent Patients: the Impact of Treatment Discontinuation on Relapse Rates and Health Care Costs in us Public Health Insurance Claims Clay E 1, Zah V 2, Kharitonova E 1, Ruby J 3, Aballéa S 1, Khemiri A 4 . . . . . Objectives: To compare the treatment cost of paliperidone palmitate (PP) versus risperidone long acting (R-LA), both indicated for the treatment of schizophrenia in Brazil. Methods: In Brazil, both (PP and R-LA) long acting 2nd generation antipsychotics are approved for the treatment of schizophrenia. Published literature shows no difference in safety and efficacy between them; therefore, a cost-minimization analysis was performed. Yearly treatment costs were calculated for an average dose of 37,5 mg per patient in the case of R-LA and 75 mg in the case of PP. The two initial treatment doses were considered: for PP, 150 mg on the 1st day and 100 mg on the 8th day, and for R-LA 21 days oral supplementation with 3 mg of risperidone, according to dosing intervals defined in the product label. Prices were gathered from the official price list (CMED – Apr’14). Results: PP has the lowest cost of treatment, at R$ 12,739 per patient in the 1st year – against R-LA R$ 18,165 – and R$ 11,359 in the 2nd year (R-LA has R$ 17,971). Treatment with PP compared to R-LA may bring important savings to the payers (HMOs or Government), with potential to reduce the cost of treatment by 30% in the 1st year, and 37% in the 2nd year - allowing a higher number of patients to be treated at the same budget level. Conclusions: Although both molecules, PP and R-LA, have demonstrated similar efficacy, PP offers a cost reduction from the perspective of the Brazilian private health care system compared to R-LA. In addition, PP offers advantages that can have additional value for public and private payers alike such as a monthly injection and no need for cold chain. PP can therefore be considered a cost-saving therapeutic option for schizophrenia compared to R-LA. . . . Kellar J 1, Mittmann N 2, von Heymann C 3, Zingaro J 3, Kuriakose B 4, Li A 4 1Centre for Addiction and Mental Health, Toronto, ON, Canada, 2Sunnybrook Health Sciences Centre, Toronto, ON, Canada, 3Cubic Health Inc., Toronto, ON, Canada, 4Janssen Inc, Toronto, ON, Canada . Tay-Teo K 1, Pezzullo L 1, Violin B 2, Dias T 2, Sardi P 2, Delatorre R 2, Pititto L 3, Guarniero F 3 Economics, MELBOURNE, Australia, 2Monitor Deloitte, Sao Paulo, Brazil, 3Janssen Cilag, São Paulo, Brazil . . PMH19 Costs of Employees with Treatment-Resistant Depression Based on a Canadian Private Claims Database Mental Health – Cost Studies . Outcomes Research Inc., Belgrade, Serbia and Montenegro, Benckiser Pharmaceuticals, Inc. /NA, Richmond, VA, USA, 4Creativ-Ceutical, Tunis, Tunisia 3Reckitt . . . . . Approximately 10-20% of individuals with Major Depressive Disorder (MDD) fail to respond to antidepressant monotherapy. These individuals with treatment resistant depression (TRD) have been found to be frequent users of health care services, thus incurring significantly greater costs than those without TRD. Objectives: To investigate the cost of Treatment-Resistant Depression from a private payer perspective in Canada. Methods: An employer-sponsored benefits plan database (2011/2012) was used to define a cohort of Non-TRD and TRD claimants. TRD claimants are defined as those on their third antidepressant monotherapy; or combination antidepressant therapy; or antidepressant augmented with lithium, thyroid hormone or an antipsychotic medication. The cost of prescription medication utilization, short-term disability (STD), and long-term disability (LTD) benefits for employees was calculated (2011 and 2012 $CAN) for both Non-TRD and TRD groups. Descriptive statistics were used to characterize the cohort of claimants and employees, as well as resources and costs for employees. Results: There were 55,324 and 61,028 employee claimants in 2011 and 2012, respectively. 717 (1.3%) and 798 (1.3%) were TRD claimants; 4,744 (8.6%) and 5,137 (8.4%) were NonTRD claimants in 2011 and 2012, respectively. In 2011, the medication costs for treating depression was $774 per TRD employee claimant compared to $303 per Non-TRD claimant. STD costs were $6,263 for TRD (n= 79) and $5,855 for Non-TRD (n= 276). LTD costs were $13,598 for TRD (n= 80) and $12,272 for Non-TRD (n= 119). In 2012, the medication costs for treating depression per TRD employee claimant was $794 compared to $293 for Non-TRD claimants. STD costs were $7,832 for TRD (n= 86) and $4,001 for Non-TRD (n= 248). LTD costs were $13,927 for TRD (n= 89) and $12,901 for Non-TRD (n= 121). Conclusions: Claimants identified with TRD had higher medication, STD and LTD costs than those with Non-TRD. Limitations include lack of diagnostic information for claimants and small sample sizes for STD and LTD subgroups. A457 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PMH20 The Societal Costs of Schizophrenia in Switzerland Pletscher M 1, Mattli R 1, Reich O 2, Von Wyl A 3, Wieser S 1 Institute of Health Economics, Zurich University of Applied Sciences, Winterthur, Switzerland, 2Helsana Insurance Group, Dübendorf, Switzerland, 3School of Applied Psychology, Zurich University of Applied Sciences, Zürich, Switzerland . . . . . 1Winterthur Objectives: The objectives of this study are to estimate the prevalence of schizophrenia in Switzerland and to assess its burden on patients, caregivers and society as a whole. Methods: A hospital registry was combined with a physician survey and health insurance claims data to capture all patients living in the northern part of the canton of Zurich. Total costs included direct medical and non-medical costs and lost production. All costs were calculated for the year 2012 from a societal perspective using a prevalence-based bottom-up approach. Intangible costs were expressed as quality adjusted life years (QALY) lost and were calculated from Swiss life tables, standardized mortality ratios and utility weights from the literature. Uncertainty and its sources were addressed in univariate and probabilistic sensitivity analysis. Results: The point prevalence of schizophrenia in 2012 was estimated at 0.39% of the Swiss population. The average annual costs of schizophrenia amounted to EUR 39,408 per patient and consisted of direct medical costs of EUR 9,507 (24%), the costs of care by relatives and in residential homes of EUR 4,793 (12%) and lost production of EUR 25,108 (64%). Inpatient hospital care accounted for EUR 6,242 per year or 66% of direct medical costs. The estimated reduction in life expectancy of 10.46 years and the utility decrement of 22.05 percentage points lead to intangible costs of 19.02 QALY per incident chronic case. Conclusions: The results of this study show the high burden of schizophrenia on patients, caregivers and society as a whole. The high costs of inpatient hospital care demonstrate the benefits of an effective prevention of relapse associated with hospitalization. Programs for the reintegration of schizophrenic patients into the labor market have a high potential to reduce the costs of schizophrenia considering the high burden of lost production and the early onset of the disease. PMH21 A Model to Estimate the Health System Burden of Prescription Opioid Abuse in Europe Shei A 1, Hirst M 2, Kirson N Y 1, Enloe C J 1, Birnbaum H G 1, Dunlop W 2 1Analysis Group, Inc., Boston, MA, USA, 2Mundipharma International Limited, Cambridge, UK . . . . . . . . . Objectives: Prescription opioid (“RxO”) abuse has not been regarded as a major problem in Europe so far, but a lack of reliable data hinders the assessment of this problem. This study aimed to derive estimates of the prevalence and excess costs of RxO abuse in the five largest European countries (France, Germany, Italy, Spain, and UK; “EU5”). Methods: Data from the European Monitoring Centre for Drugs and Drug Addiction and the UN Office on Drugs and Crime, on the prevalence of problem opioid abuse and the share of opioid abuse patients who report using non-heroin opioids, were used to estimate the prevalence of RxO abuse in the EU5. The costs of RxO abuse were calculated by applying published estimates of the excess health care costs of RxO abuse to country-specific estimates on the costs of chronic pain. Sensitivity analyses varied assumptions surrounding the prevalence of opioid abuse patients in the general population and the estimates of the excess costs of RxO abuse in the EU5. Results: The prevalence of RxO abuse, in the general population, varied between the EU5 countries, ranging from 0.7 per 10,000 to 13.7 per 10,000. In the base case scenario, the total annual health system costs of RxO abuse across all EU5 countries were estimated to be €323 million; results of sensitivity analyses ranged from € 98 million to € 730 million. These cost estimates included health system costs only; indirect costs were not included. Conclusions: RxO abuse imposes a burden on EU5 health systems. Future research should examine trends in the prevalence and total economic burden of RxO abuse in Europe over time and assess the potential benefits of abuse-deterrent formulations, which published research suggests have been associated with a significant relative reduction in rates of diagnosed opioid abuse. PMH22 Productivity Loss and Resource Utilization in Individuals Providing Care for Adults with Schizophrenia in the 5eu Gupta S 1, Isherwood G 2, Jones K 3, Van Impe K 4 1Kantar Health, Princeton, NJ, USA, 2Kantar Health, Epsom, UK, 3European Federation of Associations of Families of People with Mental Illness, B-3000 Leuven, Belgium, 4Janssen-Cilag GmbH, Neuss, Germany . . . . Objectives: This study aimed to understand the impact of providing care for adults with schizophrenia on productivity, daily activities and resource utilization in the 5EU. Methods: Data from the 2010-2011 and 2013 5EU (France, Germany Italy, Spain, UK) National Health and Wellness Survey, an online questionnaire of a nationwide sample of adults (18+ years) was analyzed. Schizophrenia caregivers (n= 398) were matched to non-caregivers (n= 158,989) and other caregivers (n= 14,341) on baseline characteristics (sociodemographics, BMI, comorbid status) via propensity scores (1: 2). Outcome measures included health care utilization (type/number of resources used within the past 6 months) and Work Productivity and Activity Impairment questionnaire-based scores. Chi-square tests and ANOVAs were used to determine significant differences between schizophrenia caregivers vs. non-caregivers and other caregivers (e. g., cancer, Alzheimer’s). Results: The average age of schizophrenia caregivers was 45.3 (SD= 15.8 years), 59.6% were female, and 52.5% were currently employed. After matching, schizophrenia caregivers reported greater activity impairment (38.4% vs. 26.1%), more health care provider visits (8.0 vs. 5.7), emergency room visits (0.9 vs. 0.2) and hospitalizations (0.8 vs. 0.1) than non-caregivers, all p< 0.001. Amongst employed respondents, schizophrenia caregivers reported greater absenteeism (12.4% vs. 5.6%), presenteeism (29.9% vs. 17.5%), and overall work impairment (35.0% vs. 20.7%) than non-caregivers, all p< 0.001. Comparing schizophrenia caregivers and other caregivers, schizophrenia caregivers reported more activity impairment (38.4% vs. 32.3%) and health care provider visits (8.0 vs. 6.6), both p< 0.05. A greater proportion of schizophrenia caregivers reported at least one emergency room visit (26.1% vs. 20.2%) and hospitalization (20.4% vs. 14.3%) than other caregivers, both p< 0.05. No significant difference was found on work-related impairment, probably due to the small sample of employed respondents. Conclusions: Schizophrenia caregivers reported greater activity impairment and more resource use than non-caregivers and caregivers of adults with other conditions. Better family and social support systems may help reduce the burden for schizophrenia caregivers. PMH23 Medication Usage Pattern, Health Resource Utilization and Economic Burden for Patients with Mdd in Beijing, China Zhang H 1, Yue L 1, Chen Y 1, Ding H 2, Zhao K 3, Montgomery W 4 1Lilly Suzhou Pharmaceutical Company, Ltd., Shanghai, China, 2Dalian Medical University, Dalian, China, 3Beijing Brainpower Pharma Consulting Co. Ltd., Beijing, China, 4Eli Lilly Australia, Sydney, Australia . . . . . . Objectives: To investigate medication usage patterns, health care resource utilization and direct medical costs of patients with Major Depressive Disorder (MDD) in Beijing, China. Methods: Data were randomly extracted from Beijing Urban Employee Basic Medical Insurance Database. Patients who were aged ≥ 18 years, with at least 1 primary diagnosis of MDD and 12-month continuous enrollment after their first observed MDD diagnosis between 2012 and 2013 were identified. Those with a diagnosis of schizophrenia, bipolar disorder or cancer within the study period were excluded. Descriptive statistics were used to describe patient profiles, medication usage, health care resource utilization and costs. Results: A total of 8484 patients were included with mean (±SD) age of 57.15 (±15.34) years, 63.02% female and 94.47% having co-morbidities. 71.35% of patients were treated with antidepressant medications, including 60.53% of patients with SSRIs, followed by NaSSA (8.96%) and SNRIs (8.26%). Concomitant medications were prescribed for 76.78% of patients. Only 0.42% of patients experienced ≥ 1 MDDrelated hospitalizations during the 1-year follow up and the average annual number of hospitalization was 1.22 (±0.64) for those hospitalized patients. The length of stay was 33.38 (±30.6) days per hospitalization and 36.61 (±40.04) days per patient-year. All patients had ≥ 1 MDD-related outpatient visits. The mean annual number of outpatient visits was 3.06 (±2.99). The mean annual direct medical cost for all MDD patients was 1694.05 (±2513.71) RMB with 48.54% for antidepressant medications, and that for hospitalized patients was 21290.97 (±16121.61) RMB with 15.03% for antidepressant medications and 66.45% for non-drug medical costs. Conclusions: In Beijing China, most MDD patients also had comorbid conditions and were mainly treated in the outpatient setting. SSRIs were the most commonly used antidepressants. The economic burden of MDD was considerable. PMH24 Atomoxetine for the Treatment of Newly Diagnosed Adults with Adhd - a Cost Effectiveness Analysis in Spain Tockhorn A 1, Televantou F 1, Dilla T 2 Lilly UK, Windlesham, Surrey, UK, 2Eli Lilly Spain, Madrid, Spain . . . 1Eli Objectives: Atomoxetine is the first medication to receive marketing authorization in Spain for the treatment of newly diagnosed adults with Attention-Deficit/ Hyperactivity Disorder (ADHD). The aim of this analysis was to assess if treatment with atomoxetine in adults with ADHD was cost-effective vs. placebo from the Spanish Healthcare System perspective. Methods: A Markov state transition model was developed for a theoretical cohort of newly diagnosed adult patients with moderate-severe ADHD. Key input data (response and discontinuation) were derived from the atomoxetine trial program. Patients enter the model at the age of 18 and receive atomoxetine (initiated at 40mg for a week and then titrated to 80mg or 100mg) or placebo (in the absence of another authorized medication for the treatment of newly diagnosed adults with ADHD). Treatment success has been defined as response to treatment, showing improvements in both symptoms and functioning as measured by the CAARS and CGI-S scales, respectively. Treatment, non-specific health state utilities were populated with estimates from a vignette study in adults conducted in the UK. Drug and direct medical costs were obtained from local databases. In accordance with other published ADHD models, a 1-year time horizon was used. To check the model for robustness, probabilistic and deterministic sensitivity analyses were performed. Results: Atomoxetine was found to be cost-effective with an ICER of € 24,248/QALY despite patients in placebo arm only accumulating cost of physician visits. In addition, a QALY gain of 0.023 was projected, due to greater proportion of patients responding to treatment in the atomoxetine arm. Results from a probabilistic sensitivity analysis indicated that atomoxetine has a 57% probability of being more cost-effective than placebo at a willingness to pay threshold of € 30,000/QALY in the Spanish setting. Conclusions: Atomoxetine is a costeffective option versus no active medical treatment for newly diagnosed adults with ADHD in Spain. PMH25 Aripiprazole Once-Monthly is a Cost-Effective Therapeutic Option in the Maintenance Treatment of Schizophrenia: Results from a Markov Model Sapin C 1, Tempest M J 2, Gaughran F 3, Beillat M 1, Robinson P 4, Treur M 5 SAS, Issy les Moulineaux, France, 2Pharmerit Ltd, York, UK, 3South London and Maudsley NHS Foundation Trust, London, England, 4Otsuka Pharmaceutical Europe Ltd, Wexham, UK, 5Pharmerit International, Rotterdam, The Netherlands . . . . . . . 1Lundbeck Objectives: Schizophrenia is a heterogeneous chronic disease with enormous economic consequences for the society. This study aimed at building a conceptual framework to evaluate the cost-effectiveness of Aripiprazole Once-Monthly (AOM) versus other atypical long-acting injectable (LAI) antipsychotics: Risperidone LAI (RLAI), Paliperidone Palmitate (PP) and Olanzapine Pamoate (OP) in the mainte- A458 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 nance treatment of schizophrenia. Methods: The schizophrenia Markov model developed by the National Institute for Health and Care Excellence (NICE) was adapted to the context of LAI antipsychotics. Effectiveness was measured through Quality-Adjusted Life Years (QALYs) and number of relapses. The economic analysis was conducted over a ten-year time horizon, including cost of managing stable schizophrenia, relapse and treatment-emergent adverse events (TEAEs). Probabilities of relapse, discontinuation due to adverse events, and due to other reasons came from a mixed treatment comparison of pivotal clinical trials; disutilities associated to TEAEs and other non-drug-specific inputs came from various epidemiological sources. Results: AOM was associated with higher number of QALYs (7.26 vs 7.17, 7.18 & 7.19 for PP, RLAI and OP respectively) over a 10-year time horizon. Assuming a theoretical parity price between AOM and PP, the base case analysis showed that AOM was the dominant strategy as compared to RLAI, PP and OP. Deterministic sensitivity analyses confirmed these overall Conclusions, the main drivers of cost-effectiveness being both probability and cost of relapse. In the probabilistic sensitivity analysis, AOM demonstrated a higher probability of being cost-effectiveness than RLAI, PP and OP at a willingness to pay threshold of £20,000 (52%, 89% and 90%, respectively). Conclusions: Although model outcome may vary according to local data and settings, and assuming a theoretical parity price with PP, AOM was found to provide clinical benefits at lower total costs compared to other atypical LAI antipsychotics, showing its value in the maintenance treatment of schizophrenia. PMH26 Prescribing Anti-Depressants by Baseline Severity: Evidence Synthesis, Economic Model and Value of Information Analysis Thom H 1, Welton N 1, Lewis G 2 1University of Bristol, Bristol, UK, 2University College London, London, UK . . . Objectives: Aim to determine the most cost-effective threshold of depression severity above which to prescribe anti-depressants to patients, in England and Wales, presenting with depression and under consideration for anti-depressants. Also aim to evaluate the cost-effectiveness of a new trial of anti-depressants in a population of wider range of depression severity than in previous trials. Methods: Meta-regression of existing study results to estimate a proportional treatment effect on depression severity, which is then extrapolated to a wider range of severity than in included trials. An economic model which consists of a continuous outcome for the initial 12 weeks of treatment, followed by a Markov model with states for depression category and treatment. Treatment effects on Hamilton Depression Rating Scale (HAMD) were mapped to EQ5D. Expected value of partial perfect information (EVPPI) was used to determine an upper bound on the value of collecting ffurther evidence. Results: Patients on anti-depressants had an additional 12% (CrI 3-21%) decrease in 6-week HAMD versus placebo. Treating patients with a severity > 2 on HAMD had the highest probability (> 65%) of being cost-effective at £20,000 willingness-to-pay threshold. However this assumes that the relationship with severity can be extrapolated beyond the range of HAMD included in the systematic review. A short-term trial investigating the relation between treatment effect and severity and quality of life in depression patients had EVPPI= £67.7 million over a 10 year timehorizon. Conclusions: On the basis of available evidence, our model suggests it may be cost-effective to treat patients with lower severity of symptoms than have been included in the majority of existing RCTs. However, evidence of treatment efficacy in those with low HAMD is lacking, and there is likely to be value in conducting a trial on this lower severity population. PMH27 Economic Evaluation of Nalmefene for the Treatment of Alcohol Dependence in Greece Relakis J 1, Paparrigopoulos T 2, Kourlaba G 3, Maniadakis N 1 1National School of Public Health, Athens, Greece, 2Athens University Medical School, Athens, Greece, 3Collaborative Center for Clinical Epidemiology and Outcomes Research (CLEO), Athens, Greece . . . . Objectives: To assess, from the perspective of national health insurance, the cost-effectiveness of psychosocial support plus nalmefene versus psychosocial support alone, for the treatment of adult patients with alcohol dependence who have a high drinking risk level, without physical withdrawal symptoms, and who do not require immediate detoxification. Methods: A cost-effectiveness Markov model, originally developed for the Single Technology Appraisal by the National Institute for Health and Care Excellence of Nalmefene, was adapted to the Greek health care setting to evaluate the health effects and associated costs of compared therapeutic options. The model consisted of a short-term phase (1 year: based on clinical trials assessing nalmefene) and a long-term phase (2-5 years) evaluating patient progression and a second-line abstinence treatment option. Clinical inputs were derived from ESENSE1 (NCT00811720), ESENSE2 (NCT00812461) and SENSE (NCT00811941) clinical trials and epidemiological data from the national statistical authority and published literature. All direct costs with respect to treatment of alcohol dependence and the management of alcohol-attributable harmful events were considered reflecting the year 2014. Results: Over a 5-year horizon, the addition of nalmefene to psychosocial support led to the avoidance of 3,071 alcohol-attributable diseases/injuries and 851 deaths per 100,000 patients. Nalmefene plus psychosocial support reduced the proportion of high-risk and very high-risk drinkers (23% versus 37% with psychosocial support alone) and increased the number of controlled drinkers (61% versus 45%). In the base-case, the nalmefene plus psychosocial support arm cumulated a 5-year incremental cost of € 160 and an incremental QALY of 0.024. The incremental cost-effectiveness ratio of € 1,928 was considerably low with respect to the decision threshold of € 16,000 per QALY gained (equal to the GDP per capita of Greece). Conclusions: Treatment with nalmefene is cost-effective in the Greek health care setting leading to significant reductions of alcohol dependent patients and alcohol-attributable harmful events. PMH28 The Effect of Attention-Deficit/Hyperactivity Disorder on Functioning and Resource Utilization by Psychiatric Outpatients in Europe Televantou F 1, Sobanski E 2, Kan C C 3, Lebrec J 4, Kraemer S 5, Dieteren N A H M 6, Deberdt W 7 1Eli Lilly UK, Windlesham, Surrey, UK, 2Central Institute for Mental Health, Mannheim, Germany, 3Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands, 4Eli Lilly and Company, Bad Homburg, Germany, 5Eli Lilly and Company Ltd, Bad Homburg, Germany, 6Lilly Nederland BV, Houten, The Netherlands, 7S.A. Eli Lilly Benelux N.V., Brussels, Belgium . . . . . . . . . . . Objectives: Attention-deficit/hyperactivity disorder (ADHD) can have a significant negative impact on health outcomes in adults. This study was designed, in part, to determine the functional outcomes and health care utilization of adult psychiatric outpatients with ADHD in several European countries. Methods: This was a multinational observational study. All eligible outpatients (excluding patients with any psychotic disorder) from a variety of outpatient settings were invited to participate. ADHD diagnosis was established with the Diagnostic Interview for ADHD in Adults (DIVA) based on the Diagnostic and Statistical Manual of Mental Disorders, 5thEdition. All patients were further evaluated with the Sheehan Disability Scale (SDS) and the EuroQol-5 Dimensions (EQ-5D) questionnaire, which was also used to assess anxiety/depression. Results: Of 5662 patients approached, 2284 (40.3%) enrolled, of whom 1986 patients (87.0%) completed the study. Patients were 17 to 72 (median= 42) years of age, and the majority were women (58.8%). Based on the DIVA, 17.4% (95% CI 15.7%-19.0%) of patients were diagnosed with ADHD. Patients with ADHD had moderate to severe overall impairment (mean SDS total score 18.9 [SD= 6.6, n= 348] versus 11.6 [SD= 8.6, n= 1659] in patients without ADHD). On the EQ-5D, a majority of patients with ADHD indicated having problems performing their usual activities (66.2% versus 41.2%) and many reported being “extremely anxious or depressed” (24.6% versus 16.0%). However, compared to patients without ADHD (n= 1660), patients with ADHD (n= 349) were less often prescribed antidepressants (57% versus 71.9%). The proportions of patients who visited a primary care physician, psychiatrist, or psychotherapist during the previous 6 months were similar between the 2 groups. Conclusions: Adult psychiatric outpatients with ADHD in our sample reported more overall functional impairment and psychiatric comorbidities compared to outpatients without ADHD. The use of medical resources was similar between the 2 groups. PMH29 Paliperidone Versus Atypical Long-Acting Antipsychotics for Relapsed Chronic Schizophrenia: an Economic Analysis Einarson T R 1, Letchumanan M 2, Pudas H 3, Van Impe K 4 of Toronto, Toronto, ON, Canada, 2PIVINA Consulting Inc., Mississauga, ON, Canada, 3Janssen, Espoo, Finland, 4Janssen-Cilag GmbH, Neuss, Germany . . . . . 1University Objectives: To determine the cost-effectiveness of atypical long-acting injectable (LAI) antipsychotics in treating relapsed chronic schizophrenia from the viewpoint of the Finnish National Health Service. Methods: A 1-year decision tree was adapted for use with patients in relapse, guided by an expert panel. Drugs included available atypical long-acting antipsychotics: paliperidone (PP-LAI), risperidone (RIS-LAI), olanzapine (OLZ-LAI) and aripiprazole (ARI-LAI). Rates of adherence, success, relapse and hospitalization were taken from the literature. Prices were obtained from standard lists and expressed in 2014 euros: drugs, psychiatrists/ physicians, psychiatric nurse, inpatient and outpatient hospital care. Outcomes included expected cost/patient treated, QALYs, rates of re-hospitalization, emergency room (ER) visits and days in relapse. The primary analysis was the incremental cost per QALY. These preliminary results were tested with 1-way sensitivity analyses on important inputs. Results: Over the 1-year time horizon, PP-LAI had the lowest total cost of 34,446€ per patient, RIS-LAI cost 37,338€ , ARI-LAI cost 37,433€ and OLZ-LAI cost 41,384€ . PP-LAI had the highest number of QALYs (0.686), followed by OLZ-LAI with 0.680, RIS-LAI with 0.674 and ARI-LAI with 0.671. PP-LAI also had the lowest rates of all negative outcomes. Re-hospitalization rates were 10.1%, 12.5%, 12.4% and 12.2% for PP-LAI, RIS-LAI, ARI-LAI and OLZ-LAI, respectively, Respective ER visits were 20.6%, 23.2%, 25.4% and 21.7%. Patients receiving PP-LAI experienced 224.0 relapse days, as opposed to 234.8 with RIS-LAI, 237.5 with ARI-LAI and 234.8 with OLZ-LAI. In 1-way sensitivity analyses, costs were robust (required changes > 20%) against changes in drug price, primary rate of success, rates of relapse, dropouts and adherence. The driver of the model for all drugs was hospitalization, comprising from 70%-81% of the total cost; drug costs constituted 12%-22% and medical care 8%-15%. Conclusions: PP-LAI was shown to have the lowest cost and best clinical outcomes, and hence should be the atypical LAI of choice. PMH30 The Cost Effectiveness of Group Art Therapy for Patients with NonPsychotic Mental Health Disorders in England And Wales Stevenson M , Rawdin A , Uttley L , Sutton A University of Sheffield, Sheffield, UK . . . . Objectives: Art therapy provides an alternative to standard forms of psychological therapy. We estimated the cost-effectiveness of group art therapy for people with non-psychotic mental disorders. Methods: A de novo area under the curve model was constructed with the following assumptions that: the maximum treatment effect would be associated with the time at which treatment ended; there would be a linear increase in treatment effect, from zero at baseline to the time at which treatment ended; there would be a residual effect of treatment with a linear decline in benefit until there was zero benefit at 52 weeks; given the short assumed duration of benefit, discounting was not necessary. Two RCTs identified in an accompanying clinical review provided data from which EQ-5D values could be estimated via mapping allowing comparisons to be made of group art therapy with wait-list control and with group verbal therapy. Scenario analyses altering the cost per patient and the assumed residual benefit were conducted. Results: Art therapy compared with wait-list control had a mean cost per quality adjusted life year (QALY) below A459 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 £6000 for all scenarios and a 100% probability of being cost-effective at a willingness to pay of £20,000 per QALY. Verbal therapy appeared more cost-effective than art therapy with a cost per QALY below £1000 but there was considerable uncertainty in the decision and a sizeable probability (20%) that art therapy was dominant. In neither comparison was the art therapy intervention similar to that employed in England and Wales, furthermore in the wait list comparison patients were not explicitly diagnosed with non-psychotic mental disorders. As such, the generalisability of the results to practice in England and Wales is uncertain. Conclusions: Art therapy appears cost effective versus wait-list but of uncertain value compared with verbal therapy. Confirmatory studies are required to allow more definitive statements to be made. PMH31 Cost-Effectiveness of Lithium Versus an Atypical Anti-Psychotic (AAP) used to Augment Treatment with a Selective Serotonin Reuptake Inhibitor (SSRI) in Treatment Resistant Depression (TRD) Edwards S J , Nherera L , Trevor N , Wakefield V BMJ, London, UK . . . . . Objectives: To estimate the cost-effectiveness of augmentation of SSRI antidepressant therapy with either lithium or an AAP in TRD, defined as failure to respond to two or more antidepressants. Methods: CENTRAL, EMBASE, MEDLINE, PsycINFO and NHS Economic Evaluation Database (NHS EED) were searched from inception to August 2011. Additional data were obtained from manufacturers. Systematic reviews of the economic and quality of life (QoL) literature were executed. Studies were assessed, independently by two reviewers, for quality against predefined criteria. A de novo probabilistic economic model was developed to synthesise the available data on costs and clinical effectiveness from UK NHS perspective; time horizon 1-year (8 weeks of acute treatment and 10 months of maintenance treatment). Results: Four economic evaluations (none directly addressing the review question) and 17 QoL studies were identified and summarised in narrative reviews. Model results indicate that augmentation of an SSRI with lithium dominates augmentation with AAP (i.e. Results in cost savings of £905 per person per year and generates more health benefits, estimated to be 0.03 quality-adjusted life-years). However, sensitivity analyses showed that the model was highly sensitive to changes in acute treatment efficacy (response and remission) or discontinuation. The model was not sensitive to changes in other parameters. Conclusions: Costeffectiveness analyses suggest that augmentation with lithium is less expensive and more effective than augmentation with AAP. However, the uncertainty in the clinical estimates of discontinuation and treatment response is reflected in the model results. An RCT comparing the two augmentation strategies, reporting relevant outcomes, including QoL, is needed. PMH32 Computerised Cognitive Behaviour Therapy for Depression Management: A Cost-Effectiveness Analysis Duarte A 1, Walker S 2, Littlewood L 1, Gilbody S 1, Palmer S 1 1University of York, York, UK, 2University of York, Heslington, York, UK . . . . . Objectives: Computerised cognitive behaviour therapy (cCBT) forms a core component of stepped psychological care within primary care in the UK and other countries. However, the existing clinical effectiveness evidence for cCBT comes from developer-led trials and independent research is needed which evaluates the clinical and cost-effectiveness of cCBT. Methods: A cost-effectiveness analysis was undertaken comparing two cCBT software packages, free-to-use MoodGYM and a commercial pay-to-use Beating the Blues (BtB), in addition to usual general practitioner care (UGPC), with UGPC alone, for the treatment of depressed adults. The analysis was based on data collected on the Randomised Evaluation of the Effectiveness and Acceptability of Computerised Therapy (REEACT). REEACT was a large (n= 691), pragmatic multicentre study, independently conducted in a primary care setting. Outcomes were assessed using EQ-5D and used to estimate quality-adjusted life-years (QALYs). Resource use and costs were estimated from a NHS and Personal Social Services perspective. Scenario analyses were performed to determine the impact on cost-effectiveness of alternative assumptions, and uncertainty was characterised using cost-effectiveness acceptability curves. Results: BtB was both more expensive and generated lower QALYs than UGPC alone (dominated) and MoodGYM yielded lower QALYs but at lower cost, resulting in an ICER of £6,933 per additional QALY for UGPC alone versus MoodGYM. UGPC alone was the most cost-effective intervention in the majority of scenario analyses, and the intervention most likely to be cost-effective at a £20,000 per QALY threshold (probabilities ranging across scenarios from 0.545 to 0.619). Conclusions: Our findings indicate that commercially-produced products were no more effective than free-to-use cCBT programmes. Importantly, neither BtB or MoodGYM appeared cost-effective compared to UGPC alone. Practice recommendations such as those offered by NICE and other countries supporting the use of cCBT within stepped models of care for depression will need to be reconsidered in light of these results. PMH33 Cost-Utility Analysis of Long-Acting Paliperidone in Comparison with Oral Risperidone, Oral Paliperidone and Long-Acting Risperidone in the Maintenance Treatment of Schizophrenia in the Czech Republic Kolek M 1, Duba J 1, Vesela S 2, Pasztor B 1, Doleckova J 1 1OAKS Consulting s.r.o., Prague 9, Czech Republic, 2Janssen-Cilag s.r.o., Prague 5, Czech Republic . . . . . Objectives: The number of patients with schizofrenia in the Czech Republic amounts annually to approximately 126,000. Schizophrenia causes significant increases in mortality, shortening life expectancy by 25 years compared to the general population which implies high disease burden. The aim was to estimate the cost-effectiveness of long-acting paliperidone in the treatment with schizofrenia compared to oral risperidone, oral paliperidone and long-acting ris- peridone. Methods: Cost-utility analysis was performed using a Markov model. The primary outcome was ICER/QALY. Oral risperidone, oral paliperidone and long-acting risperidone were selected as comparators. The basic components of the model include probabilities of relapse, individual hazard ratios for non-compliance by medication type and switch of treatment probabilities. Specific utilities for each health state were considered. Among relevant costs, reflecting payer’s perspective, drug acquisition costs, monitoring costs, costs of relapses, follow-up care and adverse events were considered. Results: Long-acting paliperidone reached ICER of EUR 16,233/QALY compared to oral risperidone, EUR 15,058/QALY to oral paliperidone and EUR 335/QALY to long-acting risperidone. The robustness of the model was supported by one-way deterministic analysis and probabilistic sensitivity analysis, which gave stable results. Long-acting paliperidone was cost effective in 97% of the simulations compared to oral risperidone. Long-acting paliperidone treatment gained incremental 0.903 QALYs on average compared to oral risperidone. Conclusions: The treatment of schizophrenia using long-acting paliperidone is associated with increased QALYs. It reduces incidence of adverse events, results in better prevention of relapses and can be considered a cost-effective treatment in the Czech Republic. PMH34 Cost-Utility of Vortioxetine in the Treatment of Major Depressive Disorder: Comparison with Agomelatine, Bupropion, Sertraline and Venlafaxine in the Finnish Setting Soini E J 1, Hallinen T 1, Brignone M 2, Despiégel N 3, Aalto-Setälä M 4, Danchenko N 2, Kolasa K 4 1ESiOR Oy, Kuopio, Finland, 2Lundbeck SAS, Issy-les-Moulineaux, France, 3Optum, Nanterre, France, 4Oy H. Lundbeck Ab, Turku, Finland . . . . . . . . Objectives: Switching to vortioxetine (a new antidepressant) after inadequate response to selective serotonin re-uptake inhibitor/serotonin–norepinephrine reuptake inhibitor (SSRI/SNRI) resulted in a significant and clinically relevant improvement versus agomelatine (REVIVE head-to-head clinical study) and also better efficacy over sertraline, venlafaxine and bupropion (indirect comparison). The aim of this study is to assess the Finnish cost-utility of vortioxetine versus these antidepressants in patients who switch due to inadequate response to previous treatments. Methods: A one year cost-utility analysis was performed using a decision tree model for the second line and a Markov model for subsequent lines of MDD treatment. Undiscounted payer and societal perspectives were considered. Three health-states (depression, remission, recovery) and two treatment phases (2-month acute, 6-month maintenance) were defined. The relative efficacy of antidepressants was derived from the REVIVE trial and indirect comparisons at 2 months. Efficacy in subsequent treatment steps was derived from the STAR*D study. Adverse events and their consequences were included and derived from REVIVE, indirect comparisons, literature and expert opinion. Utilities were derived from REVIVE and the literature using Finnish preference weights. Finnish costs in 2013/2014 value were considered. Sensitivity analyses were conducted to assess the robustness of the findings. Results: Vortioxetine was dominant from the payer’s and societal perspective versus all comparators. It was projected to result to QALY gain of 0.013, 0.017, 0.025 and 0.028, and € 223 (€1074), €128 (€957), €110 (€ 720) and € 238 (€ 1390) direct (total) annual cost saving compared to agomelatine, bupropion, venlafaxine and sertraline respectively. These results were confirmed to be robust in several sensitivity analyses. Conclusions: Vortioxetine dominated agomelatine, bupropion, venlafaxine and sertraline in Finland and appears to be a relevant treatment option for MDD patients who need a therapy switch. PMH35 Cost-Effectiveness of Paliperidone Palmitate Versus other Antipsychotics for the Treatment of Schizophrenia in France Doutriaux A 1, Cognet M 2, Druais S 2, Lançon C 3, Samalin L 4, Levy P 5, Godet A 6, Guillon P 7 1Amaris Consulting, London, UK, 2Amaris, London, UK, 3Hopital Sainte-Marguerite CHU, Marseille, France, 4CHU Clermont-Ferrand, Clermont-Ferrand, France, 5University of Dauphine Paris, Paris, France, 6Janssen Cilag, Issy-les-Moulineaux, France, 7Janssen, Issy-les-Moulineaux, France . . . . . . . . Objectives: To estimate the cost-effectiveness of paliperidone palmitate (PLAI), a once-monthly long-acting injectable (LAI) atypical antipsychotic, compared to the most common antipsychotic strategies in France. Methods: A Markov model was developed to simulate the progression of a cohort of schizophrenic patients through four health states (stable treated, stable non-treated, relapse and death) and up to three lines of treatment. PLAI was compared to risperidone LAI (RLAI), aripiprazole LAI (ALAI), olanzapine LAI (OLAI), haloperidol decanoate (HLAI) and oral olanzapine (OO). Costs, quality-adjusted-life-years (QALYS) and number of relapses were assessed over five years based on three-month cycles, and discounted at 4%, from a health insurance perspective. Patients were supposed to be stabilised after a clinical decompensation and entered the model into an initiation phase, followed by a relapse prevention phase in case of success. In the prevention phase, relapse rates were derived from hospitalisation risks based on French real-life data in order to capture the adherence effects. Safety and utility data were derived from international publications. Costs came from French health insurance databases and publications. Robustness of results was assessed through deterministic and probabilistic sensitivity analyses. Results: PLAI was the less costly LAI and associated with an incremental cost-effectiveness ratio (ICER) of € 1,988/QALY gained and € 2,267/relapse avoided versus OO. RLAI and PLAI were associated with the highest number of QALYs (ICER of € 2,421,386/ QALY gained between PLAI and RLAI). PLAI dominated all other LAIs in terms of relapse but OLAI. Nevertheless, PLAI was highly cost-effective versus OLAI (ICER of € 1,575,217/relapse avoided). Conclusions: This analysis is the first to assess the cost-effectiveness of antipsychotics based on French observational data. PLAI was found to be the least expensive LAI antipsychotic from French payer perspective. Oral therapies were less expensive but associated with lower levels of QALYs and more relapses compared to all atypical LAIs. A460 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PMH36 Economic Evaluation of Agomelatine for Major Depressive Disorders Relative to other Antidepressants in the Italian Setting in terms of longer length of stay, more readmissions and higher consumption of drugs. The average quality of the studies was moderate. Further research is needed to establish the degree of burden of agitation and containment borne by hospitals and the health care system. Lanati E P , Lidonnici D MA Provider, Milano, Italy . . . Objectives: The purpose of the present study is to conduct an economic evaluation of Agomelatine vs the current alternatives in daily clinical practice for treating patients with major depression disorders (MDD) in Italy (Venlafaxine, Fluoxetine, Sertraline, Escitalopram and Duloxetine). Methods: Using a Markov model-based cost-effectiveness analysis, Agomelatine was compared with other therapies used for the treatment of MDD commonly prescribed (Venlafaxine, Fluoxetine, Sertraline, Escitalopram and Duloxetine), chosen on the basis of market shares, and compared with placebo. The population considered in the model consists of patients suffering from MDD and with an average age of 45 years. The perspective of the third party payer (Italian National Healthcare Service) and the societal perspective were considered. Results: The study shows that Agomelatine administration is linked with higher direct and indirect costs only when compared with Duloxetine (respectively € 4,365 vs. €4,253 and €5,553 vs. €5,484). Nevertheless, Agomelatine has the higher efficacy in terms of QALY gained (1.477) in comparison to all comparators considered in the analysis. According to the societal perspective, Agomelatine is dominant against Venlafaxine, Escitalopram, Fluoxetine and Sertraline, since it is less expensive and more effective and cost-effective compared to Duloxetine since the incremental cost per QALY gained is €12,461. According to the perspective of the Italian NHS, Agomelatine is dominant versus Venlafaxine, Fluoxetine and Sertraline and is cost-effective in comparison to Duloxetine (ICER 6,101 € /QALY) and Escitalopram (3,336 € /QALY). Conclusions: The present economic evaluation indicates that Agomelatine provides greater benefit and is less costly compared to generic Venlafaxine, generic Escitalopram, generic Fluoxetine and generic Sertraline and that Agomelatine is cost-effective compared to Duloxetine. In conclusion, according to its favorable tolerability profile and its proven efficacy, Agomelatine represents a powerful tool for many patients suffering from MDD, which may lead to both clinical and economic advantages. PMH37 Retrospective Database Study on Health Care Resource Utilization of Patients Initiating Long-Acting Olanzapine in Sweden Tockhorn A 1, Johansson S 2, Borgeke H 3 1Eli Lilly UK, Windlesham, Surrey, UK, 2IMS Health Sweden, 113 46 Stockholm, Sweden, 3Eli Lilly and Company, Stockholm, Sweden . . . Objectives: The Swedish national payer, TLV, was interested in understanding how OLAI is used in routine clinical practice regarding dosing and its impact on psychiatric-related hospitalization. Methods: Three Swedish nationwide health registers: the patient register, the drug register and the mortality register were linked. Patients with ≥1 prescription of OLAI and one diagnosis of schizophrenia or schizoaffective disorder prior to initiation of OLAI were included in this retrospective patient mirrorimage study. The minimum follow-up was six months. The study period was from 03/2010 until 12/2011. The average number and duration of hospitalizations were compared before and after initiation with OLAI using a Student’s t-test. Results: 70 patients met the inclusion criteria. The proportion of patients being hospitalized ≥1 was 77% in the pre-index period and 67% in the post-index period. The number of outpatient visits increased from 45% prior to OLAI to 77% post initiation. The results showed a significant reduction in the mean length of stay per hospitalization (19.6 days vs 3.9 days [p<0.001] ), and in the mean total number of days spent in hospital per patient (52.3 days vs. 16.2 days [p< 0.001] ). No significant difference was seen in the number of hospital visits between the pre- and post-initiation periods, although there was a numerical decrease observed within the post-index period. The average dose of OLAI was 18.1 mg per day (95% CI: 16.6 mg; 19.7 mg) and the prescription refill period was 19.6 days (95% CI: 17.7 days; 21.5 days). Conclusions: This study provides evidence that initiation with OLAI significantly reduces the length of stay per hospitalization and the total days spent in hospital. PMH38 Use of Services and Cost of Agitation and Containment in Psychiatric Hospitals: a Systematic Review Rubio-Valera M 1, Luciano-Devís J V 1, Ortiz J M 1, Salvador-Carulla L 2, Haro J M 1, Gracia A 3, Serrano-Blanco A 1 1Parc Sanitari Sant Joan de Déu, Sant Boi de Llobregat, Spain, 2Centre for Disability Research and Policy, Lidcombe, Australia, 3Ferrer Internacional, Barcelona, Spain . . . . . . . . . . Objectives: The aim of this study was to evaluate the use of services and costs related to agitation and containment of adult patients admitted to a psychiatric hospital. Methods: Systematic review through searches of Pubmed, CINHAL and Web of Knowledge (using a wide variety of terms related to agitation; inpatient care and use of services/costs); bibliographic references in retrieved studies and expert consultation. Studies published since 1998 were selected in duplicate by reviewing abstracts and full-text papers. Results: After removing duplicates, 372 papers were reviewed and 11 included in the review. Four studies were of high quality, 4 of moderate-high to moderate-low quality and three of low quality. Eight of the studies evaluated the impact of agitation on the length of stay and 6 showed that it was associated with longer stays. Four studies evaluated the impact of agitation on readmission and showed a statistically significant increase in the probability of readmission of agitated patients in comparison with non-agitated patients. Two studies evaluated medication, one showed that the mean medication dose was higher in agitated patients and the other found higher costs of treatment compared with non-agitated patients in the unadjusted analysis. Another estimated the costs of conflict and containment related to acute inpatient psychiatric care in UK. The total annual cost in England for all conflict was £72.55 million (£145,177 annual conflict cost per ward) and £106 million for containment (£212,316 annual containment cost per ward). Conclusions: Studies on use of services and costs of agitation are scarce. Overall, agitation has an effect on health care use and costs Mental Health – Patient-Reported Outcomes & Patient Preference Studies PMH39 Which Adverse Effects Influence the Dropout Rate in Selective Serotonin Reuptake Inhibitor (Ssri) Treatment? Kostev K 1, Ehlken B 2, Rex J 1, Engelhard J 1, Altmann V 1, Heilmaier C 3 1IMS Health, Frankfurt am Main, Germany, 2IMS Health Germany, Munich, Germany, 3FOM University of Applied Sciences, Essen, Germany . . . . . . Objectives: Nowadays selective serotonin reuptake inhibitors (SSRIs) are the most frequently prescribed antidepressants due to their better clinical efficacy, effectiveness, tolerability, and safety, when compared to tricyclic antidepressants or monoamino oxidase inhibitors. However, despite this, especially at the beginning of treatment SSRIs are associated with side effects, which may lead to premature discontinuation of therapy in some cases. Assessment of these factors was the aim of the present study. Methods: This retrospective database analysis used data from 50,824 patients first time treated with SSRIs for major depressive disorder selected from a Electronic Medical Records (EMR) database (IMS Disease Analyzer) in Germany, providing information on SSRI side effects and their influence on premature treatment discontinuation calculated by regression analysis. In addition to that, presence of certain co-morbidities was registered. Results: Mean age was 54.5 ± 19 years with two-thirds of study population being female. Most frequently mentioned adverse effects were “discomfort” of the digestive system (10%), sleep disorders (8.6%), and heart rhythm disorders (4%); however, these were of tolerable severity as they did not significantly influence dropout rate. Contrary to that, especially somnolence and younger age (≤ 50 years) increased the chance of premature treatment discontinuation, while patients suffering from cardiovascular risk factors or osteoporosis tended to adhere to therapy. Conclusions: Overall, the findings indicate a good tolerability of SSRIs at the beginning of treatment, whereas occurrence of somnolence leads to incompliance. PMH40 General Beliefs about Medicines Among Depressed Patients in Saudi Arabia Aljumah K 1, Hassali A A 2, Al Mutari A 1, Al Zaide N 1 Sains Malaysia, Penang, Malaysia . . . . . 1MOH, Riyadh, Saudi Arabia, 2Universiti Objectives: The aim of this study to explore patients’ general and specific beliefs about medicines among depressed patients and effect on adherence Methods: a cross-sectional design used to measure patients’ general and specific beliefs among depressed patients, using BMQ general and specific scale. Patients were recruit from outpatient clinic at AL-Amal hospital in Riyadh (psychiatric hospital) between 2013 and January 2014. Results: A total of 403 patients meet the inclusion criteria and were participated in this study. Two hundred three representing 50.37 % of the total study sample, were female, while the remaining 200 (49.6 %) were male, with average 39 years. Half of the patients (52.9%) report low adherence to antidepressant medication. Both low and high adherence group scored high in the necessity beliefs (18.02 (SD 3.91) -18.32 (SD 3.9) respectively with no statistically different. contrariwise patients with high adherence had significantly lower level of concerns belief about antidepressants medication and less harmful belief also the same finding with general overuse belief. Conclusions: General patients beliefs either general overuse or general harm about medication influence patients taking medication behavior and have negative correlation with adherence to medication on another hand only specific concerns belief to antidepressant have a positive correlation with adherence to antidepressant this finding will help psychiatric to improve adherence and clinical outcome by addressing medications taking behavior using a systematic approach based in this finding. PMH41 Adherence to Psychotropic Medications by Outpatients in Psychiatric Hospital, Uselu Benin City, Nigeria Arute J E 1, Eniojukan J F 2, Eboigbe N P3 1Delta State University, Abraka, Nigeria, 2Niger Delta University, Wilberforce Island, Nigeria, 3Delta State University, Abraka, Nigeria, Abraka, Nigeria . . . . . Objectives: Patients adherence studies are essential for evaluating the quality of care delivery of a health facility and patients’ role in improving their conditions. The objective of this study is to determine the level of adherence of outpatients to psychotropic drugs and evaluate the impacting factors. Methods: The study was a cross-sectional study done at the psychiatric hospital, Uselu, Benin city, Nigeria from April to September, 2013. Convenient sampling method was used in population size determination for data collection. The participants were adult patients (18 years and above) attending the outpatients psychiatric clinic of the hospital with diagnosis of various psychiatric illnesses. A total of 250 patients participated in the study and a well-structured self-report 10-item questionnaire using the medication adherence rating scale (MARS) was used. Additional information was patients’ socio-demographic profile and clinical variables that affect patients’ adherence to medications. Adherence to medication data were analyzed with respect to gender and level of adherence and factors that impact patients’ adherence. Results: The level of patients’ adherence to psychotropic was 63.6% and factors found to significantly affect adherence include amount spent per clinic visit, perception of social support, intake of alcohol, medication side effects, and existing denial of illness and use of traditional medicine. Conclusions: The level of adherence to psychotropic medications was fairly high and factors that were significantly related to adherent status were amount spent per clinic visit, perception of social support, intake of alcohol, medication side effects, existing denial of illness and use of traditional A461 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 medicine. The study provides a baseline data for further studies on patients’ adherence to psychotropic medications. PMH42 Treatment Continuation and Treatment Characteristics of 3 Long Acting Antipsychotic Medications (Paliperidone Palmitate, Risperidone Microspheres and Haloperidol Decanoate) in Belgium Sermon J 1, Geerts P 1, Decuypere F 2, Weglewski T 2, Rijntjes R 3, De Hert M 4 NV, Beerse, Belgium, 2IMS Health, Vilvoorde, Belgium, 3Janssen-Cilag BV, Tilburg, The Netherlands, 4Katholieke Universiteit Leuven, Kortenberg, Belgium . . . . . . 1Janssen-Cilag Objectives: Treatment continuation of 3 long acting, injectable, antipsychotic drugs: paliperidone palmitate, risperidone microspheres and haloperidol decanoate, was evaluated in the Belgian outpatient setting using panel data from public pharmacies. Drug dosage, age distribution and frequency of co-prescribed antipsychotic medications were investigated. Methods: IMS LifelinkTM Treatment Dynamics database was used, applying appropriate selection criteria. Three patient cohorts that started paliperidone palmitate, risperidone microspheres or haloperidol decanoate treatment respectively, between 1 December 2011 and 31 August 2012, were analyzed. All cohorts included at least 13 months of follow up. Treatment continuation was investigated. Results: After 90 and 180 days, more patients continued treatment with paliperidone palmitate (60,71% and 42,41% respectively) than with risperidone microspheres (39, 07% and 26, 49%) or haloperidol decanoate (34,23% and 17,57%). Within 3 months after discontinuation, more patients restarted their treatment when using paliperidone palmitate (41%) compared to risperidone microspheres (27%) or haloperidol decanoate (17%). For all therapies, dosing was comparable between treatment initiation and discontinuation. Patients treated with paliperidone palmitate were generally younger (patients ≤ 32 years: paliperidone palmitate, 26%; risperidone microspheres, 17%; or haloperidol decanoate, 5%). Over 1 year, on average 62% of patients used paliperidone palmitate or risperidone microspheres in monotherapy. In contrast, haloperidol decanoate in monotherapy declined over time (from 49% to 28% in 1 year). Medication against extrapyramidal symptoms was on average more used with haloperidol decanoate (37%) than with paliperidone palmitate (16%) or risperidone microspheres (18%). Conclusions: Results of the database research indicate that more patients treated with paliperidone palmitate continued their therapy, restarted therapy and were of younger age than patients receiving risperidone microspheres or haloperidol decanoate. Monotherapy was more frequently observed with paliperidone palmitate and risperidone microspheres while co-medication against extrapyramidal symptoms was less frequently used compared to with haloperidol decanoate. PMH43 Describing the Health Status of Schizophrenia Caregivers in the 5E.U Gupta S 1, Isherwood G 2, Jones K 3, Van Impe K 4 1Kantar Health, Princeton, NJ, USA, 2Kantar Health, Epsom, UK, 3European Federation of Associations of Families of People with Mental Illness, B-3000 Leuven, Belgium, 4Janssen-Cilag GmbH, Neuss, Germany . . . . Objectives: Research indicates schizophrenia is a cause of burden for patients and caregivers. This study examined health-related quality of life (HRQoL) and comorbidities experienced by schizophrenia caregivers compared to non-caregivers and caregivers of other conditions. Methods: Data were obtained from the 2010-2011 and 2013 5E. U. (UK, Germany, France, Italy, Spain) National Health and Wellness Survey, an online questionnaire that’s representative of the total adult (18+ years) population. Respondents provided information on HRQoL (SF-36v2: mental and physical component summary (MCS, PCS) and SF-6D (health utility) scores), and comorbidities (sleep-difficulties, insomnia, pain, headaches, heartburn, anxiety, depression) experienced in the past 12 months. Schizophrenia caregivers (n= 398) were matched to non-caregivers (n= 158,989) and other caregivers (n= 14,341) on baseline characteristics (age, gender, income, etc.) via propensity scores (1:2). Chisquare tests and ANOVAs were used to determine significant differences across the groups. Results: The average age of schizophrenia caregivers was 45.3 (SD= 15.8 years), 59.6% were female, and 14.8% reported an income of € 50,000/£40,000+. After matching, schizophrenia caregivers reported significantly lower MCS (40.3 vs. 45.9), PCS (46.8 vs. 49.0) and health utilities (0.64 vs. 0.71), compared to non-caregivers (all p< 0.001). Schizophrenia caregivers reported experiencing significantly more sleep difficulties (42.7% vs. 28.5%), insomnia (32.4% vs. 18.5%), pain (39.7% vs. 30.4%), headaches (48.0% vs. 42.0%), heartburn (31.7% vs. 22.9%), anxiety (37.9% vs. 23.6%), and depression (29.4% vs. 19.4%) than non-caregivers. Comparing schizophrenia caregivers and other caregivers, schizophrenia caregivers reported lower MCS (40.3 vs. 42.7, p< 0.001), and health utilities (0.64 vs. 0.67, p< 0.001). Schizophrenia caregivers reported significantly more sleep difficulties (42.7% vs. 36.8%), insomnia (32.4% vs. 26.0%), pain (39.7% vs. 31.5%), and anxiety (37.9% vs. 29.8%) than other caregivers. Conclusions: Schizophrenia caregivers reported worse HRQoL and more comorbidities than non-caregivers and caregivers of other conditions. Providing caregivers with support services to better manage patients effectively may improve caregiver’s health status. PMH44 Does Giving Carers a Break Improve their Wellbeing? Results from an Evaluation using the EQ-5D-5L Gladwell D 1, Moule P 2, Moody E 3, Pollard K 2, Watson C 4, Hatswell A J 1 1BresMed, Sheffield, UK, 2University of the West of England, Bristol, UK, 3NHS Bristol Clinical Commissioning Group, Bristol, UK, 4Bristol City Council, Bristol, UK . . . . . . caring. It provided direct payments and aid to allow carers to take a break from caring (with or without the person cared for). Here we investigate the effects of the scheme on carers’ HRQL. Methods: In order to conduct the evaluation, a quality of life survey was conducted. Carers were asked to complete questionnaires at Week 0 and Week 12. One of the questionnaires selected for inclusion in both surveys was the EQ-5D (5 level). Results: Of the 155 patients completing the Week 0 survey, 97 completed at least some of the Week 12 survey. There was some evidence to suggest a small improvement in HRQL via the EQ-5D-5L (0.62 to 0.66, n=86). Both pre- and post-intervention the estimated mean utility of the cohort was significantly lower than 0.81; the value that would be predicted in the general population given the age and gender characteristics of the cohort. Conclusions: Informal carers have significantly lower HRQL than age-matched controls. While it is not possible to draw firm conclusions around the benefit derived from the intervention, due to the lack of control arm, the evaluation of a carer breaks service indicates a potentially modest benefit. PMH45 The Impact on Work and Social Activities Among Carers of Children with Adhd in Sweden Relative to Other Nordic Countries Romero B 1, Gajria K 2, Dietrich C N 1, Flood E 1, Gustafsson P A 3, Ahnemark E 4, Sikirica V 2 Commercialisation and Outcomes, Bethesda, MD, USA, 2Shire, Wayne, PA, USA, 3Linköping University, Linköping, Sweden, 4Shire, Danderyd, Sweden . . . . . . . . . 1ICON Objectives: To understand social and work impacts of caring for children/ adolescents with attention-deficit/hyperactivity disorder (ADHD) in Sweden relative to a combined cohort of other Nordic countries (Denmark, Finland and Norway). Methods: Carers in Sweden and other Nordic countries completed the Caregiver Perspective of Pediatric ADHD (CAPPA) online survey capturing carer impacts, including work, social activities and family relationships, due to their child’s ADHD. Impacts were explored when the child was “on” and “off” medication (e. g. days medication not taken). Comparisons of “on” and “off” medication were examined using the Wilcoxon Signed-Rank test. No statistical comparisons of impacts were made between countries. Results: 219 Swedish and 249 other Nordic carers of ADHD children aged 6–17 years completed the survey. 37% of Swedish carers reported employment changes (e. g. resigned, changed shift, reduced hours) due to their child’s ADHD; 52% of these changes occurred when the child was “on” medication. In the past 4 weeks, 60% of Swedish carers reported missing work and 45% reported being late for work. After excluding outliers (n= 15), mean number of hours missed was 4.32 (n= 91, SD 2.53) and mean number of times late was 2.91 (n= 69, SD 1.35). Swedish carers reported fewer “moderate” to “tremendous” impacts on social life when their child was “on” versus “off” medication (partner relationship strain: 37% vs 67%; relationship strain with other children: 29% vs 57%; social activity interference: 40% vs 59%). Relative to other Nordic countries, more Swedish carers reported being late for work (36% vs 45%) and more “moderate” to “tremendous” interference with social activities while the child was medicated (29% vs 40%). All other impacts described were similar between the two cohorts. Conclusions: While medication helped, it did not completely alleviate child ADHD-related impacts on work and social activities among carers from Sweden and other Nordic countries. PMH46 Conceptual Comprehensiveness of Anxiety Instruments in Chronic Obstructive Pulmonary Disease: Exploring the Potential for Confounding Somatic Items Willgoss T G 1, Humphrey L 1, Fatoye F 2, Yohannes A M 2 International, Manchester, UK, 2Manchester Metropolitan University, Manchester, UK . . . . . . 1Abacus Objectives: Clinically relevant anxiety is a highly prevalent co-morbidity in chronic obstructive pulmonary disease (COPD), affecting up to 74% of patients. However, despite its prevalence, co-morbid anxiety remains under-recognised and undermanaged. Furthermore, its identification and measurement can be confounded by the overlap of somatic symptoms between anxiety and COPD. This study sought to evaluate the ability of existing patient-reported outcome (PRO) measures to assess anxiety in COPD through conceptual mapping, with particular attention on the coverage of non-somatic anxiety symptoms. Methods: To determine conceptual comprehensiveness, the content of 12 extant anxiety PROs was mapped to a conceptual model of anxiety in COPD, developed through a qualitative literature review and in-depth qualitative interviews (n= 15) of COPD patients with anxious symptomology. Results: The conceptual model contained 29 concepts within five domains (somatic [15 concepts]; psychic tension [5 concepts]; apprehension [5 concepts]; panic [3 concepts]; behavioural [1 concept]). The most comprehensive conceptual coverage was found in the Mind Over Mood Anxiety Inventory which assesses 18/29 (62%) concepts across all five domains. Concept mapping revealed the majority of PROs are biased toward assessing somatic symptoms of anxiety, with no measure providing comprehensive assessment of non-somatic concepts. Indeed, the two most widely used anxiety PROs in COPD research and clinical practice (Beck Anxiety Inventory [BAI] and Hospital Anxiety and Depression Scale) provide sub-optimal coverage of anxiety concepts. In particular, the BAI is heavily weighted toward assessing somatic concepts, with little focus on psychic, apprehensive, panic and behavioural concepts of anxiety. Conclusions: In light of the sub-optimal content validity of extant instruments, including those commonly used in research and practice, there is a need for a comprehensive COPD-specific anxiety PRO. Such an instrument can be utilised in clinical trials for evaluating new products and enhance the accuracy of anxiety screening and measurement in clinical practice. . Objectives: Informal carers are widely acknowledged to be an important part of health care provision, as well as a source of substantial cost savings to the health care system. It has been observed, however, that carers have worse health-related quality of life (HRQL) than the general population. As a result of this, the wellbeing of carers has been set as a government priority. In Bristol, UK, the National Health Service and council jointly funded a 12-week scheme to facilitate breaks from PMH47 The Anxiety Inventory for Respiratory Disease (Air): an Exploration of the Air’s Psychometric Properties Through Rasch Analysis Willgoss T G 1, Humphrey L 1, Fatoye F 2, Yohannes A M 2 International, Manchester, UK, 2Manchester Metropolitan University, Manchester, UK . . . . . . 1Abacus Objectives: The Anxiety Inventory for Respiratory disease (AIR) is a novel, nonsomatic patient-reported outcome (PRO) measure of anxiety among patients with chronic obstructive pulmonary disease (COPD). Traditional psychometric meth- A462 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 ods have demonstrated promising properties of the tool. However, traditional psychometrics may have clinically relevant limitations, including scale and sampling dependency. The objective of this study was to evaluate the AIR’s strengths and weaknesses using more sophisticated Rasch Measurement Theory (RMT) Methods. Methods: Previously collected data from an observational study of 56 patients with COPD (mean age 70 years, 52% female) were analysed using RMT (conducted using RUMM2030 software) to evaluate scale-to-sample targeting, measurement adequacy and person reliability. Results: Person severity was well spread (-5.47 to 2.01 logits) but there was evidence of sample-to-scale mis-targeting at the lower end of the anxiety severity spectrum. Monotonic ordering of response categories for all 10 items suggest that the response options work as expected. No evidence of misfit on any of the items was identified (fit residual range -0.85 to 2.28) and this was supported by item characteristic curves, which demonstrated good fit to the Rasch model. Correlations among item residuals identified 11 item pairs which exceeded recommended values (-0.30 to 0.30) indicating local dependency (range -0.48 to 0.31). The AIR had a good person separation index (0.90) indicating that the sample were well separated by the instrument. Conclusions: RMT analysis demonstrated that the AIR has promising measurement properties. Mis-targeting at the lower end of the severity spectrum is of minimal concern given the mild severity of anxiety symptoms in this sample. Any modifications to address this issue should be carefully considered in light of the excellent screening properties previously identified. Evidence of local dependency between some items can be explored by reviewing qualitative data used in the development of the AIR. PMH48 Quality of Life and the Predictors of Thai Depressive Disorders Patients Kongsuk T , Leejongpermpoon J , Primtra S Ministry of public health, Ubonratchathani, Thailand . . . Objectives: To study the impact of patients with depressive disorder on quality of life and factors predicting quality of life in depressive disorders patients. Methods: This is a cross-sectional survey research. The sample consisted of depressive disorder patients, having depressive disorder diagnosed by psychiatrists using standard criteria of DSM-IV during January - July 2012 and aged 18 years and above, approximately 400 participants were selected using quota sampling method. The instruments used in this study were questionnaires on SF-36 Health Survey. Comparative analysis of average multiple regression analysis imported variables stepwise, partial correlation and colinearity statistics. Results: The number of patients with major depressive disorder found that respondents were 397 people. Quality of life of depressive disorder patients in 8 scales found that physical function (mean 67.9) role limited by physical problems (mean 40.4) role limited by emotional problems (mean 38.0) bodily pain (mean 66.6) vitality (mean 48.5), mental health (mean 53.7) social-functioning (mean 65.6) general health (mean 47.8) and female who had experienced to be depressive disorder shown score in quality of life lower than male in Physical Function (t=3.608, p<0.000). A multiple regression analysis, the independent variables by stepwise analysis found the equation that predicts the optimal quality of life consists of 6 independent variable were the severity of symptoms, age, marital status, unemployed widower, primary school and female which is related to the quality of life of patients with depression is high (R = 0.706) could explain the change in quality of life was 49.8 percent. Conclusions: Quality of life in depressive disorder Thai patients is lower than that of general population. Increased unemployment and marital divorce will negatively affect the quality of life of depressive disorder patients. Therefore care depressive disorder patients in addition to medical treatment should pay attention to the social economy of the patient. PMH49 Cultural Adaptation of the Alcohol Quality of Life Scale for Use in Japan, China, and Korea Whalley D 1, Crawford S R 1, Laramée P 2, Higuchi S 3, Hao W 4, Kim S G 5, Luquiens A 6, Aubin H J 6 1RTI Health Solutions, Manchester, UK, 2Lundbeck S.A.S., Issy-les-Moulineaux cedex, France, 3National Hospital Organization Kurihama Medical and Addiction Center, Yokosuka, Japan, 4Second Xiangya Hospital of Central South University, Changsha, China, 5Pusan National University, Yangsan, South Korea, 6Hôpital Paul Brousse, INSERM 669, Université Paris-Sud, Villejuif, France . . . . . . . . . . . Objectives: The Alcohol Quality of Life Scale (AQoLS) is a new measure to assess health-related quality of life (HRQL) in alcohol use disorder (AUD). The measure, developed in the UK and France, has 34 items over 7 hypothesised domains: relationships, activities, looking after self, emotional impact, control, living conditions, and sleep. The aim of this study was to adapt the measure for use in Japan, China and South Korea. Methods: The cultural adaptation process included three stages: initial translation of the UK-English AQoLS using forward-back translation; lay assessment of the initial translations with members of the general population in the target countries to ensure cultural appropriateness of the language register and content; cognitive debriefing interviews with AUD patients in the target countries to assess face and content validity. Results: Initial translations of the AQoLS into the three target languages were produced. Lay assessments were conducted with 8 individuals in Japan, 5 in China, and 5 in Korea, and modifications made to each new language version. Cognitive debriefing interviews were conducted with 8 AUD patients in Japan (3 male, age 34-71 years), 10 in China (8 male, age 25-45 years), and 12 in Korea (9 male, age 44-65 years). Patients found the AQoLS easy to understand and relevant to their experiences. Further modifications were made to maintain consistency with the source version, clarify the intended meaning, or to improve acceptability of individual word or phrases. Conclusions: The new adaptations of the AQoLS are appropriate instruments to assess HRQL in AUD patients in Japan, China, and Korea. The adaptation methodology was comprehensive, involving translation experts, members of the general public, and patients with AUD in each target country. This ensured that the new language versions were conceptually equivalent to the source instrument and appropriate to the target culture. PMH50 The Patient Experience of Alcohol use Disorder Whalley D 1, Luquiens A 2, Crawford S R 1, Laramée P 3, Doward L 1, Price M 4, Hawken N A 5, Dorey J 6, Owens L 7, Llorca P M 8, Falissard B 2, Aubin H J 2 Health Solutions, Manchester, UK, 2Hôpital Paul Brousse, INSERM 669, Université Paris-Sud, Villejuif, France, 3Lundbeck S.A.S., Issy-les-Moulineaux cedex, France, 4RTI Health Solutions, Research Triangle Park, NC, USA, 5Creativ-Ceutical, Luxembourg, Luxembourg, 6Creativ-Ceutical USA, Chicago, IL, USA, 7Institute of Translational Medicine, University of Liverpool, Liverpool, UK, 8Centre Hospitalier Universitaire de Clermont-Ferrand, Clermont-Ferrand, France . . . . . . . . . . . . . . . . 1RTI Objectives: Alcohol consumption carries a risk of adverse personal, social, and health effects. Alcohol use disorder (AUD) is a problematic pattern of alcohol use leading to clinically significant impairment or distress. The aim of this study was to determine the patient-perceived impact of AUD. Methods: Focus groups were conducted with a sample of patients with current or remitted AUD in the UK and France to identify key areas of impact from the patient perspective. The groups were audio-recorded and transcribed. Thematic analysis of the data was undertaken. Results: Ten focus groups were conducted with a total of 38 patients (20 current AUD, 18 remitted AUD; 26 male, 12 female; age range 23-69 years). All patients met the diagnostic criteria (Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition) for alcohol dependence. Patients characterised their relationship with alcohol as an ongoing battle for control. The cycle of consumption and dependence permeated most areas of patients’ lives, with often devastating consequences. Seven key areas of impact of AUD were identified: social and personal relationships; household, family and social activities; selfcare and personal safeguarding; emotional well-being; control over life, self and alcohol; financial and housing situation; and sleep disturbance. The impact of AUD continued even after patients stopped drinking; remitted patients described the ongoing battle to remain abstinent, and the lasting legacy of damaged relationships and feelings of low self-worth. Conclusions: AUD has considerable impact on many areas of patients’ lives, particularly relationships, and feelings of control and self-worth, which can continue even after abstinence has been achieved. From the patient perspective, drinking cessation or reduced consumption may not be the most relevant outcome to determine treatment benefit. There is a clear need for measures that can quantify the humanistic outcomes associated with AUD. PMH51 Societal Costs and Qol of Children with Adhd and their Parents: a Comparison to a Reference Group from the General Population Van Der Kolk A 1, van Agthoven M 2, Bouwmans C 3, Buitelaar J K 4, Hakkaart L 3 1Janssen-Cilag BV the Netherlands, Tilburg, The Netherlands, 2Janssen-Cilag BV, Tilburg, The Netherlands, 3Erasmus University Rotterdam, Rotterdam, The Netherlands, 4Donders Institute for Brain, Cognition and Behaviour, Nijmegen, The Netherlands . . . . . . Objectives: To compare Quality of Life (QOL) and societal costs in an ADHD population to a control group representing the general population. Methods: An online cross-sectional retrospective study among parents of children with ADHD was conducted via Balans, a Dutch ADHD organization. Because of large differences in responder groups in the initial study, comparison with a control group was added to understand the impact of ADHD compared to the general population in greater detail. Therefore, a separate online research among parents of children without ADHD or any other chronic disease was performed. The control group was selected on the basis of pre-defined characteristics in order to match the ADHD group. QoL (EQ-5D for children and parents and KIDSCREEN-10 for children) and societal costs (TiC-P) were surveyed based on proxy reporting. In this study, monthly resource use of patients and their parents was identified. Descriptive statistics, independent samples t-test and Chi-square for p-values (sig. 2-tailed) were used for analysis on significance in differences between samples. Results: The ADHD sample (n= 618) was compared to the control group (n= 704). Both groups primarily contained boys with a mean age of 11 years. QOL differed significantly between children with ADHD and the control group, both for EQ-5D (ADHD 0.80 vs. controls 0.96; p< 0.000) and KIDSCREEN-10 (ADHD 41.67 vs. controls 55.46; p< 0.000). The same is true for their parents (ADHD 0.83 vs. controls 0.88; p< 0.000). Total monthly costs associated with ADHD differed significantly as well, both for children (ADHD € 518 vs. controls € 91; p< 0.000) and parents (ADHD € 288 vs. controls € 49; p< 0.000). Conclusions: Although this was not a direct comparative study, the magnitude of differences in absolute results seem to justify the conclusion that ADHD accounts for significantly higher costs and lower QoL for patients, parents and society compared to a control group representing the general population. PMH52 Alcohol Consumption: the Burden of Entourage Touboul C 1, Hoertel N 2, Limosin F 2, Crochard A 3, Rouillon F 4 1Kantar Health, Montrouge, France, 2Hôpital Corentin-Celton, Issy-les-Moulineaux, France, 3Lundbeck S.A.S., Issy-Les-Moulineaux, France, 4Centre Hospitalier Sainte-Anne, Paris, France . . . . . Objectives: The purpose of this study, carried out on a general sample in France in 2013, was to measure the burden of alcohol consumption on the entourage (family members and friends) of people with alcohol problems and to identify the consequences of excessive alcohol consumption on the caregivers. Methods: The ENTOURAGE survey was conducted among individuals from KANTAR HEALTH’s panel. A self-administered questionnaire was sent to a representative sample of 10,000 adults (aged 18+) based on the quota method. The level of burden was assessed using the Zarit Burden Scale (ZBI) and the level of alcohol consumption using the AUDIT-C. Results: Of the 7,813 individuals who filled out their questionnaires, 13% (n= 1,018) stated that a member of their entourage has a problem with alcohol. Of this 13%, the alcohol consumption level was ‘not at risk’ in 2% (n= 18), ‘excessive’ in 39% (n= 406) and 52% (n= 533) were ‘alcohol-dependent’ according to AUDIT-C scores; 7% did not specify the quantity of alcohol consumed. 37.0% of these 1,018 respondents rated burden as ‘minimal’, 40.7% as ‘mild’, 18.3% as ‘moderate’ and 3.9% considered it as ‘severe’. The average score of the ZBI was 28.5 A463 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 (SE= 16.0). Burden felt by the entourage was positively correlated with the physical and affective proximity between the interviewee and the drinker, particularly when the interviewee lives with him, and with the presence of social, behavioral and medical consequences of alcohol consumption (on average the number of medical comorbidities was 2.1 for ‘severe’ burdens vs 1.4 for ‘minimal’ burdens, p< 0.01). Conclusions: Excessive drinking has deleterious effects on drinkers and his entourage. Our results demand greater recognition of caregivers’ burden in the management of patients with alcohol dependence and further support the need for improving alcohol dependence management and prevention of its development to reduce the global burden of alcoholism. PMH53 Quality of Life in Hospitalized Seniors with Psychiatric Disorders Poliakova N , Grochalova M, Mastiliakova D , Gerlichova K , Bielik J Trencin University, Trencin, Slovak Republic . . . . PMH56 Work Productivity Loss with Depression, Diagnosed and Undiagnosed, among Employed Respondents in an Internet-Based Survey Conducted in Japan Asami Y 1, Goren A 2, Okumura Y 3 1Pfizer Japan Inc., Tokyo, Japan, 2Kantar Health, New York, NY, USA, 3Institute for Health Economics and Policy, Tokyo, Japan . . . Objectives: In Japan, estimated numbers of patients who received medical treatment with major depressive disorder have recently increased and are reported almost 708,000 patients according to the results of the Patient Survey 2011. In addition, the potential magnitude of disease burden associated with undiagnosed depression is suggested by the low mental health service utilization in Japan, which is even lower than many western countries. The aim of this study is to investigate whether depression severity was associated with work impairments, regardless of diagnosis. Methods: Employed respondents (n=17,820) of an internet-based crosssectional survey (the 2011 National Health and Wellness Survey fielded in Japan) were categorized according to depression diagnosis (diagnosed or undiagnosed) and severity (Patient Health Questionnaire-9 score= 10+ or < 10). Work Productivity and Activity Impairment questionnaire-based measures were predicted by diagnosis, severity, and their interaction, adjusting for covariates. Results: Among employed respondents (n= 17,820) in the Japan NHWS, 3.8% were diagnosed with depression within 12 months (n= 678). Among those with diagnosis, 51.0% (n = 346) had PHQ-9 scores of 10 and above, while among those without a diagnosis, 7.8% (n = 1,336) had scores of 10+. In other words, 7.5% among all employed respondents (1,336/17,820) self-reported that they were undiagnosed as depressed but had PHQ-9=10+. Among the undiagnosed, high severity respondents had greater overall work impairment (33.3% vs. 14.8%), absenteeism (5.9% vs. 2.2%), presenteeism (30.4% vs. 13.4%) than low severity respondents. Significant interactions between diagnosis and severity indicated greater impairments among undiagnosed vs. diagnosed respondents (except on absenteeism). Conclusions: Depression severity was associated with work productivity loss, even among the undiagnosed, suggesting a need for early detection, referral, and treatment of depression in the workplace. Objectives: The number of seniors in Europe is increasing. The aim of our research study was to determine quality of life in the long term hospitalized seniors with psychiatric diseases and to determine to what extent the age and education affects the quality of life. Methods: Two standardized questionnaires - WHOQOL-BREF (World Health Organization Quality of Life Assessment-Brief) and WHOQOL-OLD (World Health Organization Quality of Life Assessment-Old) were used in the study. The research sample consisted of 100 patients hospitalized in geriatric psychiatric department of a Woman’s Psychiatric Hospital in Kroměříž. Patients were from 65 to 95 years old. Duration of hospitalization was from 2 to 12 months. Data were collected was in October 2012 - February 2013. Results: Average score WHO-QOL BREF was lower than of the population norms. Satisfaction with health was lower than the assessment of quality of life. Comparing the results to respondents 79 years old and over 80 years old it has been found lower scores in all domains of quality of life by younger patients. Worsening of physical health and quality of social relations were patients with lower level of education. Other domains were worse evaluated by patients with higher level of education. Assessment domains WHOQOL OLD values were lower than of population norms. Very low score, regardless of age and education, appeared in ratings independence. Conclusions: Long-term hospitalization is reflected in the perception of quality of life, particularly in the physical domain and the feeling of dependence. Chronic health problems, social isolation, reduction in the activity disturb seniors wellbeing and reduce subjective assessment of quality of life. The effort to engage of hospitalized seniors in society and leisure activities can increase the quality of their life. Early diagnosis and treatment reduce the price of treatment and length of hospitalization. PMH57 Description of Agitation and Crisis Interventions in a Psychiatric Hospital in Spain: a Qualitative Study PMH54 A Potential Gender Bias in the Quality of Life - an Exploratory Standard Gamble Experiment Among Economics Students Rubio-Valera M 1, Ortiz J M 1, Baladón L 1, Luciano-Devís J V 1, Salvador-Carulla L 2, Haro J M 1, Gracia A 3, Serrano-Blanco A 1 1Parc Sanitari Sant Joan de Déu, Sant Boi de Llobregat, Spain, 2Centre for Disability Research and Policy, Lidcombe, Australia, 3Ferrer, Barcelona, Spain Mental Health – Health Care Use & Policy Studies . Mahlich J 1, Al Obaidi L 2 1University of Vienna, Vienna, Austria, 2University of Nottingham, Nottingham, UK . . Objectives: To estimate determinants of quality of life of undergraduate Economics students at the University of Vienna. Methods: Quality of life values were elicited by means of the standard gamble approach. The impact of several variables such as gender, side job, length of study and living arrangements on the quality of life were identified using different types of regression techniques (OLS, GLM, Betafit). Results: Significant evidence was found that female gender is associated with a higher quality of life in all specifications of our estimations. Conclusions: The observed gender differences in quality of life can be attributed to a higher degree of risk aversion of women. A higher risk aversion leads to a higher valuation of given health states and a potential gender bias in health economic evaluations. PMH55 Economic Evaluation of Dialectical Behavior Therapy (Dbt) Amongst those with Borderline Personality Disorder (Bpd) who Engage in Self-Harm in Ireland Murphy A M , Bourke J University College Cork, cork, Ireland . . . Objectives: This study examines the cost effectiveness of Dialectical behavior therapy (DBT) amongst those with borderline personality disorder (BPD) who engage in self-harm compared to treatment as usual (TAU) in Ireland. DBT is a comprehensive cognitive–behaviour treatment, which has been demonstrated to be effective in reducing suicidal behaviour, hospitalisations, etc. The programme consists of group therapy and individual therapy, running for 24 weeks in total. Methods: A decision analytical model is developed to conduct the economic evaluation. The perspective taken is that of the health care payer/provider (HSE). Costs of the treatments were identified, measured and valued. This included direct costs of DBT and TAU, as well as additional resources consumed (emergency department, outpatient and inpatient attendances). Relevant health effects of the treatments were measured using a variety of condition specific and generic measures including the Beck Depression Inventory, Borderline Symptom Checklist and EQ-5D. These measures were collected at baseline, six months and 18-month intervals. To account for uncertainties a probabilistic sensitivity analysis, employing a Monte Carlo simulation, is performed. Results: Preliminary results indicate there are additional costs associated with DBT compared to TAU. With regard to effectiveness DBT was associated with lower presentations at emergency departments, output and inpatient stays; higher utilities and better scores on condition specific measures. The reduction in health service utilisations represents cost savings to the health care provider and payer. Conclusions: This is first economic evaluation of DBT in the Irish setting. A review by Brazier et al in 2006 suggested further evidence was required to demonstrate the cost effectiveness of DBT. This study contributes greatly to international literature. While DBT is an expensive therapy, compared to TAU, this study finds that the cost savings owing to reduction in utilisations of acute services is noteworthy. . . . . . . . . . . Objectives: The aim of the study was to define and characterize the agitation states present in usual medical practice in the acute and emergency units of a psychiatric hospital from the viewpoint of psychiatrists and nurses. We also aimed to describe the typical care packages (crisis intervention) for each of the agitation states described. Methods: Two nominal groups, one with 7 nurses and the other with 10 psychiatrists, from the Parc Sanitari Sant Joan de Déu emergency and acute care psychiatric wards were established. Results: The nurses described two main states forming the endpoints of a spectrum: from mild (pre-agitation) to severe (agitation). A third state was outlined in which agitation was characterized by disorganized behavior problems. Agitation subtypes were described according to the characteristics and severity of the clinical situation. Various care packages were described for each agitation state. The care packages were divided into first, second and third line approaches. The first line approaches (i. e., verbal containment) were used on every (pre) agitated patient. If the first line approach was not effective, the second and third line approaches were implemented, culminating with physical restraint. Every episode was followed by behavioral observation. The psychiatrists described 3 states: a mild initial state (anxiety and irritability), moderate (pre-agitation without aggressiveness) and a severe state of agitation with aggressiveness and/or violence. Again, every state was associated with specific treatment. Conclusions: In order to avoid progression to a severely agitated state, both groups agreed on the importance of appropriate verbal containment for all states. This would be followed by environmental measures, medication and mechanical restrain depending on the severity of the state. Mechanical restrain is stressful for the patient, requires considerable health care resources and should be avoided whenever possible. PMH58 Comparison of Health Care Resource use and Costs in Patients with Opioid Prescription Drug Dependence (Opd) Treated with Buprenorphine/Naloxone and Patients without Pharmacological Treatment: Retrospective Analysis of us Public Insurance Claims Khemiri A 1, Clay E 2, Kharitonova E 2, Aballéa S 2, Zah V 3, Ruby J 4 . . . . . . 1Creativ-Ceutical, Tunis, Tunisia, 2Creativ-Ceutical, Paris, France, 3ZRx Belgrade, Serbia and Montenegro, USA 4Reckitt Outcomes Research Inc., Benckiser Pharmaceuticals, Inc./NA, Richmond, VA, Objectives: The buprenorphine/naloxone (BUP/NAL) combination is used in the treatment of prescription opioid dependence (OPD). The objective of this study was to determine if there were health economic advantages related to treatment compared to no pharmacological treatment. Methods: A retrospective cohort analysis was performed using insurance claims extracted from the US Truven Health MarketScan® Medicaid database from January 2007 to December 2012. Two groups were considered: 1) patients with opioid prescription drug (OPD) dependence treated with buprenorphine/naloxone and 2) patients with opioid prescription drug (OPD) dependence and no pharmacological treatment. Final study groups were selected with one-to-one matching on demographic characteristics, comorbidities at baseline and cost of outpatient and inpatient care over six months before index date. A464 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Resource use (pharmacy claims, outpatient claims, emergency room admission and hospital admission) and corresponding costs over twelve months after index date were compared between groups. Results: Each group included 362 patients. Patient characteristics at baseline, resource use, and health care costs before index date were comparable between the two groups. At twelve months after the index date, patients with no pharmacological treatment had higher resource utilization in every category but medication. Total costs over 12 months were $14,983 and $15,692 in groups with and without pharmacological treatment, respectively (p = 0.67). Patients with pharmacological treatment had a higher pharmacy cost, but this was offset by the higher cost of outpatient visits in patients with no pharmacological treatment. These visits were mostly related to mental disorders, nervous system, skin and musculoskeletal disorders and injuries and poisonings. Conclusions: While the treatment of opioid dependence with buprenorphine/naloxone is associated with higher medication acquisition cost, it is outweighed by cost savings in other categories, especially outpatient care. Patients without pharmacological treatment use more health care resources and have higher total costs. age group of 21 -30 years. In study population 141 (61.3%) were married and majority of Housewives 57 (24.8%) followed by Govt. service 28 (12.2%). Amongs study population 136 (59.1%) patients received the second generation antipsychotics and 8 (3.5%) patients received first generation antipsychotics Resperidone was the most commonly prescribed antipsychotic given to 36.5% of the patients followed by clozapine 26.5% and olanzapine. Monotherapy was received by 22.2%. 106 (46.08%) patients and 89 (38.69%) received dual drug regimen. Conclusions: The utilization pattern of antipsychotics, revealed that atypical antipsychotics were prescribed more commonly when compare to typical antipsychotics. Among the atypical antipsychotics, Resperidone was commonly used during Schizophrenia as compared to other atypical antipsychotic drugs. PMH59 Drug Utilization Study of Antipsychotics used for the Hospital Treatment Of Schizophrenia in Russia 1Market Access PMH62 Comparison of Resource use and Health Care Costs in New Initiators of Long-Acting Injectable (Lai) and Oral Second Generation Antipsychotics Seetasith A 1, Burudpakdee C 2 Solutions LLC., Raritan, NJ, USA, 2MKTXS, LLC, Raritan, NJ, USA . Vaskova L , Tyapkina M 1st Moscow State Medical University named after I.M. Sechenov, Moscow, Russia . . Objectives: The goal was to examine utilization of typical (TA), atypical antipsychotic (AA) and depot antipsychotic s (DA), used for treatment of patients with schizophrenia in Russian hospital during 1 year. Methods: A retrospective study, carried out in Moscow hospital from January 2012 to December 2012. Patient data on demography (age, sex), medicines used (dose, duration of treatment), length of hospital stay and clinical outcome were recorded and analyzed. Drug consumption was calculated using defined daily dose (DDD) methodology. Results: Total 227 patients were included in the study. Among study population 121 (53.3%) patients were male and mean age was 33.2 (male) and 45.5 (female). Out of 227 patients 219 (96.5%) were improved and 1 (0.4%) person were recovery. Mean length of hospitalization was 75.3 days. Total 27 antipsychotic were used for the treatment schizophrenia among these patients. Among used antipsychotic consumption (DDD/100 bed days) was highest for AA clozapine oral (194.25) followed by TA haloperidol oral (52.39), DA fluphenazine parenteral depot (20.04) and DA haloperidol parenteral depot (19.59). However the cost of treatment for AA clozapine was higher, than for TA haloperidol. The total consumption (DDD/100 bed days) in the antipsychotics group was: 224.3 for AA, 73.38 for TA and 39.91 for DA. Conclusions: This study provides estimate of consumptions different antipsychotics used for the hospital treatment of schizophrenia. Atypical antipsychotic clozapine oral is highest consumed among 27 antipsychotics. Total drug utilization for AA was three times higher than for TA and five times higher than DA, however the costs of treatment for AA was substantially higher than for TA. PMH60 Drug Utilization Pattern of Lisdexamfetamine Dimesylate in Germany . Objectives: To measure health care utilization and costs among new initiators of LAI and oral second generation antipsychotics with schizophrenia or bipolar disorder. Methods: A large database of a commercially insured US population was used to index patients on their first treatment between 1/1/2011 and 12/31/2011. Patients were required to have ≥ 12 months pre-index and ≥ 12 months post-index, were new users of a second generation antipsychotic, and diagnosed with schizophrenia or bipolar disorder during their pre-index period. LAI and oral patients were matched 1: 3 using propensity scores. Mean differences in annual resource use and costs were compared across groups in an unadjusted difference-in-difference analysis: [ (LAI post - LAI pre) - (Oral post - Oral pre)]. Results: Initial selection identified 250 LAI and 8,356 oral treatment patients. Matching resulted in balanced cohorts of 204 LAI and 612 oral initiators. Annual hospitalizations and ER visits from pre-index to post-index was significantly lower in LAI initiators compared to oral initiators. Mean annual hospitalizations per LAI patient reduced from 1.09 to 0.51 (p < 0.0001) while that of the oral cohort reduced from 0.53 to 0.39 (p = 0.0011). This resulted in a net reduction of 0.45 annual hospitalizations per patient in the LAI cohort, using the oral cohort as a reference (p < 0.0001). Mean annual ER visits reduced from 1.72 to 1.03 per LAI patient compared to no change in the oral cohort, resulting in a net difference of 0.72 ER visits between the two groups (p < 0.0001). The unadjusted difference-in-difference analysis showed a relative reduction in total health care costs of $4,997 in the LAI cohort compared to the oral cohort. Conclusions: Initiating treatment with an LAI resulted in greater reductions in hospitalizations and ER visits compared to oral second generation antipsychotic medications in patients with schizophrenia or bipolar disease. PMH63 Antidepressant use and Suicide Rate in England: the Geographic Divide Anderson R , Wilson T , Griffiths M Costello Medical Consulting Ltd., Cambridge, UK . . . Objectives: - The objective of the study was to provide utilization data in Germany for the ADHD stimulant medication lisdexamfetamine dimesylate (LDX) in the first six months following its launch in June 2013. Methods: - This drug utilization study (DUS) analysed a longitudinal electronic medical record database (IMS Disease Analyzer - Germany), and was part of a larger DUS providing data for 8 European countries for up to 5 years. The study included records of all patients who had been prescribed LDX from June 2013 to December 2013 in the paediatrician panel (PP) or neurologist/psychiatrist panel (NPP) of the German database. Results: - The analysis included 123 patients (348 prescriptions) from the PP and 296 patients (710 prescriptions) from the NPP. In both panels, 91% of patients had a documented diagnosis of ADHD. When initiated on LDX treatment, most patients (PP, 98%; NPP, 91%) were between 6 and 18 years of age; up to 1% of patients in both panels were below 6 years of age and 2% and 8% of patients in the PP and NPP, respectively, were above 18 years of age. The majority of patients in both panels were male (PP, 77%; NPP, 79%). The average prescribed daily dose of LDX was within the recommended range (30–70 mg) for all patients in the NPP and for 98.4% in the PP, with a mean daily dose across patients of 42 mg in both panels. Conclusions: - The findings of this analysis of electronic medical records indicate that, during the first six months after launch, LDX was mainly prescribed in Germany within the EMA-approved Summary of Product Characteristics (SmPC) with regard to the indicated patients, age group and dose regimen. Objectives: Mental illness is widespread, with 1 in 3 people worldwide reporting symptoms indicative of a psychiatric disorder at some point in their lives. The use of antidepressants has risen globally and within Europe has been reported to be correlated with a reduced suicide rate. The aim of this research is to analyse the use of antidepressants in England and identify any trends. Methods: Antidepressant prescribing data for National Health Service (NHS) England from 2003–2012 and from all Primary Care Trusts (2010/11–2012/13) and Clinical Commissioning Groups (2013/14) were obtained from the Health & Social Care Information Centre. Data were collated for the four NHS regional area teams (North, Midlands & East, London, and South) and analysed against population size and suicide rates. Results: Antidepressant use in England has increased dramatically in recent years and coincides with a year-on-year drop in ingredient costs, with 27.7 million prescriptions in 2003 and a net ingredient cost of £395.2 million, to 50.2 million prescriptions in 2012 and a net ingredient cost of £211.1 million. From 2010 to 2014, almost £1 billion has been spent on antidepressants by NHS England, of which almost a third is accounted for by the North region. Over 4 years, the average number of prescriptions per 1,000 population was 1,140.7, 987.4, 888.2 and 540.5 in the North, Midlands & East, South and London, respectively. These figures were correlated with a suicide rate of 9.87 and 7.05 per 100,000 people in the North and London, respectively. Conclusions: There is a clear divide within regions of England regarding antidepressant use and suicide rate, and the correlation between these two measures was found to be opposite to that reported for Europe generally. These findings highlight the importance of understanding mental illness and the underlying reasons for the wide disparity in England. PMH61 Utilization Patterns of Antipsychotics Usage in Tertiary Care Hospital Patients with Schizophrenia PMH64 Analysis of Prescribing Patterns of Atypical Antipsychotics in Lhu Caserta Pethekar A J 1, Nair S 1, Kunhikatta V 1, Sharma P S V N 2, Thunga G1 1Manipal College of Pharmaceutical Sciences, Manipal University, Manipal, India, 2Manipal University, Manipal, India Farina G 1, Menditto E 2, Manna S 1, Pagliaro C 1, Troncone C 1, Putignano D 2, Orlando V 2, Linguiti C 1, Buffardi G F 1, Tari M G 1 1LHU Caserta, Caserta, Italy, 2CIRFF- Center of Pharmacoeconomics, Naples, Italy Objectives: The main objective of the study was to find the utilization pattern of antipsychotics in schizophrenia patients in a tertiary care teaching hospital South India. Methods: A retrospective study has been carried on schizophrenic patients admitted in a tertiary care teaching hospital for a period of one year. The diagnosis of schizophrenia was based on ICD-10 (Tenth revision) criteria. Patients of both sexes who diagnosed with schizophrenia were included in the study. Other mental illness or drug induced psychosis patients were excluded from the study. All demographical and clinical characteristic including treatment pattern were collected and entered. Data were analyzed in SPSS 20.0. Results: Out of 230 patients, 144 (63%) were males and the majority of patients were 65 (28%) in the Objectives: The Local Health Unit Caserta has made web-platform (SANIARP) available to specialists and pharmacists to enter the diagnostic and therapeutic information of the patient with each prescription. The advantage provided by this platform is to make available the information on the analysis of the profiles of prescriptive drugs in a large population sample. The aim of study was to evaluate prescribing patterns of atypical antipsychotics in LHU Caserta for the years 20112013. Methods: This retrospective cohort study was carried out from the data of pharmaceutical prescriptions and of plans therapeutic in Saniarp in the LHU Caserta in the 2011-2013. Information about users of atypical antipsychotics were analyzed. Information about the diagnosis and treatment plans were obtained through the Cook M 1, von Bredow D 2, Rex J 3, Keja J 2, Maxwell T 1, Varughese S 1 1Shire, Wayne, PA, USA, 2IMS Health, Munich, Germany, 3IMS Health, Frankfurt am Main, Germany . . . . . . . . . . . . . . . . . . . . . . . . . . A465 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 linkage between the pharmaceutical database and the SANIARP database, using anonymized patient code. Based on the date of the first prescription (index date) the following prescribing patterns have been defined: continuers (subjects with a gap < 30 days between two prescriptions on-going); intermittent (subjects with a gap> 30 days but that receive another prescription at index date); switchers (discontinuation of index drug and prescription of a new antipsychotic); add-on (addition of a new antipsychotic); as-needed (addition of a new antipsychotic for a limited period). Results: We identified 2,768 patients (44.5% females) with at least one prescription of atypical antipsychotics and with a diagnosis coded in the study period. Schizophrenia is the most frequent indication (31.1%) the most prescribed drug is olanzapine (29.1%), followed by risperidone (17.7%), quetiapine (13.4%), aripiprazole (12.5%), clozapine (10.3%) and asenapine (3.1%). About 70% of schizophrenic patients is treated with the same drug, 7.9% switch and 23.6% is in polytherapy. Conclusions: The use of SANIARP, web- platform able to allow the systematic monitoring of prescribing patterns of drugs, is of primary importance for better health planning. PMH65 Maintenance Daily Dose of Venlafaxine and Duloxetine in the Monotherapy of Patients with Major Depressive Disorder Resistant to Selective-Serotonin-Reuptake-Inhibitors in Routine Clinical Practice in Spain Sicras Mainar A 1, Rejas Gutiérrez J 2, Blanca Tamayo M 1, Navarro Artieda R 3, De Lossada Juste A 4 1Badalona Serveis Assistencials SA, Badalona (Barcelona), Spain, 2Pfizer S.L.U., Alcobendas/ Madrid, Spain, 3Hospital Germans Trias i Pujol, Badalona (Barcelona), Spain, 4Pfizer, S.L.U, Alcobendas/Madrid, Spain . . . . . Objectives: Major Depressive Disorders (MDD) guidelines recommend using antidepressants with dual mechanism-of-action (venlafaxine, duloxetine) when resistant to a prior course of Selective-Serotonin-Reuptake-Inhibitors (SSRI). Dose to use should be close to the DDD recommended by WHO. Routine clinical practice may be frequently far from guidelines. The aim was to ascertain the average maintenance daily dose (DD) of venlafaxine and duloxetine in the monotherapy of patients with MDD who showed resistance to a previous SSRI in routine medical practice in Spain. Methods: Retrospective analysis extracting consecutively electronic medical records (EMR) of the BSA, a provider which health plan coverage includes near 120,000 inhabitants in Badalona (Spain). EMR of male/female patients, > 18 years, included in the chronic prescription follow up program, with a MDD ICD-9-CM code (296.2x/296.3x), and who were resistant to a previous SSRI course, were extracted for analysis. Resistant was defined as persistence of symptoms (score >17 in the Hamilton-Depression scale and/or reduction lower than 30% of the baseline score). Maintenance DD was considered the dose repeated (refills) at least two time consecutively during the study period (years 2012-2013). Results: Threesixty-eight EMR [81% women, 60.3 (15.2) years] were extracted; 160 of duloxetine and 208 of venlafaxine. Average maintenance DD were 65mg/day and 117 mg/day for duloxetine and venlafaxine, respectively. Demographics, number of comorbidities or previous SSRI were not related with average dose. 86% of duloxetine EMR were prescribed the WHO DDD for this drug (60mg), while only 42% of venlafaxine received its WHO DDD (100mg), p< 0.001. Number of DDD per day were significantly higher with venlafaxine; 1.17 (1.10-1.23) vs. 1.09 (1.05-1.12), p=0.049. Conclusions: Routine medical practice average maintenance DD of venlafaxine and duloxetine in SSRI resistant subjects with MDD are different in terms of both their recommended doses in labelling or guidelines and in number of DDD per day. PMH66 The Health Economic Impact of Resource use in Dementia: the Erlanger Dementia Registry (EDR) Schaller S U 1, Marinova-Schmidt V 1, Gobin J 1, Luttenberger K 2, Richter-Schmidinger T 3, Gräßel E 2, Maler J M 3, Kornhuber J 3, Kolominsky-Rabas P L 1 1Centre for Health Technology Assessment (HTA) and Public Health (IZPH), Friedrich-AlexanderUniversity Erlangen-Nürnberg, Erlangen, Germany, 2Centre for Health Services Research in Medicine, Psychiatric and Psychotherapeutic Clinic, University Hospital, Friedrich-AlexanderUniversity Erlangen-Nürnberg, Erlangen, Germany, 3Psychiatric and Psychotherapeutic Clinic, University Hospital Erlangen, Friedrich-Alexander-University Erlangen-Nürnberg, Erlangen, Germany . . . . . . . . . . . . Objectives: Dementia patients are in need of more extensive personal care compared to other long-term care users. This results in a high economic impact of dementia on patients, families and health care systems. Due to the increasing prevalence of dementia worldwide, combined with limited health care expenditures, a better understanding of resource use in dementia care is needed. Therefore the aim of our study is the assessment of resource use in dementia in the most common setting: home-based care (informal caregivers). Methods: The Erlanger Dementia Registry structure was set up in 2013. Both dementia patients and informal caregivers are interviewed separately with internationally approved valid instruments. Follow-up takes place after 6, 12 months and afterwards annually. Resource use in dementia is assessed via the ‘Resource Utilization in Dementia (RUD) instrument’. Results: A total number of 50 informal caregivers (mean age=63, 61% female, 23% employed, 72% live together with the patient) were interviewed after the initial dementia diagnosis at baseline, and 22 study participants took part at the 1st follow-up. Informal caregivers were mainly spouses (72%) and children (22%). Main support was provided for instrumental activities of daily living (t0:77%; t6:86%), followed by activities of daily living (t0:37%; t6:52%) and supervision (t0:26%; t6:33%). Average hours for support 6 months after diagnosis were: IADL= 4.3h/day (min= 1.0, max=16.0), ADL= 5,2h/day (min=2.0, max=16.0), and supervision=12.9h/day (min=1.0, max=24.0). The average monthly costs for informal caregivers 6 months after diagnosis (medication; additional disease-related costs) are 76 €. Conclusions: Our results highlight the significant impact of informal costs (time provided for care) in dementia care, occurring early in the disease course. For dementia patients cared for at home, informal costs put an additional economic burden on families. For future health policy planning in dementia, the perspective and inclusion of informal costs is essential. The research is funded by the European Commission, ICT FP7, project ID 287509. PMH67 Economic Burden of Major Depressive Disorder (Mdd) in Five European Countries: Description of Resource use by Health State Painchault C 1, Brignone M 2, Lamy F X 2, Diamand F 2, Saragoussi D 2 1Keyrus Biopharma, Levallois-Perret, France, 2Lundbeck SAS, Issy-les-Moulineaux, France . . . . . . Objectives: Estimating resource use [RU] in real life is an important part of health economic evaluations. RU data should reflect how patients are actually treated. In MDD, RU data are mostly obtained from expert opinion. Variability in RU may lead to uncertainty in health economic evaluations, but few published studies report these data in the detail needed. The present analysis reports RU data by depression health state from an observational study. Methods: PERFORM (Prospective Epidemiological Research on Functioning Outcomes Related to Major depressive disorder) is a 2-year prospective observational study conducted in 5 Western-European countries. Two- and six-month RU were estimated by health state: remitters, non-remitters, patients in relapse or not. RU included visits to different health care professionals, hospitalization and sick leave. Results are reported for the whole study population and are also available by country (including the UK, for which EQ5-D-derived utilities are also available) and for subgroups (e. g., patients who switched antidepressants at baseline). Results: Of the 819 analysable patients at 2 months, 29% were in remission. Among patients with at least one visit, the frequency of visits to general practitioners, psychiatrists and psychotherapists was consistently lower for remitters versus non-remitters (1.8 vs. 2.4, 2.2 vs. 2.4 and 2.6 vs. 3.1 respectively). Fourteen patients had at least one hospitalisation. Sick leave was less frequent (14% vs. 27%) and shorter (34 vs. 41 days) for remitters versus non-remitters respectively. At 6 months, 19.3% of patients relapsed. RU were higher with more visits to psychiatrists, psychotherapists (4.0 vs. 2.7, 7.8 vs. 5.5) for relapsed versus non-relapsed patients. Conclusions: This first analysis provides European RU data in MDD. More information is expected at completion of the two-year follow-up and this study offers the possibility to describe RU by health states, countries and subgroups and assess their transferability to other countries. PMH69 The Impact of Economic Crisis on Suicide Rates in Greece Skroumpelos A 1, Zavras D 1, Kyriopoulos I I 1, Nikolaidis G 2, Kyriopoulos J 1 School of Public Health, Athens, Greece, 2Institute of Child Health, Athens, Greece . . . . . . 1National BACKGROUND AND Objectives: Economic crisis in Greece has several social implications, as unemployment and poverty have largely increased during the past years. Since the onset of the economic crisis, suicides have marked a significant increase. Therefore, aim of this study is to investigate the relationship between suicides and the economic crisis and certain macroeconomic indices. Methods: Annual suicide rates were obtained from the Hellenic Statistical Authority. Multiple linear regression analysis with Newey-West standard errors was carried out in order to examine the relationship between gender and age specific suicide rates and unemployment, GDP per capita and economic crisis (binary variable). Additionally, several statistical tests were conducted in order to examine the properties and the robustness of the model. Results: Unemployment appears as the major factor affecting suicides of men and women above the age of 15. However, gender and age-related differences are being observed. Unemployment is positively associated with the suicides of men aged 15-24, 35-44, 55-64 years. Female suicides are also affected by unemployment, excluding the age groups of 35-44 and 45-54 years old. Interestingly, suicides of women aged between 45-54 and 55-64 were negatively associated with economic crisis. In the total population, unemployment has impact on suicides for 15-24, 35-44, 55-64 age groups, while economic crisis affects suicide rate in the age group of 25-34. In addition, GDP per capita is negatively associated with suicide rates for young men (aged under 24). The same effect is also observed for the young population in general. Conclusions: The current economic turmoil in Greece affects suicides deaths. According to this analysis, unemployment is the main factor that determines age-specific rates and essentially point to the direction where measures should be taken in order to control suicides’ incidence and lessen the effect of economic crisis on health. PMH70 Population Health: Mental Health of us Veterans by Benefits Enrollment Status Richardson T 1, Claeys C 2, Sastry P 1 1KJT Group Inc., Honeyoe Falls, NY, USA, 2KJT Group Inc., Honeoye Falls, NY, USA . . . Objectives: Population-based approaches to improving health are critical to controlling rising health care costs. US Veterans represent an identifiable population of significant interest due to their unique occupational exposures. The objectives of this study is to characterize the mental health of a representative US sample specifically comparing: non-Veterans, Veterans receiving VA benefits and Veterans not receiving VA benefits. Methods: A representative (U.S. ) sample of 2,000 individuals completed an online survey assessing their mental health (PHQ-2), Veteran status and receipt of Veteran Administration health care benefits. We conducted bivariate analyses among the entire population and multivariate logistic regression among 352 Veterans to assess the relationship between ones mental health (PHQ-2 score > =3) and Veteran benefit enrollment status controlling for ones self-reported physical health, history of combat, awareness of Veteran Crisis Line and sociodemographic factors. Results: Overall 28% of Veterans scored positive for depressive symptom based on a score of 3 or greater on the PHQ-2 compared to 11% of non-Veterans. However, breaking Veterans into those enrolled and those not enrolled revealed that 38% of enrolled Veterans were positive compared to just 17% of non-enrolled Veterans. After removing insignificant and/or collinear variables from the logistic regression model, the final set of independent variables included Vets enrolled in VA benefits OR=3.91 (1.93- 8.44), poor/fair physical health OR= 4.32 (1.98-9.68), age OR= 0.93 (0.91-0.95) and awareness of Veteran crisis line 3.12 (1.69- 5.97). Conclusions: Younger veterans in poorer physical health whom are receiving VA benefits and are aware of the Veteran Crisis line are more likely to have depressive symptoms. Population based approaches to A466 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 identifying Veterans at risk for poor mental health may be working to provide health care benefit coverage but the persistence of symptoms, amongst those enrolled, may suggest a need for improved treatment or surveillance. PMH71 History of Antidepressant use Among Primary Care Depressed Patients Switching Treatments in the United Kingdom Lamy F X 1, Quelen C 1, Brignone M 1, Ferchichi S 2, Vataire A L 2, Rive B 1, Saragoussi D 1 1Lundbeck SAS, Issy-les-Moulineaux, France, 2Creativ-Ceutical, Paris, France . . . . . . . . . Objectives: Major depressive disorder is characterised by the presence of one or more major depressive episodes. Up to one-third of patients do not adequately respond to first-line therapies. In case of treatment failure, the most common strategy is to switch to another antidepressant drug (AD). However, patients may have been exposed in prior episodes to one or more other ADs, which could impact the efficacy of subsequent treatments. This study describes historical AD use in depressed patients undergoing treatment switch. Methods: This retrospective longitudinal cohort study used a database of medical records from general practitioners located throughout the UK (CPRD). Adult patients with a depression diagnosis undergoing an AD switch between 01/01/2012 and 30/06/2013 and with no diagnosis of schizophrenia or bipolar disorder were included. Historical AD use was assessed on all available prescriptions prior to switch and was grouped by treatment class (SSRI, SNRI, TCA and others). Results: 11,611 patients were identified. Their mean age was 44.5 years (SD= 16.18) and 66.7% were women. Prior to switch, patients used 2.65 different classes of ADs on average (SD= 1.89); at least two different classes of ADs were prescribed for 64.8% of patients. Before the switch, SSRIs were the most common (1.65/patient) followed by TCAs (0.51/patient), other ADs (0.31/patient) and SNRIs (0.18/patient). Significant proportions of patients had previously been prescribed several AD classes: 34.1% of patients received at least one SSRI and one TCA, 24.0% at least one SNRI and one TCA, 15.0% at least one SSRI and one SNRI, and 13.7% at least one TCA and one other AD prior to AD switch. Conclusions: This analysis showed that for patients initiating an AD switch, historical use of ADs was very common, and that a substantial proportion of patients had already been prescribed several ADs, thereby reducing the treatment options for future switches. PMH72 Treatment Patterns and Health Care Costs in Patients with Depression Treated with Antidepressant only or Combined with Benzodiazepine: Results From a Japanese Claims Database Analysis Jamotte A 1, Clay E 1, Onishi Y 2, Aballéa S 1, Toumi M 3 K.K., Tokyo, Japan, 3University of Marseille, Marseille, France . . . . . 1Creativ-Ceutical, Paris, France, 2Sanofi Objectives: To describe and compare treatment patterns and health care costs in patients with depression initiating treatment with antidepressant (AD) (in monotherapy or in combination) and patients initiating AD treatment combined with benzodiazepine (BZD) using claims database in Japan. Methods: Using data from the Japan Medical Data Center (JMDC) database, a retrospective longitudinal cohort study was conducted, including adults aged 18-65 years with a first prescription of AD (index date) between January 2009 and September 2013, and diagnosed with major depression around index date. Time to treatment discontinuation (gap of 90 days without AD treatment) and total costs at 6 and 12 months were estimated for each group (AD vs. AD+BZD), and compared, with adjustment on potential confounders (gender, age, insurance status, index year and baseline costs) using Cox proportional-hazards regression and log-linear regression models. Results: 7,723 patients were included (43.7% initiating treatment with AD+BZD and 56.3% initiating with AD). Mean age of the population was 36.5 years and 59.1% were males, with no significant differences found between the groups (p= 0.14 and p= 0.39 respectively). The total health care costs at 6 months were significantly higher in AD+BZD patients (¥176,946 vs. ¥151,992; p< 0.0001). After adjustment, AD+BZD treatment was associated with higher costs at 6 months (+5.2%, IC95%: [+4.7 %; +5.8 %], p< 0.0001) and at 12 months (+6.4%, IC95%: [+5.7%; +7.1%], p< 0.0001). Time to discontinuation was longer for the AD+BZD group (HR= 0.758, p< 0.0001). Conclusions: A large proportion of patients with depression were treated with AD+BZD at initiation, a practice not recommended in Japanese guidelines. Patients treated with AD+BZD tend to have higher costs than patients treated with AD. The combination with BZD is associated with longer treatment duration and further analyses are recommended to determine if this is related to improved persistence or longer time to remission. PMH73 Level of Testing for Potential Medication-Related Co-Morbidities for Patients Taking Antipsychotics Rascati K L 1, Richards K M 2 1The University of Texas at Austin, College of Pharmacy, Austin, TX, USA, 2The University of Texas at Austin, Austin, TX, USA . . . 13 years (children) [22.7%], 13-17 years (adolescents) [17.3%], 18-40 years [30.0%], and >40 years [30.0%]. The prevalence of HbA1c, FBS, and lipid panel testing significantly differed among the groups overall, and increased with increasing age, for both FGAs and SGAs. Conclusions: It is recommended that patients taking antipsychotics be tested regularly for glucose and lipid changes. Only about one-third of patients taking antipsychotic medications had a lipid panel test, and less than one-fifth had an HbA1c or FBS test during the 12 months of analysis. . Objectives: Patients taking antipsychotics are at higher risk for other co-morbid diseases, including metabolic-related conditions. The objective of this study was to determine the level of testing for potential medication-related co-morbidities, looking at rates of Hemoglobin A1c (HbA1c) and fasting blood sugar (FBS) tests for diabetes as well as lipid panel tests for hyperlipidemia. Methods: Using Texas Medicaid claims for 2012, a total of 135,757 patients filled at least one prescription for an antipsychotic medication. Claims data on lab testing were extracted for these patients. Results: Overall, 36.0% of patients taking antipsychotics received a lipid panel test, 18.7% received an HbA1c test, and only 6.5% received a FBS test. Nine percent and 95.8% of patients had claims for first- and second-generation antipsychotics (FGAs and SGAs), respectively [some patients had both prescribed]. A total of 53.5% were male and 46.5% were female. A larger proportion of patients taking FGAs had HbA1c, FBS, and lipid panel tests compared to patients taking SGAs: 28.9% vs. 18.4% (HbA1c); 8.1% vs. 6.4% (FBS); 49.8% vs. 35.6% (lipid panel). A larger proportion of females had HbA1c, FBS, and lipid panel tests compared to males: 23.4% vs. 14.8% (HbA1c); 7.6% vs. 5.6% (FBS); 40.7% vs. 32.3% (lipid panel). In addition, patients were divided into four age groups: < PMH74 Treatment Patterns and Health Care Costs in Patients with Schizophrenia Initiating with First- or Second-Generation Antipsychotic: Results from a Japanese Claims Database Analysis Jamotte A 1, Clay E 1, Aballéa S 1, Onishi Y 2, Toumi M 3 K.K., Tokyo, Japan, 3University of Marseille, Marseille, France . . . . . 1Creativ-Ceutical, Paris, France, 2Sanofi Objectives: To describe and compare treatment patterns and health care costs in patients with schizophrenia initiating treatment with first-generation antipsychotic (s) (FGA) and patients initiating treatment with second-generation antipsychotic (s) (SGA) using claims database in Japan. Methods: Using data from the Japan Medical Data Center (JMDC) database, a retrospective longitudinal cohort study was conducted, including adults aged 18-65 years with a first prescription of antipsychotic (index date) between January 2009 and September 2013, and diagnosed with schizophrenia around index date. Time to treatment discontinuation (gap of 90 days without antipsychotic treatment), time to switch, time to psychiatric hospitalisation and total costs at 6 and 12 months were estimated for each group (FGA vs. SGA), and compared, with adjustment on potential confounders (gender, age, insurance status, index year and baseline costs) using Cox proportional-hazards regression and log-linear regression models. Results: 2,615 patients were included (14% initiating treatment with FGA and 86% initiating with SGA). 55.0% were female, with no significant difference between the groups (p= 0.22). FGA patients were older than SGA patients (40.1 years vs. 34.1, p< 0.0001). The total health care costs at 6 months were significantly higher in FGA patients (¥1,489,680 vs. ¥324,595; p< 0.0001). After adjustment, FGA treatment was associated with higher costs at 6 months (+121%, IC95%: [+93%; +152%], p< 0.0001), and at 12 months (+87%, IC95%: [+60%; +119%], p< 0.0001). Time to discontinuation as well as time to hospitalisation were shorter for the FGA group but did not reach significance (HR=1.146, p=0.09; HR=1.406, p=0.07 respectively). The probability of switching was higher in the FGA group (HR: 5.281, p< 0.0001). Conclusions: A large proportion of patients with schizophrenia were treated with SGA at initiation, consistently with previous studies conducted in Japan. Patients treated with SGA tend to have longer treatment duration and lower costs than patients treated with FGA. Urinary/Kidney Disorders – Clinical Outcomes Studies PUK1 Comparative Efficacy And Tolerability Of Solifenacin 5mg Versus Oral Antimuscarinic Agents In Overactive Bladder (Oab): A Systematic Literature Review (Slr) And Mixed Treatment Comparison (Mtc) Kelleher C 1, Aballea S 2, Maman K 3, Nazir J 4, Hakimi Z 5, Chambers C 4, Odeyemi I A 4 and St. Thomas’ NHS Foundation Trust, London, UK, 2Creativ-Ceutical, Paris, France, Ceutical, Paris, France, 4Astellas Pharma Europe Ltd, Chertsey, UK, 5Astellas Pharma Global Development, Leiden, The Netherlands . . . . . . . . 1Guy’s 3Creativ Objectives: To compare the efficacy and tolerability of oral first-line antimuscarinic agents recommended for the treatment of OAB. Methods: Literature searches were undertaken to identify published randomised controlled trials (2000–2012), which reported efficacy and/or tolerability in adults receiving pharmacological treatment for OAB. Bayesian MTC methodology was used to assess changes from baseline in micturition frequency per 24h, urgency, incontinence, and urge incontinence (UI). Safety outcomes included dry mouth, constipation, and blurred vision. Results: Forty-one eligible trials involving 25,118 patients were included in the MTC. Solifenacin 5mg was significantly more effective than darifenacin 7.5mg based on incontinence (mean difference [MD]= 0.531/day), fesoterodine 4mg based on micturition (MD= 0.374/day), and tolterodine 4mg based on micturition (MD= 0.387/day), urgency (MD= 0.434/day), incontinence (MD= 0.309/day), and UI (MD= 0.416/day). However, the starting dose of solifenacin 5mg was significantly less effective than solifenacin 10mg according to micturition (MD= –0.343/day), and urgency (MD= –0.391/day). No other significant efficacy differences were reported. Solifenacin 5mg had a significantly lower risk of dry mouth compared to oxybutynin (intermediate-release) IR 9mg (odds ratio [OR]= 2.732) or 10mg (OR= 3.534), oxybutynin extended-release (ER) 10mg (OR= 1.714), tolterodine IR 4mg (OR= 1.763), and solifenacin 10mg (OR= 2.470). Tolterodine ER 4mg (OR= 0.481) or IR 4mg (OR= 0.440), oxybutynin IR 9mg (OR= 0.429) or ER 10mg (OR= 0.440), and fesoterodine 4mg (OR= 0.463) were associated with a significantly lower risk of constipation than solifenacin 5mg. Solifenacin 10mg was associated with a higher risk of constipation (OR= 1.804) and blurred vision (OR= 1.730) than solifenacin 5mg. Conclusions: This MTC suggests that the 5mg starting dose of solifenacin is more effective than the 4mg starting dose of tolterodine in reducing symptoms of OAB; solifenacin 5mg also has similar or better efficacy than the starting doses of other antimuscarinic agents across the spectrum of symptoms analysed. The lower risk of dry mouth with solifenacin 5mg may deliver improved treatment persistence. PUK2 Comparative Effectiveness Of Automated Versus Continuous Ambulatory Peritoneal Dialysis On Patients With End-Stage Renal Disease In Taiwan Tang C H , Sue Y M , Wu Y T , Huang S Y Taipei Medical University, Taipei, Taiwan . . . . . . . . A467 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Automated peritoneal dialysis (APD) has been increasingly used since payment scheme was expanded to cover APD machine in addition to the continuous ambulatory peritoneal dialysis (CAPD) in the benefit package of National Health Insurance Program in May of 2008. This study aims to compare the health outcome between patients who used APD and CAPD. Methods: The including criteria were patients treated by APD or CAPD identified in National Health Insurance Research Database (NHIRD) during 2001-2010. The excluding criteria were patients who were treated by hemodialysis for more than 3 months before PD, or younger than 18 years old, or received kidney transplant before. The CAPD patients and APD patients were identified and matched according to their propensity score predicted by age, gender, comorbid conditions, Charlson Comorbid Index, medication history, and premiums wages in the year of treatment initiation. There were 2,287 APD and 2,287 CAPD patients entered the final analysis. The Kaplan-Meier curve and the Cox proportional hazard regression were performed to examine the differences in mortality rate, technique failure rate and incident rate of peritonitis between APD and CAPD. Results: There are significant differences in mortality rate (0.76% vs. 0.65% per patient month, p= 0.02), in technique failure rate (1.07% vs. 0.87% per patient month, p< 0.001), and in incidence rate of peritonitis (1.94% vs. 2.14% per patient month, p< 0.01) between APD and CAPD cases. When compared with CAPD, patients treated by APD posed higher risk in death and technique failure but no difference in the incidence of the 1stperitonitis with hazard ratios of 1.18 (CIs: 1.031.34), 1.22 (CIs: 1.09-1.37) and 0.98 (CIs: 0.88-1.10), respectively. Conclusions: The APD patients seemed to have higher mortality rate and technique failure rate than CAPD patients, however, APD patients had lower incident rate of peritonitis than CAPD patients in Taiwan. PUK3 A Study To Assess Disease Progression To Esrd Within A Year In Patients With Advanced Ckd Kumar P , Nair M , Prabhu R , Bairy M Manipal University, Manipal, India . . . . Objectives: Aimed at determining the time period for the progression to ESRD and also to find out the major risk factors for early progression to ESRD. Methods: A retrospective cohort study was conducted in a tertiary care teaching hospital. The data was collected from medical record dept. for the last two years (2012 and 2013) in hospital. Demographic details and clinical parameters of ESRD patients with major risk factors for early progression to ESRD were collected using the descriptive statistics feature of SPSS v20.0. Results: A total of 240 patients were included in the study. The mean age of the population was found to be 54.6±14.2 years. Majority of the population were males (74.83%). Hypertension 67.1%, Diabetes 47.1% and anemia 40.83% were the most prominent risk factors present in the study population. More than half of the population (72.5%) took more than a year to progress to ESRD. Conclusions: The study revealed that males are at a higher risk of ESRD with half of the study cohort consisted of elderly patients (above 50 years of age). The study results concluded that hypertension is the most common risk factor for ESRD followed by diabetes and anemia. Most of the CKD patients took more than one year to progress to ESRD (72.5%). This information help physicians and patients inform decisions regarding preparation for renal replacement therapy in patients with advanced CKD. PUK4 Be Cautious Of Triple Whammy!!! . . PUK6 Analysis Of Budget Impact Of Anemia Correction In Russian Patients With Chronic Kidney Disease Ryazhenov V V , Gorokhova S G I. M. Sechenov First Moscow State Medical University, Moscow, Russia . . . . Objectives: To analyze budget impact of strategies of anemia correction with different stimulators of erythropoiesis in patients with chronic kidney disease (СКD). Methods: Pharmacoeconomic analysis included cost modelling for a new strategy for patients with CKD and anemia that includes the use of continuous erythropoietin receptor activator (CERA) compared with traditional darbepoetin alfa (as described in ARCTOS study). The model included two stages: 1. Estimation of costs in 18 week correction period (phase 1 study), and 2. estimation of costs and effectiveness within 10 week period (phase 2 study). The costs were evaluated in two groups of 100 people, the first with CERA and the second with darbepoetin alfa. According to ARCTOS study, fewer patients treated with continuous erythropoietin receptor activator (CERA) required transfusion procedures compared with darbepoetin alfa (2.5% and 6.8%, respectively). The drugs in the study were to be administered under medical supervision in a day hospital. The time horizon of the study was 28 weeks. Results: The costs of a phase 1 was 53 187.21 in CERA group and 104 528.88 RUB in darbepoetin alfa group. The costs were almost identical in the second phase. The costs in darbepoetin alfa group were 2.7 times higher compared with CERA in respect of blood transfusion and 2 times higher in respect of drug administration. The cost of pharmacotherapy in CERA group was significantly lower than in group of darbepoetin alfa. Total costs in CERA group were 1, 7 times lower than those for darbepoetin alfa. Conclusions: The study demonstrates that administration of CERA is the most economically effective strategy for the treatment of patients with chronic kidney disease in Russia. It is associated with considerably lower costs compared to darbepoetin alfa. PUK7 Budget Impact Evaluation Of Treatment With A Low Protein Diet And Ketoanalogues Of Essential Aminoacids For Predialysis Patients In Russian Federation Serpik V G , Kulikov A I.M. Sechenov First Moscow State Medical University, Moscow, Russia . Zulkifly H , Abdul Wahab M S , Shaharuddin S , Chiau Ming L , Mat Zaid H Universiti Teknologi MARA, Selangor, Malaysia . to estimate associated direct health costs comparing the use of paracalcitol and cinacalcet plus low dose vitamin D over a 5-years horizon and using the Italian National Health System perspective. The model was developed using parameters from literature and assumption discussed with clinicians. National tariffs and costs from literature were used to value drug use, dialytic treatment, hospitalizations and transplant. One-way sensitivity analyses for model inputs were conducted. Costs and effects were discounted at 3% annum. Results: Considering 13.311 candidate subjects for each treatment strategy, results from the model showed a decrease in direct health care costs from 1.782.921.351 Euro to 1.622.357.209 Euro in favour of paricalcitol over 5 years. Particularly, paricalcitol produced an overall saving in drug costs for more than 51 millions Euro while the other direct health costs related to dialysis, hospitalization and transplant were reduced by approximately 109 millions Euro. Conclusions: In light of the high economic burden of end stage renal disease mainly associated with dialysis and transplant the use of paricalcitol for the treatment of IPTS in these patients represents a valid alternative not only from a clinical point of view but also from an economic point of view. . . . Objectives: This study was aimed to identify the occurrence of concomitant prescribing of NSAIDs, ACE Inhibitors (ACEI) and diuretics known as triple whammy received by out-patients at a Malaysian teaching hospital. It also aimed to identify the relationship between prescriptions of triple whammy with specific age. Methods: A retrospective, observational study was performed in a general teaching hospital. The patients’ prescriptions (January-March 2012) from the outpatient pharmacy department that were prescribed with NSAIDs, ACE Inhibitors (ACEI) and diuretics were reviewed and recorded. The association between the prescriptions with age was investigated. Statistical analysis was done using SPSS with significance difference determined by p value of < 0.05. Results: Four hundred and twenty four patients (56.1% male) were included. Four hundred and twenty two patients were taking one or more of NSAIDs, ACEI and diuretic and only 2 patients were taking all three. Majority of our patients (40.1%) received the combination of ACEI and diuretics. Combination of ACEI and diuretics were mainly (21.7%) prescribed to patients above 65 years old (p=0.362). Conclusions: The occurrence of triple whammy at a teaching hospital during the period of data collection is low. This is indeed a good predictor of safe prescribing of drugs among physicians as concomitant use of these three medications may impair renal function especially in the elderly and dehydrated patients. Majority of the patients that were prescribed with combination of diuretics and ACEI are above 65 years old, therefore proper monitoring of their renal function and the hydration status should be performed to reduce the risk of renal insufficiency in the future. . . Objectives: To evaluate budget impact of low protein diet (LPD) and ketoanalogues supply of predialysis patients in Russian Federation from Pharmacoeconomic perspective. Methods: Analytic decision-making model and budget impact analyses (BIA) were performed. Direct costs (ketoanalogues, hemodialysis, peritoneal dialysis, etc.) were considered. Two patients groups were compared in the model: ketoanalogues and LPD was administrated for the first arm in predialysis stage, while second arm started in the model of dialysis patients. Results of retrospective efficacy analysis state, that administration of keto-analogues and LPD delays start of dialysis at least for one year. The following prices were used: rates of State medical insurance fund, Sechenov First Moscow State Medical University hospitals price lists, NHS medication prices. Exchange rate: 1EURO = 40 RUB. Results: The first year annual costs were 23186 EURO per patient for the patients group on dialysis and 8620 EURO for the group of ketoanalogues. The results of BIA have shown that administration of ketoanalogues and LPD provides cost-saving of 14562 EURO per patient in the first year compared with dialysis group. Total cost-saving per patient in the group of ketoanalogues and LPD over 5-year period was 12895 EURO compared with the dialysis group. Conclusions: Budget impact assessment has shown, that ketoanalogues and LPD in Russian Federation is a preferable technology and provides cost-saving of 12895 EURO over five years per patient. PUK8 An Economic Model To Investigate The Budget Impact In Spain Of Onabotulinumtoxina To Manage Urinary Incontinence In Patients With Idiopathic Overactive Bladder Ruff L 1, Zeidman R 1, Aracil J 2, Loveman C 3 Group, London, UK, 2Allergan S.A.U., Madrid, Spain, 3Allergan Holdings Ltd., Marlow, UK . . . . Urinary/Kidney Disorders – Cost Studies 1Medaxial PUK5 A Budget Impact Analysis (Bia) Of The Use Of Paricalcitol For The Treatment Of Secondary Hyperparathyroidism (Shpt) In End Stage Renal Disease Patients Objectives: Treatment options for patients for whom urinary incontinence (UI) due to idiopathic overactive bladder (OAB) is inadequately managed by anticholinergic therapy are limited, and can be expensive, invasive and inefficacious. This can lead to a significant economic burden to hospitals and health care systems. OnabotulinumtoxinA may provide an effective and minimally-invasive treatment option for OAB. An economic model was developed to investigate potential cost savings associated with the use of OnabotulinumtoxinA – as an adjunct to bestsupportive care (BSC) – to manage OAB rather than other treatment options in Spain. Methods: A prevalence-based, deterministic budget impact model with a five-year time horizon was developed from the perspective of the Spanish health care Lorenzoni V , Pierotti F , Turchetti G Scuola Superiore Sant’Anna, Pisa, Italy . . . Objectives: Budget impact analysis (BIA) of the use of paricalcitol versus alternative treatment for the management of secondary hyperparathyroidism (SHPT) in end stage renal disease patients. Methods: A Markov model was used to simulate the evolution of end stage renal disease patients trough transplant and death and A468 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 system. The model took into account the cost of drugs, disposables (e. g. incontinence pads) and devices, procedures, monitoring costs, and the cost of managing adverse events. Model input data were derived from the Spanish Ministry of Health, published and unpublished clinical studies, clinical guidelines, and expert opinion. Results: In the Spanish population, an estimated 96,360 individuals were eligible for treatment with OnabotulinumtoxinA for OAB. Compared to a current treatment pattern in which 5% of patients received OnabotulinumtoxinA, increasing OnabotulinumtoxinA usage annually from 10% in year 1 to 30% in year 5 resulted in an estimated cost saving of €26.4 million over five years. This saving represented a 4.7% decrease in overall spending compared to the current treatment pattern. Increased drug acquisition costs were entirely offset by savings due to decreased use of incontinence pads and anticholinergic therapy. Conclusions: Costs associated with inadequate management of UI due to OAB are significant and avoidable. Our model estimated that managing patients with OnabotulinumtoxinA may reduce the economic burden to the Spanish health care system, with increased acquisition costs of OnabotulinumtoxinA completely offset by savings due to decreased resource use. PUK10 The Impact Of Cardiovascular Disease And Type 2 Diabetes Mellitus On Social Cost In Chronic Kidney Disease Patients In Italy Bellelli S , Turchetti G Scuola Superiore Sant’Anna, Pisa, Italy . . Objectives: Chronic kidney disease (CKD) is leading condition of several comorbidities with additional social economic burden. The study aims to estimate the economic impact of cardiovascular disease (CVD) and type 2 diabetes mellitus (T2DM) on the social cost of a patient with CKD (stage IV and V pre-dialyses) in Italy. Methods: All adult outpatients in charge of 14 main Hospitals Centers in Tuscany Region have been enrolled during 7 week in the cross sectional study. Direct medical costs have been estimated using tariff for laboratory test, diagnostic exams, visits and hospitalization and price for drugs. Non medical costs included the cost of diet, patients and caregivers travel expenses, domestic help and informal care. The loss of productivity of patients and caregivers have been estimated as indirect costs using the human capital approach. The incremental effects of having comorbidities on social cost of CKD were estimated by multivariate Generalized Linear Models (log link, Gamma family) adjusting for gender, age and stage of CKD. Costs are expressed in Euro 2012. Results: Among 484 CKD patients enrolled, CVD and T2DM have been found respectively in 214 (44%) and 171 (35%) patients. The raw estimated mean annual social costs were €11,375 (± €7,480) per patient with CKD-CVD and €11,627 (± €7,657) per patient with CKD-T2DM. Direct non medical costs and indirect costs accounted respectively for 31% and 23% of social cost for CKD-CVD and 30% and 22% for CKD-T2DM. The incremental effects of having comorbidities on the overall social cost of CKD were € 2,928 (95% CI: € 1,680-€4,176, p=0.000) for CVD and € 2,640 (95% CI: € 1,301-€3,979, p=0.000) for T2DM. Conclusions: CVD and T2DM significantly increase the social cost of CKD due to the rise of the medical component. However the burden of the disease is mainly sustained by patients, their families and the productivity system. PUK11 The Economic Cost Of Urinary Tract Infections In The Community: Results From Ireland Callan A , O’Shea E , Galvin S , Duane S , Corry O , SIMPle Team T , Vellinga A NUI Galway, Galway, Ireland . . . . . . . Objectives: Urinary tract infections (UTIs) are the second most common bacterial infection in primary care and are often treated empirically with antibiotics. However, outside of the United States, there has been a lack of systematic cost-of-illness studies to assess the economic cost of UTIs in the community. The objective of this study was to estimate the economic cost of UTIs in Ireland from the perspective of the health service. Methods: We conducted a secondary analysis of a dataset which contained information on adult patients (excluding pregnant women) presenting to their General Practitioner (GP) with a suspected UTI. Data was collected from 22 GP practices in the West of Ireland over a 9 month period from September 2009 to May 2010. Results from microbiological analysis to determine a positive urine culture were obtained from the University Hospital Galway laboratory. Health care costs were estimated on the basis of GP consultation costs, antibiotic costs, laboratory costs and secondary inpatient and outpatient costs. The latter was acquired from national clinical data on discharges in acute hospitals. Results: Of the 2,850 GP consultations, almost 70% of the urine samples showed no laboratory evidence of a UTI. 57% of the patients were prescribed an antibiotic and 11% of patients re-consulted within one month. Total primary care costs (including antibiotics) to treat suspected UTIs across the 22 practices was approximately €343,000 annually. Extrapolating results to a national level, the annual estimated primary care costs are approximately € 19.2 million. The annual cost of UTIs in secondary care is low at approximately €155,000. Conclusions: Suspected UTIs are a common reason to attend primary care services in Ireland. From the health service perspective, the overall economic cost of UTIs is mainly driven by primary care costs with comparatively low secondary care costs. PUK12 A Review Of Cost Of Illness Studies In Patients With End Stage Renal Disease bases (MEDLINE, EMBASE, SCIENCE CITATION INDEX, NHSEED, HEED, CEA REGISTRY, RePEc, HTA database) for manuscripts published between January 2003 and November 2013. Results were assessed for relevance by two reviewers. For eligible studies, data extracted included objectives, costing methodology, sources of data, disaggregated and aggregated costs, and conclusions. Results: Of 2094 de-duplicated references identified from the combined search, 57 manuscripts remained after review of titles/ abstracts and 40 after full-text review. The majority (n=35) of studies used retrospective data; the most common study country was the US (n=16); most studies (n=36) reported direct costs only. There were four national level costing studies. Societal costs for ESRD in Canada totalled $CAN 1.857 billion (2000); cost of RRT in England was £780 million (2009–2010); total public budgetary impact of Spain’s RRT programme was €1,829 million (including indirect costs) in 2010; health costs of ESRD patients with diabetes was $3,611 million in the US and £184.5 million in the UK in 2001. Annual perpatient ESRD health care costs ranged from $US 36,917 (PPP, 2004) in Australia, to $US 96,014 (2002) in US patients with diabetes. Studies also reported national and patient level costs for transplant and dialysis. Conclusions: A number of COI studies have quantified the cost of ESRD and demonstrate the substantial economic burden associated with the management of these patients. PUK13 Cost-Utility And Value Of Information (Voi) Analyses On The Feasibility Of A Future Randomised Controlled Trial (Rct) Of Invasive Urodynamic Testing Prior To Surgery For Stress Urinary Incontinence In Women Homer T 1, Vale L 1, Shen J 1, Hilton P 2 University, Newcastle, UK, 2Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle, UK . . . . 1Newcastle Objectives: To assess the feasibility and VOI of conducting a definitive RCT to determine whether invasive urodynamic testing (IUT) is cost-effective compared with clinical assessment in women prior to surgery for stress (SUI) or stress predominant mixed urinary incontinence (MUI) and to rehearse the economic evaluation for a definitive RCT. Methods: Cost-utility analysis was performed alongside a pilot RCT. 222 participants were randomised to receive IUT (‘IUT’ arm) or clinical assessment (‘no IUT’ arm) before surgery. Health service resource use, costs, utility values (from EQ-5D-3L and SF-12) and quality-adjusted life years (QALYs) were calculated. Expected value of sampling information (EVSI) analysis was used to determine the expected net gain (ENG) of additional information and the optimal sample size needed to maximise ENG in a future trial. Results: At 6 months the average cost incurred in the ‘IUT’ arm was £154 less than the ‘no IUT’ arm (95% bootstrapped CI -315 to 24) and there was no evidence of a difference in effects (-0.019 QALYs, 95% bootstrapped CI -0.0258 to 0.0133). ‘IUT’ generated an incremental cost per QALY of £8090. A stochastic analysis showed that at a zero-willingness to pay threshold the IUT was 96% likely to be cost-effective. The VOI analysis suggested there would be added value from additional research to confirm which treatment is more efficient. The ENG was estimated to be £91m and to maximise ENG a sample size of 105 complete cases in each treatment arm is required. Conclusions: There was no significant difference in QALYs between study arms, the ‘no IUT’ arm incurred a higher average cost but only short-term NHS costs were considered. Our results should be interpreted with caution due to the limitations within our analysis. The EVSI analysis confirmed that there is additional value to be gained from a definitive study in this area. PUK14 Cic Users’ Preference For Catheters Reducing The Uti Frequency Neovius K E , Lundqvist T Wellspect HealthCare, Göteborg, Sweden . . . Objectives: To investigate the preference among individuals practicing clean intermittent catheterization (CIC) on a daily basis for urinary catheters that can reduce the frequency of urinary tract infections (UTIs). Methods: A questionnaire was sent by e-mail to 769 catheter users from Germany, Italy, Sweden, the UK and the USA through a database held by Wellspect HealthCare. The participants were asked to assume a situation in which they use their current catheter but have the choice to switch to a similar catheter, which hypothetically would reduce the frequency of UTIs including possible complications. Either every fourth or every second UTI could be avoided. The participants’ willingness-to-pay for the new catheter was collected by letting them choose to spend either one of eleven explicit monthly amounts from € 0-€ 100 or any other amount in an open answer. The participants also reported their UTI frequency. Only individuals stating that they were “certain” or “very certain” regarding their answers were included in the analyses. Results: 429 (response rate 56%) individuals returned the questionnaire, of which 278 (65%) were certain or very certain regarding their answers. The proportion of respondents with ≤ 1 UTI/y and > 1 UTI/y were 47% and 53%, respectively. The respondents were willing to spend on average € 16 each month to avoid every fourth UTI and € 22 each month to avoid every second UTI. The willingness-to-pay was higher among the users with > 1 UTI/y both to avoid every fourth and every second UTI (€ 21 vs € 11 and € 30 vs € 15, respectively). These differences in willingness-to-pay were statistically significant (p< 0.05).Conclusions: The CIC users in this study, who practice intermittent catheterization on a daily basis, expressed a clear preference for catheters that could help to decrease the frequency of urinary tract infections (UTI). Paczkowski R 1, Norrbacka K 2, Van Brunt K 3, Kennedy-Martin T 4 1Eli Lilly and Company, Inc., Indianapolis, IN, USA, 2Eli Lilly Finland, Helsinki, Finland, 3Eli Lilly and Company, UK, Windlesham, Surrey, UK, 4Kennedy Martin Health Outcomes, East Sussex, UK PUK15 Cost Consequence Analysis Of Darbepoetin Alfa For The Treatment Of Anemia Due To Chronic Kidney Disease (Ckd) In Greece Objectives: End-stage renal disease (ESRD) is a debilitating medical condition of chronic kidney failure. In ESRD, the kidneys are permanently impaired and patients require renal replacement therapy (RRT). In order to understand the economic burden of ESRD, this scoping literature review was undertaken to identify published cost of illness (COI) studies in ESRD. Methods: The search strategy identified studies quantifying the economic burden of ESRD in selected countries (USA, Canada, Australia, Japan, UK, France, Germany, Spain, and Italy). We searched electronic data- Kourlaba G 1, Boletis I 2, Goumenos D 3, Iatrou C 4, Papagiannopoulou V 5, Tritaki G 5, Maniadakis N 6 1Collaborative Center for Clinical Epidemiology and Outcomes Research (CLEO), Athens, Greece, 2Medical School of Athens, National and Kapodistrian University of Athens School of Medicine, Athens, Greece, 3Nephrology and Renal Transplantation Department, Patras University Hospital, Patras, Greece, 4General Hospital of Nikaia, Athens, Greece, 5AMGEN Hellas, Marousi, Greece, 6National School of Public Health, Athens, Greece . . . . . . . . . . . A469 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Conduct an economic evaluation of darbepoetin alfa for the treatment of anaemia due to chronic kidney disease (CKD) compared to other erythropoiesis stimulating agents (ESAs) in patients on haemodialysis (HD) or peritoneal dialysis (PD) in Greece. Methods: A decision tree model was developed to compare overall costs per patient in control of darbepoetin alfa to other ESAs under different haemoglobin (Hb) concentrations and CKD patients. It estimates the economic impact of treating CKD patients on dialysis with darbepoetin alfa, originator and biosimilar short-acting ESAs and pegylated epoetin beta to different Hb target strategies, 10 (±1) g/dL and 11 (±1) g/dL, over 1 year. The effectiveness of “% patients under anaemia control” was defined as the number of patients being alive, not hospitalized, nor transfused during the analysis period. Clinical inputs depend on Hb target strategy and not on ESAs as extracted from published studies. Cost inputs are drug acquisition, administration and clinical event costs, the last comprising hospitalisation and blood transfusion costs (2014). Analysis was conducted from public third-party-payer perspective. Results: For both Hb target strategies in patients on HD or PD, darbepoetin alfa demonstrated the lowest overall costs per patient in control followed by short-acting originator ESAs, whereas pegylated epoetin beta and short-acting biosimilar ESAs showed the highest overall costs per patient in control. As clinical event costs are identical across all ESAs, the overall costs per patient in control are mainly affected by drug acquisition costs and by ESAs doses. Treatment with darbepoetin alfa at Hb target 11 (±1) g/dL is associated with the lowest overall costs per patient in control at € 8,210 and € 6,689, for patients on HD or PD, respectively. Conclusions: Darbepoetin alfa may be the most cost saving treatment compared to other ESAs for the management of anaemia (CKD) in patients on HD or PD in Greece. PUK16 Cost Effectiveness Of Solifenacin Compared With Oral Antimuscarinic Agents For The Treatment Of Patients With Overactive Bladder (Oab) In The Uk Kelleher C 1, Aballea S 2, Maman K 3, Nazir J 4, Hakimi Z 5, Chambers C 4, Odeyemi I A 4 and St. Thomas’ NHS Foundation Trust, London, UK, 2Creativ-Ceutical, Paris, France, 3Creativ Ceutical, Paris, France, 4Astellas Pharma Europe Ltd, Chertsey, UK, 5Astellas Pharma Global Development, Leiden, The Netherlands . . . . . . . . 1Guy’s Objectives: To evaluate the cost effectiveness of solifenacin 5mg compared with other oral antimuscarinic agents in adults with OAB from a UK NHS payer perspective. Methods: A Markov model was developed to model major symptoms of OAB and the efficacy and tolerability of antimuscarinics. A 5-year time-horizon with monthly cycles was used in the base case. Five levels of symptom severity were used for micturitions and incontinence (total: 25 health states). Treatment with at least two antimuscarinic agents was assumed before patients received botulinum toxin. For each antimuscarinic, there was a probability of continuing or discontinuing medication, or switching to an alternative antimuscarinic with or without adverse events. Efficacy and tolerability inputs were based on a phase 3 study of solifenacin (Chapple et al. BJU Int 2004; 93: 303–10) and a mixed treatment comparison (results reported separately). Other model inputs were determined from the literature or expert opinion. All costs were based on GBP 2012/2013. Utilities were based on EuroQol-5D. Incremental cost-effectiveness ratios (ICER) were expressed as cost/quality-adjusted life year (QALY). Deterministic (DSA) and probabilistic (PSA) sensitivity analyses were performed. Results: Solifenacin 5mg was dominant versus tolterodine extended-release 4mg, darifenacin 7.5mg, fesoterodine 4mg and solifenacin 10mg, and cost-effective versus darifenacin 15mg (ICER= £982.89/ QALY), fesoterodine 8mg (ICER= £232.53/QALY), oxybutynin extended-release 10mg (ICER=£1117.32/QALY) and trospium 60mg (ICER=£3674.25/QALY). Both DSA and PSA showed that results were robust at a willingness-to-pay threshold of £20,000/QALY. At a cost-effectiveness threshold of £20,000/QALY, the probabilities that solifenacin 5mg was cost-effective versus comparators were: 95.4% (tolterodine extendedrelease 4mg), 92.1% (solifenacin 10mg), 97.3% (fesoterodine 4mg), 99.6% (fesoterodine 8mg), 99.4% (oxybutynin extended-release 10mg), 99.4% (darifenacin 7.5mg), 99.2% (darifenacin 15mg), and 97.3% (trospium 60mg). Conclusions: Solifenacin 5mg appears to be a cost-effective strategy compared with other oral antimuscarinic agents for the treatment of adults with OAB from a UK NHS perspective. PUK17 Cost-Effectiveness Of Mirabegron 50mg Compared To Tolterodine Er 4mg In The Treatment Of Patients With Overactive Bladder In Canada Herschorn S 1, Vicente C 2, Nazir J 3, Ramos B 4, Hakimi Z 5 1University of Toronto, Toronto, ON, Canada, 2PIVINA Consulting Inc., Mississauga, ON, Canada, 3Astellas Pharma Europe Ltd, Chertsey, UK, 4Astellas Pharma Canada, Inc, Markham, ON, Canada, 5Astellas Pharma Global Development, Leiden, The Netherlands . . . . . Objectives: Mirabegron, a β 3-adrenoceptor agonist approved for the treatment of overactive bladder (OAB), has been shown to reduce OAB symptoms and improve HRQoL relative to placebo and to cause a lower incidence of dry mouth than tolterodine. We investigated the cost-effectiveness of mirabegron 50mg compared to tolterodine ER 4mg. Methods: A Markov model was developed to simulate management, symptom changes (micturitions, incontinence), and adverse events in OAB patients. The model predicted costs and QALYs over 1 year in a hypothetical cohort of Canadian patients previously treated for OAB. The analysis compared treatment with mirabegron 50mg to tolterodine ER 4mg, administered in a reference arm in the pivotal study SCORPIO and one of the antimuscarinic options most widely prescribed and funded in Canada for the treatment of OAB, from both a payer (Ontario Ministry of Health and Long-term Care [MOH]) and societal perspective. Clinical and health state utility data for the model were based on the previously treated OAB population of SCORPIO. Transition probabilities between symptom severities were obtained using multinomial logistic regression models estimated from SCORPIO. Probabilities of discontinuation and switching therapy were derived from the literature. Resource utilisation estimates were based on clinical trial data, the literature and expert opinion. Unit costs were derived from official schedules for Ontario and published Canadian literature. Deterministic and probabilistic sensitivity analyses were performed. Results: Mirabegron 50mg dominated tolterodine ER 4mg: incremental cost savings were CA$118.91 (MOH perspective) and CA$140.48 (societal); and 0.005 QALYs (MoH and societal) were gained. ICERs were robust over a wide range of sensitivity analyses, but were most sensitive to micturition symptom levels for tolterodine and to costs of subsequent therapy. Conclusions: Treatment with mirabegron 50mg is cost-effective compared with tolterodine ER 4mg in a population previously treated for OAB from Canadian health care and societal perspectives. PUK18 Economic Evaluation Of Percent Free Psa For Prostate Cancer Detection In Taiwanese Men LAI A Y H ESSEC Business School Asia Pacific, Singapore . . . Objectives: The prevalence of prostate cancer in the Chinese population with abnormal prostate-specific antigen (PSA) levels of 4.0-10.0 ng/ml is low. The diagnosis can only rely on transrectal ultrasound-guided prostate biopsies (TRUS-Bx) which may cause major complications. We assess the cost-effectiveness of two diagnostic strategies for prostate cancer detection in Chinese men with abnormal PSA levels of 4.0-10.0 ng/ml and normal digital rectal examination of prostate (DRE). Methods: Using a decision tree model, we performed a cost-effectiveness analysis to compare the two strategies: 1) direct prostate biopsies (TRUS-Bx) and 2) percent free PSA testing prior to TRUS-Bx. A systematic review of 855 patients with PSA levels of 4.0-10.0 ng/ml and normal DRE enrolled in a single medical institute from Jan 2002 to December 2005 was conducted. The outcome measures were the incremental cost-effectiveness ratio, and costs were calculated through activity-accumulation costing based on National Insurance Scheme Bill Size. A one-way sensitivity analysis was undertaken. The effectiveness was measured by means of the number of detected cases and actual cases (detected cases minus lost cases). A threshold analysis is used to illustrate the value of a given variable of which the two strategies have equal outcomes or costs. Results: The strategy of percent free PSA with TRUS-Bx was dominant and found to be the most cost-effective. The incremental cost-effectiveness ratio for free PSA + TRUS-Bx compared with TRUS-Bx was USD 3,871.58. Strategy 2 (TRUS-Bx) would be more cost-effective if the cost of percent free PSA increased to USD 36.78 or if prostate cancer prevalence increased to 42%. Conclusions: The use of percent free PSA prior to TRUS-Bx is the most cost-effective diagnostic strategy and will become more cost-effective as prostate cancer prevalence increases in the ageing population and the free PSA test costs down. PUK19 Cost Effectiveness Of Extracorporeal Shock Wave Lithotripsy Against Ureteroscopic Laser Lithotripsy For Treatment Of Ureteral Calculi Caballer V 1, Vivas D 1, Reyes F 2, Budia A 3 Politecnica de Valencia, Valencia, Spain, 2Universidad de Vigo, Vigo, Spain, 3Hospital La Fe Valencia, Valencia, Spain . . . . 1Universitat Objectives: To evaluate the cost-effectiveness of extracorporeal lithotripsy against ureteroscopic laser in the treatment of ureteral stones in the La Fe Hospital in Valencia. Methods: A decision tree was performed to evaluate the costs and effectiveness of two treatment strategies (starting with ESWL and second line URS, versus directly start with URS). It was considered as a parameter of effectiveness, the ratio of patient with stone-free status or the presence of insignificant residual fragments (< 3 mm) at 3 month of follow up. Subsequently, to consolidate the results of the decision tree, a sensitivity analysis of Monte Carlo with 1000 iterations was developed. Were available from a database of 162 patients treated for ureteral stones at the Hospital La Fe de Valencia. Of these, 77 were treated in lithotripsy sessions (27 % of whom had upper stones 1 cm), while 85 were referred directly to URS (the percentage of these patients with upper stones centimeter was 32.4 %). The unit cost of each treatment was calculated from data provided by the economic management Department of Hospital La Fe and were 286.06 for each session in the case of ESWL and 1409.89 for URS. Results: The overall effectiveness after four sessions of lithotripsy plus second line URS was 99,76%, while in the case of URS was 98.81%. The average cost of ESWL plus URS alternative was 881.59 euros, while in the case of the strategy of starting with URS was 1,496 euros. Therefore, the ESWL plus URS showed dominant versus first line URS. For both group upper and less than 1 cm also ESWL plus URS was dominant. Conclusions: Thus, the combination of first line lithotripsy plus second line ureteroscopic laser is better than first line ureteroscopic laser for treatment of ureteral calculi in terms of in terms of cost effectiveness. PUK20 Cost-Effectiveness Of Mycophelonate Sodium On Patients With Renal Transplant At The Public Mexican Health Care System Ruiz Miranda C I , Ubiarco Lopez V Novartis Mexico, Mexico, Mexico . . . Objectives: To perform a cost-effectiveness analysis of Mycophenolate Sodium (MPS) vs Mycophenolate Mofetil (MMF) with empirical treatment of rejection, both options available at IMSS*. Methods: A Decision Model was designed to analyze effectiveness and treatment cost of MPS and MMF for renal transplant patients. The outcome was renal graft survival related to gastrointestinal tolerance in the first year. The model starts with a renal transplant performed, in a period of three months the patient may experience gastric intolerance or tolerance. For the patients showing gastric intolerance, there are two alternatives: switching to azathioprine or reducing the dose of mycophenolates (MPS or MMF). Then, the patient moves to the next state which may be rejection (humoral or cellular) or not. The absorbent state is the graft losses which include the cost of the intervention of renal transplant plus the cost of replacement therapy: Ambulatory Peritoneal Dialysis (APD), Intermittent Peritoneal Mechanical Dialysis (IPMED), Intermittent Peritoneal Manual Dialysis (IPMAD) and Hemodialysis. A probabilistic sensitivity analysis (PSA) was A470 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 performed using Monte Carlo technique. Results: Annual patient costs of MMF were: $7,746.75, $7,993.03, $7,694.19 and $7,939.14 USD. For MPS were: $7,673.35, $7,989.92, $7,605.80 and $7,920.64 USD with an incremental efficacy of 0.07 less graft rejection in APD, IPMED, IPMAD and hemodialysis respectively in one year horizon. PSA shows consistency on model results. Conclusions: MPS was a dominant alternative having lower costs and more effectiveness than MMF. These results show possibilities to achieve cost-savings and a potential clinical benefit in renal transplants, from the perspective of the Mexican public health system, in specific from IMSS. *IMSS (Mexican Institute of Social Security) PUK21 A Cost-Effectiveness Analysis Of Onabotulinumtoxina VersUS Best Supportive Care (bsc) For The Treatment Of Anticholinergic Treatment-Refractory Neurogenic Detrusor Overactivity (ndo) Hamid R 1, Loveman C 2, Millen J 2, Colayco D 3, Stanisic S 4, Gultyaev D5 1Department of NeuroUrology, Royal National Orthopaedic Hospital & University College Hospital, London, UK, Stanmore, UK, 2Allergan Holdings Ltd., Marlow, UK, 3Allergan, Inc., Irvine, CA, USA, 4The LASER Group, Milano, Italy, 5The LASER Group, Loerrach, Germany . . . . . Objectives: Uncontrolled NDO may lead to medical sequelae, such as upper urinary tract complications and renal failure. Treatment choices include BSC (comprised of behavioural therapy and pads, alone or in combination with clean intermittent catheterisation, and possibly with anticholinergics), onabotulinumtoxinA, and surgery. The study’s objective was to determine the cost-effectiveness of onabotulinumtoxinA 200 U vs. BSC among patients inadequately managed with anticholinergics in a UK setting. Methods: A Markov model was developed to compare onabotulinumtoxinA + BSC to BSC alone, with surgery as a downstream option. Efficacy and safety inputs were based on Phase 3 trials. Utility values were derived from a UK preference elicitation study. Costs were obtained from various NHS sources. Model uncertainty was examined through deterministic and probabilistic sensitivity analyses. Results: The base case incremental cost-effectiveness ratio (ICER) was £3,856, with an incremental cost of £1,692 and incremental benefits of 0.4387 quality-adjusted life-years (QALYs) for onabotulinumtoxinA + BSC compared with BSC alone over 5 years. A lifetime horizon yielded an ICER of £2,739 per QALY. Univariate sensitivity analyses indicated that the main cost drivers are mean monthly use of catheters and treatment administration costs. Probabilistic sensitivity analysis suggested there would be 100% probability of the ICER being ≤ £10,000. Conclusions: Our analysis suggests that onabotulinumtoxinA + BSC is a cost-effective treatment option, compared with BSC alone for patients with NDO who are inadequately managed with anticholinergics in the UK. PUK22 Cost-Effectiveness Comparison Of Botulinum Toxin Type A Plus Best Supportive Care Versus Best Supportive Care Alone In The Treatment Of Idiopathic Overactive Bladder With Urinary Incontinence Among Patients Not Adequately Managed By Anticholinergic Therapy In France Ruffion A 1, Velard M 2, Loveman C 3, Khalaf K 4, Roze S 5, Pignata M 5, Stanisic S 6, Lister J 7 Hospitalier Lyon Sud, Pierre Bénite, France, 2Allergan France SAS, Courbevoie, France, 3Allergan Holdings Ltd., Marlow, UK, 4Allergan, Inc, Irvine, CA, USA, 5HEVA HEOR, Lyon, France, 6The LASER Group, Milano, Italy, 7Laser Analytica, Loerrach, Germany . . . . . . . . 1Centre Objectives: To assess the cost-effectiveness of botulinum toxin type A (BTXA; BOTOX®) 100 U in the treatment of idiopathic overactive bladder (OAB) with urinary incontinence (UI) among patients inadequately managed by anticholinergic therapy in France. Methods: A 10 year Markov model divided into 3-month cycles was developed to predict the long-term costs and health outcomes of BTXA + best supportive care (BSC; comprising behavioural therapy, incontinence pads, continued anticholinergic therapy for some patients and, occasionally, catheters) versus BSC alone from a societal perspective (excluding productivity loss) in France. Health states were determined by daily number of UI episodes. Patients discontinuing BTXA and a proportion of patients receiving BSC alone were eligible to receive downstream sacral nerve stimulation (SNS). Costs and health outcomes were discounted at 4% annually. The modelled cohort comprised patients from two phase 3 clinical trials of BTXA and a long-term extension study. Literature, published guidelines and expert advice informed all other model assumptions. Incontinence Quality of Life (I-QOL) data from the trials were mapped to 5-dimension EuroQol questionnaire (EQ-5D) utility values using a published algorithm. Sensitivity analyses assessed the impact of varying model parameters as well as providing a direct comparison between BTXA + BSC and SNS. Results: BTXA + BSC was economically dominant compared with BSC alone in the base case (quality-adjusted life-year [QALY] gain: +0.198; cost difference: –€1937). BTXA + BSC was also economically dominant when compared directly with SNS (QALY gain: +0.143; cost difference: –€ 8973). Probabilistic sensitivity analysis indicated that the incremental cost-effectiveness ratio has approximately a 90% likelihood of being below €20 000 per QALY gained. Conclusions: In France, BTXA + BSC is economically dominant over BSC alone for patients with OAB, symptoms of UI and an inadequate response to anticholinergic therapy. PUK23 Early Versus Late Ketoanalogs Supplementation In Patients With Chronic Kidney Disease In Taiwan – A Cost-Effectiveness Analysis You J 1, Ming W 1, Lin W 2, Tarn Y H 2 1The Chinese University of Hong Kong, Shatin, Hong Kong, 2Taiwan Pharmacist Association, Taipei, Taiwan . . . . . Objectives: In Taiwan, chronic kidney disease (CKD) patients with estimated glomerular filtration rate (eGFR) < 15 mL/min/1.73m2 are suggested to be managed with low-protein diet (LPD) (≤0.6 g/kg/day) plus ketoanalogs (KA) supplement. Recent clinical findings showed early KA initiation with LPD at eGFR 15-29 mL/min/1.73m2 would significantly slow down eGFR decline. We compared cost-effectiveness of KA initiation at eGFR 15-29 mL/min/1.73m2 versus eGFR <15 mL/min/1.73m2 in CKD patients on LPD from Taiwan health care payer’s perspective. Methods: A Markov was designed to simulate outcomes of two options in a hypothetical cohort of adult CKD patients with eGFR 15-29 mL/min/1.73m2: (1) Initiation of LPD plus KA, and (2) watchful-waiting on LPD and initiation of KA at eGFR < 15 mL/min/1.73m2. The Markov states included CKD stage 4 and 5, hemodialysis, and death. Total direct medical cost and qualityadjusted life-years (QALYs) gained were calculated over a maximum period of 10 years. Model inputs were derived from literature. Sensitivity analyses evaluated the impact of uncertainty in all model variables. Results: In base-case analysis, early KA initiation group (3.926 QALYs and USD548,191) gained higher QALYs and cost less than the watchful-waiting group (3.787 QALYs and USD887,608) (USD1=NTD30). Sensitivity analysis indicated that early KA initiation at eGFR at 17-29 mL/min/1.73m2 would be the preferred cost-effective option if reduction of eGFR decline associated with LPD plus KA was 4% or above. When KA was initiated at eGFR 15-17 mL/min/1.73m2, it would remain cost-effective if the reduction of eGFR decline associated with LPD plus KA was 13.5% or above. 10,000 Monte Carlo simulations showed early KA initiation group to be less costly with higher QALY gained than watchful-waiting group by USD333,655 (95% CI 332,174-335,137) and 0.160 (95% CI 0.140-0.180) QALYs, respectively. Conclusions: KA Initiation with LPD in CKD patients as early as eGFR 15-29 mL/min/1.73m2 seems to be cost-effective in Taiwan. PUK24 COST- MINIMIZATION ANALYSIS OF THE DIRECT COSTS OF SEVELAMER CARBONATE AND LANTHANUM CARBONATE IN THE TREATMENT OF CKD-ND PATIENTS Petrov M K 1, Dimitrova M 2, Petrova G I 3 Bulgaria, Sofia, Bulgaria, 2Medical University of Sofia, Faculty of Pharmacy, Sofia, Bulgaria, 3Medical University, Faculty of Pharmacy, Sofia, Bulgaria . . . . . 1Sanofi Objectives: Hyperphoshatemia or elevated phosphorus in the blood is prevalent in patients with chronic kidney disease - mineral and bone disorder (CKD-MBD) and independently and significantly contributes to morbidity and mortality. The objective of this study is to perform cost - minimization analysis of the newly available medicines sevelamer carbonate (SC) and lanthanum carbonate (LC), for the treatment of hyperphosphatemia in CKD patients not on dialysis (CKD-ND) in Bulgaria. Methods: The results of the head-to-head clinical trial conducted by Spraque (2009) demonstrated equivalent efficacy and safety profiles between the two treatment options. To differentiate the cost in high dose and low dose therapeutic regimes was performed a cost-minimization analysis. Based on that was forecasted the expected cost savings for four years period. Discounting rate of 3.5% was applied. The robustness of the Results was tested through sensitivity analysis (SA) using Tornado diagram. Results: The estimated treatment cost per patient/ per year with SC and LC was 1441,75€ and 1569,50€ respectively at the low dose regimen (4000 mg of SC vs. 2000 mg of LC), while within the high dose regimen (6400 mg of SC vs. 3000mg of LC) it was 2306,80€ and 2354,25€ respectively. Expected cost savings (discounted) for the four years period within the assumed market shares was between 1 348 794€ and 2 696 431€ at the low dose regimen, while at the high dose regimen the estimated cost savings was between 501 593€ and 1 001 532€ respectively. The results of SA (discounted) show that the major cost drivers in the treatment of hyperphosphatemia were the unit costs of SC and LC. Conclusions: The equal efficacy and lower cost of sevelamer carbonate than lanthanum carbonate when used for treatment of hyperphosphatemia in patients with CKD -ND in Bulgaria should made the sevelamer carbonate a preference alternative. PUK25 A Spanish Cost-Effectiveness Analysis Of Sevelamer Versus Calcium Carbonate In Nondialysis-Dependent Chronic Kidney Disease (Ckd) Patients Subirà R 1, Rubio M 1, Rodríguez-Carmona A 2, Pons R 3, del Pino M D 4 1Sanofi, Barcelona, Spain, 2Hospital Juan Canalejo, A Coruña, Spain, 3Servicio del Hospital de Castellón, Castellon de la Plana, Spain, 4Hospital Torrecardenas de Almería, Almería, Spain . . . . . . Objectives: In a 36-month, open label RCT that involved 213 patients in stage 3-4 nondialysis-dependent CKD (NDD-CKD) (INDEPENDENT study), sevelamer showed lower rates of all-cause mortality and dialysis inception vs. calcium carbonate. The aim of this study was to assess the cost-effectiveness of sevelamer vs. calcium carbonate in NDD-CKD patients with hyperphosphatemia in Spain. Methods: A Spanish National Health System perspective and lifetime horizon was chosen for the analysis. A Markov model was developed considering health states of “alive with NDD-CKD”, “alive with dialysis-dependent CKD”, and “dead”). All-cause mortality, dialysis inception, hospitalization (frequency and length of stay [LOS]), and drug dosage data were taken from the INDEPENDENT study. All-cause mortality and dialysis inception were extrapolated beyond 36 months using Weibull regression analysis. Local costs (euros, 2014) were applied to pharmaceutical, hospitalization and dialysis utilization. Health utility data was taken from the published literature. Costs and effects were discounted at a rate of 3%. Results: In the base case analysis sevelamer was associated with increased survival, delay in dialysis inception, fewer hospitalizations, shorter LOS, 2,12 life years gained (LYG) and 1,61 quality-adjusted life years gained (QALYG) vs. calcium carbonate. Increased survival translated into more treatment time and dialysis sessions vs. calcium carbonate, resulting in an incremental cost of 33.687 €. The incremental cost per LYG for sevelamer vs. calcium carbonate was 15.897 € and the incremental costs per QALY gained was 20.883 €. Sensitivity analysis showed that sevelamer was more effective and less costly (i.e., dominant) vs. calcium carbonate in time horizons < 6 years. Conclusions: The Spanish analysis showed that sevelamer is a cost-effective strategy vs. calcium carbonate for the treatment of hyperphosphatemia in patients with NDD-CKD, with cost-effectiveness ratios well below the accepted thresholds of 30.000-45.000 €/QALY gained. PUK26 Burden On Secondary Care Of Overactive Bladder Patients Who Are Inadequately Managed With Anticholinergics In England Hamid R 1, Loveman C 2, Morton R 3, Millen J 2, Hassan Y 4 . . . . . A471 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1Department of NeuroUrology, Royal National Orthopaedic Hospital & University College Hospital, London, UK, Stanmore, UK, 2Allergan Holdings Ltd., Marlow, UK, 3Allergan Holdings Ltd, Marlow, UK, 4HealthIQ, London, UK Objectives: Anticholinergics drugs (ACHD) are an established first-line pharmacological therapy for overactive bladder (OAB). Despite the fact that many patients cycle through multiple ACHD, there is limited evidence on the relationship between the burden on secondary (hospital) care and these patients whose OAB is inadequately managed with ACHD. The objective of this study was to analyse the variation in health care burden in secondary care of patients who cycle through multiple ACHD. Methods: A retrospective observational study was conducted to examine the relationship between health care burden and number of prior ACHD with each ACHD switch. Data was extracted from the Hospital Episode Statistics (HES) database and 3,059 GP practices for patients with a diagnosis for OAB, or any of the symptoms of OAB (frequency, urgency, incontinence or nocturia) and with at least one prescription for an ACHD between April 2007 and March 2013. Treatment activity and cost burden (including burden associated with comorbid conditions) were analysed following initiation of the second and the third ACHD and compared to the same cohort from the point at which they had received the first ACHD. All costs were calculated by applying national health service (NHS) tariff prices to treatment activity. Results: Overall, the number of patients identified was 13,117. Our analysis showed that there were increases of 70%, 40% and 10% in inpatient, outpatient and emergency settings respectively from the initial ACHD prescription to 3+ ACHD. This led to an increase in overall health care costs of 30% from the first to 3+ ACHD or £197 (£642 to £839) per patient over the investigation period. Conclusions: OAB patients who are inadequately managed with ACHD place an increased burden on hospital resources. These findings emphasize the importance of identifying alternative ways to treat these patients to address the cumulative burden they place on health care systems. Urinary/Kidney Disorders – Patient-Reported Outcomes & Patient Preference Studies PUK27 Accept® Questionnaire: Relation Between Acceptance And Compliance In Liver- And Kidney-Transplanted Patients Converted To Once-Daily Tacrolimus Bourhis Y 1, Chretin S 1, Cantarovich D 2, Gilet H 1, Bugnard F 1, Arnould B 1 NANTES, Nantes Cedex 1, France . . . . . . 1Mapi, Lyon, France, 2CHU Objectives: ACCEPT® is a 32-items self-administered questionnaire recently validated to measure patient acceptance to treatment. Acceptance was defined as the balance between treatment advantages and disadvantages as rated by the patient and may help predict adherence. The objective was to evaluate the relation between early Acceptance and Compliance in liver- and kidney-transplanted patients converted to once-daily tacrolimus (TAC-OD). Methods: 6-month observational, prospective, longitudinal, multicenter study conducted by 23 hepatologists and 56 nephrologists in France. 1106 adult patients with kidney and/or liver transplant, initiating TAC-OD during post-transplant follow-up were included. Acceptance and compliance were assessed 3 and 6 months after TAC-OD initiation using ACCEPT® and Compliance Evaluation Test questionnaires. Results: Data from 271 liver-, 824 kidney- and 11 liver+kidney-transplanted patients were analyzed. Mean age was 52.4 (±13.2) years. 61.5% of patients were male. Mean time between graft and TAC-OD initiation was 5.0 (±4.9) years. Mean general acceptance score (range: 0-100) at month 3 was 75.4 (±26.5). At month 6, 25.5% of patients had good treatment compliance, 68.0% minor non-compliance and 6.5% were non-compliant. Higher general acceptance score at month 3 was significantly associated with better compliance at month 6 (good compliance: 85.2±20.7, minor non-compliance: 74.5±26.7, non-compliant: 68.0±29.6, p< 0.001). 3-month general acceptance score was particularly low in patients who specified at month 6 having ‘omitted to take their treatment this morning’ (N=4, mean: 41.7±44.1) or ‘ever forgotten to take their treatment because their memory was failing’ (N=53, mean: 69.5±29.7). Although not significant, the 17 patients who discontinued TAC-OD before 6 months had lower 3-month mean general acceptance score (64.6±36.5 vs. 75.7±25.9 in patients still treated with TAC-OD). Conclusions: This study highlighted a strong association between early Acceptance and late Compliance to TAC-OD. Further investigations are needed to explore how early detection of low acceptance can help patient management and improve long-term outcomes. PUK28 Persistence And Adherence With Mirabegron, A New Beta-3 Receptor Agonist, Versus Antimuscarinics In Overactive Bladder: Early Experience In Canada Wagg A 1, Franks B 2, Ramos B 3, Berner T 2 of Alberta, Edmonton, AB, Canada, 2Astellas Scientific and Medical Affairs, Inc., Northbrook, IL, USA, 3Astellas Pharma Canada, Inc, Markham, ON, Canada . . . . 1University Objectives: To compare persistence and adherence with mirabegron versus antimuscarinic treatment for overactive bladder (OAB). Methods: This was an exploratory analysis of retrospective claims data from the largest Private Drug Plan database in Canada. Patients aged ≥18y who had a first prescription claim for a target medication during a four-month index period in 2013 were identified. A six-month look-back was used to categorize patients as ‘treatment-naïve’ (no claims for OAB medication) or ‘treatment-experienced’ (≥ 1 prior medication) during this time. Time to end of persistence (defined by a maximum gap in therapy of 30 days or switching to another medication) and adherence (calculated by medication possession ratio) were analyzed after six months. Hazard ratios (HR) with 95% confidence intervals (CI) were calculated for mirabegron versus each antimuscarinic, using Cox proportional hazards modeling. Results: Data were analyzed for 13,391 patients (mirabegron, n=993). In the treatment-experienced cohort, six-month persistence was highest with mirabegron (51%), followed by solifenacin (39%) (HR 1.383; CI: 1.109−1.726; p=0.004), ranging to 18% with oxybutynin IR (HR 2.488; CI: 1.961−3.157; p<0.001). In the treatment-naïve cohort, persistence was also highest with mirabegron (40%), followed by fesoterodine (24%) (HR 1.512; CI: 1.333−1.716; p< 0.001), ranging to 13% with oxybutynin IR (HR 2.315; CI: 2.093−2.561; p<0.001). Median number of days on therapy with mirabegron was 183 (treatment-experienced) and 129 (treatment-naïve), compared with 67−120 and 34−90 days with antimuscarinics, respectively. Mean adherence overall was 68% with mirabegron vs 39−55% with antimuscarinics (each p<0.05 vs mirabegron). Conclusions: Patients treated with mirabegron had improved persistence and adherence over antimuscarinics. While differences in patient characteristics among the OAB drugs were accounted for within the model, results should be viewed in the light of the likely characteristics of patients started on any new medication to the market; hence this sample may not be reflective of more mature usage. PUK29 Treatment Discontinuation In Patients With Urinary Incontinence Suffering From Glaucoma Kostev K 1, Rex J 1, Engelhard J 1, Altmann V 1, Ehlken B 2, Rockel T 1, Kalder M 3 1IMS Health, Frankfurt am Main, Germany, 2IMS Health Germany, Munich, Germany, 3University Hospital of Gießen and Marburg, Marburg, Germany . . . . . . . Objectives: The frequency of side effects in the treatment with anticholinergic drugs are well described in a number of previous studies. However, little is known about the impact of side effects on therapy discontinuation. The aim of the present study was to estimate the frequency of glaucoma in association with urinary incontinence therapy begin and the impact of glaucoma diagnosis on the therapy discontinuation based on real life data. Methods: Data from Disease Analyzer database including 988 general, 95 urologist and 203 gynecologist practices were used. 26,834 patients (17,125 female and 9,709 male) were identified to have received a first-time anticholinergic prescription of UI, namely darifenacin, fesoterodine, oxybutynin, propiverine, solifenacin, tolterodine or trospium. Co-variates studied included demographic data, concomitant diagnoses and potential drug-induced side-effects. Glaucoma (H40) was defined as strict indication for the use of anticholinergic drugs. A Cox proportional hazard regression model was used to estimate the relationship between non-persistence and the diagnosis of glaucoma for up to 36 months. Results: The proportion of patients that were diagnosed with glaucoma during the time of treatment was very similar in each of the study substances. 32 - 38 % of patients received a referral to an ophthalmologist and 0.3 - 1.2 % of patients were first time diagnosed with glaucoma. Not surprisingly, there was a highly increased risk for treatment discontinuation in patients having glaucoma (HR: 1.46; p < 0.0001). Conclusions: Overall, the potential side effects including the aggravation of diagnosed glaucoma that were registered in the database were rarer than in clinical trials; most likely they were under-reported due to the nature of the registry. However, there was a significant impact of glaucoma on therapy discontinuation. This finding should be taken into account in clinical practice for the use of anticholinergic drugs in patients suffering from glaucoma. PUK30 Health-Related Quality Of Life (Hrqol) Of Asian Patients With EndStage Renal Disease (Esrd) In Singapore Yang F 1, Griva K 2, Lau T 3, Vathsala A 3, Lee E 3, Ng H J 2, Mooppil N 4, Newman S P 5, Chia K S 1, Luo N 1 1Saw Swee Hock School of Public Health, National University of Singapore, Singapore, 2Department of Psychology, Faculty of Arts and Social Sciences, National University of Singapore, Singapore, 3Division of Nephrology, University Medicine Cluster, National University Health System, Singapore, 4National Kidney Foundation, Singapore, 5City University London, London, UK . . . . . . . . . . . . . Objectives: This study aimed to assess the health-related quality of life (HRQOL) of multiethnic Asian end-stage renal disease (ESRD) patients treated with dialysis and to identify factors associated with the HRQOL of those patients. The role of dialysis modality was also explored. Methods: Data used in this study were from two cross-sectional surveys of Singaporean ESRD patients on haemodialysis [HD] or peritoneal dialysis [PD]. In both surveys, participants were assessed using the Kidney Disease Quality of Life (KDQOL) instrument and questions assessing socio-demographic characteristics. Clinical data including co-morbidities, albumin level, haemoglobin level, and dialysis-related variables (e. g. dialysis vintage and dialysis adequacy) were retrieved from medical records. The 36-item KDQOL (KDQOL-36) was used to generate three summary scores (physical component summary [PCS], mental component summary [MCS], kidney disease component summary [KDCS]) and two health utility scores (Short Form 6-Dimension [SF-6D] and EuroQol 5-Dimension [EQ-5D]). Multivariate analysis was performed to examine the association of demographic, social and clinical variables with each of the HRQOL scores. Results: Five hundred and two patients were included in the study (HD: 236, PD: 266; mean age: 57.1 years; female: 47.6%). The mean (standard deviation [SD]) were PCS 37.9 (9.7), MCS 46.4 (10.8) and KDCS 57.6 (18.1); the mean (SD) of the health utility were 0.66 (0.12) for SF-6D and 0.60 (0.21) for EQ-5D. In multivariate regression analysis, factors found to be significantly associated with better HRQOL included: fewer co-morbidities, higher albumin level, and higher haemoglobin level with PCS and EQ-5D; higher albumin level with SF-6D; longer dialysis vintage with MCS; and Malay ethnicity, PD modality, and longer dialysis vintage with KDCS. Conclusions: Socio-demographic and clinical factors are both associated with HRQOL in ESRD patients on dialysis in Singapore. Dialysis modality has no impact on the health utility of those patients. PUK31 Health-Related Quality Of Life And Subjective Happiness Of Patients With Benign Prostatic Hyperplasia: First Results Of A CrossSectional Survey From Hungary Rencz F 1, Kovács Á 1, Gulacsi L 1, Majoros A 2, Nyirády P 2, Tenke P 3, Németh Z 3, Nagy G J 4, Nagy J 5, Buzogány I 6, Böszörményi-Nagy G 7, Brodszky V 1 1Corvinus University of Budapest, Budapest, Hungary, 2Semmelweis University, Budapest, Hungary, 3Jahn Ferenc South-Pest Hospital and Clinic, Budapest, Hungary, 4Saint Borbala Hospital of Tatabánya, Tatabánya, Hungary, 5Szentgotthárd Clinic, Szentgotthárd, Hungary, 6Péterfy Sándor Hospital, Budapest, Hungary, 7Bajcsy-Zsilinszky Hospital, Budapest, Hungary . . . . . . . . . . . . . A472 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To assess health-related quality of life (HRQOL) and subjective happiness of men with lower urinary tract symptoms associated with benign prostatic hyperplasia (BPH/LUTS). Methods: In 2014 a multicenter cross-sectional questionnaire survey was conducted in Hungary. Inclusion criteria were diagnosis of BPH/LUTS at least 12 months before the study and patient had never been undergone any prostate surgery. HRQOL was assessed indirectly by the EQ-5D and visual analogue scale (EQ VAS) and directly by matching indifference time trade-off (TTO) where patients’ subjective life expectancy was applied as time frame. Subjective happiness was measured on a 10-cm-long visual analogue scale. Results: In total 64 men completed the questionnaire, mean age and disease duration were 70.8±8.8 years and 6.9±6.5 years, respectively. At the time of the survey, 75% of the patients received alpha-blockers (AB), 11% were treated by 5-alpha-reductase inhibitors (5ARI), and 11% by combination therapy (AB+5ARI). Median EQ-5D, EQ VAS, TTO and happiness scores were 0.85, 70, 0.73 and 6.5, respectively. Utilities assessed by EQ-5D were found significantly higher compared to those elicited by TTO (p<0.001). Mean utilities decreased with clinical severity measured with EQ-5D (mild: 0.90±0.1, moderate to severe: 0.79±0.2, p= 0.046) whereas increased when assessed with TTO (mild: 0.46±0.3, moderate to severe: 0.63±0.3, p=0.336). Patients with moderate to severe disease reported significantly lower happiness scores than those with mild disease (5.85±2.2 and 7.02±2.3, p=0.044). Subjective happiness correlated moderately with EQ VAS (rs=0.64, p<0.001) but only weakly with EQ-5D (rs=0.28, p=0.029). Conclusions: BPH/LUTS patients experience notable reduction in their HRQOL and subjective happiness measured by any of the applied outcome measures. Utilities for BPH/LUTS ranged from 0.46 to 0.90 depending on clinical severity and assessment method. Considerable discrepancy was found between utilities elicited by EQ-5D and TTO that raises concerns about which method is feasible to use for utility evaluation in BPH/LUTS. PUK32 A Mixed Methods Approach (Mma) To Understanding Men’s Attitudes Toward The Management Of Lower Urinary Tract Symptoms (Luts) Associated With Benign Prostatic Hyperplasia (Bph) Ikenwilo D 1, Watson V 1, Heidenreich S 1, Chambers C 2, Newman C 2, Nazir J 2, Ryan M 1 1Health Economics Research Unit, University of Aberdeen, Aberdeen, UK, 2Astellas Pharma Europe Ltd, Chertsey, UK . . . . . . . Objectives: To understand men’s attitudes, preferences and values for LUTS/BPH management and its impact on quality of life (QoL) using a MMA. Methods: The innovative MMA approach combined qualitative (online discussion groups [ODG]) and quantitative methods (discrete choice experiment [DCE]). ODGs and a selfadministered online DCE survey were conducted with men in the UK aged ≥ 45 years with an International Prostate Symptom Score (IPSS) ≥ 8. A series of five asynchronous ODGs (or bulletin boards) were live for 5 days (moderated by the research team). The ODGs collected information about men’s experiences of LUTS/BPH and its impact on their QoL; data were analysed using an inductive thematic analysis. A literature review informed the ODG design, and both informed the DCE design. The DCE survey elicited men’s preferences for LUTS/BPH medical management. Please see a separate meeting abstract for more information on the DCE (Ikenwilo, et al). Results: 48 men were invited to the ODGs; 22 participated and had a mean age of 59.8 years and mean IPSS of 16.4 (standard deviation 7.1). The ODGs found that LUTS/BPH affects how men organise their day at work and at home, and prevents men from leading their preferred daily activities. Symptoms most adversely affecting QoL were urgency (especially urge incontinence) and day-time and night-time urinary frequency. There appears to be a lack of awareness about LUTS amongst the men: many had not discussed LUTS with their doctor, family or friends, and were unaware that LUTS/BPH is treatable. The DCE informs that men are willing to pay at least £23.95/month to improve urgency, and £29.70/month to avoid erectile dysfunction. Conclusions: This is the first MMA to evaluate men’s preferences in LUTS/BPH. The ODG and DCE results together expand our understanding of men’s attitudes toward the management of LUTS/BPH. PUK33 Prevalence Of Symptoms And Cluster Analysis In Dialysis Patients Using Kdqol-36 Galain A I 1, Alvarez R 2, Dapueto J J 1, Varela A 1 1Facultad de Medicina, Universidad de la República, Uruguay, Montevideo, Uruguay, 2Facultad de Ciencias Economicas, Universidad de la República, Uruguay, Montevideo, Uruguay . . . . . . Objectives: This study assessed the prevalence of symptoms, and symptom clusters in a sample of dialysis patients in Uruguay. Methods: A longitudinal study aimed to assess HRQL in hemo (HD) and peritoneal dialysis patients (PD) was conducted in thirteen centers in Uruguay. Data of the first assessment was included here. The 12 items of the KDQOL-36 Symptom Subscale were analyzed for symptom prevalence. For cluster analyses the ICLUST procedure was followed, using an algorithm to hierarchically cluster items to form composite scales. Alpha, the mean split half correlation, and beta, the worst split half correlation, are estimates of the reliability and general factor saturation of the test. Clusters are combined if coefficients increase in the new cluster. Results: Of a total of 751 patients, 486 accepted to participate and completed the forms (64.7%), with mean age 60.5 years, SD: 15.5), 211 (43.4%) were women; 407 (83.7%) were HD, and 79 (16.3%) were PD patients. Most prevalent symptoms were muscle sores (59.0%); cramps (57.5%); “washed out” (50.9%), dry skin (48.5%), and itchy skin (43.8%). Five clusters of symptoms were identified; “cutaneous” (itchy and dry skin, α .68), ”cardiac” (chest pain and shortness of breath, α .69), ”digestive” (nausea and lack of appetite, α .62), ”neurophatic” (numbness and cramps, α .50), ”energy” (“washed out”, faintness, α .59). A second block called “locomotive” included neuropathic plus muscle sores (α .64), and a third of “neurological symptoms” included locomotive plus energy (α .70). Cutaneous symptoms remained as a separated cluster. Cronbach alpha of the total Symptom Scale was .93. Conclusions: Dialysis patients experience a high number of symptoms. The study of clusters identified 5 groups of symptoms; cutaneous, cardiac, neuropathic, digestive and lack of energy. More complex associations included neurological, digestive and cardiac complaints. The identification of symptom clusters can help to understand common underlying pathways, and the use of drugs targeted to associated symptoms. PUK34 Men’s Preferences For The Treatment Of Lower Urinary Tract Symptoms (Luts) Associated With Benign Prostatic Hyperplasia (Bph): A Discrete Choice Experiment (Dce) Ikenwilo D 1, Watson V 1, Ryan M 1, Heidenreich S 1, Newman C 2, Nazir J 2, Chambers C 2 1Health Economics Research Unit, University of Aberdeen, Aberdeen, UK, 2Astellas Pharma Europe Ltd, Chertsey, UK . . . . . . . Objectives: To quantify men’s preferences and willingness to pay for attributes of medications to treat LUTS/BPH using a DCE. Methods: An online DCE was conducted with men in the UK aged ≥ 45 years with a self-reported International Prostate Symptom Score (IPSS) ≥ 8. Men were asked to consider different scenarios associated with moderate-to-severe LUTS, and then choose between two medical alternatives and a no-treatment option, according to seven attributes: day-time urinary frequency, night-time urinary frequency, urgency, sexual and non-sexual side effects, number of tablets/day and cost/month. These attributes were identified from a literature review and online discussion groups (ODGs). Men’s marginal willingness to pay (MWTP) and 95% confidence intervals for improvements in each attribute were calculated. Please see a separate meeting abstract for more information on the ODGs (Ikenwilo, et al). Results: The DCE survey was conducted 19–26 March 2014. 5212 individuals were invited to participate: 1097 opened the hyperlink, 586 were ineligible or excluded because the quota was full, and 247 completed the final survey used for this analysis. The mean age was 62.3 years and the mean IPSS was 15.7 (standard deviation 6.0). MWTP was £23.95/month and £25.42/month for an improvement from urge incontinence to moderate urgency or mild urgency, respectively; £5.80/month for one less night-time urination, and £1.30/month for one less day-time urination. Sexual side effects reduced MWTP by £29.70/month for erectile dysfunction, £18.37/month for “decreased sexual desire”, and £16.92/ month for “no fluid during ejaculation”. Non-sexual side effects reduced MWTP by £17.25/month for dizziness and £16.75/month for headaches. Neither dry mouth nor the number of tablets/day significantly affected MWTP.Conclusions: Improving urgency is perceived as the most important benefit of medication to men with LUTS/ BPH; improving frequency is also valued. The perceived value of a medication is reduced by sexual and non-sexual side effects. Urinary/Kidney Disorders – Health Care Use & Policy Studies PUK35 Incidence Of Unlicensed And Off-Label Prescription In Urologic Cancers Therapy In Turkey: Assessment Of Legislative And Regulatory Policy Tanyeri M H 1, Kockaya G 2, Tanyeri P 3, Yenilmez F B 4, Buyukokuroglu M E 3, Vural I M 5, Akbulat A 5, Artiran G 5, Kerman S 5 1Yenikent Governmet Hospital, Sakarya, Turkey, 2Health Economics and Policy Manager, Ankara, Turkey, 3Sakarya University, Sakarya, Turkey, 4Hacettepe University, Ankara, Turkey, 5Turkish Medicines and Medical Devices Agency, Ankara, Turkey . . . . . . . . . . . . . Objectives: “Off-label” is defined by the Turkish Ministry of Health (MoH) as the use of licensed pharmaceutical products in doses outside the scope of the registered indication and the use of unlicensed medicinal products that are imported for the purpose of individual treatment. Off-label use of medications is extremely common especially in oncology. The aim of the study is to evaluate the usage of off-label or unlicensed medicines in urologic cancers for understanding of Turkey’s perspective within this area of health care provisions. Methods: This study involved patients (n= 105) with metastatic bladder CA who received paklitaxel or other medicine application off-label, (n= 194) with metastatic renal cell CA who received sorafenib, sunitinib or other medicine application off-label, (n= 44) with metastatic testis cancer who received paklitaxel, gemsitabin or other medicine application off-label in Turkey. A computer search was performed using the TITCK’s (Turkish Drug and Medical Devices Institution) database. The patient base was searched for off-label medicine applications between 1st May 2008 to 1st May 2011. Results: The data obtained from TITCK’s database showed that 86, 136, and 44 applications for off-label metastatic bladder CA medicine, metastatic renal cell CA and metastatic testis CA medicine usage, respectively. Ninety five (90.47%) of all off-label medicine usage applications for metastatic bladder CA, one hundred twenty two (62.88%) for metastatic renal cell CA and thirty nine (88.63%) for metastatic testis CA were approved. University hospitals were created the vast majority of applications (79.89%) for metastatic bladder CA, (64.76%) for metastatic bladder CA and (79.89%) for metastatic bladder CA. The most preferable off-label drug medications for bladder CA, renal CA and testis CA were paklitaksel (84.04%), sorafenib (68.42%) and paklitaksel (43.24%), respectively. Conclusions: Off-label urology medicine use is rising in Turkey. If off-label use increasing parallel to the off-label urology medicines, it is needed to define new pathways to evaluate the applications. Research Poster Presentations - Session Iii Disease–Specific Studies CARDIOVASCULAR DISORDERS – Clinical Outcomes Studies PCV1 Readmittance To Hospital Within 6 Months After A Venous Thromboembolism Event: Prefer In Vte Registry Bauersachs R 1, Gitt A K 2, Mismetti P 3, Monreal M 4, Willich S N 5, Wolf W P 6, Agnelli G 7, Cohen A T 8 . . . . . . . . . . . . A473 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1Max-Ratschow-Klinik für Angiologie, Gefäßzentrum Klinikum Darmstadt GmbH, Darmstadt, Germany, 2Herzzentrum Ludwigshafen, Ludwigshafen, Germany, 3Centre Hospitalier Universitaire Saint-Etienne, Hopital Nord, Saint Etienne, France, 4Hospital Universitari Germans Trias I Pujol, Barcelona, Spain, 5Charité - Universitätsmedizin Berlin, Berlin, Germany, 6Daiichi Sankyo Europe GmbH, Munich, Germany, 7University of Perugia, Santa Maria della Misericordia Hospital, Perugia, Italy, 8King’s College, London, UK Objectives: Data are limited on the course of patients after an acute venous thromboembolism (VTE) event under clinical practice conditions. Methods: We present current data on the re-hospitalization of patients with deep venous thrombosis (DVT) or pulmonary embolism (PE) from the PREFER in VTE registry in 7 Western European countries (France, Italy, Spain, Germany, UK, Austria and Switzerland). The current interim analysis includes data of 2863 patients. 1689 patients had DVT alone and 1174 had PE (±DVT) as qualifying event for inclusion. Results: At baseline, 72.2% of the patients received heparin, 11.9% fondaparinux, 48.0% vitamin K antagonists, 25.4% non-VKA oral anticoagulants, 7.0% acetylsalicylic acid, 2.1% other antiplatelets and 1.6% thrombolysis agents. Within the first 6 months, 14.7% of VTE patients were hospitalized for any reason (DVT: 12.9%, PE: 17.5%). Hospitalization rates varied between countries (Spain 10.7%, Germany 10.8%, France 12.5%, Italy 19.3%, UK (at 3 months) 7.5%). The mean number of hospitalizations was 1.3 ±0.85 (1.1 to 1.5 in the various countries; median 1, range 1-6). The documented reasons for hospitalization, among others, were VTE (15.3%), surgery/trauma (13.5%), bleeding (7.4%), or stroke/TIA (3.7%). Mean duration of all combined hospital stays was 9.8 ±12.61 days (4.0 days in Germany, 4.7 in France, 9.0 in Spain, 14.0 in Italy). The overall median duration of hospitalization was 6 days (interquartile range 2-12), for DVT cases it was 6 days (IQR 3-13), for PE cases it was 5 days (IQR 2-11). Conclusions: Under real-life conditions, one in seven patients had to be readmitted to hospital in the first 6 months after the DVT or PE event. However, the majority of hospitalizations were not due to the thromboembolic disease or bleeding. The duration of hospital stays showed a wide range across Western European countries. PCV2 Discontinuation And Hospitalisation Rates In Patients With Atrial Fibrillation: Follow-Up Results Of The Prefer In Af Registry Brüggenjürgen B 1, Schliephacke T 2, Darius H 3, De Caterina R 4, Le Heuzey J Y 5, Pittrow D 6, Reimitz P E 2, Schilling R J 7, Zamorano J L 8, Kirchhof P 9 1Steinbeis University Berlin (SHB), Berlin, Germany, 2Daiichi Sankyo Europe GmbH, Munich, Germany, 3Vivantes Hospital Neukölln, Berlin, Germany, 4G. d’Annunzio University, Chieti, Italy, 5Hôpital Européen Georges Pompidou, Université René Descartes, Paris, France, 6Technical University Carl Gustav Carus, Dresden, Germany, 7Barts and St Thomas Hospital, London, UK, 8University Hospital Ramón y Cajal, Madrid, Spain, 9University of Birmingham Centre for Cardiovascular Sciences and SWBH NHS Trust, Birmingham, UK . . . . . . . . . . . . . . Objectives: The great majority of patients with atrial fibrillation (AF) require lifelong antithrombotic therapy for prevention of stroke. For optimal treatment, it is important to investigate treatment discontinuations, as those might be associated with increased hospitalization rates. As potential indicator for unstable anticoagulation efforts we assessed hospitalization rates for non-vitamin K antagonist oral anticoagulants (NOAC) and for warfarin under real-life practice conditions. Methods: PREFER in AF (The PREvention oF thromboembolic events – European Registry in Atrial Fibrillation) is a prospective non-interventional disease registry of patients with AF in 7 countries in Western Europe. Discontinuation rates were assessed looking at all patients (n= 6412), whereas hospitalization rates reported at Baseline (BL) and 1 year Follow-Up (FU) focused on two groups: patients treated with warfarin (BL= 1379, FU= 1571) and patients treated with NOAC (BL= 194, FU= 424). Descriptive statistics were applied. Results: Out of 6412 patients, 9 to 18% of patients treated with vitamin K antagonists (VKA) discontinued therapy (warfarin 10.3%, phenprocoumon 9%, acenocoumarol 10.4%, fluinidone 18%). The discontinuation rates for patients on NOAC were 9.4% for rivaroxaban and 10.3% for dabigatran, respectively. Reported hospitalization rates (irrespective of reason) were 19.8%/10.0% for NOAC, and 24.1%/11.6% for warfarin at BL or FU, respectively. Mean duration of hospitalization was 1.3±4.0/ 0.9 ±5.1 days for NOAC, and 2.0±6.1/ 1.0 ±4.2 days for warfarin. Conclusions: No major differences were observed in the discontinuation rates between VKA and NOAC on the class level nor the various available NOAC drugs. Hospitalization rates and corresponding number of days in hospital reported at BL were substantially higher compared to FU. Patients on NOAC had lower hospitalization rates and less days in hospital compared to those on warfarin. Further analyses are needed to explore the reasons for hospitalizations. PCV3 The Additional Costs Of Clinical Complications In Patients Undergoing Transcatheter Aortic Valve Replacement In The German Health Care System Kaier K 1, Gutmann A 2, Sorg S 2, Beyersdorf F 2, Vach W 1, Zehender M 2, Bode C 2, Reinöhl J 2 of Freiburg, Freiburg, Germany, 2Heart Center Freiburg University, Freiburg, Germany . . . . . . . . 1University Objectives: Transcatheter aortic valve replacement (TAVR) is a relatively new alternative to surgical replacement of the aortic valve. Recent development of third generation systems and new products have shown reductions in the frequency and severity of complications associated with the procedure. The aim of this study was to identify the cost impact of complications for patients undergoing TAVR in a German hospital. Methods: Data was derived from a prospective observational study, whereby a total of 163 consecutive patients were treated either with transfemoral (TF-, n= 97) or transapical (TA-) TAVR (n= 66) between February 2009 and December 2012. Predefined clinical endpoints were analyzed, “in-hospital” costs determined from the hospital perspective (2012=100) and results reported within the seven most relevant cost categories. Results: TF-TAVR patients experienced more minor access site bleeding (p= 0.017), major non-access site bleedings (p= 0.026), minor vascular complications (p=0.002), stage 2 acute kidney injury (AKI, p=0.043) and permanent pacemaker implantation (p< 0.001) compared with TA-TAVR. However, total in-hospital costs did not differ between groups (mean € 40,348; SD €15,851). Costs were proportioned in either categories (staff = 26%, materials = 62% and infrastructure = 12%) or units (normal ward = 13%, ICU = 15%, catheter laboratory or operating room = 62% and other = 10%). The average additional cost of any single complication was € 3,438 (p<0.01). Life-threatening non-access site bleeding was associated with the highest additional costs (€47,494; p<0.05), followed by stage 3 AKI (€20,468; p<0.01), implantation of a second valve (€16,767; p<0.01) and other severe cardiac dysrhythmia (€10,611 p<0.05). Interestingly, censoring cases of in-hospital mortality is associated with a substantial decrease in the additional costs. Conclusions: Bleeding complications, severe kidney failure, and implantation of a second valve are the most important cost drivers in TAVR patients, and strategies to reduce those complications may have the potential to generate significant in-hospital cost reductions. PCV4 Clotting Factor (Cf) Product Use And Same-Day Risk For Thrombotic Adverse Events (Tes), As Recorded In Large Health Care Database During 2008-2013 Study Period Ekezue B F 1, Sridhar G 1, Izurieta H S 2, Forshee R A 2, Selvam N 1, Ovanesov M V 2, Jain N 2, Mintz P D 2, Anderson S A 2, Menis M 2 1HealthCore Inc., Alexandria, VA, USA, 2U. S. Federal Drug Administration, Silver Spring, MD, USA . . . . . . . . . . . . . . . . Objectives: There has been an increase in the number of clotting factor (CF) products available in the U. S. in recent years. Thrombotic events (TEs) are serious adverse events that can occur following administration of CFs. The objective of this study is to assess the same-day TE risk following exposure to CF products. Methods: A retrospective cohort study of individuals exposed to CF products during January 2008 through June 2013 was conducted using HealthCore Integrated Research Database (HIRDSM). CF products were identified by Health care Common Procedure Coding System (HCPCS) codes, and TEs were ascertained via ICD-9-CM diagnosis codes. Crude sameday TE rates (per 1,000 persons exposed) were estimated overall, by congenital clotting factor deficiency status and by specific CF products, age, and gender. Results: Of 3,801 individuals exposed to CFs, 117 (30.8 per 1,000 persons) had TEs recorded on the sameday as CF exposure. The crude same-day TE rate (per 1,000) was higher for CF users without congenital CF deficiency, 70.2 (102 of 1,452), as compared to those with congenital CF deficiency, 6.4 (15 of 2,349), unadjusted rate ratio of 11.0 (95% CI 6.4-18.9). For individuals without congenital CF deficiency, the crude TE rates (per 1,000) were 15.9 for under 15 years of age, 16.1 for 15 to 44 years, 62.8 for 45 to 64 years, and 160.9 for 65 years and older. The unadjusted same-day TE rates (per 1,000) ranged from 12.8 for Factor VIII to 204.1 for Factor IX complex product(s). Multivariable analyses are underway to control for potential confounders and identify recipient risk factors. Conclusions: The study shows an increased risk of same-day TEs for CF users without congenital CF deficiency and suggests a potentially substantial off-label use of CFs, which needs further investigation. In addition, study findings suggest elevated same-day TE rates for specific CF products with additional multivariable investigation ongoing. PCV5 The Effect Of Atriala Fibrillation In Acute Myocardial Infarction Patients In Taiwan Li C Y 1, Chang C J 2, Chu P H 3, Fann C S J 4 Gung University, Taichung, Taiwan, 2Chang Gung University, Kwei Shan, Tao Yuan, Taiwan, 3Chang Gung University, Taoyuan, Taiwan, 4Academia Sinica, Taipei, Taiwan . . . . . . . . . 1Chang Objectives: Acute myocardial infarction (AMI) is a major cause of mortality and disability worldwide. AMI occurs when blood flows irregularly into heart and thus heart muscle is injured due to insufficient constant oxygen received. It will cause severe complications or co-morbidities if the condition is lasting. Hence, we hypothesized the atrial fibrillation (AF) is an independent risk factor to the major severe cardiovascular events (MACE) after AMI occurred. Methods: The AMI patient is defined by the patient treated in the emergency room at the beginning of the illness. Frequencies and costs of AMI data were extracted from the National Health Insurance Research Database for this observational retrospective cohort study between 2007 and 2012 in Taiwan. ICD-9-CM 410 was used to extract the AMI patients. Fisher’s exact test and categorical data analytic method were utilized to assess the AF as a risk factor to MACE in the AMI patients. Results: We mainly focused on the AMI adults without any prior MACE occurred. As a result, there were 3,452 AMI who can be divided into 2 groups: 2,939 AF patients and 513 non-AF patients. The average medical cost was USD$ 3142.8 and the mean LoS was 10.4 days with st. d. 29.0 days. There were 1,791 MACE identified among the AF patients (60.9%), while there were 251 MACE among the non-AF patients (48.9%). In consequence, the AMI with AF resulted more MACE than those in non-AF (RR = 1.63, C.I. = (1.35, 1.97), p < 0.0001). The difference of cost was not significant between both groups. The mean LoS in AF was 16.6 days, which was significantly smaller than that in non-AF (19.1 days). The difference was probably due to higher fatalities in AF. Conclusions: This study has demonstrated that AF risk is associated with MACE in patients after AMI occurred. PCV6 Development Of A Collaborative European Pharmacoepidemiologic Post-Authorization Safety Study (Pass) Programme Examining Rivaroxaban Use In Routine Clinical Practice Brobert G 1, García Rodríguez L A 2, Garbe E 3, Bezemer I D 4, Layton D 5, Friberg L 6, SuzartWoischnik K 1, Alderson J 7, Winchester C 7, Herings R M C 4, Jobski K 3, Schink T 3, Shakir S 5, Soriano-Gabarró M 1, Wallander M A 8 1Bayer Pharma AG, Berlin, Germany, 2Ceife - Centro Español de Investigación Farmacoepidemiológica, Madrid, Spain, 3Leibniz Institute for Prevention Research and Epidemiology - BIPS, Bremen, Germany, 4Pharmo Institute for Drug Outcomes Research, Utrecht, The Netherlands, 5Drug Safety Research Unit, Southampton, UK, 6Friberg Research AB, Stockholm, Sweden, 7Oxford PharmaGenesis™ Ltd, Oxford, UK, 8Uppsala University, Uppsala, Sweden . . . . . . . . . . . . . . . . . . . . Background: Proactive, post-authorization monitoring of drug safety and effectiveness is of increasing importance. Rivaroxaban is a Factor Xa inhibitor with multiple indications, including: treatment of venous thromboembolism (VTE) and prevention of recurrent VTE; stroke prevention in atrial fibrillation; and prevention A474 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 of atherothrombotic events (when combined with antiplatelet therapy) following an acute coronary syndrome. Use, safety and effectiveness of rivaroxaban in real-life settings need to be monitored to understand its value and comply with regulatory requirements. Objectives: To develop a pharmacoepidemiologic PASS programme to characterize post-authorization rivaroxaban use and relevant outcomes related to safety (intracranial, gastrointestinal, and urogenital haemorrhage) and effectiveness (myocardial infarction, ischaemic stroke and thromboembolic events). Methods: European sources of longitudinal observational health care data were identified that contained population-based data on demographics, comorbidities and comedications. Additional studies were designed to capture drug utilization, safety and effectiveness data in primary and secondary care via physician questionnaire completed based on medical chart review. With an emphasis on the consistent definition of endpoints, protocols were developed and tailored for each study for submission to regulatory authorities and relevant ethics committees. Results: A PASS programme of 7 studies was designed. Rivaroxaban drug utilization, and safety and effectiveness, in comparison with standard of care, are being assessed using The Health Improvement Network (UK), the PHARMO Database Network (The Netherlands), the German Pharmacoepidemiological Research Database and the National Swedish Registries. The programme also includes 2 Specialist Cohort Event Monitoring studies to collect data on the initial treatment period in secondary care settings, and a complementary Modified Prescription-Event Monitoring study that evaluates long-term use in primary care. Conclusions: This pharmacoepidemiologic PASS programme, with its unique and complementary approach, will monitor and characterize the real-life benefit–risk profile of rivaroxaban. A flexible design allows the accommodation of any new indications for rivaroxaban approved during the study. PCV7 Coronary And Cardiovascular Disease Risks In Migraine Patients: Evidence From National Health And Nutrition Examination Survey 1999–2004 Yu J , Zhang W , Mezzio D , Vo T Touro University California, Vallejo, CA, USA . . . . Objectives: (1) To assess the prevalence of migraine using the National Health and Nutrition Examination Survey (NHANES) conducted in the general United States population; (2) To calculate the 10-year Framingham cardiovascular disease (CVD) and coronary heart disease (CHD) risks and compare the risks between migraine patients (MP) and non-migraineurs (NM). Methods: MP were identified using the survey question: “During the past 3 months, did you have severe headaches or migraines?” from the 1999-2000, 2001-2002, and 2003-2004 NHANES. Independent chi-square tests and t-tests were used to compare differences in CVD or CHD risk factors between MP and NM. Multivariant ordinal logistic regressions were used to examine the odds ratios of low, moderate, and high CHD and CVD risks between MP and NM accounting for the complex sampling design used by NHANES. Significance was defined as p≤ 0.05. Results: Amongst the total of 14,204 study subjects (6,731 men and 7,473 women), 2,895 (20.38%) subjects reported having migraines or severe headaches in the past three months. The average age of MP was 41.8 (±16.8) years, compared to 47.4 (±19.8) years in the NM group. MP had a significantly higher percentage of females, with 1,962 (67.77%) females. Subjects in the migraine group were found to have an odds ratio of 1.208 (95% CI, 1.033-1.412, p-value< 0.001) for moderate CHD risk, and an odds ratio of 0.975 (95% CI, 0.732 – 1.412, p-value< 0.001) for high CHD risk. Regarding CVD risk, migraine patients were found to have an odds ratio of 1.173 (95% CI, 0.818 – 1.683, p-value<0.001) for moderate risk, and an odds ratio of 1.112 (95% CI, 0.909 – 1.582, p-value< 0.001) for high risk. Conclusions: Migraine appears to be more prevalent among younger subjects with an overall mean age of 42 years old. These findings also suggest a positive association between migraine and CHD risks. PCV8 Effectiveness And Costs Of Different Strategies For The Diagnosis Of Stable Coronary Artery Disease Results From The Evinci Study Turchetti G 1, Lorenzoni V 1, Bellelli S 1, Pierotti F 1, Rovai D 2, Caselli C 2, Underwood R 3, Knuuti J 4, Neglia D 2 1Scuola Superiore Sant’Anna, Pisa, Italy, 2National Research Council, Pisa, Italy, 3Imperial College London, London, UK, 4University of Turku and Turku University Hospital, Turku, Finland . . . . . . . . . Objectives: The analysis of different non-invasive imaging strategies in a European population of patients with stable angina could help the identification of the best approach for the diagnosis of significant CAD. Methods: In 475 pts (291 males, 60±9 yrs) with stable angina enrolled in the EVINCI multicenter study, CT coronary angiography (CTCA) and stress imaging were performed before invasive coronary angiography (ICA). Significant CAD was defined as > 50% stenosis in the left main or > 70% stenosis in a major coronary vessel or 30-70% stenosis with fractional flow reserve ≤ 0.8. Nine non-invasive imaging strategies including CTCA or stress imaging (ECHO, CMR, SPECT or PET) alone or in combination with CTCA performed as first line examinations were evaluated. Combinations were positive if both CTCA and the stress test were positive and their performance was evaluated in terms of accuracy (AUC). Centres specific reimbursements were collected for invasive and non-invasive examination and the costs associated with different strategies were obtained at individual patient level by summing up reimbursement of the examinations involved. Results: Significant CAD was diagnosed at ICA in 140 patients (29%). CTCA had the highest diagnostic performance among strategies involving single imaging modalities [0.91 (0.88-0.94)] and CTCA-PET among combinations [0.84 (0.74-0.93)]. Costs vary significantly among the strategies involving single non invasive modalities with value ranging from 425 Euro for ECHO to 1245 Euro for PET as well as for those involving combinations of imaging modalities ranging from 508 Euro for CTCA-ECHO to 870 Euro for CTCA-SPECT. Conclusions: The diagnostic workflow for CAD detection shows variable effectiveness and costs according to the use of different single or combined non invasive imaging modalities. Data collected in the EVINCI study offer the opportunity to evaluate several diagnostic workflow from a combined clinical and economic perspective using a cost-effectiveness analysis. PCV9 An Assessment Of The Current Literature On Apheresis Use In The Treatment Of Familial Hypercholesterolemia Wang A 1, Richhariya A 2, Gandra S R 2, Calimlim B 1, Kim L 1, Nordyke R 1 Segundo, CA, USA, 2Amgen, Inc., Thousand Oaks, CA, USA . . . . . . . 1ICON, El Objectives: Although apheresis is an important treatment for reducing LDL-C in familial hypercholesterolemia (FH) patients, little is known about its treatment patterns. We conducted a systematic review to assess the efficacy/effectiveness, practice patterns, cost, and clinical guidelines for apheresis in FH patients. Methods: Electronic databases were searched for publications of apheresis in FH patients. Inclusion criteria include: articles in English published 2000-2013, description of practice patterns, efficacy/effectiveness, and costs. Data were stratified by country and FH genotype where possible. Results: Thirty-seven studies met the inclusion criteria: 8 open-label clinical trials, 11 observational studies, 16 reviews/guidelines, and 2 health technology assessments. Of the 19 clinical and observational studies, 6 assessed only homozygous FH (HoFH) patients, 4 assessed only heterozygous FH (HeFH) patients, 6 included HoFH and HeFH patients, and 3 did not specify type of FH. The prevalence of FH is not well characterized by country and underdiagnosis is a barrier to optimal FH treatment. Apheresis guidelines recommend weekly/biweekly treatments conducted at apheresis centers that may last ≥ 3 hours per session. Apheresis may be recommended as first-line treatment in HoFH patients and after drug therapy failure in HeFH patients. Studies reported a range of LDL-C reduction after apheresis: HoFH: 57-75%; HeFH: 58-63%. Eight studies reported apheresis costs. Cost (USD 2013) per apheresis session ranged from $2,200-$4,300 in the US, $2,150-$2,600 in the UK, $1700-$1850 in Germany and France, and $2,350-$2,750 in Australia. Calculated annual costs may reach $88,400-$225,000 per patient for weekly treatment. Conclusions: LDL-C apheresis treatment is necessary for FH patients when drug therapy is inadequate. While apheresis reduces LDL-C, high per-session costs and the frequency of guideline-recommended treatment result in substantial annual costs. The costs and the inconvenience of apheresis sessions are barriers in optimal treatment of FH. PCV10 Estimated Added Benefit Of Catheter-Based Renal Denervation For Moderate Treatment-Resistant Hypertension: Impact Of Age And Cardiovascular Risk Factors Pietzsch J B 1, Geisler B P 1, Esler M D 2 1Wing Tech Inc., Menlo Park, CA, USA, 2Baker IDI Heart and Diabetes Institute, Melbourne, Australia . . . . . . Objectives: Our objective was to estimate the impact of catheter-based renal denervation plus standard of care (RDN) versus standard of care alone (SoC) on European stage-1 hypertension patients with different cardiovascular risk factor profiles. Methods: We simulated resistant hypertension cohorts with different cardiovascular risk factor profiles (CVRPs) who had stage-1 hypertension despite adequate treatment with SoC. Six different cohorts with permutations of starting ages of 40, 55, or 70 years; and low and high CVRPs based on JNC7 guidelines were modeled. Interventions were RDN plus SoC with three or more anti-hypertensives including a diuretic at full doses or SoC alone. As observed in a recent prospective uncontrolled pilot study, the impact of a 13 mmHg reduction in systolic blood pressure (SBP) – from a baseline SBP of 151 mmHg – was evaluated. Undiscounted life year (LY) and quality-adjusted life year (QALY) gain was computed using a published life-time Markov simulation model based on multivariate risk equations. Results: Across the studied age groups and cardiovascular risk profiles, catheter-based renal denervation was associated with gains in unadjusted (0.51-1.48 LYs) and qualityadjusted life expectancy (0.39-1.20 QALYs). Younger age and higher cardiovascular risk profile led to numerically higher increments. The cohort-specific LY and QALY gains were projected as follows: 1) 40 year old cohorts: low CV risk profile: 0.84 LYs/ 0.78 QALYs; high CV risk profile: 1.48 LYs/ 1.20 QALYs; 2) 55 year old cohorts: low CV risk profile: 0.69 LYs/ 0.60 QALYs; high CV risk profile: 1.29 LYs/ 0.95 QALYs; 3) 70 year old cohorts: low CV risk profile: 0.51 LYs/ 0.39 QALYs; high CV risk profile: 0.99 LYs/ 0.68 QALYs. Conclusions: RDN, when used to treat moderate treatment-resistant hypertension, is associated with clinically significant increases in life expectancy. These model-based findings need to be confirmed in clinical trials. PCV11 The Use Of Minimally Invasive Surgery (Mis) And Intraoperative Imaging Modalities In The Treatment Of Intracerebral Hemorrhage (Ich): A Systematic Review Of The Literature Karamalis M 1, Langer K 1, Demessinov A 2, Birinyi-Strachan L 3 Kazakhstan, Almaty, Kazakhstan, 3Medtronic Australasia, Sydney, Australia . . . . 1Medtronic, Tolochenaz, Switzerland, 2Medtronic Objectives: The objective of this review was to investigate the use of MIS and intraoperative imaging technologies in the treatment and management of persons with ICH. Methods: A systematic search of the published literature was undertaken in February 2014, using the Embase, PubMed and CRD York bibliographic databases. Key society websites were also searched for relevant ICH practice guidelines. Studies and guidelines reporting on MIS techniques in persons eligible for treatment for Primary ICH were included. Results: 1042 citations were retrieved, of which 190 full-text articles were reviewed. Six practice guidelines, 4 Systematic reviews, 10 RCTs and 3 Economic evaluations were included. All 4 systematic reviews of RCT data showed consistent results in favour of MIS versus medical management (MM) or craniotomy (CR) for both efficacy (mortality and/ or mortality and dependent living benefit) and safety (lower rebleeding and higher hematoma volume reduction) vs. CR. A statistically significant reduction in the relative risk (RR) of death ranging from 34-71% for patients treated with MIS as opposed to CR or MM was reported in all 4 reviews. Three ICH guidelines (Europe, Japan and USA) recommended the use MIS in specific circum- A475 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 stances. All 3 economic studies found MIS in the treatment of ICH was associated with lower costs compared to MM or CR. In one study MIS resulted in a $USD44,329 saving in acute care costs compared to MM, primarily driven by lower hospital bed days (70 vs. 98.9 days). Conclusions AND IMPLICATIONS OF KEY FINDINGS: MIS in the treatment of ICH presents convincing evidence for improved efficacy over MM and CR. The latest clinical practice guidelines for the treatment of ICH do not yet accurately reflect these latest findings and as such, current treatment practices may be lagging behind what the highest level of evidence suggests should be standard of care. PCV12 Pharmacoeconomic Grounding Of Using Polypill Amlodipine With Atorvastatin Versus Monodrugs In Patients With Hypertension And Dyslipidemia In Ukraine Mishchenko O , Bezditko N , Adonkina V , Tkachova O National University of Pharmacy, Kharkiv, Ukraine . . . . Objectives: One reason of the low efficiency of cardiovascular diseases (CVD) treatment in Ukraine is the low adherence of patients. Modern trends improving the quality of treatment and increase patients compliance is use of polypills (PP). The aim: pharmacoeconomic assesment the feasibility of PP amlodipine and atorvastatin versus monodrugs in patients with hypertension and dyslipidemia (DYS) from the Ukrainian perspectives point of view. Methods: The results of clinical studies AVALON (Granger C. B., McMurray J. J., Yusuf S. et al., 2003) were used. Cost minimization analysis of three regimens of patients treatment with hypertension and DYS during 8 weeks: PP amlodipine 5 mg + atorvastatin 10 mg; amlodipine 5 mg; atorvastatin 10 mg. Results: The results of the clinical research AVALON found, that the use of PP amlodipine+atorvastatin provides significant clinical benefit: the largest number of patients reached target levels of blood pressure (BP) and low-density lipoprotein cholesterol (LDL-C) (45.5%), versus amlodipine (8.3%), atorvastatin (28.6%), placebo (3.5%). The scheme using amlodipine is the most expensive (cost for course of treatment (CCT) € 20.28), the regimen of atorvastatin 10 mg (CCT € 10.46) and the PP amlodipine+atorvastatin (CCT € 17.72) are less costly. This PP is more cost effective versus amlodipine monotherapy (CER = € 38.95 versus € 244.34 per patient with target levels of BP and LDL-C) and less cost effective compared atorvastatin monotherapy (CER = € 36.57 per patient with target levels of BP and LDL-C). The cost of an additional unit of effectiveness (ICER) showed that the use of PP amlodipine+atorvastatin instead amlodipine provides for the treatment of each 100 patients additional 37.2 patients achieved target levels of BP and LDL-C and saving € 6.88 per patient. Conclusions: Pharmacotherapy of patients with hypertension and DYS based on PP amlodipine+atorvastatin provides significant clinical benefit versus monodrugs and pharmacoeconomic advantages versus amlodipine. PCV14 Lifetime Clinical Events Avoided And Resource Utilization With Apixaban Compared To Low-Molecular-Weight Heparin Followed By A Vitamin K Antagonist For The Treatment And Prevention Of Venous Thromboembolism Hamilton M 1, Phatak H 1, Lanitis T 2, Mardekian J 3, Rublee D A 3, Leipold R 4, Quon P 4, Browne C 2, Cohen A T 5 1Bristol-Myers Squibb Company, Princeton, NJ, USA, 2Evidera, London, UK, 3Pfizer, Inc., New York, NY, USA, 4Evidera, Bethesda, MD, USA, 5Guy’s and St Thomas’ NHS Foundation Trust, London, UK . . . . . . . . . PCV15 The Effectiveness Of Carotid Artery Stenting Compared With Endarterectomy In Symptomatic Patients With Carotid Stenosis In Korean Multi-Center Setting You J H 1, O h S H 1, Lee J Y 2, Park J J 1, Shin S 3 1National Evidence-based Healthcare Collaborating Agency, Seoul, South Korea, 2National Evidence-based Healthcare Collaborating Agency (NECA), Seoul, South Korea, 3National Evidencebased healthcare Collaborating Agency, Seoul, South Korea . . . . . . . . . PCV16 Real-Time Assessment Of Medication Taking And Activities Of Daily Living In Patients With Uncontrolled Hypertension DiCarlo L , Kim Y A , Young J, Bezhadi Y Proteus Digital Health, Redwood City, CA, USA . . . . Objectives: For patients with uncontrolled hypertension, differentiation of pharmacological resistance from inadequate or improper medication use is key to clinical management. Proteus Digital Health has developed a unique digital feedback system, which utilizes an Ingestible Sensor (IS) to determine medication-taking patterns. A wearable sensor in the form of an adhesive patch collects timing and taking of IS ingestions, and physiological and behavioral metrics such as heart rate, and patterns of activity and rest, providing insights into the patient’s dayto-day lifestyle. This study evaluates the utility of the Proteus system in patients with uncontrolled hypertension. Methods: Patients with a history of uncontrolled hypertension (BP> 140/90) at 5 primary care centers in the United Kingdom were prescribed the Proteus system for 14 days. Patients co-ingested the IS along with their prescribed BP medications while simultaneously wearing the patch. BP was measured on days 1 and 14 by a clinician, and all other parameters, such as adherence, activity and rest patterns, were collected via the Proteus system. Results: Of the 190 patients, 21 patients had incomplete data. In the remaining 169 patients (89%), mean medication adherence was 88%, and mean BP decrease was -7.6 mm Hg systolic and -3.8 mm Hg diastolic. The system data provided diagnostic insight differentiating non-response vs. non-adherence. One hundred forty-eight (78%) patients had ≥ 70% adherence; 100 (53%) achieved blood pressure control on their prescribed therapy, and 48 (25%) remained uncontrolled and required modification of their therapeutic regimen. The remaining 21 patients (11%) were identified as needing intervention to support medication adherence. Conclusions: In patients with a history of uncontrolled hypertension, 53% achieved BP control within 2 weeks, and 25% received an informed therapeutic intervention using the Proteus system. Thus, Proteus can identify specific individual needs for progressing through the recommended treatment pathway and for advancing toward treatment goals. . . Objectives: The AMPLIFY trial compared apixaban to low-molecular-weight heparin (LMWH) followed by a vitamin K antagonist (VKA) for treatment and prevention of recurrent venous thromboembolism (VTE). The AMPLIFY-EXT trial compared extended treatment with apixaban to placebo in previously treated patients. This analysis evaluated the potential lifetime implications of apixaban treatment versus LMWH/VKA. Methods: A Markov model was developed to evaluate the lifetime impact of treatment and prevention of VTE with apixaban (5 mg BID for 6 months, then 2.5 mg BID) versus LMWH/VKA. Clinical event rates were taken from AMPLIFY, AMPLIFY-EXT, and indirect treatment comparison. Length of stay for hospitalizations was taken from AMPLIFY for recurrent VTE (median 5 days for apixaban, 6 days for LMWH/VKA) and major bleeds (median 5 days for apixaban, 7 days for LMWH/VKA). Sixty percent of patients with recurrent VTE and all patients with major bleeding were assumed to be hospitalized. Outcomes evaluated were events and hospital bed days avoided, number needed to treat to avoid a recurrent VTE, number needed to treat to harm with an additional bleed, and life years gained. Results: In a cohort of 1,000 patients, lifetime treatment with apixaban versus LMWH/VKA resulted in 6 fewer recurrent VTE events, 191 fewer major bleeds, 707 fewer clinically relevant non-major bleeds, and 1,730 hospital bed days avoided. On average, a patient treated with apixaban gained about 3 months of life expectancy due to avoidance of VTE events and major bleeds. These results translated to one recurrent VTE event avoided for each 157 patients treated and one major bleed avoided for each 5 patients treated with apixaban versus warfarin. Conclusions: Apixaban for treatment and prevention of VTE appears to be a superior alternative to LMWH/VKA, leading to fewer recurrent VTEs, bleeding events, and hospital bed days resulting in a projected increase in life-expectancy. . compare the effectiveness with CAS and CEA in 677 patients with symptomatic carotid artery stenosis in korean clinical practice. Methods: From January 1 2008 to December 31 2011, retrospective cohort study was conducted in 677 symptomatic carotid stenosis patients with more than 50% stenosis) (CAS=346, CEA=331) in the Korean hospitals (Asan medical center, Samsung medical center, Severance hospital, Inha university hospital, Chonnam university hospital). The primary outcome was stroke, myocardial infarction, or death during periprocedural (30-day) and postprocedural period. Results: For 677 patients over 2-year follow-up period, All death, major stroke, minor stroke were higher in CAS group than CEA (1.45% vs. 0.30%, 4.05% vs. 1.81%, 3.47% vs. 3.02%, 0.58% vs. 0%). All outcomes were higher in CAS than in CEA within 30-day after treatment and in subsequent years, except the incidence of 30 days-minor stroke. Conclusions: CEA was superior to CAS in symptomatic patients with carotid stenosis. This study suggests that CEA can be considered the first-line therapy for symptomatic carotid artery stenosis in South Korea. . Objectives: Carotid endarterectomy (CEA) has been recommended as the gold standard for the management of carotid disease in many clinical guidelines. But, in Korean clinical practice, carotid artery stening (CAS) was conducted more than CEA (21.6%) based on the national claims-data. The purpose of this study was to PCV17 Use Of Computer Simulation To Generate Evidence To Aid Health Care Decision Making: An Example Using The Archimedes Model To Compare Rosuvastatin With Atorvastatin Colivicchi F 1, Sternhufvud C 2 San Filippo Neri, Rome, Italy, 2AstraZeneca, Mölndal, Sweden . . 1Ospedale Objectives: Randomized controlled trials (RCTs) provide the most robust evidence source for making patient health care decisions. When RCT data are lacking, however, complementary evidence sources may also be needed. As an example of this, three clinical trials comparing rosuvastatin with atorvastatin were simulated using the Archimedes model, a validated, individual-based simulation of human pathophysiology and behaviours, treatment interventions and health care systems. Methods: Comparison A assessed clinical outcomes in patients receiving available doses of the two drugs. Comparison B assessed the impact of initial treatment decisions, with individuals randomized to receive various doses of either rosuvastatin or atorvastatin and eligible for treatment intensification for up to 5 years if target lipid levels were not met. Comparison C assessed the effect of switching patients’ treatment from rosuvastatin to atorvastatin. Results: In comparison A, rosuvastatin was estimated to result in greater reductions than atorvastatin in major adverse cardiac events (MACEs) at 5 and 20 years at all doses examined (relative risk [RR]: 0.907,0. 892 and 0.931 at 20 years for rosuvastatin 20 mg versus atorvastatin 40 mg, rosuvastatin 40 mg versus atorvastatin 80 mg, and rosuvastatin 20 mg versus atorvastatin 80 mg, respectively; P< 0.05 in all cases). In comparison B, outcomes were significantly better in patients initially prescribed rosuvastatin relative to atorvastatin (RR of MACE at 10 years: 0.919; P< 0.001). In comparison C, risk of MACE was significantly greater in patients who switched from rosuvastatin to atorvastatin, relative to those who remained on rosuvastatin (RR at 10 years: 1.115; P< 0.001). Conclusions: In this example using the well-validated Archimedes model, better outcomes were predicted in patients receiving rosuvastatin than in those receiving atorvastatin in a variety of different settings. This provides an example of the utility of robust modelling approaches to generating evidence that is not available from clinical trials. PCV18 Critical Appraisal Of Network Meta-Analyses Evaluating The Efficacy And Safety Of New Oral Anticoagulants In Atrial Fibrillation Stroke Prevention Trials Cope S 1, Clemens A 2, Hammès F 3, Noack H 4, Jansen J 5 1Mapi, Inc., Toronto, ON, Canada, 2Boehringer Ingelheim Pharma GmbH & Co KG, Corporate Devision Medicine, Ingelheim, Germany, 3Boehringer Ingelheim, Paris, France, 4Boehringer Ingelheim Pharma GmbH & Co KG, Medical Data Services, Ingelheim, Germany, 5Tufts University School of Medicine, Boston, MA, USA . . . . . A476 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To critically appraise the published network meta-analyses (NMAs) evaluating the efficacy or safety of the new oral anticogulants (NOACs) dabigatran, rivaroxaban and apixaban for the prevention of stroke in patients with non-valvular atrial fibrillation (AF). Methods: A systematic literature review was performed to identify the relevant NMAs using MEDLINE®, EMBASE®, Cochrane Library, Database of Abstracts of Reviews of Effects, and Health Technology Assessment. The synthesis studies were evaluated using the ‘Questionnaire to assess the relevance and credibility of the NMA’. Results: Eleven NMAs evaluating NOACs among adults with non-valvular AF were identified. Most NMAs included three large phase III RCTs, comparing NOACs to adjusted-dose warfarin (RE-LY, ROCKET-AF, ARISTOTLE). The main differences identified related to potential treatment effect modifiers regarding the mean time spent in therapeutic range (TTR) in the warfarin arm, the risk of stroke or systemic embolism across the trials (mean CHADS2 score: Cardiac failure, Hypertension, Age ≥ 75 years, Diabetes mellitus, Stroke, 2 two points for stroke) or primary versus secondary prevention, and type of populations used in the analysis. Kansal et al. appropriately adjusted the ROCKET-AF TTR to match the RE-LY population based on individual patient data. Meta-regressions are not expected to minimize confounding bias given limited data, whereas subgroup analyses had some impact on the point estimates for the treatment comparisons. Conclusions: Results of the synthesis studies were generally comparable and suggested the NOACs had similar efficacy, although some differences were identified depending on the outcome. The extent to which the differences in the distribution of TTR, CHADS2 or primary versus secondary prevention biased the results remains unclear. PCV19 Targeted Literature Review Of Unmet Need In The Hyperlipidaemia Population With High Risk Of Cardiovascular Disease Mitchell S E 1, Roso S 2, Samuel M 1, Woods M S 1, Pladevall-Vila M 3 1RTI Health Solutions, Manchester, UK, 2Pfizer Ltd, Surrey, UK, 3RTI Health Solutions, Barcelona, Spain . . . . . . . Objectives: To examine recommended target levels of low-density lipoprotein cholesterol (LDL-C) for hyperlipidaemia patients at high risk (i.e., with two or more risk factors or coronary heart disease or its risk equivalents) for cardiovascular disease (CVD); to determine the proportions of patients who do not achieve targeted LDL-C levels in real-world setting studies. Methods: A targeted literature review identified guidelines and real-world studies that analysed hyperlipidaemia patients who were not at goal (as defined by study). MEDLINE, Embase, the Cochrane Library, and BIOSIS databases were searched. Guideline publications were searched from 2008; observational studies were searched from January 2005 to December 2013. There were no language or geographical restrictions. Results: 17 guidelines and 70 observational studies were included in the review. While country-specific guideline recommendations vary slightly, the commonly used European Atherosclerosis Society and European Society of Cardiology (EAS/ESC) guidelines recommend a LDL-C target of < 2.5 mmol/L for patients with high CVD risk. Most studies reported that between 61.8% and 95.4% of high-risk patients did not reach this target. 3 studies from North America reported lower proportions, between 18.9% and 42.3%. The EAS/ESC guidelines recommend a LDL-C target of < 1.8 mmol/L for patients with very high CVD risk. Studies reported that 68.1% to 96.0% of patients do not achieve this goal. Conclusions: Patients in higher cardiovascular-risk categories tend to have more stringent LDL-C target levels, which may contribute to failure to achieve target levels. This suggests several unmet needs: large numbers of patients who fail to achieve LDL-C targets, reducing the patients’ risk for CVD, and consequently reduce the occurrence of cardiovascular events. Based on recently published American College of Cardiology and American Heart Association guidelines, which do not recommend a treatment target LDL-C level, further research is needed to re-evaluate the unmet need in hyperlipidaemia patients. PCV20 Study On Drug Utilization And Assessment Of Stroke Risk Using Chads2 And Cha2ds2-Vasc Scoring In Elderly Patients With NonValvular Atrial Fibrillation Raj T 1, Bonthu S 1, Mallayasamy S R 2 University, Manipal, India, 2Manipal College of Pharmaceutical Sciences, Manipal University, Manipal, India . . . . 1Manipal Objectives: Stroke Risk Stratification in AF patients of can be done using CHADS2 (Congestive heart failure, Hypertension, Age ≥ 75, DM, prior Stroke/TIA [2 points]); or CHA2DS2VASc2 (Congestive heart failure/left ventricular ejection fraction ≤ 35%, Hypertension, Age ≥ 75 [2 points], DM, prior Stroke/TIA/thromboembolism [2 points], Vascular disease, Age 65–74, Sex- female). Treatment options for Prevention of stroke includes Anti-coagulants (Vitamin K Antagonist-Warfarin, Acenocoumarol; and Newer Oral Anticoagulant- Dabigatran) and anti-platelets (Aspirin and Clopidogrel). The objective of this study was to assess better tool for Stroke Risk Stratification; CHADS2 vs CHA2DS2VASc2 and to observe utilization pattern of antithrombotics with stroke as the outcome. Methods: Elderly patients (Age> 65yrs) with Non-Valvular Atrial Fibrillation admitted in the hospital within span of 2yrs (2012-13) were selected excluding patients with comorbidities like Atrial flutter, DVT, PFO, Endocarditis and/or ARF (after approval of ethical committee). Total of 160 patients were segregated based on stroke risk and percentage of patients experiencing thromboembolic event in each group was observed and CHADS2 and CHA2DS2-VASc were compared. The efficacy of antithrombotics in prevention of thromboembolic event in patients with AF was studied. Results: For stroke risk stratification, CHA2DS2-VASc was observed to be a better tool than CHADS2 to predict ‘truly low risk’, ‘moderate risk’ and ‘high risk’ patients. A shift of AF patients from ‘low-moderate risk’ by CHADS2 to ‘high risk’ by CHA2DS2VASc was noticed, 95% of patients required anticoagulation (either VKA or NOACs) as per CHA2DS2VASc, whereas, only 60% required OACs as per CHADS2. Most patients who experienced CVA belonged to ‘No antithrombotics prescribed’ group (25%). Dabigatran showed no incidence of CVA outcome, followed by VKA-, (Warfarin-28% and Acenocoumarol-18%) and least efficacy was seen by Antiplatelets-30%. Conclusions: CHA2DS2-VASc showed better prediction than CHADS2 for stroke risk prediction. Dabigatran was observed to have better outcome followed by VKA and Anti-platelets. PCV21 Management Of Chronic Thromboembolic Pulmonary Hypertension: Clinical And Reported Outcomes From A Referral Hospital In Spain Escribano P 1, Del Pozo R 1, Cuervo J 2, Rebollo P 2, Alvarez M P 3, Espinós B 3, Vieta A 3, LópezGude M J 1, Cortina J 1 1Hospital Universitario 12 de Octubre, Madrid, Spain, 2LASER ANALYTICA, Oviedo, Spain, 3Bayer Hispania, Barcelona, Spain . . . . . . . . . . . Objectives: To evaluate the management of Chronic Thromboembolic Pulmonary Hypertension (CTEPH) in a referral hospital by assessing clinical variables, patientreported outcomes and caregivers’ burden. Methods: An observational, retrospective study was conducted. All patients (aged > 18 years) attending the specialised unit on CTEPH at the 12 de Octubre Hospital (Spain), between January 2010 and November 2012, were offered to participate. Clinical variables were recorded at the clinical session for treatment decision (Pulmonary endarterectomy –PEA- if operability was confirmed or medication therapy –MT- if inoperable), and after one year. Outcomes considered: The New York Heart Association Functional Class (FC), 6-Minute Walking Distance, pulmonary arterial pressure, pulmonary vascular resistance and pro-brain natriuretic peptide. Participants completed the EQ-5D and caregivers’ fulfilled the Zarit Burden Interview. Differences between groups were studied (Chi-squared, Mann-Whitney U and ANCOVA). Results: A total of 64 CTEPH cases (57.8% males) were included. Mean (SD) age at diagnosis was 55.8 (14.9) and 67.2% had an III-IV FC at diagnosis. At the moment of treatment prescription, differences in clinical variables were not found (all p> 0.4) between PEA (n= 35-54.7%-) and MT groups (n= 29-45.3%-). After 12 months, 8 patients died (2 in PEA group and 6 in MT). Among survivors, FC was significantly better in PEA group (93.9% improved at least one level). Regarding EQ-5D, patients undergoing PEA showed significant higher utilities (0.83-0.17- vs. 0.53-0.31-p= 0.007) and VAS values (80.22-14.24- vs. 49.47-20.68-p< 0.001). Furthermore, mean VAS values in PEA group were comparable to general population (adjusted by sex and age). Finally, formal care was needed by just 4.8% of patients in PEA versus 33.3% in MT. Reported caregivers’ burden were relatively low in both groups (p= 0.87). Conclusions: The positive outcomes obtained, especially in those patients undergoing PEA, suggest the experienced management of CTEPH by this referral hospital and highlights the importance of detecting candidates for PEA. PCV22 The 3.5-Year Mortality Impact Of Drugs In Secondary Prevention Of Myocardial Infarction In Real-Life (Interim Analysis Of The Eole Cohort) Droz C 1, Dureau C 2, Thomas D 3, Danchin N 4, Tricoire J 5, Benichou J 6, Paillard F 7, Hercberg S 8, Sibon I 9, Rouanet F 9, Rambelomanana S 2, Maizi H 2, Bernard M A 2, Blin P 2, Moore N10 1INSERM CIC Bordeaux CIC 1401, Univ. Bordeaux, INSERM U657, Bordeaux, France, 2INSERM CIC Bordeaux CIC1401, Univ. Bordeaux, Bordeaux, France, 3Hôpital Pitié Salpétrière, Paris, France, 4Hôpital Européen Georges Pompidou, Paris, France, 5Clinique Ambroise Paré, Toulouse, France, 6CHU de Rouen, INSERM U657, Rouen, France, 7CHU de Pontchaillou, Rennes, France, 8INSERM U557, Bobigny, France, 9CHU de Bordeaux, Bordeaux, France, 10INSERM CIC Bordeaux CIC1401, Univ. Bordeaux, INSERM U657, CHU Bordeaux, Bordeaux, France . . . . . . . . . . . . . . . Objectives: Few studies have assessed the real-life impact of secondary prevention drugs on all-cause mortality post-myocardial infarction (MI), especially in countries with low incidence of MI. The objective of this interim analysis after 3.5-year of followup was to assess the real-life all-cause mortality impact of drugs reimbursed for MI secondary prevention in France: acetylsalicylic acid (ASA), anti-platelet agents (APA), beta-blockers (ß-), angiotensin converting enzyme inhibitors (ACEI), statins, and omega-3 supplementation (Om3). Methods: Cohort study of patients with recent (≤3 months) acute MI included by hospital and non-hospital cardiologists, with 6-year follow-up. Vital status was obtained from the National death registry, and failing that by patient/relatives/physicians investigation. Drug exposure was defined using both physician and patient reports at inclusion. Cox proportional hazard model was used to estimate for each drug, mortality hazard ratio (HR) of exposed versus non exposed patients, adjusted for gender, age, cardiovascular risk factors, other MI prevention drugs, and propensity score to be exposed at inclusion. Results: Between May 2006 and June 2009, 596 physicians included 5538 patients: mean age 62.1 years, 77.6% male, 9.6% current smokers, 14.5% diabetic, 44.6% hypercholesterolemic, 43.6% hypertensive, 8.2% with LVEF < 40%. At inclusion, 97.5% were exposed to ASA, 91.0% to APA, 89.7% to ß-, 71.1% to ACEI, 92.0% to statins, and 15.7% to Om3. The 3.5-year mortality was 7.8% (95%CI [7.1%-8.5%]) with an incidence rate of 23.2 per 1000 patient-years. Adjusted HR were: 0.98 [0.60-1.61] for ASA, 0.86 [0.60-1.24] for APA, 0.84 [0.63-1.11] for ß-, 0.80 [0.61-1.03] for ACEI, 0.67 [0.45-1.00] for statins, and 0.82 [0.58-1.16] for Om3. Conclusions: The 3.5 year interim all-cause real-life death reduction point estimates were close to those of large randomized controlled trials, except for ASA, for which almost all patients were exposed. The study’s statistical power will be sufficient to confirm or not these trends at the final 6-year analysis. PCV23 A Database Analysis Of Patients Eligible For Second-Line LipidLowering Treatment For Hypercholesterolaemia In England Amber V 1, Jameson K 1, Das R 1, Baxter C 1, Watson L 2 1MSD Ltd., Hoddesdon, UK, 2Epi Pharmaco Ltd., Buxton, UK . . . . . Objectives: In 2012, the NHS Health and Social Care Information Centre (HSCIC), with support from NICE, reported on the eligible population for ezetimibe as a second-line lipid-lowering therapy (LLT) in England. Several populations were omitted from this analysis, including some very high-risk Type 2 Diabetes Mellitus (T2DM) patients with CVD. We re-evaluated the eligible population for ezetimibe indicated for treatment intensification in a retrospective analysis. Methods: Patients with ≥ 1 total cholesterol (TC) measure in each year of interest were iden- A477 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 tified from the Clinical Practice Research Datalink (CPRD). From this cohort, three groups of patients were identified and counted for 2010, 2011 & 2012: Group 1 – very high-risk T2DM patients with CVD, TC ≥ 4.0 to < 5.0mmol/l and low-density lipoprotein cholesterol ≥ 2.0mmol/l despite statin treatment; Group 2 – untreated patients, with a TC ≥ 5.0mmol/l who were not prescribed any LLT after ceasing statin therapy; Group 3 – patients prescribed atorvastatin, rosuvastatin and/or ezetimibe with TC <5.0mmol/l who previously had TC ≥5.0mmol/l. Numbers were extrapolated to the population in England. Results: Of the general population in England in 2012, 8, 200, 699 (15.4%) patients had a TC test recorded, with similar proportions in 2010 & 2011 (15.4% and 15.3% respectively). Among the three groups defined in this analysis, a total of 305, 261 patients eligible for ezetimibe were not included in the 2012 estimates by HSCIC. This represents an 80.0% increase of the original estimate of 381, 797 patients using the original HSCIC methodology, corresponding to 33, 753, 141, 619, and 129, 889 patients in Groups 1, 2 and 3, respectively. Hence a year-on-year increase in the estimated eligible population were observed compared to the original HSCIC estimate; 64.8% increase in 2010 and 75.7% increase in 2011. Conclusions: A significant and increasing number of high-risk patients eligible for ezetimibe were missed in the HSCIC estimates during 2010-2012. PCV24 More Than One In Two Instances Of Venous Thromboembolism (Vte) Treated In French Hospitals Could Have Occurred During The Hospital Stay Allaert F A 1, Benzenine E 2, Quantin C 3 Biotech/CEN Nutriment, Dijon, France, 2university hospital, Dijon, France, 3University hospital, dijon, France . . . . 1CEN Objectives: describe the prevalence of venous thromboembolism (VTE), pulmonary embolism (PE) and deep vein thrombosis (DVT) without PE among all hospitalized patients and the percentages of those occurring during the hospital stays. Methods: Statistics are issued from the national PMSI MCO databases which are encoded using the CIM10. The codes used for VTE are I801 to I809 for DVT and codes I260, I269 for PE... Any stay with a the ICD-10 codes selected regardless of the Principal Diagnosis of Medical Unit Summaries and whatever its position (Principal, Related or Associated Diagnosis) was considered as a hospital-occurred thrombosis unless it was the Principal Diagnosis of the first Medical Unit Summary of the stay. To eliminate outpatient consultations or in day care, stays of < 48 hours were excluded. The term of hospital-occurred is preferred to hospital-acquired VTE suggesting a nosocomial origin which can be the case or not. Results: The results bear on the 18 683 603 hospital stays in 2010-2011. Out of 100 hospital stays involving VTE, for 40.3% VTE was the cause of hospitalization whereas 59.7% can be considered to have occurred during hospital-stay. These distributions are of 25.6% and 74.4% for DVT respectively 53.8% and 46.2% for PE. The age of patients varies little with whether VTE, DVT, and PE were hospital-occurred or not and are similar in men and women. The percentage of mortalities of these VTE is high and reaches 6.58% and the mortality from VTE, DVT, and PE is multiplied by a factor of 3 to 4 (p< . 0001) when hospital-occurred. Conclusions: The high proportion of hospital-occurred VTE is an alarming situation that should question the quality of prevention and/or its effectiveness. VTE prevention policies must be strengthened in hospitals for the sake of patients and health care savings alike. PCV25 Risk Factors Associated With Venous Thromboembolism Recurrence In A European Population Hamilton M 1, Gupta S 2, Goren A 3, Auziere S 4, Claflin A B 5, Reboul R 4, Phatak H 1 Squibb Company, Princeton, NJ, USA, 2Kantar Health, Princeton, NJ, USA, 3Kantar Health, New York, NY, USA, 4Kantar Health, Montrouge, France, 5Pfizer, Inc., New York, NY, USA . . . . . . . . 1Bristol-Myers Objectives: This study estimated venous thromboembolism (VTE) recurrence and associated risk factors in a European population, given limited data in this region. Methods: This retrospective cohort study included data from physicians (376 general practitioners and 307 specialists) in France, Spain, Italy, and Germany, who completed case report forms (2,184 patient records) for the next 3-4 patients seen in consultation for any reason who had an initial VTE event 3 to 24 months prior (i.e., patients surviving 90 days since initial VTE). All Anderson & Spencer individual risk factors (plus gender, bleeding, and initial VTE type, but excluding previous VTE and bed rest > 3 days) were entered into a Cox proportional hazard model accounting for censored data and predicting VTE recurrence. Backward stepwise regression was used to select a reduced final model. Results: Patients’ mean age was 61.3 years (SD= 14.3) and 47.4% were female. Of 2,184 patients, 379 developed recurrent VTE over 1,298 person-years of follow-up. The final model included: age=40+ (91.6%) vs. 18-39 years, varicose veins (26.4%), history of heart failure (5.9%), congestive respiratory failure (1.3%), arthroscopic knee surgery (2.2%), central venous line (0.9%), chemotherapy (6.1%), and orthopedic (3.1%) surgery. Significant, independent predictors of recurrence were: varicose veins (hazard ratio [HR]: 1.4; 95% confidence interval [CI] 1.1-1.8), central venous line (HR: 3.2; CI 1.5-6.8), congestive respiratory failure (HR: 2.4; CI 1.2-4.6), and heart failure (HR: 1.5; CI 1.1-2.2). Other factors, including age= 40+, knee surgery, chemotherapy, orthopedic surgery, and type of VTE (e.g., deep vein thrombosis vs. pulmonary embolism) did not exhibit significant associations with recurrence of VTE. Conclusions: VTE recurrence is high in this European population and associated with several independent risk factors. Targeted anticoagulant treatment post-initial VTE plus longer term prevention of recurrence are needed, including attention to risk factors that help differentiate patients more likely to experience recurrence. PCV26 Real World Incidences And Hospital Cost Of Venous And Pulmonary Thromboembolic Events In France Bouee S 1, Emery C 2, Samson A 3, Bailly C 4, Cotté F E 4 1Cemka, Bourg la Reine, France, 2Cemka, Bourg La Reine, France, 3Paris-Dauphine University, Paris, France, 4Bristol-Myers Squibb, Rueil Malmaison, France . . Objectives: To estimate the cumulative incidence and hospital cost for venous and pulmonary thromboembolic events in a real world setting in France. Methods: We conducted a retrospective analysis of the EGB database, a 1/97th random sample of the whole National health insurance database records linked to hospitalizations. All patients hospitalized in 2010 and 2011 with a diagnosis of deep vein thrombosis (DVT) or pulmonary embolism (PE) were included. Inpatients were identified through principal diagnosis of hospitalization stay. Outpatients with a DVT were identified by 1) an echo Doppler exam, 2) preceded or followed by a low molecular weight heparin or fondaparinux delivery (+/-7 days), and 3) a subsequent Vitamin K antagonists delivery (0 to 7 days). Incidences and annual hospital cost of DVT and PE were estimated and extrapolated to the overall French population, and cumulative proportions of recurrences were calculated. Results: For 2011, the estimated crude incidences were 141/100,000 (91,650 patients) for DVT, and 79.4/100,000 for PE (51,610 patients in France). Mean age of patients was 67.0+/-17.2 years for PE and 64.1+/-17.7 years for DVT. A majority of patients were females (57% in both groups). After index event (PE/DVT), the cumulative proportions of venous thromboembolic recurrences were 2.6% at 1 month, 3.7% at 3 months, 5.1% at 6 months and 6.7% at 12 months. The cumulative proportions of death after a PE and a DVT first event were 0.2% at 1 month, 1.1% at 3 months, 2.6% at 6 months and 6.2% at 12 months. Annual hospital cost of venous and pulmonary thromboembolic events was estimated at 712 million € (362 million € for DVT and 350 million € for PE). Conclusions: In 2011, around 143,000 patients suffered from venous and pulmonary thromboembolic events in France. Hospitalized events accounted for an important burden in France. . . . . PCV27 Cognitive Function And Non-Adherence To Antihypertensive Medications Li S S , Brondolo E , Dalrymple N , Schupf N , Kronish I M Columbia University, New York, NY, USA . . . . . . . Objectives: Non-adherence to blood pressure medications is common, present in 30-50% of patients, and known to be associated with an increased risk for major cardiovascular events and increased health care costs. Prior research suggests that impaired cognitive function is associated with medication non-adherence. Our aim was to determine if easily administered measures of cognitive function can be used to identify hypertensive patients at increased risk of medication nonadherence. Methods: A convenience sample of 101 primary care patients (n= 101) with uncontrolled hypertension was enrolled from two hospital-based clinics in ethnically diverse communities of New York City. Patients with overt dementia as noted by their primary care doctors were ineligible. Subjects completed three brief cognitive tests (≤ 5 minutes to complete each one): Trail Making Test-A, Trail Making Test-B, and the Symbol Digit Modalities Test. The primary outcome was adherence based on percentage of doses taken as prescribed, measured by a 4-compartment electronic pillbox (MedSignals). Multivariable logistic regression was used to determine if impaired cognitive function was associated with poor adherence after adjusting for age, gender, ethnicity, education, and total blood pressure medications. Results: Patients who were classified as impaired when screened by Trail Making Test-B had a three-fold (OR= 2.91,95% CI, 1.02-8.35) increased likelihood of non-adherence compared with those who were not impaired, adjusting for age, education, gender, ethnicity, and number of BP medications (p= . 05). Trail Making Test-A and Symbol Digit Modalities Test were non-significant predictors of adherence in both adjusted and unadjusted analyses. Conclusions: Trail Making Test-B, a measure of executive function, may be a useful screening tool to identify patients without overt dementia who are at risk for non-adherence to anti-hypertensive medications. The findings from this study may provide an opportunity to identify a tailored approach to medication adherence interventions. PCV28 Retrospective Analysis On Hospitalization And Health Care Costs, According To Serum Uric Acid Levels In Patients From A Sample Of Italian Local Health Units Degli Esposti L 1, Saragoni S 1, Buda S 1, Desideri G 2, Borghi C 3 1CliCon Srl, Ravenna, Italy, 2L’Aquila University, L’Aquila, Italy, 3Policlinico S. Orsola, University of Bologna, Bologna, Italy . . . . . Objectives: Hyperuricemia is an independent risk factor for gouty arthropathy, renal disease, atherosclerosis and cardiovascular diseases (CVD). The objective of this study was to explore the relationship between serum uric acid (SUA) levels and hospitalization events and assess health care costs. Methods: A retrospective analysis using a large administrative database and a clinical registry containing laboratory results was performed. Subjects, aged ≥ 18, were assigned to one of the 4 groups based upon the first SUA measurement between October 1, 2010 and September 30, 2011: ≤ 6 mg/dl [good-control], > 6 mg/dl and ≤ 7 mg/dl [fair-control], > 7 mg/dl and ≤ 8 mg/dl [poor-control], > 8 mg/dl [very-poor-control]. We calculated incidence rates to estimate the risk of hyperuricemia-related and CVD hospitalizations occurred until December, 2012. A Poisson regression model was used to assess the relationship between the number of CVD hospitalizations and SUA level. Total annual costs included all the pharmacological treatments and the direct costs due to hospitalizations and outpatient services. Results: Of 52,822 patients included, SUA level was ≤ 6 mg/dl -good-control- for 33,638 (63.7%) patients and > 6 mg/dl –suboptimal-control- for 19,184 patients (36,3%), of whom 60,7% with fair-control, 25,8% poor-control and 13,5% very-poor-control. Compared with good-control group, suboptimal-control patients showed an increased risk of hyperuricemia-related hospitalizations (unadjusted rate was 1,02 vs 0,43 per 1000 person-years, p< 0.001) and of CVD hospitalizations (unadjusted rate 5,09 vs 3,17 per 100 person-years, p< 0.001; adjusted incidence rate ratio 1.22, p< 0.001). Over one year, the mean total cost was: € 2,077.43 in good-control patients; € 2,079.87 in fair-control patients; € 2,296.39 in poor-control patients; € 3,295.41 in very-poor-control patients. Conclusions: The 36,3% of the patients in this study sample were at sub-optimal SUA control (>6mg/dl). This analysis indicates that higher hyperuricemia-related and CVD hospitalizations as well as total health care costs resulted associated with higher SUA levels. A478 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PCV29 The Association Between The Nature And Timing Of Dental Visits And C-Reactive Protein Levels PCV34 Statin Use And Risk Of Developing Diabetes In Cardiovascular Disease: Systematic Literature Review And Meta-Analysis Candrilli S D 1, Blackwood J 2 1RTI Health Solutions, Research Triangle Park, NC, USA, 2Danbury Hospital, Danbury, CT, USA Thakker D , Nair S R , Shukla H, Shaikh J S Capita India Pvt. Ltd., Mumbai, India Objectives: Evidence suggests an association between dental disease and cardiovascular disease (CVD). C-reactive protein (CRP), an inflammatory marker, has been implicated as a risk factor for CVD, and dental disease can affect CRP levels. Our study examined the relationship between the timing and nature of dental visits and CRP. Methods: Using data from the US-based 1999-2000, 2001-2002, and 2003-2004 National Health and Nutrition Examination Surveys, we examined the relationship between time since and reason for most recent dental visit and CRP among adults ≥ 20 years old. Participants were excluded if they were pregnant at the time of the survey, did not take part in the examination component of the survey, or were missing covariates for a logistic regression model: age, sex, race, BMI, HbA1c, WBC count, CRP measure, time since and reason for last dental visit, smoking status, cholesterol-lowering medication use, and history of asthma, cancer, rheumatoid arthritis, chronic bronchitis, or recent illness. A dichotomous elevated CRP measure was used, defined as CRP > 0.30 mg/dL. Time since last dental visit was categorized as < 6 months, 6 months to < 1 year, 1 year to < 2 years, and 2+ years ago; respondents who reported never visiting a dentist were placed in the 2+ years category. Results: A greater proportion of the normal (≤ 0.30 mg/dL) CRP group last visited a dentist < 6 months ago (P= 0.0460), and last visited the dentist for a “preventive” visit (P< 0.0001), while a greater proportion of the elevated CRP group last visited the dentist for a “symptom-driven” visit (P< 0.0001). Regression model results demonstrated that preventive visits are associated with a reduced likelihood of elevated CRP (OR= 0.722; P= 0.0120), regardless of the time since last visit. Conclusions: Given the apparent association between risk of elevated CRP and reason for the last dental visit, medical and dental providers should consider interventions specifically around appropriate dental care. Objectives: Statins are widely used for the primary and secondary prevention of cardiovascular diseases (CVDs). Studies have shown that statins may induce diabetes in non-diabetic CVD patients, as class effect. This systematic literature review aims to evaluate the risk of developing diabetes in CVD patients receiving statins. Methods: Randomized controlled trials (RCTs), which used any statin as an intervention for non-diabetic CVD patients, were identified from August 2010 to June 2014 in databases such as, Embase, PubMed, and Cochrane. The timeframe of the searches was selected post the study conducted by Mills et al. in 2011. This review will also include relevant studies from the Mills et al. study until August 2010. Two researchers will independently review studies as per the Cochrane methodology for systematic reviews. The primary outcome is the incidence of diabetes. Subgroup analyses will also be performed to assess whether statin type, age, ethnicity or patient groups contribute to the intensity of the risk of developing diabetes. Results: In total, 5238 potentially relevant studies were retrieved from the databases and are being screened for inclusion in the review. The data extraction and analyses (both qualitative and quantitative) are being performed and the full results will be presented in the poster. Conclusions: This systematic literature review is an update of the findings of a previous study and will hopefully throw more light on the association between statin use and the risk of diabetes in CVD patients, with special emphasis on subgroups. . . . . . . . . PCV35 A Retrospective Study Of Mortality In Risk Patients With High Dose Statin Usage And No Statin Usage Rockberg J 1, Jørgensen L 1, Taylor B 2, Kilpatrick R D 2, Sobocki P 1 Health, Stockholm, Sweden, 2Amgen Inc, Thousand Oaks, CA, USA . . . . . . 1Pygargus/IMS PCV30 Traditional And Non-Traditional Risk Factors For Cardiovascular Disease In Type 2 Diabetes: Systematic Review Of Longitudinal Studies Smith-Palmer J 1, Bae J P 2, Boye K S 2, Perez-Nieves M 2, Valentine W J 1 Health Economics and Communications, Basel, Switzerland, 2Eli Lilly and Company, Indianapolis, IN, USA . . . . . . . . 1Ossian Objectives: A systematic literature review of longitudinal studies across a broad range of both general and disease specific populations was performed to understand characteristics of patients with type 2 diabetes (T2D) with risk factors for cardiovascular (CV) events and whether the magnitude of CV risk varies across different T2D populations. Methods: MeSH term based literature searches were performed in the PubMed, EMBASE and Cochrane Library databases. For inclusion, studies (in any disease area) were required to have a minimum of 1,000 T2D patients and minimum 5 years follow up. After 2 rounds of review a total of 52 articles and 2 meta-analyses were included. Results: Twenty nine articles were described as prospective studies; 13 articles were from Europe, 20 from Asia Pacific, 13 from North America, 2 from the Middle East and 4 were multinational. Six articles were sourced from general or nondiabetes specific data and 46 articles were based on diabetes studies. Several publications were from large scale diabetes registries (e.g. Hong Kong Diabetes Registry [n=6] and the Swedish National Diabetes Register [n=4]). The review also identified several post-hoc analyses from clinical trials and large scale retrospective observational/ database analyses. Risk factors investigated included well characterized markers in T2D patients (e.g. HbA1c, blood pressure, lipids, proteinuria [n= 20]), less well characterized factors including dietary components (n=7), biochemical measures (e.g. serum uric acid, fibrinogen; n= 6), depression (n= 6) and poorly characterized risk factors including oral health (n=1), erectile dysfunction (n=1) and presence of H. pylori (n=1). Conclusions: Analysis of the literature showed that in addition to traditional risk markers, factors such as depression, erectile dysfunction, poor oral health are also independent risk factors for CV disease. These factors should be taken into account when estimating the CV risk profile for patients with T2D. PCV31 Relation Of The Time In Therapeutic Range (Ttr) Of Warfarin To Bleeding Incidences In Patients With Atrial Fibrillation Jasmi M A S 1, Mazlan M N 1, Shaharuddin S 1, Zulkifly H H 1, Long C M 1, Hashim R 2, Abdul Wahab M S 1 1Universiti Teknologi MARA, Selangor, Malaysia, 2Cyberjaya University College of Medical Sciences, Cyberjaya, Malaysia . . . . . . . . . . . . . Objectives: Warfarin use in atrial fibrillation has been established for preventing occurrence of stroke in patients with atrial fibrillation. However, safety and clinical monitoring of warfarin use is crucial due to its risk of bleeding complications. This study aims to compare and establish relation of time in therapeutic range (TTR) of warfarin in patients with atrial fibrillation in the first 6 months and 6 month thereafter of anticoagulation therapy. Methods: This is a retrospective study carried out at a tertiary-care hospital with anticoagulation clinic in the state of Selangor, Malaysia. Data collected included patients’ demographics, co-morbidities, and international normalized ratio (INR). TTR were determined using Rosendaal method based on records found in database (INR Desk 4.0 system) and patients’ hemorrhage events were also recorded. Samples of the study were patient who started warfarin from January 2009 until March 2013. Results: A total of 167 patients with atrial fibrillation were enrolled and only 6% (n= 10) achieved TTR of more than 75% for the first 6 months of warfarin use as compared to 16.8% (n= 28) of TTR more than 75% 6 months thereafter. As for bleeding incidences, 29% (n= 45) of patients in the group of TTR less than 75% in the first 6 months had bleeding complications as compared to 18.7% (n= 26) in patients of TTR less than 75% 6 months after. Conclusions: A more regular follow up is necessary during the first 6 months of new warfarin users as they tend to be out of the TTR and have a higher bleeding risk. Objectives: The aim of this retrospective cohort study was to calculate hazard rates of CV/non-CV related death in risk patients stratified by statin dose and to generate data for health-economic analyses. Methods: Anonymous retrospective electronic medical records were extracted from a 10% sample of the Swedish population > 18 years in primary care and merged with hospital records, prescribed drugs, and death data by National Board of Health and Welfare. Cohorts were defined as diabetic (DM), Clinically-evident CardioVascular Disease (CeCVD), Acute Coronary Syndrome (ACS), Heart Failure (HF) and Ischemic Stroke (IS) patients. CV/non-CV related death was investigated. The population was divided into high dose statin users (> 40 mg simvastatin or equivalent) and non-statin users. Royston Parmar (RP) spline analysis was used to calculate hazard rates of both outcomes in each cohort to enable a smooth hazard function with good fit to data compared to ordinary Cox regression. Initial AIC-based optimization demonstrated that modeling of the log cumulative odds as a spline function of log time and three nodes gave the best fit. Crude rates were calculated for each cohort and event, and hazard rates adjusted for gender, diabetes, LDL (where applicable) and age were calculated. Results: The total database consisted of 1.3 million patients; there were 55,778 DM, 47,581 CeCVD, 49,857 ACS, 82,835 HF and 38,949 IS patients. High dose statin users showed after initial decrease a constant/marginally increasing hazard rate, whereas the no statin group after initial decrease had a strong increase in hazard rate versus time. The overall hazard rate was always higher for the no statin group. Independent of statin status diabetics and males demonstrated an at least numeric elevated hazard rate. Conclusions: RP splines could be used to generate hazard rate functions for cost-effectiveness models. The death rate was lower in patients with high dose statin usage. Cardiovascular Disorders – Cost Studies PCV36 Budgetary Impact Analysis Of Reimbursement Varenicline In The Smoking Cessation Treatment Of Patients With Cardiovascular Diseases, Chronic Obstructive Pulmonary Disease Or Type-2 Diabetes Mellitus: A National Health System Perspective In Spain Rejas Gutiérrez J 1, Sicras Mainar A 2, Navarro Artieda R 3, De Lossada Juste A 4 1Pfizer S. L. U., Alcobendas/Madrid, Spain, 2Badalona Serveis Assistencials SA, Badalona (Barcelona), Spain, 3Hospital Germans Trias i Pujol, Badalona (Barcelona), Spain, 4Pfizer, S. L. U, Alcobendas/Madrid, Spain . . . . Objectives: Varenicline is indicated for smoking cessation. Around 60.000 people die every year in Spain because of tobacco related diseases, particularly cardiovascular diseases (CVD), chronic obstructive pulmonary disease (COPD) and type-2 diabetes mellitus (t2DM). At present, varenicline is not reimbursed in the Spanish National Health System (NHS). The objective was to estimate the budgetary impact (BI) for the NHS of the reimbursement of varenicline in smoking cessation in patients with CVD, COPD or t2DM. Methods: BI was estimated comparing the actual not-reimbursed scenario versus a reimbursed scenario using the Spanish NHS perspective. A hybrid BI model was designed using epidemiological data to estimate size of populations and a Markov modelling simulating until four quitting attempts to estimate smoking cessation rates with varenicline during a 5-year horizon. Costs of cessation attempts were considered in the reimbursement scenario only, and included varenicline, medical visits and counselling. Effectiveness of varenicline was expressed as one year of abstinence rate following a 12-weeks course of standard doses, and were derived from clinical trials. Cost savings due to smoking cessation were extracted from local published cost-of-illness studies in such populations. Results showed incremental cost-savings of reimbursed versus not reimbursed scenario. Univariate sensitivity analysis was also applied. Results: Five-years cumulated cessation attempts increased from 140,795 in the not-reimbursed to 354,631 in the reimbursed scenario, yielding to 52,127 extra subjects A479 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 quitting smoking. Cost of 5-year reimbursement varenicline was estimated to be € 63.0 millions, while smoking cessation avoided costs reached € 99.9 millions, which compared with € 21.1 millions savings in the not-reimbursed scenario: a net incremental cost-saving of € 15.9 millions. Savings were observed since 3rd year of modelling. Conclusions: The BI of the reimbursement of varenicline in smoking cessation is a cost-effective health policy in the Spanish NHS, and could produce cost-savings since the 3rd year of implementation. PCV37 Budget Impact Analysis Of Hypertensive Treatment With Indapamide And Amlodipine Single-Pill Combination In The Polish Setting Kawalec P 1, Stawowczyk E 2, Holko P 2, Borowiec L 3, Filipiak K J 4 University Medical College, Krakow, Poland, 2Centrum HTA Sp z o. o. Sp. komandytowa, Kraków, Poland, 3Medical Department, Servier Poland Ltd, Warsaw, Poland, 41st Chair and Department of Cardiology, Medical University of Warsaw, Warsaw, Poland . . . . . . 1Jagiellonian Objectives: The aim of this study was to calculate and compare public payer and patients’ costs of hypertensive treatment with indapamide 1.5 mg and amlodipine 5 mg or 10 mg single-pill combination (SPC) and free combination (FC), in the Polish setting. Methods: The analysis compared two scenarios: existing and new. The existing one assumed treatment with FC of indapamide 1.5 mg and amlodipine 5 mg or 10 mg. The new one also included treatment with SPC of indapamide 1.5 mg +amlodipine 5/10 mg. Population and market shares were estimated on the basis of published reimbursement data, experts’ opinion and validated with available epidemiological data. Cost data were analysed from the public payer perspective (National Health Fund) and from patient perspective, in a three-year horizon. SPC cost is based on average pharmacy price reported in April 2014 (18.13PLN and 19.75PLN respectively for 1.5+5mg and 1.5+10mg /30 tabs); 30% patient copayment was assumed. The cost of FC was calculated as an average cost of reimbursed indapamide and amlodipine products in corresponding doses. All costs present 2014 values, and are expressed in Polish zloty (PLN). Average monthly exchange rate of May 2014 was applied (1EUR=4.1790PLN). Difference in clinical effectiveness between SPC and FC was also included, in the form of cardiovascular events risk. Results: Introduction of indapamide/amlodipine SPC on the reimbursement list next to FC brought savings from public payer perspective and from patient perspective amounting to: 509,255PLN (121,860EUR) and 5,893,941PLN (1,410,371EUR) in first year, 689,239PLN (164,929EUR) and 7,833,005PLN (1,874,373EUR) in second year, 725,965PLN (173,717EUR) and 8,328,480PLN (1,992,936EUR) in third year. Additionally it resulted in avoidance of 808 cardiovascular events in the three-year horizon. Conclusions: Treatment with indapamide/amlodipine SPC in comparison to FC generates significant savings both from the public payer perspective and from patient perspective in contemporary Polish setting, and reduces cardiovascular events. PCV38 Modeling The Impact Of A Digital Health Feedback System In Uncontrolled Hypertensive Patients Kim Y A , Virdhi N , Raja P , DiCarlo L Proteus Digital Health, Redwood City, CA, USA . . . . . Objectives: Despite the availability of numerous therapeutic agents and management tools, half of all hypertensive patients do not have their blood pressure (BP) at goal. A model was developed to estimate the incremental costs of uncontrolled vs. controlled hypertension and the impact of shifting patients to controlled status via a unique digital health feedback system. This Proteus system utilizes an Ingestible Sensor to determine medication-taking patterns, and a wearable 7-day sensor in the form of an adhesive patch to collect physiological and behavioral metrics such as heart rate, step count, and patterns of activity and rest, providing a means of determining non-response vs. non-adherence to prescribed medications and recommendations for daily routine. Methods: The additional costs of outpatient services, monitoring, and cardiovascular complications were calculated for uncontrolled vs. controlled hypertensive patients from a US plan perspective for a 1 year time horizon. The clinical and utilization assumptions were derived from the literature and expert opinion, and costs were derived from the Medicare Fee Schedule and AHRQ databases. The impact of the Proteus system on BP control was based on a real-world study evaluating this technology in 164 patients with a history of uncontrolled hypertension. Results: In a health plan of 1 million members, 7.9% (78,656) were uncontrolled hypertensive patients receiving care who were eligible for the Proteus system. The direct annual medical costs of uncontrolled hypertension were estimated to be $60.9 million over the costs for controlled disease. The Proteus system was estimated to result in $7.3-18.3 million in savings ($328-$717 per BP at goal), and lead to a 3-9% reduction in the number of coronary artery disease and stroke events in one year. Conclusions: Even in the short-term, a digital health feedback system appears to provide an effective way to mitigate the substantial costs of uncontrolled hypertension. PCV39 Budget Impact Analysis Of Apixaban Versus Other Noacs For The Prevention Of Stroke In Italian Non-Valvular Atrial Fibrillation Patients Pradelli L 1, Calandriello M 2, Di Virgilio R 3, Bellone M 1, Tubaro M 4 HE&OR, Turin, Italy, 2HE OR Unit - Bristol-Myers Squibb S. r. l., Rome, Italy, 3Pfizer Pharmaceuticals, Rome, Italy, 4San Filippo Neri Hospital, Rome, Italy . . . . . 1AdRes Objectives: This study aims to perform a budget impact analysis of the use of three available novel oral anticoagulant agents (NOACs) for preventing thromboembolic events in Italian patients with non-valvular atrial fibrillation (NVAF). Methods: Estimated Italian population of patients is run through a decision tree/Markov model simulating their treatment with the available therapeutic options: dabigatran at two dose levels (110 mg/bid for the over 80 years old, 150 mg/bid for younger NVAF patients), rivaroxaban, and apixaban. Effectiveness estimates derive from an adjusted indirect treatment comparison using warfarin as link. Epidemiological data and unit costs are collected from Italian published sources. The budget impact analysis evalu- ates the financial impact of apixaban introduction by comparing expected 1,2, and 3 years costs in hypothetical scenarios: with and without apixaban. Italian NVAF patient population estimation is based on official apixaban reimbursement criteria, applying the characteristics of the trial population to national epidemiologic data. Sensitivity analysis is performed on an alternative non-experimental population of NVAF patients. Results: Among available NOACs, apixaban is expected to be the least expensive at 1,2, and 3 years in an estimated patient population of 364,000 Italian patients, allowing for savings of over 5 million € by the third year. Results of the simulation run on an alternative non-experimental population of NVAF patients yields comparable estimates. Exclusive use of apixaban for three years in the identified population would allow for savings of € 8,832,500, € 14,446,551 and € 27,282,998 when compared with dabigatran (110 mg), dabigatran (150 mg) and rivaroxaban, respectively. Conclusions: The different safety and effectiveness profiles of the available NOACs emerging from the adjusted indirect comparison indicate that the introduction of apixaban could improve health care expenditure control while maintaining or increasing therapeutic appropriateness in the Italian NVAF population. PCV40 The Budget Impact Of New Generation Ct Scanners For Difficult-ToImage, Low-Risk Patients With Suspected Cad Shields G E , Chapman A M BresMed, Sheffield, UK . . . . Objectives: The National Institute of Health and Care Excellence (NICE) issued diagnostic guidance on new generation computed tomography (CT) scanners recommending them as an option for the first-line imaging of coronary arteries in patients with suspected low-risk coronary artery disease (CAD) in whom imaging with old generation scanners is difficult (e.g. obese patients). The capital investment for a new generation scanner is considerably more than a standard scanner, which could hamper implementation. Based on the NICE guidance, a model was designed for use as a planning tool for rapid access chest pain clinics (RACPCs) looking to replace their current scanner with a new generation scanner. Methods: An Excel® model was developed to estimate up to a 10-year impact of acquiring a new generation scanner. It was assumed that under standard care low-risk, difficult-to-image patients would be referred for a diagnostic invasive coronary angiography (ICA). Under the new diagnostic pathway these patients can be scanned with a new generation scanner. Whenever possible the NICE guidance was used to guide assumptions and populate default values. Results: The model estimates that for each difficultto-image patient a new generation scanner has the potential to save approximately £946.62 in diagnostic costs. Considering the capital investment required, a RACPC looking to replace their standard scanner and considering implementing a new generation CT scanner only need 53 difficult-to-image patients per year to see a positive return on investment over a 10-year period. Conclusions: The model is likely to be conservative as it focuses on difficult-to-image patients only, yet the scanner is available for all patients who will likely benefit from the better sensitivity and specificity associated with the new scanners. However, it highlights that even a low number of these difficult-to-image patients will result in a positive return on investment over the expected life-time of the scanner. PCV41 Simvastatin Plus Fenofibrate As A Fixed Dose Combination In The Treatment Of Mixed Dyslipidemia In Greece: Budget Impact Analysis Relakis J 1, Kourlaba G 2, Maniadakis N 1 School of Public Health, Athens, Greece, 2National and Kapodistrian University of Athens School of Medicine, Athens, Greece . . . 1National Objectives: To evaluate the affordability of switching patients already treated with the multi-pill therapy of simvastatin and fenofibrate to the first simvastatin and fenofibrate fixed dosed combinations (FDC) product, for the management of mixed dyslipidemia in the Greek health care setting. Methods: A budget impact model was locally adapted. The analysis was conducted from a third-party payer perspective over a time horizon of 3 years. The population with mixed dyslipidemia in Greece, the market shares of available treatments and the corresponding drug acquisition costs were combined to estimate the total budgetary impact that will result from the penetration of FDC in the Greek market. Data on population with mixed dyslipidemia were derived from the National Statistical Service and published literature. Estimates of the current and future market shares were obtained from Abbott Hellas market research. Drug acquisition costs were calculated using the latest price bulletin issued and the corresponding reimbursement prices. Reimbursement prices were reduced by the patient’s relevant co-payment and relative rebates. Since market prices for the FDC are not available yet in Greece, estimated retail prices provided by Abbott Hellas were considered [FDC 20/145: € 13.70 (€ 12.22-16.27) and FDC 40/145: € 16.71 (€ 15.22-19.23)]. Results: Savings in pharmaceutical reimbursement on year 1 were estimated at 146,974 (€181,084-€88,503) decreasing the relevant cost by 5.87% (7.23%-3.53%). On year 2, savings of € 250,544 (€ 322,536-€ 136,474) were attributed to the penetration of FDC, lowering the budget by 9.93% (12.78%-5.41%). On the 3rdyear, savings were estimated at € 403,405 (€ 509,489-€ 221,565) reducing spending by 15.82% (19.98%-8.69%). On average, over the 3-year time horizon of the analysis, the addition of FDC was found to decrease reimbursement costs by 10.54% (13.33%-5.88%) generating savings of € 266,974 (337,703€ -€ 148,147). Conclusions: The introduction of FDC reimbursement will result in a predictable budget impact which is expected to decrease the relevant pharmaceutical cost for national health insurance. PCV42 Budget Impact Analysis Of Botulinum Toxin A Therapy For Upper Limb Spasticity In Germany Zoellner Y F 1, Gerwe M 2, Richter M 2 1Hamburg University of Applied Sciences, Hamburg, Germany, 2Ipsen Pharma, Ettlingen, Germany . . . . Objectives: Upper limb spasticity (ULS) secondary to stroke has a considerable patient and caregiver burden, particularly with regards to pain, activities of daily A480 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 living and mobility. Botulinum neurotoxin-A (BoNT-A) injections are effective in treating ULS. We aimed to calculate annual BoNT-A treatment cost of ULS on a per-patient basis, and the expected overall annual budget impact of BoNT-A treatment in Germany using static and dynamic market share scenarios. Methods: A budget impact model for BoNT-A use, adopting a German health care system perspective, was developed. Two market-share scenarios were modelled over 5 years. While the static scenario assumed current market shares (abobotulinumtoxinA, 36%; onabotulinumtoxinA, 26%; incobotulinumtoxinA, 38%) to remain constant over time, the dynamic scenario assumed market share of abobotulinumtoxinA to rise up to 76% across 5 years. Epidemiologic data inputs were sourced from the most recently published literature, unit costs for BoNT-As from the Lauertaxe (pharmacy purchase price [PPP] level), and market share assumptions from IMS Health (Market Sizing). Equivalence of the 3 BoNTAs in terms of efficacy and safety is assumed and therefore no differential use of medical services or other goods. Results: Annual BoNT-A drug costs per ULS patient were 3 463€ , 5 603€ , and 5 410€ , respectively, with prescribing patterns following SmPC recommendations for abobotulinumtoxinA, onabotulinumtoxinA, and incobotulinumtoxinA. The total drug costs were decreased by between 1 231 861€ (-4%) in year 2 and 5 133 684€ (-17%) in year 5 by shifting the market share to abobotulinumtoxinA. Sensitivity analyses showed that the number of patients treated with BoNT-As, time to re-injection, and dose per injection were the most influential parameters on budget impact, impacting both drug acquisition costs and physician visits. Conclusions: Increased use of abobotulinumtoxinA compared with incobotulinumtoxinA and onabotulinumtoxinA for ULS in Germany could potentially reduce the total cost of treatment. PCV43 Renal Denervation With The Symplicity Catheter System For Treatment-Resistant Hypertension: A Budget Impact Analysis Naclerio M 1, Corbo M 2, Beccagutti G 2 1Medtronic Italia S. p. a, Sesto San Giovanni (MI), Italy, 2Medtronic Italia, Sesto San Giovanni (MI), Italy . . . Objectives: Renal denervation (RDN) is a new treatment option for patients with uncontrolled resistant hypertension. Studies have shown RDN to have a potential reduction in systolic blood pressure and therefore reduce risk for different cerebrocardiovascular disease (stroke, ischemic heart disease, heart failure, etc.). The aim is to assess the economic impact of renal denervation in resistant hypertensive patients based on the potential clinical benefit demonstrated in literature. Methods: A budget impact model was developed to quantify RDN-related cost-savings accomplished through lower rate of cerebro-cardiovascular events compared with the medical treatment cohort (with 3+ anti-hypertensive medications including a diuretic). The model was developed from the Italian National Healthcare Service (NHS) perspective with a 10-year time horizon. The risk of events was calculated using different multivariate equations. A literature review was carried out to collect the treatment costs for both acute and chronic phase of hypertension-related events. The cost of RDN procedure was approximated to the related DRG tariff. Results: Renal denervation, considering the blood pressure reduction observed after procedure, substantially reduces cerebro-cardiovascular event probabilities by 30% if compared to the medical treatment cohort. Assuming a yearly 2% RDN adoption rate, approximately 2,000 events would be avoided in 10 years allowing total savings up to 9 Million euro. The higher cost of new technology was offset from potential saving associated to the avoided events. Compared to the total costs associated to the medical treatment cohort in 10 years the overall expenditure for RDN was greater than 0.8%. The annual incremental cost for patients treated with surgical approach was estimated at 16 euro per patient. Conclusions: The economic impact of hypertension-related complications is significant in the Italian setting. Consequently the potential savings according to the NHS perspective derived from the implementation of strategies aimed at improving the level of hypertension should be considered. PCV44 Dabigatran versUS Dicoumarins In Non-Valvular Atrial Fibrillation. Budget Impact Study In The Extremadura Public Health System Gemio Zumalave P 1, Parejo Hernández E 1, Alvárez Fernández M 1, Montaño Pérez L M 1, Sánchez Chorro J L 1, Hidalgo A 2 1Servicio Extremeño de Salud, Mérida, Spain,, 2Instituto Max Weber, Majadahonda, Spain . . . . . . of the study is that the safety data were obtained from clinical trials rather than actual clinical practice with these drugs. PCV45 Estimating The Value Of Cangrelor From Eliminating Preloading In Coronary Artery Bypass Graft (Cabg) Patients Bay C , Cyr P L , Jensen I ICON plc, Cambridge, MA, USA . . . . Objectives: 2011 AHA and 2012 ESC guidelines recommend antiplatelet therapy at presentation for suspected ACS patients. For CABG patients (approximately 10% of ACS presentations) this strategy creates a dilemma: delay CABG to washout antiplatelet therapy or perform surgery and risk bleeding. Bridging strategies have been proposed using GPIs. Cangrelor, a novel reversible, IV P2Y12 inhibitor with rapid onset/offset, demonstrated in the CHAMPION PHOENIX trial a reduction in ischemic events vs. clopidogrel in patients undergoing PCI. Adoption of cangrelor for angiography in ACS patients eliminates the need to washout an oral P2Y12inhibitor in case of unanticipated CABG. The aim of our analysis was to quantify the annual value of this pathway change to a US hospital. Methods: A decision analytic model based on the current CABG patient pathway was developed to quantify the value of a reduction in bridging by adopting cangrelor from a US hospital perspective. Premier hospital database informed the timing of CABG after washout; demographics, drugmix, LOS, blood product utilization, and hospital ward costs. Ischemic event rates and costs were informed by published sources. Drug costs were 2014 wholesale acquisition costs. Cangrelor cost was set to $0. Results: For a hypothetical US hospital treating 195 CABG patients/year (patient mix: 4% STE-ACS, 26% NSTE-ACS, 70% SA), shifting away from bridging with GPI by adopting cangrelor resulted in a total annual cost reduction of $120,000. Total costs in the base-case are estimated to be $2.95MM vs. the scenario case of $2.83MM. Cost savings were derived from eliminating GPI utilization, lowering washout drug costs and reducing hospitalization days. Conclusions: Removal of the need to washout oral antiplatelet therapy prior to CABG through adoption of cangrelor is estimated to deliver a clinical value of ~$615/CABG patient with fewer total days spent hospitalized. PCV46 Budget Impact Of The Introduction Of New Oral Anticoagulants (Noac) For No Valve Atrial Fibrillation (Nvaf) In Extremadura ParejoHernández E D P 1, Gemio Zumalave P 2, Alvárez Fernández M 1, Montaño Pérez L M 1, Sánchez Chorro J L 2, Hidalgo A 3 1Servicio Extremeño de Salud, Mérida, Spain, 2Extremadura Health System, Mérida, Spain, 3Instituto Max Weber, Majadahonda, Spain . . . . . . . . . . Objectives: To analyse the budgetary impact (BI) of the introduction of NOAC in patients with NVAF compared to warfarin treatment scenario (DC) from the perspective of the public health system of Extremadura (PHSE). Methods: A budget impact study was performed during 2013. It was based on anticoagulant therapy (AO) on patients with NVAF. The data came from the electronic clinical history called JARA®. Pharmaceutical costs associated to the control and monitoring of patients with DC were included. Specifically those related to the incidence of stroke (CVA), gastrointestinal haemorrhage (HG), intracranial haemorrhage (ICH) and acute myocardial infarction (AMI) as the RE-LY, ROCKET and ARISTOTLE trials. For DC, there were taken into account the best and worst scenario. The pharmaceutical cost considered was the sum of the official price and the VAT. To other public health costs, official prices from the PHSE and the Spanish National Health Care System were considered. Results: Treatment distribution among patients was 92.63% DC; 4.02% Dabigatran (DAB); 3.07% rivaroxaban (RIV) and 0.28% Apixaban (APX). The total expenditure for the AO patients with VKA in the best scenario is € 3,036,604.28 (average cost per patient: € 342.11, CI 95%: 341.89 - 342.34) and the worst would be € 3,727,169 (€ 419.92, CI 95%: 419.70 - 420.15). For NOAC the results are € 572,515.39 (€ 810.93, CI 95%I: 786.71 - 835.16). NOAC represent the 71.44% of the pharmaceutical expenditure (PE), the 23.47% of PE’s control and monitoring of DC and the 15.86% of the total expenditure from AO patients with NVAF, in the best scenario. Conclusions: The market access of NOAC has been modest and similar to the rest of Spain. AVK alternative has the lowest average cost per patient. Limitation: safety data has been obtained from clinical trials instead of using clinical practice. . . Objectives: To estimate the costs associated with the use of dabigatran (DB) vs. dicoumarins (DC) in non-valvular atrial fibrillation (NVAF) from the perspective of the Extremadura Public Health System (EPHS) in 2013. Methods: A budget impact study of anticoagulant therapy (AT) with DB and DC in patients with NVAF was carried out during 2013. The data were processed using the JARA® electronic record application. Pharmaceutical costs and costs associated with patient followup and monitoring were included. Costs relating to incidence of stroke, gastrointestinal bleeding, intracranial haemorrhage and acute myocardial infarction were also included based on RE-LY clinical trials. For pharmaceutical cost RRP + VAT of the medicinal products was used, and other health care costs were based on the public prices of the EPHS and the National Health System. Results: Three hundred eighty-five patients were treated with DB (4.01 %) and 8876 with DC (92.63%). Mean age was 77 (95 % CI: 76.82-77.19). Total expenditure for AT with AVK for the best case scenario would be € 3,036,604.28 (mean cost per patient: € 342.11; 95% CI: 341.80-342.34) and the worst case € 3,727,169.83 (€ 419.92; 95% CI: 419.70-419.15) and for DB € 318,077.15 (mean expenditure: € 826.27; 95% CI: 796.81-855.57). For a DB market penetration of 50%, expenditure on anticoagulation of these patients would be € 3,958,659.57 while for 75% it would be € 5,937,989.36. Conclusions: Dabigatran is a more costly option than classic treatment with DC. In 2013 this drug’s market penetration was limited and was closer to other estimates. Estimated mean cost per patient for DCs is similar to the estimates of other communities (Basque Country: € 300-700 and the Valencian Community: € 300) and are somewhat lower for DB (Basque Country: € 1,197 and the Valencian Community: € 1,350). A limitation PCV47 The Move Towards Full Implementation Of The Nice Guidelines For Stroke Prevention In Atrial Fibrillation: The Potential Cost And Clinical Impact Shields G E , Chapman A M BresMed, Sheffield, UK . . . . Objectives: Updated treatment guidelines for atrial fibrillation (AF) have been released by the National Institute for Health and Care Excellence (NICE) in the United Kingdom and highlight a current shortfall in the prescribing of anticoagulants to patients with AF despite the importance of stroke prevention. A model was designed for use as a planning tool for Clinical Commissioning Groups (CCGs) looking to budget for the future move towards full implementation of the NICE guidelines. Methods: An Excel model was developed to estimate the 5-year impact of gradually treating all eligible AF patients who are currently not being prescribed anticoagulants, both in terms of the effect on clinical outcomes (strokes, major bleeds and mortality) and the financial impact associated with treating more patients. Assumptions had to be made in order to simplify the model and populate default values. Risks of clinical outcomes were taken from a literature search. Costs were taken from published national sources. Results: In an average sized CCG (population of 226,000) the model estimated that in order to gradually treat the full eligible AF patient population, an additional budget of £103,899 will be required in Year 1 to treat an additional 248 patients, rising to £791,076 in Year 5 to treat an additional 1,889 patients. Cumulatively over the 5-year timeframe this could lead to the prevention of 8 ischemic strokes and 3 deaths, as well as an increase of 2 haemorrhagic strokes and 8 bleeds. Conclusions: The clinical benefits of appro- A481 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 priate anticoagulation are recognised, but full implementation can be difficult and costly. Therefore the development of models such as this can support the planning process allowing stakeholders to discuss how best they can reach the target of full implementation. The model is flexible and can be adapted to suit different payers. PCV48 Cost Effectiveness Analysis Of Mitraclip In Mitral Regurgitation For High Risk Patients Jamet N 1, Bourguignon S 1, Marque S 2 Santé, Evry, France, 2Capionis, Paris, France . . . 1Stratégique Objectives: Mitral Regurgitation (MR) is a cardiac disease resulting in backflow of blood from the left ventricle to the left atrium which could increase the risk of heart failure and mortality. Half of severe MR patients are not considered eligible to surgery (valve repair or replacement) and receive a medical treatment. MR patient management could benefit from usage of Mitraclip, a transcatheter device, which enables percutaneous edge-to-edge repair to treat MR. The cost-effectiveness model presented here compares Mitraclip therapy versus medical standard care treatment. Methods: A four-state Markov model (Death, MR grade 0, MR grade I /II, MR grade III/ IV) has been developed. In each state patients could be hospitalized or not. A national payer’s perspective was chosen with a 5 year time horizon. Primary and secondary endpoints were respectively the number of deaths and of hospitalizations avoided. Data were obtained from the EVEREST II high-risk study and from french cost analysis. Results: Within the time horizon analyzed, 276 further deaths could be avoided by using Mitraclip strategy out of 1000 patients with MR, compared to medical treatment. The Incremental Cost Effectiveness Ratio (ICER) is estimated at € 93,000 per death averted with cumulative cost on five years. Sensitivity analysis shows that the cost of the initial surgery and the cost of the device where the two most sensitive variables. Costs of managing MR are higher for the Mitraclip option during the first year (€ 29,894 for Mitraclip compared to € 8,557 for medical treatment) option due to the cost of the device and surgery, whereas this is inversed from the second year onwards (€ 8,557 for the medical option vs. € 3,122 for Mitraclip). Therefore, an average ICER (20 720€ per death averted) has also been calculated. Conclusions: Mitraclip might represent a new economically attractive treatment option for MR patients at high-risk which increases survival. PCV49 Cost-Effectiveness And Budget Impact Analyses Of Risk Stratification Of Patients With Moderate Risk Of Cardiovascular Events Using Lp-Pla2 Testing Rinde H 1, Genser B 2, Sonntag D 2, Kleber M E 2, Stojakovic T 3, Scharnagl H 3, Maerz W 4 1BioBridge Strategies, Binningen, Switzerland, 2Mannheim Institue of Public Health, Social and Preventive Medicine, Mannheim, Germany, 3Medical University of Graz, Graz, Austria, 4Medizinische Fakultät Mannheim der Universität Heidelberg, Mannheim, Germany . . . . . . . . Objectives: 1) Analyze whether a testing strategy using the biomarker Lp-PLA2would improve clinical and economic outcomes vs. ESC-SCORE alone in Germany for 50-70 year-olds. 2) Evaluate the potential budget impact for payers. Methods: To decide the treatment strategy for patients with moderate risk for cardiovascular events is a challenge. They would usually not receive statin treatment in Germany. For further risk stratification there is a need to identify patients with vulnerable plaques. When an arterial plaque becomes unstable Lp-PLA2is released, which indicate that these patients could benefit from treatment to prevent future cardiovascular events. An integrated cost-effectiveness and budget impact model was constructed. Lp-PLA2increased the adjusted risk for CVD events in the moderate ESC-SCORE population by > 2 fold in the German LURIC Study Cohort (HR 2.23,95% CI 1.15-4.32; P= 0.018). Efficacy of statin treatment relevant costs were obtained from literature. A range of sensitivity analyses were performed. Results: The cost-effectiveness and the budget impact analyses used a theoretical population of 1 million, of which 14% were 50-70 year olds with moderate cardiovascular risk. The total 10-year discounted and adherence adjusted net cost savings from implementing the Lp-PLA2 testing strategy was € 19 million, or € 156 per Lp-PLA2tested patient. The 10-year accumulated number of deaths averted by the Lp-PLA2 testing strategy was 611, or 17 incremental discounted life-days and 2.2 incremental discounted event-free life-months per Lp-PLA 2tested patient. Projected to whole of Germany’s population aged 50-70 the potential annual discounted savings from the Lp-PLA2testing strategy would be € 180 million. The potential number of deaths averted per year would be 5,030. Conclusions: Our results indicate that the Lp-LPA2 testing strategy is both cost saving and provide reduction in mortality and morbidity. The implementation of Lp-LPA2 testing strategy should be considered in Germany. PCV50 Comparing Actual Patient Level Hospital Costs To The Canadian Cmg+ Costing Estimates For Acute Myocardial Infarction Chu F 1, Ohinmaa A 1, Jacobs P 2, Zheng Y 2, Kaul P 1 of Alberta, Edmonton, AB, Canada, 2Institute of Health Economics, Edmonton, AB, Canada . . . . . 1University Objectives: This study compares differences in actual hospital costs and Case Mixed Group (CMG+) costs (Canadian version of the Diagnosis Related Group), for patients with acute myocardial infarction (AMI) in Edmonton, Alberta. Methods: New AMI (ICD-10 code I21) patients (no AMI hospitalization within one year) hospitalized in Edmonton area hospitals between April 1,2006 and March 30,2009 were segmented into CMG+ categories by the Canadian Institute for Health Information. The differences between actual hospital cost and CMG+ cost were analyzed by comparing the mean and median differences between costs for each patient and trimming out 5% of high and low cost patients and excluding patients with longer than 90 days of hospitalization. 15 comorbidities were derived from secondary diagnostic codes and regressed against CMG+ costs and actual costs independently. The coefficients between the two separate regressions are then tested for statistical equivalence using the Wald test. Results: The data included 4,734 new AMI patients, and after excluding the outliers and longer than 90 days LOS, the data included 3,428 patients. The estimated mean difference using the average CMG+ estimate for the whole hospital episode costs were about $500 higher than actual costs. The median CMG+ cost were most accurate estimates for per diem costs, which was about $20 higher than actual cost. 2 comorbidities were dropped from the regression due to multicollinearity. Using average CMG+ estimate for whole hospital episode costs, 10/13 comorbidity coefficients were found to be statistically equivalent to the coefficients in a separate regression using actual cost. Conclusions: This study shows that various derivations of costing proxies using the CMG+ methodology produce relatively accurate cost estimates for AMI patients when actual cost are not available. Based on the available patient data and the context of use of the cost estimates, different methods will be optimal. PCV51 Cost Estimation Of Home Blood Pressure Monitoring Versus Combined Office And Ambulatory Measurements In Hypertension Management Boubouchairopoulou N 1, Karpettas N 2, Athanasakis K 1, Kollias A 2, Protogerou A D 3, Achimastos A 2, Stergiou G S 2 1National School of Public Health, Athens, Greece, 2Sotiria Hospital, Athens, Greece, 3Laiko Hospital, Athens, Greece . . . . . . . . . Objectives: Hypertension is a chronic condition, directly linked to cardiovascular diseases. Therefore, the monitoring of blood pressure (BP) is of utmost importance in order to avoid BP-related adverse clinical outcomes. This study aimed at comparing the health resources consumed and the subsequent costs for hypertension management using home blood pressure monitoring alone (HBPM) vs combined office measurements and ambulatory blood pressure monitoring (C/ABPM). Methods: A total of 116 previously untreated, hypertensive subjects were randomized to use either HBPM or C/ABPM for antihypertensive treatment initiation and titration. The analysis involved all health resources (BP measurements/outpatient visits, laboratory and other tests, pharmaceutical therapy) utilized within 12-months follow up, their respective costs, and efficacy (hypertension control). A 5-year projection was applied assuming (a) continuation of stable treatment as in the end of first year (for both arms), (b) single ABPM/year in C/ABPM group, (c) 2 visits/year in HBPM group and 3 in ABPM. Results: The total cost of hypertension management regardless of BP measurement method was calculated at 1,404.8€ /patient (laboratory tests: 50.4%, BP measurements+outpatient visits: 32.4%, pharmaceuticals: 17.1%). In HBPM group, total cost was 1,336.0€/patient vs 1,473.5€/patient in C/ABPM group (p<0.001). Findings suggested that the cost of treatment did not differ between the two groups (233.1 vs 247.6€/patient respectively, p=NS), while BP measurements+outpatient visits were estimated at 393.9€/patient in HBPM arm and 516.9€/patient in C/ABPM arm (p< 0.001). For subsequent years (>1), expenditures were estimated at 348.9€/patient for HBPM vs 440.2€ /patient for C/ABPM group (p< 0.001), whereas for a 5-year projection, 2,731.4€ /patient and 3,234.3€ /patient respectively (p< 0.001). Conclusions: C/ABPM strategy presented a higher first year cost compared to HBPM, while the same trend was unveiled in 5-year projection. Effective hypertension management through the appropriate strategies is of paramount importance considering its high prevalence; ergo, even small differences in the cost of applying them could have substantial impact on health expenditures. PCV52 The Cost Comparison Of Drug-Eluting Stents (Des) And Bare-Metal Stents (Bms) - A Retrospective Cohort Matched Study Lang H C 1, Chen T C 1, Chen C H 2 University, Taipei, Taiwan, 2Taipei Veterans General Hospital, Taipei, Taiwan . . . . . . 1National Yang-Ming Objectives: Literature has failed to demonstrate the clear superiority of Drug-Eluting Stenting (DES) for stable coronary artery diseases on survival as compared to the baremetal stenting (BMS). This study aimed to compare the health care utilization and the costs between drug-eluting stenting (DES) and bare-metal stenting (BMS). We also examined factors that influenced cumulative costs of these two groups. Methods: We conducted a retrospective cohort study based on the NHI program. Patients who had coronary stenting between Jan. 2007 and Dec. 2008 were recruited and followed through the end of 2010. Both groups were matched on 2: 1 by propensity score which adjusted sex, age, stent number and Charlson comorbidity index (CCI). We estimated cumulative medical cost for these two matched group by conducting the KaplanMeier Sample Average (KMSA) estimates. Regression analysis was used to explore the predictors of cost. Results: The mean age in both groups was around 66 years. After propensity score matching, we had a total of 966 patients; 644 in BMS group and 322 in DES group. KMSA estimates (discounted 3.5%) showed that DES group had a higher 3-year cumulative total outpatient cost at US$ 6,867 and heart related outpatient cost at US$ 2,548 as compared to BMS group, which were US$7,668 and US$ 3,302 respectively (1US$= 30 NTD). The heart related inpatient cost was similar between two groups. The significant predictors of heart-related outpatient costs were stent type, premium and CCI. The predictors of heart-related inpatient costs were stent type, stent number, CCI and procedure for acute coronary syndrome (ACS). Conclusions: In Taiwan, NHI reimburses DES and BMS at the same price, and hospitals can balance billing for the DES. We found that even after adding the extra national average outof-pocket payment to DES, DES still was a cost-effective procedure. PCV53 Goal Directed Perfusion (Gdp): A Differential Cost Analysis In Uk And Us Povero M , Pradelli L AdRes HE&OR, Turin, Italy . . Objectives: High oxygen delivery (DO2) during cardiopulmonary bypass (CPB) is associated with better renal outcome in cardiac surgery. Traditional perfusion (TP) techniques, targeted on body surface area and CPB temperature, achieves high DO2 in about 50% of the cases while a goal directed perfusion (GDP) approach can lead to more than 90% of cases achieving high DO2 with a consequent reduction A482 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 in Acute Kidney Injury (AKI) rate of about 40%. Aim of this study is to perform an economic evaluation of GDP strategy with respect to TP in UK and US. Methods: A Discrete Event Simulation model was developed to compare TP and GDP strategy in patients undergoing CPB. The patient’s pathways from operation to discharging from hospital was simulated: AKI incidence, in-hospital mortality, hospital length of stay, transfusions were correlated to probability to achieve high DO2 target using published correlations. National perspective was adopted to calculate costs associated to each event while GDP strategy was exploited considering card and data management system (DMS) cost per patient. Results: GDP strategy saved more than 3 days in hospital and 11% of AKI episodes. The cost-saving is 2,821 £ in UK and 3,206 $ in US; the cost of card and DMS (79 £ in UK, 110 $ in US) is completely offset by savings in hospital stay that result the main driver in cost (2,886 £ in UK, 3,222 $ in US). Deterministic sensitivity analysis shows that the total savings are mainly influenced by hospital LOS, cost per day both in ICU and in ward, and nadir haematocrit during CPB. Conclusions: GDP seems to improve significantly the main outcomes related to CPB surgery, when compared to TP techniques. Additional costs due to perform GDP strategy have no impact on the total cost since completely offset by the savings in hospital cost. PCV54 Can A Cvd Polypill Save Money In The ‘Real World’? Laba T L 1, Hayes A 2, Jan S 1, Rodgers A 1, Patel A 1, Cass A 3, Reid C 4, Tonkin A 4, Usherwood T 2, Webster R 1 1George Institute for Global Health, University of Sydney, Camperdown, Australia, 2University of Sydney, Camperdown, Australia, 3Menzies School of Health Research, Darwin, Australia, 4Monash University, Melbourne, Australia . . . . . . . . . . model was a hospital in the United States (US). Clinical data was obtained from US pivotal studies for the Medtronic stent system and the competing stent systems. The competition arm (Competition) was created by pooling the pivotal study data for the current stent systems manufactured by Gore, Endologix, and Cook. Cost data was obtained from the Premier database (2011-2012) and augmented with the published literature. All costs were adjusted to 2013 dollars. The model estimated the costs associated with the following utilization outcomes: procedure time, transfusion rate, intensive care unit (ICU) length of stay (LOS), and general ward LOS. The following adverse events were considered: myocardial infarction, respiratory failure, acute renal failure, stroke/TIA, and second endovascular procedure within 12 months of the initial procedure. Sensitivity analysis was performed to assess the impact of imputed data, and one-way sensitivity analysis was performed for each parameter. Results: The expected costs for a hospital related to the above utilization and adverse event were $8,463 for Medtronic’s stent graft system and $11,380 for the Competition. Fifty-six percent of the $2,917 difference was attributable to improved utilization associated with Medtronic’s stent graft compared to the Competition. Adverse events and secondary endovascular procedures accounted for 25% and 19% of the difference, respectively. These results were robust to alternative sensitivity analyses. Conclusions: This analysis suggested that Medtronic’s current stent graft is associated with cost savings compared to Competition for the above parameters. Future research is necessary to examine if these results are maintained based upon a head-to-head clinical study of EVAR stent systems. . Objectives: The use of polypills in the prevention of cardiovascular disease is mooted to reduce costs compared with current practice, yet there is very little prospectively-collected data to support this claim. The present study compares the ‘real-world’ costs of a polypill strategy against usual care among Australians with established cardiovascular disease or at high estimated cardiovascular risk. Methods: A ‘within trial’ cost analysis from the Australian health system perspective of polypill-based care versus usual care with separate medications was conducted using data from the pragmatic randomised controlled trial Kanyini Guidelines Adherence to Polypill (Kanyini GAP) and linked health service and medication claims data. The primary outcome, estimated with generalised linear models, was mean health service and pharmaceutical expenditure, per patient per year. All costs during the trial, conducted from 2008-2012, were inflated to $AUD 2012 prices. Results: A statistically significantly lower mean pharmaceutical expenditure of $989 (95%CI 648 to 1331) per patient per year in the polypill arm compared to usual care (P< 0.001, adjusted, excluding polypill cost). No significant differences were observed in annual non-hospital health service expenditure ($40, 95%CI -202 to 281 per patient). Conclusions: This study provides evidence that a cardiovascular disease polypill strategy has the potential to produce significant cost-savings to health systems. At an estimated reimbursement cost of $1 per day for the polypill, these savings would have amounted to over $600 per patient per year. Cost-savings would accrue to patients also, given fewer prescription charges. Linking health service and medication claims data with data from a pragmatic randomised controlled trial has provided an avenue to assess the real-world cost implications of introducing this new technology into clinical practice. PCV55 Study Of Costs Of The Cardiac And Diabetes Mellitus Patient In A Cardiology Hospital Of High Complexity Costa M G S D 1, Santos M S 2, Tura B R 2, Goulart M C 2, Cintra M A C T 2, Senna K M S 1 de Janeiro, Brazil, 2National Institute of Cardiology, Rio de Janeiro, Brazil . . . . . . . . . . . . . . . . . 1INC, Rio Objectives: There is a growing prevalence of diabetes mellitus (DM) among chronic diseases in the world. Currently there are over 135 million people with diabetes worldwide with estimates reaching 300 million in 2025. Developing countries concentrate two thirds of these patients and it is known that the economic burden of chronic diseases generate high costs for the health system and social welfare as a function of mortality and premature disability. The objective of the study was to investigate the impact on hospital costs of treating a patient with ischemic heart disease and DM, compared with cardiac patients without DM, in a cardiology hospital of high complexity Ministry of Health in Brazil. Methods: observational study of historical cohort of 421 diabetic heart disease (CD) and non diabetic (CND), from January 2009 to March 2010 in cardiology hospital of high complexity of the Unified Health System (SUS) in Brazil. Were only covered the direct medical costs of hospitalizations. The costs of the study population (CD and CND) were grouped into surgery, and clinical treatment obtained by two different approaches (top-down and bottom-up estimates), and subsequently analyzed and compared using R software version 3.0. Results: No differences between groups were observed. Cost of surgery: CND = U. S. $ 2937.55 and U. S. $ 3024.51 = CD (p = 0.319). Medical Treatment: CND = U. S. $ 685.09 and U. S. $ 304.11 = CD (p = 0.218). Values are expressed as medians. Conclusions: studies analyzing these conditions separately describe high expenses resulting from the treatment of diabetes and cardiovascular disease. We can infer from the results of this study that the diabetic patient cardiac does not generate a significant financial impact for a cardiology hospital of high complexity. PCV56 A Cost Comparison Analysis Of Medtronic’s Stent Graft System To Competition For Endovascular Aneurysm Repair For Abdominal Aortic Aneurysms Mallow P J 1, Baniewicz J 2, Williams J M 2, Au-Yeung A 2 Clinical Trial and Consulting, Cincinnati, OH, USA, 2Medtronic, Inc, Mounds View, MN, USA . . . . . . 1CTI Objectives: To perform a cost comparison analysis of Medtronic’s current stent graft system compared to currently competing stent graft systems for endovascular aneurysm repair (EVAR) of abdominal aortic aneurysms (AAA). Methods: A simulation model was constructed using Microsoft Excel. The perspective of the PCV57 Hospitalizations And Costs In Patients With Implantable Cardioverter Defibrillators: Association Of Long VersUS Standard Detection Intervals Borghetti F 1, Proclemer A 2, Arenal A 3, Kloppe A 4, Lunati M 5, Ferrer J B M 6, Hersi A 7, Gulaj M 8, Wijffels M C E F 9, Santi E 10, Manotta L 11, Beccagutti G 1, Campo C 11, Gasparini M 12 1Medtronic Italia, Sesto San Giovanni (MI), Italy, 2Azienda Ospedaliero Universitaria S. Maria della Misericordia, Udine, Italy, 3Hospital General Universitario Gregorio Marañón, Madrid, Spain, 4Berufsgenossenschaftliches Universitätsklinikum Bergmannsheil, Bochum, Germany, 5Azienda Ospedaliera Niguarda Ca’ Granda, Milano, Italy, 6Hospital de Txagorritxu, Vitoria (Álava), Spain, 7College of Medicine, King Saud University, Riyadh, Saudi Arabia, 8MSWiA Hospital, Bialystok, Poland, 9St. Antonius Ziekenhuis Hospital, Nieuwegein, The Netherlands, 10MEDTRONIC Clinical Research Institute, Roma, Italy, 11MEDTRONIC Clinical Research Institute, Sesto S. Giovanni, Italy, 12Humanitas Research Hospital, IRCCS, Rozzano (MI), Italy . . . . . . . . . . . . . . . . . . . Objectives: ADVANCEIII trial showed that a long detection programming reduces all delivered therapies as well as inappropriate shocks in patients implanted with implantable cardioverter defibrillator (ICD). The purpose of this Advance III secondary analysis was to assess the impact of long detection on hospitalizations (H), length of stay (LOS) and associated costs for the health care system. Methods: 1902 patients enrolled in the ADVANCEIII Trial: 948 patients randomized to long detection (NID 30/40) and 954 to short setting (NID 18/24). All hospitalizations were reviewed and classified according to ICD9CM codes and, consequently, to the corresponding Diagnosis-Related Groups (DRGs). Costs correspond to the specific public tariffs for the DRGs applied. The prospective was of a single-payer agent (Italian Ministry of Health). Results: Over a median period of 12 months, rates of overall and cardiovascular hospitalizations (CV) were lower in the long detection group (43.8*100 pts/years (39.6-48.4) vs 52.3*100 pts/years (47.7-57.3), IRR: 0.84 (0.73-0.96) p= 0.005,32.7*100 pts/years (29.1-36.7) vs 40.3*100 pts/years (36.2-44.6), IRR (95% CI): 0.81 (0.69-0.95) p= 0.004 respectively). Patients programmed with a long detection had shorter LOS (overall H: 407days (394-421) vs 470 days (456-484), IRR: 0.87 (0.83-0.91) p< 0.001; CV: 298 days (287-309) vs 368 days (356-381), IRR: 0.81 (0.77-0.85) p< 0.001) and lower mean hospital cost per patient-year compared with patients with nominal programming (overall H: 1.311 € (1.309 € - 1.314 € ) versus 1.528€ (1.525€ - 1.530€ ) IRR: 0.86 (0.86-0.86) p< 0.001; CV: 1.100 € (1.098 € - 1.103 € ) versus 1.339 € (1.337 € - 1.342 € ) IRR: 0.86 (0.86-0.86) p< 0.001). Conclusions: A long detection window was associated with a reduction in hospitalization rates, total length of stay and cost per patients both for all-causes and cardiovascular related events. PCV58 Economics And Clinical Evaluation Of Endovascular And Surgical Treatment Of Patients With Disability Of Superficial Femoral Artery Kamensky V , Ivlev I Czech Technical University in Prague, Kladno, Czech Republic . . Objectives: The rising cost of treatment of peripheral arterial disease and the growing incidence of this disease led to economic analysis of arterial disease. With the increasing price of modern instrumentation, it is appropriate to evaluate not only clinical efficiency, but also intervention economics. This study aims to create a recommendation for choosing the most effective treatment based on both economic data and clinical outputs of the disabled superficial femoral artery. Methods: The methods chosen were reviewed from clinical outputs for treatment effectiveness, multiple-criteria decision-making for the synthesis of treatment effects, analysis of costs at the selected interventions and cost-effectiveness analysis. Results: The four clinical outputs used in the study as a criterion by the research of the clinical studies were primary patency, technical success, patient survival and limb salvage at the year of operation. The weights of each criterion, and the preferences of the interventions were counted. The sequence of interventions was set by the AHP method: PTA (35.6%), PTA/S (33.9%) and bypass (30.6%) and method of weighted sum: PTA/S (56%), bypass (55%) and PTA (42%). From the view of both health insurance payer and health care provider, where the direct medical expenditures were included, the order of intervention was the same: PTA, bypass and PTA/S. The cost-effectiveness was calculated for both, and the PTA intervention achieved the best results. Incremental expenditures by unit of effect was calculated for each effect: ICER or the domination of one intervention over another was set. Ratio of the ICER was generally higher for PTA/S compared to bypass. In the sensitive analysis was determined the influence of A483 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 the direct medical expenditures on the final outputs. Conclusions: The treatment of disabled superficial femoral artery was compared from chosen clinical outputs, economic expenditures and cost-effectiveness. On this basis, the recommendation was set for choosing the most effective treatment. PCV59 Interest Of A Hospital Database To Analyze The Cost For Acute Stroke: The Example Of Versailles Hospital Bourguignon S 1, Milojevic K 2, Lemaire B 2, Le Lay K 3, Faille S 2, Parisse S 2, Pico F 2, Lambert Y2 1Stratégique Santé, Evry, France, 2Centre Hospitalier de Versailles, Le Chesnay, France, 3Boehringer Ingelheim France, Paris, France . . . . . . . . greater LOS and 32.3% greater ICU LOS (both p< 0.001) than patients not receiving blood. In patients who received blood, total cost of hospital stay was 22.7% greater including blood product cost and 21.7% greater excluding blood product cost (both p< 0.001). Conclusions: Preventing or rapidly controlling bleeding in patients undergoing complex cardiac surgery would likely reduce patient risk and avoid elevated costs of hospitalization. PCV62 The Cost Burden Of Syncope At A Hospital Level In Spain Egea M 1, Marti B 1, Álvarez M 2 Iberia, Madrid, Spain, 2Hospital Universitario del Henares, Coslada, Spain . . . 1Medtronic Objectives: Stroke generates an important socioeconomic burden and heavy hospital expenses. The objective of this study was to evaluate the impact of delay to manage stroke for patients transported by Mobile Intensive Care Units (MICU) in outcomes and early hospital costs. Cost of management and NHISS (National Institute of Health Stroke Score) score regression at patient discharge have been described for three different stroke management time groups. Methods: A retrospective study was conducted from a database containing records of 427 patients diagnosed with stroke. Patients were classified as “short delay” (stroke managed within 45 minutes), “medium delay” (45 minutes to 2 hours) and “long delay” (above 2 hours). 3 homogenous groups of 100 patients were established, adjusted on age, gender, stroke mechanism (ischemic or hemorrhagic), NIHSS score and comorbidity. Hospital Database (PMSI - French National Hospital Information Systems Program) recorded information between 2004 and 2013. ResultsThe number of deaths was respectively 17 for « short delay », 25 for « medium delay » and 42 for « long delay ». The average cost for one hospital stay for stroke was € 15,812 for the « short delay » group, € 20,639 for « medium delay » and € 20,184 for « long delay ». “Short delay” treated patients retrieve an average of 8.41 points compared to 1.87 for “long delay” patients. Conclusions: These results show a correlation between the delay in managing stroke and early hospital costs and improvement of NIHSS scores. These results are part of a study performed through the PMSI database aiming at calculating the average cost of retrieving one-point NIHSS impairment. They underline the importance of an early treatment of stroke. Objectives: Syncope, defined as a transient loss of consciousness that leads to a spontaneous recovery without consequences, is very prevalent. It is known that syncope related admissions (principal diagnosis) are associated with a high cost burden for the Spanish National Health System, but little is known about the number of hospitalisations due to injuries caused by the fall (secondary diagnosis). The objective of this study was to estimate the burden and cost of syncope from the Spanish Health Care System perspective. Methods: Admissions with a primary diagnosis of syncope and admissions for injuries and pacemaker/defibrillator implantations where syncope was coded as a secondary diagnosis were analysed. The source of data was the Spanish Basic Minimum Data Set and the period 2001-2010. Results: The annual incidence of syncope-related admission was 4.22 cases/10,000 inhab. The annual average number of admissions was 15,703, including 77.5% syncope as a primary diagnosis. The total annual cost of syncope related admissions was € 46mill, from which a 46.4% was associated to injuries where syncope was coded as a secondary diagnosis. The average cost of a patient admitted with syncope as a primary and secondary diagnosis was € 2,062 (min €1,556 – max €28,772) and €5,996 (min €719 – max €26,940) respectively. The average length of stay for patients with syncope as a primary and secondary diagnosis was 5.17 and 9.33 days respectively. Conclusions: Syncope has an important cost burden resulted from the high number of hospitalisations and the length of stay associated. The cost of a patient admitted with syncope as a secondary diagnosis was nearly 3 times higher than a patient with a primary diagnosis of syncope. It is important to implement health strategies to diagnose these patients in order to assure an efficient management of the available resources. PCV60 Burden Of Cardiovascular Complications In Patients With Atrial Fibrillation In France PCV63 Clnical Management Of Non-Valvular Atrial Fibrillation In Hong Kong Fauchier L 1, Samson A 2, Chaize G 3, Gaudin A F 4, Vainchtock A 3, Bailly C 5, Cotté F E 5 1Service de Cardiologie B et Laboratoire d’Electrophysiologie Cardiaque, Pôle Cœur Thorax Vasculaire, Centre Hospitalier Universitaire Trousseau, Chambray-les-Tours, France, 2ParisDauphine University, Paris, France, 3HEVA, Lyon, France, 4Bristol-Myers Squibb, Rueil-Malmaison, France, 5Bristol-Myers Squibb, Rueil Malmaison, France Lee V W 1, Chiu L Y 1, Chung S L 1, Lau C Y 1, Yousaf S 1, Wong H 1, Lam Y Y 2, Yan B P 2 1Chinese University of Hong Kong, Hong Kong, Hong Kong, 2Chinese University of Hong Kong, Shatin, Hong Kong, China . . . . . . . . . Objectives: Atrial fibrillation (AF) is associated with numerous cardiovascular (CV) complications. The objective of this study was to estimate the national annual burden of CV complications in patients with AF in hospitals. Methods: All patients hospitalized in 2012 with a diagnosis of AF were identified from the French National public/private hospital database (PMSI). Comorbidities and medical data were collected during a 5-year look-back period and used to calculate stroke risk score (CHA2DS2-VASc). Reasons for CV-related hospitalization, emergency admission, rehabilitation and death at discharge were described. Costs of acute care were determined using Diagnosis Related Groups and corresponding tariffs (2012 Euros). Results: In total, 533,044 AF patients were hospitalized for any reason. Mean age was 78.0(±11.4) years, 53% were males and mean CHA2DS2-VASc score was 4.0(±1.8). CV-related hospitalizations occurred in 267,681 patients: 34% for AF management care, 28% for heart failure, 8% for strokes, 7% for other ischemic heart diseases, 5% for vascular diseases, 4% for transient ischemic attacks and systemic embolisms, 2% for bleedings and 12% for other CV reasons. In CV-related hospitalizations, 45% of patients required emergency admission, especially patients hospitalized for strokes (75%) and for bleedings (67%). Death at hospital occurred in 6% of patients with a CV-related hospitalization: 17% in patients with strokes and 10% in patients with bleedings. Rehabilitation was needed for 34% patients with non-fatal strokes with a mean length of 48 days. The annual total cost (acute care and rehabilitation) for all hospitalized CV events during the year 2012 was €1.94 billion. Among this, heart failure represented € 518m, AF management € 306m, strokes € 291m and hemorrhages € 48m. Conclusions: A half-million AF patients were hospitalized in 2012 in France. CV-related hospitalizations involved over a quartermillion AF patients for a global burden of almost € 2 billion. PCV61 Cost Of Bleeding In Complex Cardiac Surgery Magee G 1, Whyte J 2, Zbrozek A 2 1Premier Inc., Charlotte, NC, USA, 2CSL Behring, King of Prussia, NC, USA . . . Objectives: In complex cardiac surgery there is a high demand for blood/blood products. Their administration can be costly. Our goal was to assess the cost of care and outcomes for complex cardiac surgery patients based on bleeding status. Methods: Patients >18 years, discharged between January 2010 and December 2012, were identified in the Premier Hospital Database based on ICD9 codes and categorized as having received blood/blood products or not. Patients who received blood products prior to the day of surgery were excluded. Differences in treatment costs and outcomes were assessed using univariate analysis and multivariate modeling. Results: A total of 463,734 patients (82,832 Bleeding; 380,902 Non-Bleeding) were included. Patients who received blood were older (> 65 years: 62.3% vs. 49.8%), sicker (Charlson Comorbidity Index > 3: 22.5% vs. 15.4%), with greater rates of cerebrovascular disease (10.2% vs. 4.3%), PVD (18.2% vs. 9.4%), CHF (33.0% vs. 18.6%), renal dx (20.9% vs. 13.5%) and CPD (25.3% vs. 18.7%) than those who did not. Models adjusting for patient factors showed that patients who received blood were 4.54 times as likely to be admitted to the ICU, 3.40 times as likely to die and 4.59 times as likely to be readmitted for bleeding within 30 days (all p< 0.001). They also had 49.1% . . . . . . . . . . . . . . Objectives: The present study aimed to investigate the cost of non-valvular atrial fibrillation (NVAF) management in Hong Kong. Methods: We analyzed consecutive patients with documented NVAF enrolled in the Prince of Wales Hospital Atrial Fibrillation Registry between February 2013 and February 2014.. Patient clinical characteristics, prescribed therapies and adverse clinical events were extracted from the hospital electronic Clinical Management System. One-year direct health care costs including costs of medications, accident and emergency admissions, hospitalizations, clinical visits, coagulation tests and computed tomography of brain scans were estimated from a health care provider perspective. Results: Of 534 NVAG patients, 45.9% were male with a mean age of 75.8±10.2 years. The average CHADS2, CHAsDSsVASc and HASBLED scores were 2.44±1.39,4. 12±1.77 and 3.26±1.09, respectively. 66.7% of patients were taking rate control medications only. Patients taking both rate and rhythm control medications were associated with the highest rate of hospitalizations due to poor AF control (28.2%) when compared to using neither rate or rhythm control drugs, rate control or rhythm control groups (p < 0.0005). The overall use of anticoagulants including both warfarin and new oral anticoagulants was 66.9%. The main reason for patients with CHADS2 score ≥ 2 not receiving anticoagulant was patient refusal (20%) and physicians’ opinion (40%). The one-year median cost of managing patients with stroke was higher than those without stroke (Euro$ 5521.2 versus Euro$ 1075.2) from the health care provider perspectives during the study period. Conclusions: Majority of NVAF patients in Hong Kong were prescribed rate control medications only. The clinical use of antithrombotic therapy for stroke prevention was more conservative than guideline recommendations. It was more costly (5 times more) to manage patients with stroke than patients without stroke. PCV64 The Economic Impact Of Cardiovascular Events In Patients Post Myocardial Infarction: Uk Health Care Perspective Jain M 1, Sonathi V 1, Rathi H 1, Thomas S K 2, Mollon P 3 Healthcare Pvt. Ltd., Hyderabad, India, 2Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA, 3Novaris Pharma AG, Basel, Switzerland . . . . . . 1Novartis Objectives: There is a high risk of recurring cardiovascular (CV) events in patients post myocardial infarction (MI) despite the current standard of care (SoC). The aim of this study was to estimate the current economic burden of CV events for patients after having an MI event. Methods: A lifetime markov model was developed from a UK health care perspective to capture progression post-MI. Recurrent MI, stroke and CV deaths were the major CV events captured. A comparison of cost (2012 UK£) and outcomes in terms of QALYs and life years (LYs) was made between the current scenario of patients receiving SOC and a hypothetical scenario where post-MI patients had no subsequent CV events. All cost and outcomes were discounted at 3.5% per annum. Results: The current cumulative lifetime event rate of non-fatal MI, non-fatal stroke and CV death post-MI was estimated to be 0.432,0. 06 and 0.42/patient respectively. The total lifetime cost was estimated to be £6,926/post-MI patient without CV events, which was 31% lower than the current estimated cost of £9,959/post-MI patient. At current rates of CV events there was an incremental lifetime loss of 4.3 LYs/patient and 3.5 QALYs/patient when compared to the hypothetical scenario where patients had no CV events. Considering the prevalence of MI to be 1.5 million, the economic burden posed by MI patients over lifetime if they A484 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 had no recurrent CV events was estimated to be £10.4 billion versus £14.94 billion with recurring CV events under current SOC. This translated into a 43.8% higher economic burden on health care. Conclusions: The subsequent CV events in post-MI patients pose an additional economic burden of 44% on UK health care despite the current SoC. This indicates the need to design new interventions to reduce the risk of further CV events in post-MI patients. PCV65 The Costs Of Atherosclerotic And Haemorrhagic Events Associated With Acute Coronary Syndrome (Acs) In Turkey Parali E 1, Ozdemir O 2, Aykut Aka S 3, Afsar N 4, Degertekin M 5, Ergene O 6, Ongen Z 7, Ozdemir M 8, Deger C 1, Asan Z S 1, Sumer F 1, Ozel M O 1 1Bayer Turk Kimya San. Ltd. Sti., Istanbul, Turkey, 2Yorum Consultancy, ISTANBUL, Turkey, 3Siyami Ersek Thoracic and Cardiovascular Surgery Training and Research Hospital, Istanbul, Turkey, 4Medical Park Hospital, Istanbul, Turkey, 5Yeditepe University Hospital, Istanbul, Turkey, 69 Eylül Faculty of Medicine, Izmir, Turkey, 7Istanbul University Cerrahpasa Faculty of Medicine, Istanbul, Turkey, 8Gazi University, Ankara, Turkey . . . . . . . . . . . . . . Objectives: To estimate the costs of atherosclerotic and haemorrhagic events associated with Acute Coronary Syndrome (ACS) treated with ASA + Clopidogrel, in Turkish settings. Methods: An expert panel was organized to reflect the solid data regarding disease management patterns and options to Turkish payer for reimbursement process. Due to the multidisciplinary nature of the ACS, physicians from the cardiology, neurology and cardiovascular surgery participated to panel. Cost components of bleeding events, haemorrhagic stroke, ischemic stroke, myocardial infarction and revascularization have been identified. Experts have reviewed their daily clinical practices and local & global literature for the related complications along with medical procedures. All cost components including medications, hospitalization, surgical treatments follow up procedures and rehabilitation were considered. For the prices of medications, and diagnostic and therapeutic procedures, March 2014 local prices have been taken into account. For Turkish Lira: Euro conversion, Turkish Central Bank March 2013 currency rate (was used TL/Euro = 3.06). Results: The cost of MI in the first year is 2.004€ and 376€ for the following years. (≈%70) of the cost comes from non-pharmacological treatment. Costs of ischemic and hemorrhagic stroke for the acute period and following three months are 2.266 € and 3.010€ , respectively. Annual cost of ischemic and haemorrhagic strokes are 2.951€ and 3.646€ for the first year. The main cost drivers for stroke are also non-pharmacological treatments, mainly physiotherapy for stroke. (≈%70-80 for the 1st year) Coronary revascularization costs 833€ for percutaneous interventionand 2.354€ for coronary by-pass. Bleeding costs are relatively low compared to other complications. Major and minor bleedings are 189€ /event and 17€ /event. Conclusions: Costs of the ACS related complications are quite high. Especially acute phase of MI and stroke has significant burden of the budget. More effective secondary prevention of the ACS might reduce the burden of disease in the Turkish settings. PCV66 Pharmacoeconomic Burden Of Statins In Patients With Ischemic Heart Disease In The Health Care Of Belarus Kozhanova I 1, Romanova I I 1, Gavrilenko L 1, Sachek M 2 1Belarusian State Medical University, Minsk, Belarus, 2The Belarusian Center for Medical Technologies, Computer Systems, Administration and Management of Health, Minsk, Belarus . . . . . Objectives: The aim of this study has been determination of the burden of statins in the treatment of acute coronary syndrome (ACS) and angina pectoris (AP). There is a significant number of registered statins (brands and generics) in Belarus (lovastatin, atorvastatin simvastatin and rosuvastatin). Pharmacoeconomic studies of statins are topical because of the competition at the pharmaceutical market to ensure rational use of public and private funds. Methods: Overview of available statins has been conducted. Equivalent effective doses were established on the basis of the published data. “Cost of illness” has been calculated with the help of national clinical guidelines for treatment of ACS and AP. Comparison of the each statin burden by determining the proportion of funds falling at each one in the overall cost of the patient´s care has been made. Results: Cost of statins had no significant value in the overall “cost of illness” for acute coronary pathology (ranged from 0.3% to 1.88%). The cost of statins has a significant impact on the final “cost of illness” in the case of chronic cardiac pathology and ranges from 18.54% to 39.9%. Burden of rosuvastatin (Mertenil) in this analysis seems to be the most stable and predictable because of the changes in a narrow range (23.9% - 26.48% in the first year and 28.48% - 29.49% - the second and subsequent years) vs. atorvastatin (19.35% - 35.7% in the first year and 23.3% 39.9% in the second and subsequent years) and simvastatin (18.54% - 32.62% in the first year and 22.13% - 35.89% in the second and subsequent years). Conclusions: Pharmacoeconomic evaluation of statin choosing is important for prolonged use of medicines but not for short-term use in acute coronary syndrome. PCV67 A Systematic Review Of Cost-Of-Illness Studies In Chronic Heart Failure Lesyuk W , Kriza C , Kolominsky-Rabas P L Centre for Health Technology Assessment (HTA) and Public Health (IZPH), Friedrich-AlexanderUniversity Erlangen-Nürnberg, Erlangen, Germany . . . . Objectives: The aim of this study was to perform a systematic review of recently published Cost of Illness (COI) studies related to Chronic Heart Failure (CHF) to highlight the increasing cost impact associated with the disease and identify the major cost drivers. Methods: A systematic review was conducted from 2004-2014 to identify COI studies related to CHF, searching the PubMed, Science Direct, Cochrane Library, Scopus and York CRD databases. The PRISMA guidelines were applied for the search. Results: 12 studies were included in this review and were grouped into a prevalence- and an incidence-based approach. Most articles were considering only the direct medical costs (hospitalization, medication, primary care) of CHF, while some of them also assessed the economic impact of indirect care, i.e. informal care provided by nonprofessionals. The costs of indirect care were reported to be considerably higher than the direct costs. Direct annual costs range from 1,869€ for Poland to 22,230$ for the US. The costs for inpatient care generate the main part of the direct economic burden. In the majority of cases the economic impact is referred to the severity of the CHF expressed by the New York Heart Association functional classification (NYHA) stage. The more severe NYHA stages III and IV are associated with higher costs. Conclusions: Our review highlights the considerable and growing economic burden of CHF on the health care systems of developed countries. The COI studies included in this review show large variations in methodology used and the cost results vary as a consequence. Further research is needed to highlight the economic burden of indirect costs of CHF. High quality data from COI studies with a robust methodology applied can inform policy makers about the major cost drivers of CHF and can be used as the basis of further economic evaluations. PCV68 The Cost Of Illness Of Atrial Fibrillation In Italy: A Cohort Of Hospitalized Patients Fornari C 1, Conti S 1, Ciampichini R 1, Chiodini V 1, Mantovani L G 2, Madotto F 1, Cesana G 1 of Milano - Bicocca, Monza, Italy, 2Federico II University of Naples, Naples, Italy . . . . . . . . 1University Objectives: Atrial Fibrillation (AF) is the most common sustained cardiac arrhythmia. The increasing prevalence of AF and the advances in the management of the disease in recent years highlight the importance of updating cost of illness studies to understand both the financial burden and the related health care resource use. The main purpose of this study was to perform a cost analysis of AF hospitalized cases in a large general population of Italy from the perspective of the Health System (HS). Methods: This is an observational retrospective cohort study conducted using automatic health care administrative archives of the Italian publicly funded HS. The study cohort included all subjects living in Lombardy, a region of Northern Italy, with a hospital discharge for AF or flutter (index event) in the time period 1st January 2003 - 31st December 2009. A pre-index period of eligibility shorter than 3 years was an exclusion criterion. Subjects were followed until 31stDecember 2010 or death or emigration, collecting data related to hospitalizations, drug prescriptions and outpatient visits and related direct costs. Results: The mean age of patients admitted to hospital with a diagnosis of AF was 72 years (± 14.1 standard deviation). Women were older than men and 50% of patients were male. Major comorbidities at index event were: congestive heart failure, cerebrovascular diseases, chronic pulmonary diseases and diabetes. The mean cost of index hospitalization was about 3,600 € (± 3,900 € ). Survival at one year from index event was 83%, 66% at four years and 53% at seven years. The main driver of direct medical cost during follow-up was inpatient care, followed by drug therapies. Conclusions: AF resulted in a substantial medical cost, reflecting resource-intensive and long-term treatments. These costs accompanied with increasing prevalence of AF, justify increased attention to the management of patients with AF. PCV69 Cost Of Illness: Heart Failure In Ireland Corry S E 1, McDonald K 2, Kennelly B 3 1Novartis Ltd, Dublin, Ireland, 2St Vincents University Hospital, Dublin, Ireland,, 3National University of Ireland Galway, Galway, Ireland . . . . Objectives: Map and cost the patient pathway for acute and chronic heart failure patients in Ireland. Obtain the key cost parameters using a top down approach, whilst highlighting the economic and societal impact of this non communicable disease. Methods: A prevalence based top down approach was applied to estimate the cost of heart failure in Ireland for the year 2012, taking a societal perspective. Data on primary and all case diagnoses were obtained from the National Hospital In-Patient Enquiry Scheme (HIPE). Market research identified admissions in private hospital. Patients hospitalised with heart failure have been taken as representative of patients with Acute Decompensated Heart Failure (ADHF). All treatment pathways associated with the direct cost of both prevalent and incident cases of heart failure were confirmed and validated by health care professionals in primary and secondary care settings. Indirect and economic costs consisted of informal care and Life Years Lost (LYL), respectively. These were estimated using national surveys and registries, such as the Central Statistics Office (CSO). LYL were monetised using Irish specific life tables and the value of a statistical and/or the QALY threshold. Where data were lacking or uncertain, treatment algorithms were confirmed and validated through scientific engagement with key opinion leaders. Results: The total cost of heart failure in Ireland for the year 2012 was estimated at € 662.3 million. Indirect and economic costs were identified as the largest cost component at €473.7m (76%), while the direct cost represented 24% of the cost burden. Conclusions: Once HF has developed, the pathophysiology of the condition is characterised by high morbidity and mortality, frequent hospitalisation, and a large reliance on informal care. This study identified large regional variation in available heart failure services, hence impacting the necessary follow up care. As a result, a substantial cost driver in the management and care of heart failure patients is informal care. This cost has not been considered in previous heart failure studies, making it an important area for future research. PCV70 Cost Of Bleeding In Trauma Patients Zbrozek A CSL Behring, King of Prussia, PA, USA . Objectives: Trauma patients often experience significant hemorrhage and, therefore, may require more resources during hospitalization than patients who do not hemorrhage. The cost of blood/blood products is poorly understood in trauma patients and not well documented in the literature. Our objective was to evaluate the outcomes and cost of care for trauma patients based on bleeding status. Methods: Patients with costs and length of stay (LOS) greater than zero, discharged between January 1, 2010 and December 31, 2012, were identified in the Premier Hospital Database based on select ICD9 codes. All costs were based on hospital reported costs. Patients who received blood products prior to the day of surgery were excluded. Multivariate regres- A485 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 sion models for costs and LOS used log transformation techniques. Results: A total of 62,527 patients (8,800 Bleeding; 53,727 Non-Bleeding) were included. Patients who received blood/blood products were significantly different in a number of baseline characteristics including Charlson Comorbidity Index. Unadjusted outcomes demonstrated significant differences in mean total costs ($43,375 Bleeding; $18,411 NonBleeding; p<0.001) and LOS (13.62 Bleeding; 7.53 Non-Bleeding; p< 0.001) for patients who received blood products compared with those who did not. Multivariate models adjusting for patient characteristics demonstrated that patients who received blood/ blood products had 38.9% greater LOS and 34.3% greater ICU LOS. They also were 3.53 times as likely to be admitted to the ICU, 3.75 times as likely to be readmitted for bleeding and 4.09 times as likely to die in hospital (all p<0.001). Furthermore, they had 32.9% greater total cost of care including blood product cost and 31.8% greater total cost of care excluding blood product cost (both p<0.001). Conclusions: Preventing or rapidly controlling emergent bleeding in trauma patients would likely reduce patient risk and avoid elevated costs of hospitalization. Average costs of treating unstable angina were € 2,217 - 3,644 PA. Direct medical costs of chronic HF patients were between € 3,150 - 4,792 per patient per year. Average treatment costs for hospitalized PAD in the acute phase were € 4,186 PA, € 2,138 during month 1-6 after initial hospitalization, € 1,350 in month 7-12, and € 1,172 in month 13-18. For stroke, total direct medical costs in year one were € 11,408 per patient. Total direct medical costs during the 1st year after an ischemic stroke event were € 15,573 - 18,517 per patient, € 5,280 in month 13-18, and € 5,479 per year in the subsequent 4 years. Conclusions: MI, unstable angina, HF, stroke and PAD have a high financial impact on the German health care system. Treatment costs in the first year after event were highest for MI with €11,672 – 12,713 per patient followed by stroke with € 11,408 per patient demonstrating need for improvement in preventing CVE. PCV75 Annual Cost Of Conservative Treatment Of Supraventricular Tachycardias In Poland Farkowski M M 1, Golicki D 2, Czech M 3, Pytkowski M 1, Maciag A 1, Wood K A 4, Kowalik I 1, Szwed H 1 1Institute of Cardiology, Warsaw, Poland, 2HealthQuest, Warsaw, Poland, 3Medical University of Warsaw, Warsaw, Poland, 4Duke University School of Nursing, Durham, NC, USA . PCV71 The Economic Impact Of Hypertension In Health Care System Of Pakistan . . . . . . . . . PCV72 Hospital Costs Of Ischemic Stroke And Transient Ischemic Attack In The Netherlands Objectives: Supraventricular tachycardias (SVT) are a group of arrhythmias which result in significant impairment of health-related quality of life (HRQoL). While not as common as atrial fibrillation, SVT are still a significant health care problem since an average time from the onset of symptoms to successful ablation exceeds 10 years both in highly and less developed countries. The aim of this study was to calculate an annual cost of conservative treatment of SVT in Poland. Methods: This was an economical part of The PPRA study (NCT01594814). PPRA was a single-center, prospective, cohort study conducted in tertiary care cardiology center in Poland. All data concerning health care resources utilization: outpatient and inpatient visits, emergency room (ER) consultations, drugs etc. were derived from the study. Costs of disease related groups (DRG), outpatient visits and reimbursed drugs were derived from official price lists of Polish National Health Fund (NHF). Costs of ER visits were calculated as a mean cost of visit derived from seven hospitals of different levels of care. Costs of medical transport were calculated as a mean cost derived from four different centers providing ambulance services. Cost of non-reimbursed drugs and of outpatient visits and diagnostic tests in private sector were calculated as a mean cost of those resources available on official websites of nation-wide providers. All costs are presented in Euro (EUR) as median [interquartile range]. Results: During 1.5 years, 82 patients were enrolled and the data for this analysis was available for all patients. The yearly median cost of conservative treatment of SVT was EUR 415 [237727] from the NHS perspective and EUR 468 [280-793] from the societal perspective. Those costs were mainly driven by costs of hospitalizations and ER. Conclusions: Annual cost of conservative treatment of SVT is substantial and driven mainly by hospital services. Buisman L R 1, Tan S S 1, Koudstaal P J 2, Nederkoorn P J 3, Redekop W K 1 1Erasmus University Rotterdam, Rotterdam, The Netherlands, 2Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands, 3Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands PCV76 Linking Health Care Administrative Databases And National Registry Data In Order To Monitor Icd Therapy In Italy Iqbal M S 1, Iqbal M W 2, Bahari M B B 1, Iqbal M Z 1 1Department of Clinical Pharmacy, Faculty of Pharmacy, AIMST University, Kedah, Malaysia, 2Faculty of Law, Universiti Malaya,, Kualalumpur, Malaysia . . . . . . . . . Objectives: Hypertension (HTN) is a severe public health issue across the globe and a major risk factor for various cardiovascular diseases. This study was aimed to examine the potential costs of prognostic utilities and direct and indirect health care costs of the treatment of HTN in patients visiting public hospitals in Pakistan. Methods: A cross-sectional, convenient-sampling oriented study was conducted in patients attending public hospitals in Pakistan. The direct and indirect health care costs were assessed by different variables i.e. consultation fees, cost of medicines, travelling costs, laboratory test expenses. All obtained data were analyzed using descriptive and inferential statistics. Results: The mean annual direct health care cost for a HTN patient was around PKRs. 19789.88 (US$ 201.21) and indirect health care cost was PKRs. 11990.90 (US$ 121.92). It was also observed that the mean indirect health care costs per patient were significantly varied (p < 0.001) among different public hospitals. Conclusions: Type of the HTN, distance to the hospital and length of stay in the hospital were proportional to the direct and indirect health care costs of the HTN patients. Better management of HTN can favorably impact the disease burden as untreated HTN or its comorbidities increase the overall treatment costs which affect affordability of the patient. . . . . . . . . . . Objectives: Advances in the prevention and treatment of ischemic stroke and transient ischemic attack (TIA) during the past two decades have resulted in changes of medical practice. However, the existing cost estimates of ischemic stroke and TIA are based on resource use and cost data of approximately 20 years ago. We therefore determined hospital costs of ischemic stroke and TIA in the Netherlands for 2012. Methods: A retrospective cost analysis was conducted using patient and resource use data from the Diagnosis Treatment Combination (Diagnose Behandeling Combinatie) casemix system. Four subgroups were analysed: inpatient ischemic stroke, inpatient TIA, outpatient ischemic stroke, and outpatient TIA. Additionally, we estimated the costs for carotid endarterectomy. Unit costs were based on 2012 reference prices of the Dutch Manual of Costing and tariffs provided by the Dutch Healthcare Authority. Ordinary least squares regression analysis was used to examine the association between hospital costs and various patient and hospital characteristics. Results: A total of 35,903 ischemic stroke and 21,653 TIA patients were included. Inpatient care costs for ischemic stroke were € 5,328 while the costs for TIA were € 2,470. Outpatient costs were estimated at € 495 for ischemic stroke and € 587 for TIA. Average hospital costs for carotid endarterectomy were estimated at € 6,836. The number of inpatient days was the greatest contributor to hospital costs for ischemic stroke and TIA patients. Age, type of hospital, and region were strong predictors of hospital costs. Conclusions: This study is the most recent and extensive cost analysis of inpatient and outpatient hospital costs of ischemic stroke and TIA patients. Our results may be used as input for economic evaluations and health economic models to support decision making about reimbursement, investment, and pricing of health care interventions. PCV74 Costs Of Treating Cardiovascular Events In Germany: A Systematic Literature Review Schmid T 1, Xu W 2, Gandra S R 3, Michailov G 1 GmbH, München, Germany, 2Pharmerit International, Rotterdam, The Netherlands, 3Amgen, Inc., Thousand Oaks, CA, USA . . . . . 1Amgen Objectives: Cardiovascular events (CVE) are the number one cause of death (42%) in Germany. This study evaluates available literature regarding direct medical costs related to CVE in Germany. The CVE of interest were myocardial infarction (MI), unstable angina, heart failure (HF), stroke, and peripheral artery disease (PAD). Update of a review conducted in 2012. Methods: A systematic literature search was performed in the following databases- Medline, Embase, Centre for Reviews and Dissemination, TIBORDER, and German dissertation database from 01/2003 to 10/2013. Observational studies and randomized clinical trials were considered for the review. Results: The search identified 400 publications; 13 were included in this review. For MI, average hospitalization costs during acute phase were between € 5,836 and € 7,522 per admission (PA). In the first year after a MI, direct medical costs were € 11,672 - 12,713 per patient. Madotto F 1, Fornari C 1, Chiodini V 1, Mantovani L G 2, Zecchin M 3, Proclemer A 4, Conti S 1, Cesana G 1 1University of Milano - Bicocca, Monza, Italy, 2Federico II University of Naples, Naples, Italy, 3“Ospedali Riuniti” Hospital, Trieste, Italy, 4“S. Maria della Misericordia” Hospital, Udine, Italy . . . . . . . . . Objectives: The main purpose of study was to evaluate the utilization of implantable cardioverter defibrillator (ICD) with or without cardiac resynchronization pacing (CRT-D) in Lombardy, the most populated Italian region with universal health care coverage for about 10 million inhabitants, from 2000 to 2010. The second aim was to assess health care resources utilization after a first ICD implantation, highlighting differences between implant indication (primary and secondary) or ICD type (single-chamber, dual-chamber and CRT-D). Methods: Data were extracted from: i) the data warehouse DENALI that organizes health care administrative databases concerning all subjects covered by Lombardy Health System; ii) ICD National registry data. Linking DENALI and registry data, we developed and validated (with Cohen’s kappa coefficient) an algorithm to distinguish hospitalizations for ICD in first implant and replacement in order to estimate the annual rates of first implant (per million person-years) and replacement (per hundred person-years). We selected a cohort of patients who underwent a first ICD implantation and we followed them until 12/31/2010, recording vital status and health care resources consumed: hospitalizations, drugs and outpatient claims. Results: We identified 25,358 hospitalizations for ICD implantation: 17,864 for a first ICD and 8,034 for a replacement. 86% of hospitalizations found a match in the National registry and the Cohen’s kappa coefficient showed values over 0.8 from 2005. The annual rates were 232.5 (95%CI: 228.5-236.4) and 10.5 (95%CI: 10.3-10.7) for first ICD and replacement respectively. We selected a cohort of 12,525 patients who underwent a first ICD: 55% were implanted in primary prevention and the ICD types were: 35% single-chamber, 29% dual-chamber and 35% CRT-D. Among these groups, we detected differences in survival, ICD duration and health care resources utilization. Conclusions: The combined use of information from national ICD registry and health care administrative databases could overcome the limitation of both data sources and it could improve the monitoring of ICD therapy. PCV77 Does A 12-Lead Ecg More Reliably Detect Atrial Fibrilation Than A Rhythm Strip Only Ecg? Vyas V , Duran J , Ansaripour A , Niedzielko M , Steel A , Bakhai A Barnet & Chase Farm Hospitals NHS Trust, London, UK . . . . . . Objectives: The 12-lead electrocardiogram (ECG) is the current gold standard for the diagnosis of atrial fibrillation (AF). However, undertaking a 12-lead ECG is cumbersome, expensive and time consuming compared to a rhythm strip alone. The AF detection rate using a single lead ECG compared with a12-lead ECG amongst trainees of varying degrees of experience has not yet been reviewed in depth. Methods: Five doctors of different grades and specialties and one Cardiology Specialist Nurse A486 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 were each asked to review a different batch of ECGs with differing proportions of AF and Sinus Rhythm (SR). ECGs were presented to the reviewers on two separate occasions in two variations: full 12 lead and 10-second rhythm strip only. The diagnoses (AF/SR or unsure) were compared to a Consultant Cardiologist reading the 12-lead ECG. Results: Overall, compared to the Consultant Cardiologist, the AF detection rate using a 12-lead ECG was 81% and 82% with a rhythm strip. There was a significant difference (chi squared test, p< 0.01) in AF detection rate when the reviewers were divided into 2 groups according to level of cardiology experience. The AF detection rate for the 12-lead ECG amongst specialists vs. non-specialists was 93% vs. 70% (P< 0.001) and 92% vs. 73% (P= 0.003) for the rhythm strip. Conclusions: Our findings indicate that a 10-second rhythm strip alone has a comparable AF detection rate to a 12-lead ECG in the hands of doctors and nurses when measured against the consultant’s diagnosis. With both the rhythm strip and a 12-lead ECG, the accuracy of AF detection improved with experience. Mass screening using a single strip could be acceptable and inexpensive particularly with mobile phone technology. PCV78 Cost Consequence Comparison Of Hemostatic Matrix Agents Beby A T 1, Faivre P 2, Zinck R 1, Kuntze C E 2 BV, Utrecht, The Netherlands, 2Baxter Healthcare SA, Opfikon-Glattpark, Switzerland . . . . . . clinical response to treatment and the number of QALYs per patient accrued during the study. Only direct medical costs (drug acquisition, administration and hospitalization costs) were incorporated in the model, as the analysis was conducted from a third-party payer perspective. With respect to administration cost, two alternative scenarios were considered in the base case analysis: administration in day-case unit and administration in the hospital outpatient department. Probabilistic sensitivity analysis was conducted. Primary outcomes were quality adjusted life years (QALYs) and incremental cost-effectiveness ratios (ICER). Results: In the base case analysis, QALYs of FCM treated patients were higher compared to no iron treated patients by 0.04 QALYs. The total 24-week cost of FCM was higher by € 969 and € 204, it the two scenarios respectively. This difference was mainly attributed to the administration cost and drug acquisition cost related to FMC. Incremental cost effectiveness analysis showed that treatment with FCM was a cost-effective option resulting in an ICER of € 25,506 and € 5,368 per QALY gained in the scenarios respectively. Probabilistic sensitivity analysis revealed that FCM was likely to be cost-effective in over 80% and 99% in the two scenarios respectively, at a willingness-to-pay threshold of €34,000 per QALY gained. Conclusions: Ferric carboxymaltose may be a cost – effective option in relation to no iron treatment for the management of iron deficiency of HF patients in Greece. 1Baxter Objectives: Hemostatic agents are used to stop intraoperative bleeding in the presence of actively flowing blood when applied directly to the bleeding site. Recent published literature on animal studies and real-world outcome data indicated that the hemostatic matrix agent Floseal could stop bleedings and reduce complications more effectively than Surgiflo with thrombin. The objective of this study is to quantify cost consequences in the Netherlands when using Floseal versus Surgiflo with thrombin from a hospital perspective. Methods: A cost-consequence model was built based on a large retrospective analysis of a Premier’s US Hospital database to assess the value of using Floseal to achieve hemostasis in mixed cardiac surgery procedures relative to Surgiflo with thrombin. The model captures the cost implications on the following parameters: (1) operating room time; (2) postoperative bleedingrelated complication rates; (3) the need for surgical revisions due to bleeding, and; (4) the number of intraoperative and postoperative blood transfusions. Costs (€ ) were derived from published literature, the Dutch guideline for cost research and 2014 national list prices. The model assumed a surgical case load of 100 mixed cardiac surgeries. Results: By reducing operating time, but moreover the number of complications, surgical revisions and blood products transfusion, Floseal can lead to a cost saving of € 2,933 per patient in comparison to the use of Surgiflo with thrombin. Conclusions: Our preliminary analysis indicates that using Floseal to stop intraoperative bleeding could lead to significant cost saving for the hospitals in the Netherlands. Further studies are required to confirm these findings. PCV79 Cost-Effectiveness Of Dabigatran Compared With Warfarin, Apixaban, Rivaroxaban And Low Molecular Weight Heparins For The Treatment And Secondary Prevention Of Venous Thromboembolism In Colombia Rosselli D 1, Rueda J D 2, Wolowacz S 3, Brockbank J 4, Abeysinghe S 4 1Pontificia Universidad Javeriana Medical School, Bogotá, Colombia, 2Pontificia Universidad Javeriana, Bogotá, Colombia, 3RTI Health Solutions, Manchester, UK, 4RTI Health Solutions, Didsbury, UK . . . . . . PCV81 Potential Cost-Effectiveness Of Therapeutic Drug Monitoring In Patients With Resistant Hypertension Chung O 1, Vongpatanasin W 2, Haverkamp W 1, Dorenkamp M 1 1Charite University Berlin, Campus Virchow-Klinikum, Berlin, Germany, 2University of Texas Southwestern Medical Center Dallas, Dallas, TX, USA . . . . Objectives: Non-adherence to drug therapy poses a significant problem in the treatment of patients with presumed resistant hypertension (RH). It has been shown that therapeutic drug monitoring (TDM) is a useful tool for detecting non-adherence and identifying barriers to treatment adherence, leading to effective blood pressure (BP) control. However, the cost-effectiveness of TDM in the management of RH has not been investigated. Methods: A Markov model was used to evaluate life-years, quality-adjusted life-years (QALYs), costs, and incremental cost-effectiveness ratios in RH patients receiving either TDM optimized therapy or standard best medical therapy. The model ran from the age of 30 to 100 years or death, using a cycle length of 1 year. Efficacy of TDM was modeled by reducing risk of hypertension-related morbidity and mortality. Cost analyses were performed from a payer’s perspective. Deterministic and probabilistic sensitivity analyses were conducted. Results: In the age group of 60-year olds, TDM gained 1.07 QALYs in men and 0.97 QALYs in women at additional costs of € 3,854 and €3,922, respectively. Given a willingness-to-pay threshold of €35,000 per QALY gained, the probability of TDM being cost-effective was ≥ 95% in all age groups from 30 to 90 years. Results were influenced mostly by the frequency of TDM testing, the rate of non-responders to TDM, and the magnitude of effect of TDM on BP. Conclusions: Therapeutic drug monitoring presents a potentialy highly cost-effective health care intervention in patients diagnosed with RH. Importantly, this finding is valid for a wide range of patients, independent of gender and age. PCV82 Cost-Effectiveness Analysis Of Ivabradine In Heart Failure With Reduced Left Ventricular Ejection Fraction In Spain Fernandez de Bobadilla J , Gonzalez Fernandez O , Lopez de Sa E , Lopez-Sendon J L Hospital La Paz, Madrid, Spain . . . . . Objectives: To evaluate cost-effectiveness of dabigatran and new oral anticoagulants (NOA) compared to currently reimbursed warfarin and low molecular weight heparins (LMWH) for thromboembolic events prophylaxis from a payer’s perspective in Colombia. Methods: Markov decision model based on efficacy, utilities and safety inputs from clinical trials (CT) (RE-COVER I and II; EinsteinDVT; Einstein-PE; AMPLIFY; RE-SONATE; RE-MEDY and a meta-analysis of 18 CT for LMWH. Cost of medication was obtained from SISMED, Vademecum Med®, and government reference prices; costs of events were estimated from hospital billing records, POS tariffs, SOAT Manual and local experts. Costs are reported in euros (1 € = COP 2,655) and effectiveness in terms of deaths due to bleeding and Quality Adjusted Life Years (QALYs). A 3% discount rate was used for costs and QALYs. A cohort of 1,000 patients, with similar demographic characteristics to those reported in the CT, was followed for a 5-year time horizon; with analysis for 6 months, 1 year and lifetime. A univariate and probabilistic sensitivity analysis was performed. Results: Deaths due to bleeding were 5 with dabigatran, 7 with apixaban, 9 with LMWH, 7 with rivaroxaban and 15 with warfarin, per 10,000 patients. Dabigatran is dominant compared with LMWH and apixaban. Compared with rivaroxaban, dabigatran is safer with similar costs and QALYs gained. Considering a base line scenario comparing all alternatives versus warfarin, dabigatrán demonstrates the lowest ICER (€ 32,121/QALY) compared with rivaroxaban (€ 36,388/ QALY), apixaban (€ 72,228/QALY) and LMWH (€ 604,515/QALY). Our findings were consistent in the sensitivity analyses. Conclusions: Dabigatran is the alternative with the best value for money when all alternatives are compared with warfarin, and is dominant compared with LMWH, which are reimbursed by the current Colombian National Plan POS. Objectives: SHIFT trial demonstrated that, by decreasing heart rate with Ivabradine, the incidence of events is reduced in patients with HF due to left ventricular systolic disfunction (LVSD) and heart rate (HR) over 70 bpm. A cost-effectiveness analysis demonstrating that this treatment strategy is cost-effective in United Kingdom has been published, but there are no publications with Spanish data. The objetive of this analysis is to determine whether treatment with Ivabradine in HF due to LVSD is a cost-effective strategy in Spain. Methods: A Markov model with three health states was built (baseline, admission due to HF and death), based on a simulated cohort of patients suffering HF due to LVSD and HR over 70 bpm, in order to compare the treatment strategies of the SHIFT trial, Ivabradine vs placebo, on top of standard treatment, was performed. A cost-efectiveness analysis with a life-time horizon, under the perspective of the NHS. Spanish costs were used to feed the model, and it was assumed that quality of life was worse during hospitalizations. A probabilistic sensitivity analysis was performed. The model took into account the incidence of complications related with HF (admission for HF, readmission for HF and death) and was fed with transition probabilities taken from the SHIFT trial and the SHIFT-hospitalization trial. Costs were obtained from cost data bases, the National Statistics Institute of Spain, institutions and scientific journals. Results: Cost per QALY of Ivabradine in comparison to placebo in heart failure due to left ventricular systolic disfunction is € 17,488 € , well below the acceptability threshold accepted in our enviroment (around 30,000 € ). Cost per LYG was € 13,044. Results were robust in the performed sensitivity analyses. Conclusions: In conclusion, treatment with Ivabradine in heart failure due to left ventricular systolic dysfunction is cost-effective in Spain. PCV80 Economic Evaluation Of Ferric Carboxymaltose In Patients With Chronic Heart Failure And Iron Deficiency: An Analysis For Greece Based On Fair-Hf Trial PCV83 Pharmacoeconomic Assessment Of Apixaban Versus Standard Of Care For The Prevention Of Stroke In Italian Non-Valvular Atrial Fibrillation Patients Mylonas C 1, Kourlaba G 2, Berberian K 3, Maniadakis N 1 1National School of Public Health, Athens, Greece, 2Collaborative Center for Clinical Epidemiology and Outcomes Research (CLEO), Athens, Greece, 3GENESIS Pharma SA, Athens, Greece 1AdRes Objectives: To evaluate the cost-effectiveness of iron repletion with ferric carboxymaltorse (FCM), in iron-deficient heart failure (HF) patients in Greece. Methods: An international economic model was locally adapted to evaluate the use ferric carboxymaltorse to standard care over a time horizon of 24 weeks corresponding to the duration of the FAIR-HF trial. The efficacy of therapy was evaluated based on the Objectives: The objective of this study was to evaluate the cost-effectiveness of apixaban in the prevention of thromboembolic events in patients with nonvalvular atrial fibrillation (NVAF) relatively to standard of care (warfarin or aspirin) from the Italian National Health System (SSN) perspective. Methods: A previously published Markov model was adapted for Italian NVAF patients. Clinical . . . . Pradelli L 1, Calandriello M 2, Di Virgilio R 3, Bellone M 1, Tubaro M4 HE&OR, Turin, Italy, 2HE OR Unit - Bristol-Myers Squibb S. r. l., Rome, Italy, 3Pfizer Pharmaceuticals, Rome, Italy, 4San Filippo Neri Hospital, Rome, Italy . . . . A487 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 effectiveness data were derived from head-to-head randomized trials (ARISTOTLE and AVERROES); main events considered in the model were ischemic and hemorrhagic stroke, systemic thromboembolism, bleeds (both major and clinically relevant minor) and cardiovascular hospitalizations. Expected survival was projected beyond trial duration using national mortality data adjusted for individual clinical risks and adjusted by utility weights for health states derived from literature. Unit costs were collected from published Italian sources. Costs (2013 € ) and health gains accruing after the first year were discounted at an annual 3.5% rate. Deterministic and probabilistic sensitivity analyses (PSA) were carried out to assess the effect of input uncertainty. Results: Incremental LYs (0.31/0.19), QALYs (0.28/0.20), and costs (1,932/1,104) are predicted with the use of apixaban relative to aspirin and warfarin, respectively. The incremental cost effectiveness ratios (ICERs) of apixaban were € 6,794 and € 5,607 per QALY gained, respectively. In PSA, the probability of apixaban being cost effective relative to aspirin and warfarin was 95% and 93%, respectively, for a WTP threshold of € 20,000 per QALY gained. Univariate analyses indicate that results were most sensitive to variations of the absolute risk reduction for cardiovascular events with apixaban. Conclusions: Apixaban is expected to increase life expectancy and quality-adjusted life expectancy, but also costs dedicated to Italian NVAF patients, as compared to standard of care. The resulting ICERs have high probabilities of being below the conventional thresholds of WTP for health benefits of the SSN, indicating efficient allocation of health care resources. PCV84 Assessing The Cost Effectiveness Of An Anticoagulation Clinic In Comparison With The Usual Medical Clinic In Kuala Lumpur Hospital Thanimalai S 1, Shafie A A 1, Ahmad Hassali M A 2, Sinnadurai J 3 1Universiti Sains Malaysia, Penang, Malaysia, 2Discipline of Social and Administrative Pharmacy, School of Pharmaceutical Sciences, Universiti Sains Malaysia, Penang, Malaysia,, 3Hospital Kuala Lumpur, Kuala Lumpur, Malaysia . . . . . . Objectives: Systematic anticoagulation management clinic is now recommended to manage warfarinized atrial fibrillation (AF) patient. In Malaysia, the service is recently introduced as pharmacist managed Warfarin Medication Therapy Adherence Clinic (WMTAC). The objective of the present study was to assess the cost effectiveness of anticoagulation clinic in comparison with usual medical in Kuala Lumpur Hospital. Methods: A Markov model built using the provider perspective and 20 year time horizon was used to assess the cost effectiveness. The base case analysis assumed a cohort of patients with AF 57 years of age with comorbid illnesses. Data sources include a 6 month retrospective cohort analysis of the effectiveness of the clinics, the cost of drugs, cost of personnel and space of the clinics, cost of monitoring and cost of adverse events were obtained from the local source and publications. The transition probabilities of these clinics outcomes were obtained from a literature search. Future costs were discounted by 3% to convert to present values. All costs were in Ringgit Malaysia (RM) based on year 2012. Results: The results of a 20-year period model showed that UMC was dominated by the WMTAC in the same time period. The mean cost of the WMTAC was RM 5864 whereas the UMC cost was RM 6550. The sensitivity analysis showed that clinic treatment costs and effectiveness influenced the cost-effectiveness. If the cost of WMTAC was increased by 50% of the current cost, the WMTAC would not be a dominant intervention. WMTAC was also cost effective for a willingness to pay of RM32000. Conclusions: The anticoagulation management service appears to cost less and provide greater effectiveness than usual care. In conclusion, the Markov model suggests that from the provider perspective the anticoagulation clinic is a more cost effective option than the usual medical clinic in Kuala Lumpur Hospital. PCV85 Cost-Effectiveness Of Ranolazine For The Treatment Of Angina Pectoris In Russia Gorokhova S G 1, Ryazhenov V V 1, Gorokhov V D 2, Maximkin S A 3 1I. M. Sechenov First Moscow State Medical University, Moscow, Russia, 2Scientific Clinical Center of JSC Russian Railways, Moscow, Russia, 3Center for strategic research in healthcare, Moscow, Russia . . . . . . . . Objectives: To assess cost-effectiveness of ranolazine in Russian patients with angina pectoris. Methods: A model was developed to compare a strategy of treatment of angina pectoris with ranolazine and a traditional therapy in patients with at least 3 heart attacks per week. The analysis was based on the results of ERICA (Efficacy of Ranolazine in Chronic Angina) study that included Russian patients and reflected a routine Russian practice for the treatment of the disease. The analysis included costs of drug therapy, emergency care and hospitalization. Effectiveness rate was measured as a change of frequency of angina attacks compared to initial level estimated by SAQ (Seattle Angina Questionnaire). Sensitivity analysis considered the change of parameters in different length of hospitalization. Results: Total costs in ranolazine group were insignificanlty higher than in control group (573.84 RUB per patient). At the same time, ranolazine changed the structure of costs: expenditures on medications grew while costs of emergency care and hospitalization reduced. CER value was 1640.86 in ranolazine group, and 1964.63 (1.2 times less) in group of traditional therapy. The change of hospitalization length in sensitivity analysis was also more favorable in case of ranolazine: 164 479.81 vs 197 300.02 RUB/unit, which confirms previous results. Conclusions: The use of ranolazine in patients with > 3 heart attacks of angina pectoris per week is clinically and economically substantiated. PCV86 Cost-Effectiveness Analysis Of Ticagrelor In Treating Patients With Acute Coronary Syndrome In Hong Kong Wu D B C 1, Lee B S C 2, Lee K K C 1 University Malaysia, Selangor, Malaysia, 2Prince of Wales Hospital, Shatin, Hong Kong, Hong Kong, Hong Kong . . . . . . . . . 1Monash Objectives: The multi-centered, double-blind, randomized PLATO trial on 18,624 patients from 43 countries has demonstrated that ticagrelor was superior in reduc- ing cardiovascular mortality, myocardial infarction (MI), or stroke among patients with acute coronary syndrome (ACS) compared to clopidogrel but without a significant increase in major bleedings. In Hong Kong, generic clopidogrel was introduced in 2012. This study aimed to evaluate the long-term cost-effectiveness of ticagrelor in ACS patients in Hong Kong (HK) from a public hospital’s perspective. Methods: A two-phase Markov model was used to estimate the short- and long-term costeffectiveness measured as cost per quality-adjusted-life-year (QALY) and cost per life-year-gained (LYG) over 1 year, 5 years and patients’ lifetime. Direct medical costs were HK-specific and patients’ resource use, rate of cardiovascular events (i.e. MI and stroke) and utility data were from published literature. All costs were presented as 2014 figures, cost and effectiveness were both discounted at 3% per annum. Sensitivity analyses were performed to test model robustness. Results: Despite the great difference in the daily drug cost of ticagrelor and generic clopidogrel (US2.8, vs US0.10,1US= 7.8HK), the overall cost of management between the 2 groups remains similar. Our study shows that the incremental cost-effectiveness ratios (ICER) of ticagrelor were reduced substantially from US16,071/LYG and US19,493/ QALY in the first year to US302/LYG and US357/QALY over a lifetime time horizon due to improvements in health outcomes. The ICER values were all cost-effective based on the WHO 3xGDP criteria (GDP 2013= US37,860). The results are sensitive to cost of generic clopidogrel. Conclusions: Treating ACS patients with lifetime use of ticagrelor can potentially reduce the cost of management and increase the cost-effectiveness due to better health outcomes as compared to generic clopidogrel. Ticagrelor therapy appears to be cost-effective both on short- and long-term assessment in the public health care sector of Hong Kong. PCV87 The Cost Of Increasing Physical Activity And Decreasing Body Mass Index For Mid-Life African Women Johnson T J , Wilbur J , Fogg L , Schoeny M Rush University, Chicago, IL, USA . . . . . Objectives: The purpose of this project was to evaluate the incremental costs of increasing physical activity and improving body composition for a lifestyle walking program targeting sedentary African American women. Methods: The Women’s Lifestyle Physical Activity Program was a randomized behavioral trial that included a group intervention with social support and culturally relevant, tailored content about increasing physical activity and an automated telephone response system to track physical activity for sedentary African American women ages 40-65. The principal outcomes were change in minutes of walking and moderate and vigorous physical activity per week, and body mass index (BMI) between baseline and 24 weeks. Incremental cost effectiveness ratios (ICER) were calculated for each outcome. The cost-effectiveness analysis included both program and participant costs and was calculated in 2013 US dollars. Results: For the 260 participants in the analysis, participant costs (e.g., group session attendance, logging physical activity in the ATR system) were $152 ± 42, while program costs (e.g., group session facilitation, materials and supplies) were $164 ± 21, for a total cost of $316 ± 59. Walking increased by 200 minutes per week at 24 week, with an ICER of $1.64 (95% CI, 1.63 – 1.66) per minute, moderate and vigorous physical activity increased by 117 minutes per week, with an ICER of $2.79 per minute, and BMI decreased by 0.09 points, with an ICER of $3463 (95% CI, 500 – 9,540) per 1-point reduction. Conclusions: The Women’s Lifestyle Physical Activity Program is a relatively low cost strategy for increasing physical activity. The incremental cost of increasing walking minutes is substantially lower than for moderate and vigorous physical activity. The participant costs related to time in the program were more than half of the total costs, suggesting that practitioners and policymakers should consider the participant cost when disseminating group programs into practice. PCV88 Cost Effectiveness Analysis Of Apixaban Versus Other Noacs For The Prevention Of Stroke In Italian Non-Valvular Atrial Fibrillation Patients Pradelli L 1, Calandriello M 2, Di Virgilio R 3, Bellone M 1, Tubaro M 4 HE&OR, Turin, Italy, 2HE OR Unit - Bristol-Myers Squibb S. r. l., Rome, Italy, 3Pfizer Pharmaceuticals, Rome, Italy, 4San Filippo Neri Hospital, Rome, Italy . . . . . 1AdRes Objectives: The study assessed the lifetime cost-effectiveness of apixaban in preventing thromboembolic events in non-valvular atrial fibrillation (NVAF) patients, as compared to other novel oral anticoagulant agents (NOACs), from the Italian Health System (SSN) perspective. Methods: A previously published Markov model was adapted. Baseline clinical risks were assigned based on the demographic and clinical features of the patients; effectiveness parameters derived from adjusted indirect comparison using warfarin as link. Expected survival was projected beyond trial duration using national mortality data adjusted for clinical risks and weighted by published utilities. Unit costs were collected from official and published Italian sources. Costs (2013-€ ) and health gains occurring after the first year were discounted at an annual 3.5% rate. Deterministic and probabilistic sensitivity analyses (DSA&PSA) were carried out. Results: In the short to medium term, apixaban was associated with marginal LYs and QALYs gains and slight savings. However, as apixaban extended expected survival versus dabigatran (110mg), dabigatran (150mg) and rivaroxaban (0.13,0. 08, and 0.06 LYs or 0.11,0. 07, and 0.05 QALYs), expected total lifetime costs exceeded those of these comparators (319,282,16 € ). Corresponding ICERs were estimated in € 2,911, € 3,882 and € 327 per QALY gained. In PSA, the probabilities of apixaban being cost effective with a WTP threshold of 20,000 € /QALY gained were 99%, 92% and 93% for the same comparisons. The most influential parameter according to DSA was daily cost of NOACs, but the corresponding ICERs remained well below commonly accepted WTP values. Conclusions: Apixaban is expected to be more effective than dabigatran and rivaroxaban in Italian NVAF patients, and marginally more costly due to costs in added years of life. The ICERs have a high likelihood of being below conventional thresholds of WTP for health benefits of the SSN and suggest that apixaban is cost-effective compared with other NOACs. A488 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PCV89 Novel Imaging Technology To Select Patients For Individualized Therapies: Test Performance And Cost-Effectiveness Buisman L R 1, Rijnsburger A J 1, Koudstaal P J 2, Redekop W K 1 1Erasmus University Rotterdam, Rotterdam, The Netherlands, 2Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands . . . . . . . . Objectives: Non-invasive molecular imaging tests are currently being developed to improve individual stroke risk prediction in patients with a recent transient ischemic attack (TIA) or minor ischemic stroke. We estimated the minimum performance (i.e., sensitivity and specificity) that a test must have in order to be more cost-effective than current clinical practice. Methods: The aim of imaging tests is to identify which patients should undergo surgery instead of receiving medicines. Decision modelling was used to estimate the minimum performance of a confirmatory test that is performed based on the result of an initial duplex ultrasonography. The comparators were patient management according to Dutch guidelines and three observed clinical practice strategies. Scenario analyses were performed in which the sensitivity and specificity were varied to estimate the minimum test performance needed to be cost-effective versus the comparators. Results: A perfect confirmatory test (100% sensitivity and specificity) with a cost of € 390 for a 60-year-old man is cost-effective versus all comparators. Assuming 100% sensitivity, a test must be at least 71% specific to be cost-effective versus the Dutch guidelines using a threshold of € 30,000/QALY. Similarly, assuming 100% specificity, a test must be at least 52% sensitive to be cost-effective. In addition, assuming 90% sensitivity, a test must have a specificity of at least 77% to be cost-effective. Moreover, the minimum values of sensitivity and specificity needed for a test to be cost-effective are dependent on the QALY threshold used. Conclusions: An imaging test that improves risk prediction and therefore treatment decisions for patients with a recent TIA or minor ischemic stroke has the potential to improve cost-effectiveness by reducing the risk of recurrent stroke. However, developers must consider if the minimum test performance required to be cost-effective can be achieved; other concerns of payers (e.g. budget impact) must also be considered. PCV90 Cost-Effectiveness Of Apixaban Compared To Other Anticoagulants For Lifetime Treatment And Prevention Of Recurrent Venous Thromboembolism Lanitis T 1, Hamilton M 2, Rublee D A 3, Leipold R 4, Quon P 4, Browne C 1, Cohen A T 5 1Evidera, London, UK, 2Bristol-Myers Squibb Company, Princeton, NJ, USA, 3Pfizer, Inc., New York, NY, USA, 4Evidera, Bethesda, MD, USA, 5Guy’s and St Thomas’ NHS Foundation Trust, London, UK . . . . . . . . . Objectives: Guidelines suggest only 3-6 months of anticoagulant treatment in most venous thromboembolism (VTE) patients due to concerns that the bleeding risk with vitamin K antagonists (VKAs) outweighs the reduced risk of VTE recurrence in extended treatment. However, non-VKA novel oral anticoagulants (NOACs) have been studied recently for extended VTE treatment. Apixaban demonstrated superiority to placebo in VTE reduction without increasing risk of major bleeding in the AMPLIFY-EXT trial, justifying reassessment of the potential benefit of extending treatment. This analysis reports cost effectiveness of lifetime treatment with apixaban versus rivaroxaban, dabigatran, and low-molecular-weight heparin (LMWH)/ KA from the perspective of the UK National Health Service (NHS). Methods: A Markov model was developed that includes the following health states: recurrent VTE, major bleed, clinically-relevant non-major bleed, chronic thromboembolic pulmonary hypertension, and death. Transition rates among health states were based upon AMPLIFY and AMPLIFY-EXT clinical trial data, network meta-analyses, discontinuation due to clinical events, and UK life tables. Costs were from UK NHS Healthcare Resource Group tables and utilities were from published literature. The primary outcome of interest was incremental cost per quality adjusted life year (QALY) gained. Results: Compared to other anticoagulants, lifetime treatment with apixaban was projected to result in fewer bleeds and fewer recurrent VTEs. The lower bleeding risk with apixaban led to fewer treatment discontinuations, longer time on treatment, and fewer recurrent VTEs. The reduced number of clinical events led to increased QALYs at a nominal cost increase, due primarily to longer treatment duration with apixaban. Incremental costs per QALY gained were £2,781, £619, and £10,820 for apixaban versus dabigatran, rivaroxaban, and LMWH/VKA, respectively. Sensitivity analyses indicated that results were robust to a wide range of inputs. Conclusions: Apixaban for lifetime treatment of VTE can offer substantial clinical benefits and is a cost-effective alternative to other NOACs and LMWH/VKA. PCV91 Cost-Effectiveness Of Extracorporeal Cardiopulmonary Resuscitation In Patients With Refractory Cardiac Arrest Buriskova K 1, Rogalewicz V 1, Ostadal P 2 University in Prague, Kladno, Czech Republic, 2Nemocnice Na Homolce, Praha 5, Czech Republic . . . 1Czech Technical Objectives: Extracorporeal life support (ECLS) has been recently introduced as a therapeutic option for refractory cardiac arrest (extracorporeal cardiopulmonary resuscitation - ECPR). Despite growing evidence demonstrating improved survival rate with ECPR in refractory cardiac arrest, a number of questions remains unanswered and data on cost-effectiveness of this approach are still insufficient. Methods: Retrospective cost-effectiveness analysis was performed from the provider’s perspective. Sixteen patients undergoing ECPR were included into the analysis (ECPR group) and the data were compared with 35 subjects with conventional CPR for refractory cardiac arrest (non-ECPR group). Results: In the ECPR group seven out of sixteen patients were weaned from ECMO, four of them with good neurological outcomes (CPC 1); three patients survived one year with CPC 1, one patient survived one year with severe neurological dysfunction (CPC 3) and one patient with persisting coma (CPC 4). In comparison, in the non-ECPR group all patients died within 24 hours. In the ECPR group, when using Levitronix Centrimag blood pump (Thoratec, USA), the average annual costs per patient reached CZK885,044 (~EUR32,260), with the Cardiohelp device (Maquet, Germany) CZK788,432 (~EUR28,738). The cost utility analysis revealed 3,961,970 CZK/QALY (~144,413 EUR/QALY) with Cardiohelp, and 4,447,457 CZK/QALY (~162,109 EUR/ QALY) when using Levitronix Centrimag. However, if only patients with CPC 1 were included into the analysis, the cost-utility ratio decreased to 834,616 CZK/QALY (~30,422 EUR/QALY) and 908,253 CZK/QALY (~33,106 EUR/QALY) for Cardiohelp and Levitronix, respectively. Conclusions: Our data indicate that ECPR for refractory cardiac arrest might be cost effective despite the high costs per individual patient treated with this approach. Larger studies are, however, required to confirm these observations. PCV92 Cost Effectiveness Of Ivabradine In Chronic Heart Failure Patients With Heart Rate Above Bpm In Taiwan Chang C J 1, Chu P H 2, Fann C S J 3 Gung University, Kwei Shan, Tao Yuan, Taiwan, 2Chang Gung University, Taoyuan, Taiwan, 3Academia Sinica, Taipei, Taiwan . . . . . . . 1Chang Objectives: Raised heart rate is a risk factor for cardiovascular events and death. Heart rate reduction could be an important role in management of chronic heart failure. Ivabradine is a new, pure and specific heart rate lower therapy. We assessed the cost effectiveness of Ivabradine in CHF with heart rate 75 bpm using the perspective of a middle-income country’s public health insurance system. Methods: Model-based analysis closely mirrored the SHIFT trial was performed. Micro-simulation method of CHF disease, mortality and hospitalization were used to compare Ivabradine plus standard care with standard care only. A two state Markov model has been employed to simulate the natural history of disease. Costs were estimated from national health insurance research database in Taiwan, a single-payer program that offers universal health care coverage. And another estimated costs using a medical center in Taiwan with clinical examination value were performed to validate the accuracy of costs. Incremental cost per quality adjusted life year (QALY) gained is used to express the value of Ivabradine over a lifetime horizon and 29 months of trial period. Results: The incremental cost per additional QALY for Ivabradine plus standard care versus standard care has been estimated as £14,832 for trial period and £7,634 for lifetime. Ivabradine is cost-effective according to NICE threshold. And the result was robust in sensitivity analysis. Conclusions: In a middle-income country like Taiwan, the use of Ivabradine to treat the eligible CHF patients is likely to be cost effectiveness. PCV93 Economic Analysis Of Thrombo Incode, A Clinical-Genetic Function For Assessing The Risk Of Venous Thromboembolism Rubio-Terrés C 1, Soria J M 2, Morange P E 3, Suchon P 3, Souto J C 4, Mateo J 4, Saut N 5, RubioRodríguez D 1, Sala J 6, Gracia A 6, Pich S 7, Salas E 7 de Genòmica de Malalties Complexes Sant Pau, Barcelona, Spain, 3Inserm UMR_S 1062, F-13385, and Aix-Marseille Université, Marseille, France, 4Unitat d’Hemostasia i Trombosis IIB-Sant Pau, Barcelona, Spain, 5Inserm Unité Mixte de Recherche en Santé (UMR_S) 937; ICAN Institute for Cardiometabolism and Nutrition, Université Pierre et Marie Curie, Paris, France, 6Scientific Department Ferrer inCode, Barcelona, Spain, 7Scientific Department Gendiag. exe, Barcelona, Spain . . . . . . . . . . . . . . . 1Health Value, Madrid, Spain, 2Unitat Objectives: To conduct an economic analysis of the risk assessment of venous thromboembolism (VTE) from the perspective of the Spanish National Health System with Thrombo inCode versus conventional/standard method used to date (Factor V Leiden and Prothrombin G20210A). Methods: An economic model was created from the National Health System perspective using a decision tree in patients aged 45 with a life expectancy of 81 years. The predictive capacity of VTE based on the identification of thrombophilia with Thrombo inCode and of the standard method was obtained from two case-control studies (S. PAU and MARTHA) conducted on two different populations (1,451 patients in all). Although this is not always the case, in this analysis the least favourable situation was assumed for Thrombo inCode, with patients identified as suffering from thrombophilia being subject to preventive treatment of VTE with warfarin, leading to a reduction in the number of VTE events and an increased risk of severe bleeding. The health states utilities (quality-adjusted life years, QALY) and costs were obtained from the literature and Spanish sources. Results: Based on a price of € 180 for Thrombo inCode, this would be the dominant option (more effective and with lower costs than the standard method) in both populations. The Monte Carlo probabilistic analyses indicate that the dominance would occur in 100% of the simulations in both populations. The threshold price of Thrombo inCode needed to reach the incremental cost-effective ratio (ICER) generally accepted in Spain (€ 30,000 per QALY gained) would be between € 3,950 (MARTHA) and € 11,993 (S. PAU). Conclusions: Thrombo inCode is a dominant option in assessing the risk of VTE compared to the standard method currently used. PCV94 A Cost-Effectiveness Analysis Of Interventions For Symptomatic Varicose Veins Marsden G 1, Perry M 2, Bradbury A 3, Hickey N 4, Kelley K 2, Trender H 5, Wonderling D 2, Davies A H 6 1Office of Health Economics, London, UK, 2National Clinical Guideline Centre, London, UK, 3University of Birmingham, Solihul, UK, 4Worcestershire Royal Hospital, Worcester, UK, 5Sheffield Teaching Hospital Foundation Trust, Sheffield, UK, 6Imperial College & Imperial College NHS Trust, London, UK . . . . . . . . . Objectives: Treatment of varicose veins (VV) has been shown to increase health related quality of life. However, the cost-effectiveness of such treatments has been uncertain. Indeed there is great regional variation across the UK in how, and if, VV are treated. This economic analysis formed part of the NICE clinical guideline on VV, with an objective to investigate whether such treatments should be recommended across the UK. Methods: An economic analysis was conducted to compare the A489 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 cost-effectiveness of surgery, endodermal ablation (ETA), ultrasound guided foam sclerotherapy (UGFS) and compression stockings (CS). The analysis was based on a Markov decision model, which was developed in consultation with members of the NICE guideline development group (GDG). The model had a five year time horizon, and took the perspective of the UK National Health Service. Clinical inputs were based on a network meta-analysis (NMA), informed by a systematic review of the clinical literature. Outcomes were expressed as costs and quality adjusted life years (QALYs). Results: All interventional treatments were found to be cost-effective compared to CS at a cost- effectiveness threshold of £20,000 per QALY gained. ETA was found to be the most cost-effective strategy overall, with an incremental costeffectiveness ratio of £3,161 per QALY gained compared to UGFS. Surgery and CS were dominated by ETA. Conclusions: Interventional treatment for VV is costeffective in the UK NHS. Specifically, based on current data, ETA is the most costeffective treatment in people for whom it is suitable. The results of this research were used to inform recommendations within the NICE guideline on VV. Funding: This work was undertaken by the National Clinical Guideline Centre, which received funding from the National Institute for Health and Clinical Excellence (NICE). The views expressed in this publication are those of the authors and not necessarily of the institute. PCV95 The Cost-Effectiveness Of Dabigatran Etexilate Compared With Rivaroxaban In The Treatment Of Acute Venous Thromboembolism In The Uk Jugrin A V 1, Ustyugova A V 2, Urbich M 3, Lamotte M 1, Sunderland T J 4 1IMS Health HEOR, Vilvoorde, Belgium, 2Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany, 3Boehringer Ingelheim Ltd, Bracknell, UK, 4Boehringer Ingelheim, Berkshire, UK . . . . . . . . Objectives: Venous thromboembolism (VTE) including deep vein thrombosis (DVT) and pulmonary embolism (PE) is a common cardiovascular disorder. Acute VTE has been traditionally managed with short course parenteral anticoagulation followed by 3-6 months of a vitamin-K antagonist. Novel oral anticoagulants do not require routine coagulation monitoring and dose adjustment, thus potentially providing an alternative treatment option. The cost-effectiveness of dabigatran vs. rivaroxaban over a 6 months treatment course in the UK health care setting was evaluated in this research. Methods: A life-time Markov model was developed, encompassing recurrent VTE events and VTE-related deaths, and the most common adverse events during anticoagulation therapy: major or clinically relevant bleeds (MCRB). The model was populated with data from pooled RE-COVER and RE-COVER II dabigatran trials and the 6 months treatment duration subgroup of the rivaroxaban EINSTEIN-DVT and EINSTEIN-PE trials. Long-term consequences of VTE were considered. Costs were analysed from the NHS and Public Social Services perspectives. Health outcomes were assessed in quality-adjusted life years (QALY). Utility values for modelled health states were EQ-5D data from RE-COVER studies, and published data. Probabilistic sensitivity analyses (PSA) were undertaken. Results: In patients with index PE, 6 months treatment with dabigatran dominated treatment with rivaroxaban projecting less recurrent VTE and less MCRB at lower costs. Dabigatran was likely to remain cost-effective in 70% of cases at a threshold used in the UK of £30,000/QALY gained. Dabigatran continued to dominate treatment with rivaroxaban in patients with index DVT, projecting less non-fatal PE, less intra-cranial haemorrhages and less clinically relevant bleeds, but more recurrent DVT, with 68% likelihood of remaining cost-effective. In the pooled DVT/PE group, dabigatran dominated treatment with rivaroxaban and was 62% likely to remain cost-effective. Conclusions: Dabigatran is less costly and more effective than rivaroxaban when administered for 6 months after index PE, index DVT or both. PCV96 The Cost-Effectiveness Of Dabigatran Etexilate Compared With Warfarin And Rivaroxaban In The Treatment Of Acute Pulmonary Embolism In The Uk Jugrin A V 1, Ustyugova A V 2, Urbich M 3, Lamotte M 1, Sunderland T J 4 1IMS Health HEOR, Vilvoorde, Belgium, 2Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany, 3Boehringer Ingelheim Ltd, Bracknell, UK, 4Boehringer Ingelheim, Berkshire, UK . . . . . . . . Objectives: This economic evaluation aimed to assess the cost-effectiveness of dabigatran etexilate for six months of treatment for acute pulmonary embolism (PE) compared with warfarin and rivaroxaban in the UK health care setting. Methods: A Markov state-transition cohort model was used to project the lifetime recurrence of PE and occurrence of deep vein thrombosis (DVT) in patients with initial acute PE. The incidence of recurrent venous thromboembolism (rVTE) was based on the relevant study endpoints of RE-COVER and RE-COVER II trials comparing dabigatran with warfarin, and the EINSTEIN-PE study for rivaroxaban. Intervention-specific probabilities of events within the composite efficacy endpoint and within the composite safety endpoint of major or clinically relevant bleeding (MCRB) were sourced from the very same randomised trials. Beyond the data from clinical studies, the probability of rVTE was sourced from the published literature. Long-term consequences of VTE were considered, namely chronic thromboembolic-induced pulmonary hypertension and post-thrombotic syndrome. The perspective on costs was that of the NHS and Public Social Services. Health outcomes were assessed in quality-adjusted life years (QALY). Utility values relevant to events and health staes were EQ-5D data from RE-COVER studies, and published literature. Probabilistic sensitivity analyses (PSA) were undertaken. Results: Compared with warfarin, dabigatran projected similar number of rVTE, but was associated with less MCRB. The expected lifetime incremental cost-effectiveness ratios (ICERs) were £2,004/life years (LYs) and £1,285/QALYs. PSA showed 84% likelihood for dabigatran to remain cost-effective given a willingnessto-pay of £30,000/QALY. When compared with rivaroxaban, treatment with dabigatran was projected dominant,. Dabigatran projected less rVTE, less intracranial haemorrhages and clinical relevant non-major bleeds, but was associated with a higher risk of major bleeds. PSA showed 66% likelihood for dabigatran to remain cost-effective over rivaroxaban. Conclusions: In patients treated for acute pulmonary embolism, dabigatran is a cost-effective alternative compared with both warfarin and rivaroxaban. PCV97 Cost-Effectiveness Of Apixaban Compared To Warfarin And Aspirin In Patients With Non-Valvular Atrial Fibrillation (Nvaf) In The Russian Federation Rudakova A V 1, Parfenov V A 2 1St. Petersburg Chemical Pharmaceutical Academy, Saint-Petersburg, Russia, 2I. M. Sechenov First Moscow State Medical University, Moscow, Russia . . . . Objectives: To evaluate cost-effectiveness of the novel oral anticoagulant apixaban compared to warfarin and aspirin in patients with NVAF from the Russian Federation national health care system perspective. Methods: Cost-effectiveness analysis was based on a Markov model that allowed estimation of the incremental cost-effectiveness ratio (ICER) for apixaban compared to warfarin and aspirin over lifetime horizon in vitamin K antagonists (VKA) suitable and VKA unsuitable patients with NVAF, respectively. The model enclosed cardiovascular event rates derived from the randomized clinical trials: ARISTOTLE and AVERROES. The following cardiovascular events were considered: ischemic and hemorrhagic stroke, intracranial hemorrhage, systemic embolism, other major bleeds, clinically relevant non-major bleeds, myocardial infarction and cardiovascular hospitalizations. Characteristics of the baseline patients’ cohort including quality of INR control corresponded to the local population. Direct medical costs were determined based on the rates of the compulsory national medical insurance system. The price of the antithrombotic drugs was taken as a weighted average tender price in 2013. Cost-effectiveness threshold was set at 1,4 million rubles per quality-adjusted life year (QALY) gained and corresponded to the three times GDP per capita in the Russian Federation in 2013. One-way sensitivity analyses were undertaken to examine the effects of model drivers. Results: In the base case analysis it was demonstrated that apixaban compared to warfarin and aspirin provided additional 0.187 and 0.214 QALYs, respectively. With that estimated ICER for apixaban compared to warfarin and aspirin was 603.92 and 473.02 thousands rubles per QALY gained, respectively. Sensitivity analysis indicated that results were robust over explored range of inputs. Conclusions: According to the results of the modeling study apixaban may be considered as a cost-effective alternative to warfarin in VKA suitable patients and as a cost-effective alternative to aspirin in VKA unsuitable patients for NVAF treatment from the Russian Federation national health care system perspective. PCV98 The Cost-Effectiveness Of Dabigatran Etexilate Compared With Warfarin In The Treatment And Secondary Prevention Of Acute Venous Thromboembolism In The Uk Jugrin A V 1, Ustyugova A V 2, Urbich M 3, Lamotte M 1, Sunderland T J 4 1IMS Health HEOR, Vilvoorde, Belgium, 2Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany, 3Boehringer Ingelheim Ltd, Bracknell, UK, 4Boehringer Ingelheim, Berkshire, UK . . . . . . . . Objectives: This economic evaluation aimed to assess the cost-effectiveness of dabigatran compared with warfarin, in the treatment and secondary prevention of acute venous thromboembolism (VTE) comprising deep vein thrombosis (DVT) and pulmonary embolism (PE), in the UK health care setting, based on safety and efficacy data collected during pivotal phase III trials. Methods: A life-time Markov state-transition cohort model was developed around the two primary composite endpoints in the pivotal trials: recurrent VTE and VTE-related death (rVTE), and major or clinically relevant bleeding (MCRB). Intervention-specific probabilities of events within the composite endpoint rVTE (recurrent DVT; recurrent PE) and the composite endpoint MCRB (intracranial haemorrhage; other major bleeds; non-major bleeds) were sourced from the pivotal trials. Beyond the trials follow-up period, the probability of rVTE was sourced from the literature. Longterm consequences of VTE were considered, namely chronic thromboembolic pulmonary hypertension and post-thrombotic syndrome. The perspective on costs was that of the NHS and Public Social Services. Health outcomes were assessed in quality-adjusted life years (QALY). Utility values for health states and events were EQ-5D data collected within the clinical trials, and the published literature. Probabilistic sensitivity analyses (PSA) were undertaken. Results: In patients with index DVT, the estimated incremental cost-effectiveness ratio (ICER) of treatment with dabigatran compared with warfarin was £614 per QALY gained; in patients with index PE, the ICER was £1,285/QALY; in the pooled DVT/PE group, the ICER was £862/QALY. In the treatment followed by secondary prevention analysis, ICER was £8,319/QALY. PSA suggested that the probability of dabigatran being cost-effective at a threshold of £30,000/QALY was 90%, 81% and 94% in acute treatment and 96% in secondary prevention respectively. Conclusions: In a UK setting, dabigatran appears to be a cost-effective option for treatment and secondary prevention of VTE in patients with acute DVT and acute PE compared with warfarin. PCV99 Economic Evaluation Of Valsartan Versus Olmesartan Addition To Amlodipine And Hydrochlorothiaziade Single-Pill Triple Antihypertensive Therapy Stafylas P 1, Mavrodi A 2, Kourlaba G 3, Hatzikou M 4, Rombopoulos G 4, Maniadakis N5 1Medical Research & Innovation LP, Thessaloniki, Greece, 2University of Macedonia, Thessaloniki, Greece, 3Collaborative Center for Clinical Epidemiology and Outcomes Research (CLEO), Athens, Greece, 4Novartis Hellas, Metamorfosis, Greece, 5National School of Public Health, Athens, Greece . . . . . Objectives: The aim of the study was to compare the cost-utility of the two singlepill triple combination antihypertensive therapies available in the Greek market for patients with moderate to severe hypertension; the valsartan (V) against the olmesartan (O) combination with amlodipine (A) and hydrochlorothiazide (H). Methods: A Markov model with eight health states was constructed. The short- A490 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 term effect of antihypertensive treatment on blood pressure was extracted from relative clinical trials and these data were extrapolated through the Hellenic SCORE and Framingham risk equations, estimating the long-term survival and qualityadjusted life-years (QALYs) gained. Costs and outcomes were evaluated over lifetime, divided into annual cycles and were discounted at 3.0% with 2014 as reference year. The analysis was conducted from the Greek third-party-payer perspective (EOPYY). Results: The total lifetime cost related to V/A/H combination was estimated to be lower (€ 10,970) compared to that of the O/A/H combination (€ 11,080), despite the higher drug acquisition cost. Moreover, the estimated QALYs gained with the V/A/H combination were 10.88 vs. 10.80 for O/A/H combination. Therefore, the V/A/H combination was found to be a dominant alternative over O/A/H combination, as it was associated with lower cost and greater efficacy. In a scenario when the ambulatory blood pressure measurements were taken into account, the ICER was far lower than the Greek GDP per capita (€ 6,845/QALY) vs. O/A/H combination, suggesting V/A/H combination to be a cost-effective choice. Extensive sensitivity analyses confirmed the robustness of the results. The probabilistic sensitivity analysis also demonstrated that there was about 80% probability for the V/A/H triple combination to be cost-effective at a willingness-to-pay threshold of €16,000/ QALY. Conclusions: This is the first study performed to compare the cost-utility of the two single-pill triple antihypertensive therapies. The V/A/H combination was proven dominant over the O/A/H combination for the treatment of moderate to severe hypertension. PCV100 Cost-Effectiveness Of High-Sensitive Troponin Assays For The Early Rule-Out Or Diagnosis Of Acute Myocardial Infarction (Ami) In People With Acute Chest Pain: A Nice Diagnostic Assessment Ramaekers B L T 1, Armstrong N 2, Joore M A 3, Westwood M 4, Whiting P 2, Thokala P 5, Ross J 2, Kleijnen J 2, Severens J 6, van Asselt A 7 University Medical Center,, Maastricht, The Netherlands, 2Kleijnen Systematic Reviews Ltd., York, UK, 3Department of Clinical Epidemiology and Medical Technology Assessment, Maastricht University Medical Centre, Maastricht, The Netherlands, 4Kleijnen Systematic Reviews Ltd, York, UK, 5University of Sheffield, Sheffield, UK, 6Erasmus University Rotterdam, Rotterdam, The Netherlands, 7University of Groningen, Groningen, The Netherlands . . . . . . . . . . . . PCV102 Efficiency Of Rehabilitation Programs For Patients After Traumatic Brain Injury And Acute Cerebrovascular Accident (Stroke) In Russia Omelyanovsky V 1, Avksentieva M V 2, Derkach E V 3, Dombrovskiy V S 1, Fedyaev D 1 1The Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia, 2Research Center for Clinical and Economic Evaluation and Pharmacoeconomics, Moscow, Russia, 3Russian State Medical University, Moscow, Russia . . . . . . . . Objectives: The objective of the present study was to assess efficiency of rehabilitation programs for patients after traumatic brain injury and acute cerebrovascular accident (stroke) in Russia. Methods: Short-term clinical and social outcomes (health status and disability rates) of rehabilitation were analyzed in the database of the Moscow Center of Speech Pathology and Neurological Rehabilitation. Changes in the officially registered disability rates and clinical outcomes were assessed for 3 different strategies of rehabilitation: hospital, day care and home care. The decision tree model was constructed to simulate disability rates, direct and indirect costs of rehabilitation vs “no rehabilitation” scenario under conditions when officially registered disability corresponds to real health and functional status of patients. TreeAge Pro 2009 and Microsoft Excel 2010 software were used for modeling. Results: Use of officially registered disability as an endpoint does not reflect the actual effectiveness of rehabilitation programs. 90% of patients are able to live without assistance after discharge but still are registered as 1stdegree disabled (most severe degree of disability in Russia) in order to receive social benefits. According to preliminary results of modeling total cost of rehabilitation may be less than cost of ”no rehabilitation” scenario if disability correlates with actual health and functional status of individuals, for example annual total cost is € 25,923 for home care rehabilitation and € 28,124 for “no rehabilitation” scenario. Conclusions: It’s necessary to improve approaches to official disability registration in Russia in order to make rehabilitation programs efficient. . 1Maastricht Objectives: To assess cost-effectiveness of high sensitivity troponin (hs-cTn) assays for the management of adults presenting with acute chest pain at the emergency department. Methods: An economic model was constructed to estimate lifetime costs and QALYs of five hs-cTn strategies (differing according to manufacturer, timing of the test, number of tests and cutoff point for a positive test result) compared to standard troponin (sTn) testing at presentation and at 10-12 hours, which was considered the reference standard. In the base case, it was assumed that sTn testing had perfect accuracy for diagnosing AMI and only patients with a positive test for sTn were at increased risk for adverse events and would benefit from immediate treatment. In a secondary analysis, a proportion of patients with a positive hs-cTn test and a negative sTn test were at increased risk for adverse events and would benefit from immediate treatment. Results: Base case: Strategies considered cost-effective depending upon ICER thresholds were Abbott ARCHITECT hs-cTnI 99th centile (thresholds < £6,597), Beckman Coulter hs-cTnI 99th centile (thresholds £6,597 - £30,042), Abbott ARCHITECT hscTnI optimal strategy (thresholds £30,042 - £103,194), and the sTn test (thresholds > £103,194). Secondary analysis: Strategies considered cost-effective were Abbott ARCHITECT hs-cTnI 99th centile (< £12,217), Roche Elecsys hs-cTnT 99th centile (£12,217 - £14,992) and Abbott ARCHITECT hs-cTnI optimal strategy (> £14,992). STn was dominated. Sensitivity/subgroup analyses: Main drivers are: the difference in outcomes between treated and untreated patients, and treatment costs for patients testing false-positive. Hs-cTn testing is more cost-effective in younger age, pre-existing coronary artery disease, and symptom onset < 3hrs ago. No testing is only cost-effective when pre-test prevalence is 1%. Conclusions: There is no strong evidence to prefer one hs-cTn testing strategy over another. Results do indicate that hs-cTn testing in general may be cost-effective compared to sTn testing, especially in the secondary analysis. PCV101 Cost Effectiveness Of Renal Denervation Therapy For The Treatment Of Resistant Hypertension In The Netherlands Henry T 1, De Brouwer B F E 2, Van Keep M M L 3, Blankestijn P J 4, Bots M L 4, Koffijberg H 4 Health Solutions, Sheffield, UK, 2Medtronic Trading NL BV, Heerlen, The Netherlands, 3BresMed Health Solutions, Utrecht, The Netherlands, 4University Medical Center, Utrecht, The Netherlands PCV103 Impact Of A Pharmacological Cardioversion With Vernakalant On The Management Cost Of Recent Atrial Fibrillation In Belgium Lamotte M 1, Gerlier L 1, Caekelbergh K 1, Lalji K 2, Polifka J 2, Lee E 2 1IMS Health HEOR, Vilvoorde, Belgium, 2Cardiome Pharma Corp, Vancouver, BC, Canada . . . . . . Objectives: We aimed at estimating and comparing the total management costs of patients admitted to the emergency department (ED) with recent (< 48 hours) atrial fibrillation (AF) between three types of cardioversion: direct current cardioversion (DCC), intravenous amiodarone or intravenous vernakalant. Methods: A decision analytic model was developed to mimic the pathways of patients admitted via the ED with recent AF and to calculate an average AF management cost per strategy. The cardioversion success rates were based on published observational studies (DCC 90%, amiodarone 68%, vernakalant 70%). In case of successful conversion the patient was released directly from the ED to home (DCC 50%, amiodarone 25%, vernakalant 100%) or admitted to the hospital for a median of 1 day (DCC) or 2 days (amiodarone) reflecting cardioversion-specific times to sinus rhythm. After a failed pharmacological cardioversion, patients were assumed to receive a DCC; failed DCC was followed by intravenous amiodarone. The associated inpatient costs were retrieved from the IMS Hospital Disease Database (HDD2011). ED treatments were costed using the national health care payer tariffs. Univariate and probabilistic sensitivity analyses were performed. Results: The total AF management costs from ED to discharge were estimated at € 952 (DCC), € 1,894 (amiodarone) and € 1.354 (vernakalant). Given the relative frequencies of amiodarone (55%) and DCC (45%) to treat recent FA in Belgian ED, the weighted average management cost was €1.470 (+ € 116/treatment vs. vernakalant). The break-even situation is reached when 75% of successfully treated vernakalant patients avoid the hospitalization (base case 100%). Based on a probabilistic sensitivity analysis, vernakalant was cost saving in 55% of the 1,000 simulations vs. current management. Conclusions: In patients with recent AF, the rapid mode of action of vernakalant increases the chance of obtaining sinus rhythm within 48 hours, and its use was cost saving compared to the current average Belgian practice. PCV104 Cost Model Analysis Of Gore® Propaten® Vascular Graft VersUS Standard Eptfe Vascular Graft For Infrapopliteal Bypass In Peripheral Arterial Disease (pad) Management: Spanish Scenario 1BresMed Scarpa F W. L. GORE & Associati S. r. l., Verona, Italy Objectives: Recent studies have demonstrated the safety and efficacy of catheter-based renal denervation (RDN) for the treatment of resistant hypertension. These studies have been used to estimate the cost effectiveness of this approach, however, there is no such published estimation in the Dutch health care setting. We aimed to determine the cost effectiveness of RDN from the perspective of the health care payer in The Netherlands. Methods: A Markov state-transition model previously constructed in TreeAge® was adapted to Microsoft Excel®. The Excel-based model was updated with costs and quality-adjusted life years (QALYs) relevant to the Dutch setting and the cost effectiveness of RDN was compared with standard of care (SoC) for patients with resistant hypertension. The efficacy of RDN treatment was modelled as a reduction in the risk of cardiovascular events associated with a lower systolic blood pressure (SBP). Results: Base case deterministic results showed that treatment with RDN resulted in an increase in QALYs of 0.89 at an incremental cost of € 1,315 per patient resulting in an ICER of just € 1,474 Deterministic and probabilistic sensitivity analyses (PSA) indicated treatment with RDN therapy was cost effective at conventional willingness-topay thresholds used in The Netherlands of € 10,000–80,000 per QALY. Furthermore, 9.5% of PSA iterations showed RDN to be a cost saving treatment compared to SoC. Conclusions: RDN is a cost-effective intervention for patients with resistant hypertension in The Netherlands. Objectives: Demonstrate the cost savings of using the GORE® PROPATEN® Vascular Graft compared to standard ePTFE vascular grafts in the management of PAD patients. Superior clinical outcomes in terms of primary / secondary patency and limb salvage rates result in lower average per-patient treatment costs. Background: Vascular bypass is used in patients with PAD (Peripheral Arterial Disease) to treat ischemic rest pain, to improve walking distance in patients with severe life-limiting claudication, and to save limbs that might otherwise require amputation. When patients presenting for peripheral artery reconstruction have absent or inadequate saphenous veins due to prior use, small size, or poor quality, vascular surgeons may choose a prosthetic bypass graft. The GORE® PROPATEN® Vascular Graft features a proprietary end-point covalent linkage of heparin molecules to the luminal surface of the graft that provides sustained resistance to thrombosis. Methods: A cost model analysis was developed to represent hospital treatment costs. A typical PAD patient’s treatment pathway in Spanish clinical practice was identified by a survey involving Spanish vascular surgeons. A literature review was conducted to determine patency and limb salvage rates. Spanish GRD AP27–2012-Ministerio de Sanidad tariffs were used as hospital cost of treatment inputs. A three-year patient management cycle was considered. Results: Better patency and limb salvage rates obtained using the GORE® PROPATEN® Vascular Graft result in fewer reinterventions and amputations, corresponding to lower per-patient treatment costs. The cost model demonstrates an overall cost saving for PAD patient management using the GORE® PROPATEN® Vascular Graft for infrapopliteal bypass. The cumulative . . . . . . . . . . . . A491 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 cost savings at 3 years is estimated to be 11.671€ per patient. Conclusions: The use of the GORE® PROPATEN® Vascular Graft for infrapopliteal bypass in the PAD patient population represents a safe, clinically effective, and cost-saving alternative to standard ePTFE vascular grafts. PCV105 Pharmacoeconomic Analysis Of Rosuvastatin Use In Patients With Hypercholesterolemia In The Health Care Of Belarus Kozhanova I 1, Romanova I I 1, Gavrilenko L 1, Sachek M 2 State Medical University, Minsk, Belarus, 2The Belarusian Center for Medical Technologies, Computer Systems, Administration and Management of Health, Minsk, Belarus . . . . . 1Belarusian Objectives: Pharmacoeconomic analysis of rosuvastatin use in patients with hypercholesterolemia in the health care of Belarus has been performed to determine economic advisability of its applying in Belarus. As there is own production of statins (generics of lovastatin, atorvastatin and simvastatin) in Belarus, the insclusion of new statin (rosuvastatin) in the clinical protocols requires pharmacoeconomic study. Methods: Overview of statins available in Belarus has been conducted. Equivalent effective dose to achieve target of low density lipoprotein cholesterol (CH-LPLD) values were established on the basis of published data. Costminimization analysis has been used. Model “decision tree” to achieve the target CH-LPLD values has been built on the basis of STELLAR trial. Statin doses required to achieve the target CH-LPLD values have been calculated. The costs of achieving the target CH-LPLD values have been evaluated. The cost of each statins treating during the year has been calculated. Results: The highest cost has been obtained for the equivalent dose of lovastatin ($ 0.35) compared with atorvastatin ($ 0.31) and simvastatin ($ 0.28) manufactured in Belarus. Average price rosuvastatin (Merten ®) was comparable to the cost ($ 0.21) of Belarusian generics. The average cost of achieving the target CH-LPLD level was the lowest in the case of rosuvastatin - $ 170 compared with atorvastatin ($ 200) and simvastatin ($ 286) considering available statins of all manufacturers. Due to rosuvastatin’s lower effective dosage the costs of the one-year treatment with rosuvastatin is lower (on average 94 $) than with atorvastatin (all manufacturers - $ 100) and simvastatin (202 $). Conclusions: The study has demonstrated pharmacoeconomic acceptability of rosuvastatin use in the health care of Belarus. PCV106 Cost-Utility Analysis Of Hypertensive Treatment With Indapamide And Amlodipine Single-Pill Combination In The Polish Setting Stawowczyk E 1, Holko P 1, Kawalec P 2, Borowiec L 3, Filipiak K J 4 HTA Sp z o. o. Sp. komandytowa, Kraków, Poland, 2Jagiellonian University Medical College, Krakow, Poland, 3Medical Department, Servier Poland Ltd, Warsaw, Poland, 41st Chair and Department of Cardiology, Medical University of Warsaw, Warsaw, Poland . . . . . . 1Centrum Objectives: To assess cost-effectiveness of indapamide 1.5 mg +amlodipine 5/10 mg single-pill combination (SPC) compared with free combination (FC), in the Polish setting. Methods: A Markov cohort simulation model was used. Results of meta-analysis by Gupta et al show a difference in patients’ compliance between SPC and FC. Better compliance results in lower systolic blood pressure, which influences risk of cardiovascular events. Hence, compliance is associated with life expectancy and quality of life. Cardiovascular disease risks were based on the Framingham risk equations. Life-time horizon, Polish public payer perspective and patient perspective were applied. Indapamide/amlodipine SPC cost is based on average pharmacy prices reported in April 2014 (18.13PLN and 19.75PLN respectively for 1.5+5mg and 1.5+10mg /30 tabs); 30% patient copayment was assumed. Cost of FC was calculated as an average cost of reimbursed indapamide and amlodipine products in corresponding doses. All costs present 2014 values, and are expressed in Polish zloty (PLN). Costs and effects were discounted with 5% and 3.5% rates. Results: Indapamide/amlodipine SPC compared with FC generates additional life years (LYs) and quality adjusted life years (QALYs), and is highly cost-effective from public payer perspective and dominant from patient perspective. Difference between SPC and FC in LYs and QALYs was: 0.007960 and 0.020809. Difference in total costs from public payer perspective and from patient perspective was 113.14PLN (27.07EUR) and -211.31PLN (-50.56 EUR). ICUR from public payer perspective was 5,437PLN/QALY (1,301EUR/QALY). At prices +199% vs the base-case, SPC remains a cost-effective technology from public payer perspective according to the legally defined CE threshold (111,381PLN/QALY= 26,653EUR/ QALY). At prices -9.9% vs base-case, SPC is a dominant/cost saving technology vs the FC comparator. Conclusions: From public payer perspective, indapamide/ amlodipine SPC compared with FC is a highly cost-effective treatment option for hypertensive patients in contemporary Polish setting. From patient perspective, SPC is a dominant technology. PCV107 Cost-Effectiveness Of Ldl-P-Guided Statin Therapy Folse H J 1, Rengarajan B 1, Goswami D 1, Budoff M 2, Kahn R 3 1Evidera, San Francisco, CA, USA, 2Los Angeles Biomedical Research Institute, Torrance, CA, USA, 3University of North Carolina, Chapel Hill, NC, USA . . . . . . Objectives: Numerous trials have shown that lowering LDL cholesterol (LDL-C) reduces CVD events; however, at any LDL-C level, residual risk remains. LDL particle concentration (LDL-P) may be a better predictor of events, but no studies have evaluated its cost-effectiveness. We used the Archimedes model to evaluate the cost-effectiveness of using LDL-C or LDL-P in preventing cardiovascular disease in dyslipidemic patients. Methods: Archimedes is a highly detailed, large-scale simulation model of physiology, disease and health care systems. We created a simulated population of 1,000,000 individuals age 20-84 reflective of real subjects in the NHANES dataset. Because NHANES does not contain LDL-P values, they were imputed maintaining covariance with other biomarkers. The study had three arms: •Control: subjects evaluated for therapy for elevated LDL-C and treated with statins to LDL-C goals outlined in ATP-III; •LDL-P Alone: subjects evaluated and treated based solely on their LDL-P values; •Dual Arm: subjects evaluated by LDL-C but treated to both LDL-C and LPL-P recommended goals. Results: In the general population, the costs per quality-adjusted life year (QALY) associated with the use of LDL-P alone were $76,052 at 5 years and $8,913 at 20 years and with the use of both markers were $142,825 at 5 years and $25,505 at 20 years. In high-risk subpopulations, the use of LDL-P alone was cost-saving at 5 years; whereas the cost per QALY for the use of both markers was $14,250 at 5 years and $859 at 20 years for high-risk dyslipidemics, $19,192 at 5 years and $649 at 20 years for diabetics, and $9,030 at 5 years and $7,268 at 20 years for patients with prior CHD. Conclusions: Utilizing LDL-P to guide statin therapy is cost-effective in the long term for the general population, and cost-saving or cost-effective in the short term for high-risk patients. PCV108 Cost Effectiveness Analysis Of Ticagrelor Versus Generic Clopidogrel In The Treatment Of Patients With Acute Coronary Syndrome In Spain Capel M 1, Lopez-Sendon J L 2, Heras M 3, Carrera F J 4, Brosa M 5 La Paz, Madrid, Spain, 3Clinic Hospital, Barcelona, Spain, 4Dr. Moliner Hospital, Valencia, Spain, 5Oblikue Consulting, Barcelona, Spain . . . . . . . 1AstraZeneca, Madrid, Spain, 2Hospital Objectives: The aim of this study was to evaluate the long-term cost effectiveness of ticagrelor + aspirin versus clopidogrel + aspirin in patients with acute coronary syndrome (ACS) treated for 12 months in Spain. Methods: The cost effectiveness model consisted of a decision tree (1st year) based on the PLATO study and a long-term Markov model (2nd year onwards). This allowed estimation of cardiovascular events (death, myocardial infarction and non-fatal stroke), survival, health costs, and health related quality of life. A life time horizon was applied. The daily drug cost was € 0.60 and € 2.96 for generic clopidogrel and ticagrelor, respectively. Spanish unit costs and life tables were used; outcomes and costs were discounted at 3%. A sensitivity analysis across subgroups was carried out, and probabilistic sensitivity analysis was used to validate the robustness of the model. Results: Ticagrelor compared to clopidogrel was associated with a gain of 0.1586 life years and 0.1363 years of quality-adjusted life years (QALY), with an incremental cost of € 596. The incremental cost per life year and per QALY gained was € 3,760 and € 4,374, respectively. The probabilistic sensitivity analysis showed that ticagrelor was costeffective versus clopidogrel in > 99 % of the simulations given a willingness-to-pay threshold of € 15,000/QALY. The results were consistent across different subgroups of ACS patients. Conclusions: Ticagrelor + aspirin for 12 months is a cost effective treatment compared to generic clopidogrel + aspirin in patients with ACS treated invasively or conservatively, based on the findings of the PLATO study and Spanish health care costs. PCV109 Cost-Utility Analysis Of Carotid Artery Stenting Versus Endarterectomy For Symptomatic Carotid Stenosis Patients Oh SH1, You J H 1, Lee J Y 2, Park J J 1, Shin S 3 Evidence-based Healthcare Collaborating Agency, Seoul, South Korea, 2National Evidence-based Healthcare Collaborating Agency (NECA), Seoul, South Korea, 3National Evidencebased healthcare Collaborating Agency, Seoul, South Korea . . . . . . . 1National Objectives: This study was conducted to determine the cost-effectiveness of carotid Artery stenting (CAS) versus carotid endarterectomy (CEA) in patients with symptomatic carotid stenosis (more than 50% stenosis) in Korean health care system perspective. Methods: We performed a cost-utility analysis. Costs were estimated from retrospective chart review (CAS=346, CEA=331), health insurance claims data, and other national resources and expressed in 2013 KRW. Transition probabilities were estimated from retrospective chart and systematic review. Health utility index was assessed for general population using Time Trade Off (TTO) with health scenario. We used a Markov model to project 15-year costs and quality-adjusted life years (QALYs) for the 2 treatment groups. Results: In the base case analysis, CAS produced 6.49 QALYs, compared with 6.71 QALYs for CEA. The incremental cost of stenting was 1,691,740 KRW. In the base case analysis, CEA for patients with symptomatic stenosis had a greater benefit than CAS, with lower costs. In subgroup for patients with stenosis more than 70% or patient with over 80 years old, CAS was cost-effective. Sensitivity analyses showed that the major stroke or mortality influenced the results. However the results were consistent with the base analysis. Conclusions: Under the current circumstances in Korea, CEA was dominated by CEA in symptomatic steonsis. Therefore we concluded that CEA would be cost-effective intervention for carotid stenosis. To be economically competitive, the clinical effectiveness such as mortality and major stroke rates of CAS must be at least equivalent if not less than those of CEA. PCV110 Burden Of Hyperlipidemia Resulting From Productivity Loss Estimates From Population-Based Register Data In Sweden Banefelt J 1, Hallberg S 1, Gandra S R 2, Mesterton J 1, Fox K M 3, Paoli C J 2, Johansson G 4, Levin L Å 5, Sobocki P 6 1Quantify Research, Stockholm, Sweden, 2Amgen, Inc., Thousand Oaks, CA, USA, 3Strategic Healthcare Solutions, LLC, Monkton, MD, USA, 4Uppsala University, Uppsala, Sweden, 5Linköping University, Linköping, Sweden, 6IMS Health, Stockholm, Sweden . . . . . . . . . . . . . Objectives: To estimate productivity loss and associated indirect costs in workingage patients treated for hyperlipidemia. Methods: A retrospective populationbased cohort study was conducted using Swedish electronic medical records linked to national health registers and the Social Insurance Register. Patients were included based on a prescription of lipid-lowering therapy between January 1,2006 and December 31,2011 and followed until December 31,2012 for estimation of productivity loss and cost outcomes. Patients were stratified into three cohorts based on cardiovascular (CV) risk level. Results: Total mean days lost, measured as the sum of net sick leave and net disability pension days, during the one-year period following study inclusion was highest in the CV event history cohort (n= 6,881; 159 days), followed by the CV risk equivalent (RE) cohort (n= 3,226; 131 days) and the low/ A492 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 unknown risk cohort (n= 2,497; 86 days). The annual indirect cost related to work productivity loss was estimated at 209,704 SEK for treated hyperlipidemia patients with CV event history. Corresponding data for CV RE patients and patients at low/ unknown risk were 168,517 SEK and 108,429 SEK, respectively. The higher CV risk levels were associated with greater productivity losses. Indirect costs varied within cohorts depending on past diagnoses of the patients. In patients with CV event history, a previous diagnosis of myocardial infarction was associated with the lowest annual indirect costs (189,114 SEK) while a past diagnosis of ischemic stroke was associated with the highest indirect costs (281,985 SEK). Within the CV RE cohort, a previous diagnosis of abdominal aortic aneurysm and transient ischemic attack was associated with the highest (264,441 SEK) and lowest annual indirect costs (156,254 SEK), respectively. Conclusions: The high level of productivity losses illustrates the high indirect cost burden in patients treated for hyperlipidemia. The type of past CV event affected the level of indirect costs. PCV111 Health Care Costs Associated With Cardiovascular Events In Patients With Hyperlipidemia - Estimates From Population-Based Register Data In Sweden Hallberg S 1, Banefelt J 1, Mesterton J 1, Gandra S R 2, Fox K M 3, Johansson G 4, Levin L Å 5, Sobocki P 6 1Quantify Research, Stockholm, Sweden, 2Amgen, Inc., Thousand Oaks, CA, USA, 3Strategic Healthcare Solutions, LLC, Monkton, MD, USA, 4Uppsala University, Uppsala, Sweden, 5Linköping University, Linköping, Sweden, 6IMS Health, Stockholm, Sweden . . . . . . . . . . . Objectives: To estimate annual incremental health care costs of new cardiovascular (CV) events (myocardial infarction, unstable angina, revascularization, ischemic stroke, transient ischemic attack or heart failure) in patients with hyperlipidemia or prior CV events. Methods: A retrospective population-based cohort study was conducted using Swedish electronic medical records and national registers. Patients were included in the study based on a prescription of lipid-lowering treatment between January 1, 2006 and December 31, 2006 or history of CV events (prior to 2006) and followed until December 31, 2012 for identification of new CV events and estimation of cost. Patients were stratified into three cohorts based on CV risk level. Propensity score matching was applied to compare patients with new events to patients without new events and to estimate incremental costs. Results: A new CV event resulted in increased costs during all follow-up years. Inpatient hospital stays were the main driver of the increase. The majority of the costs occurred in the first year following event when patients with CV event history (n= 6,881) had an incremental cost of 74,758 SEK. This was similar to that of CV risk-equivalent patients (n= 3,226; 75,415 SEK) and patients at low/unknown CV risk (n= 2,497; 72,635 SEK). Ischemic stroke resulted in the highest first year cost in all cohorts (88,739; 85,516; and 87,668 SEK) and transient ischemic attack the lowest (34,098; 36,042; and 29,052 SEK). Incremental costs during subsequent years remained elevated for all cohorts, with the CV event history cohort having the highest costs. Second year incremental costs varied between 10,689 and 15,082 SEK and third year costs varied between 8,828 and 11,558 SEK across cohorts. Conclusions: The direct costs of new CV events were substantial and varied considerably by type of event. Costs were increased during all follow-up years, regardless of CV risk level. PCV112 Inpatient Case-Related Treatment Costs For Different Cardiovascular Diseases In Germany Schädlich P K 1, Rosenfeld S 2, Reindl S 3, Kotowa W 3 Institut GmbH, Berlin, Germany,, 2Sanofi-Aventis Deutschland GmbH, Berlin, Germany,, 3IGES Institut GmbH, Nuremberg, Germany . . . . . 1IGES Objectives: As part of a non-interventional study, hospitalizations due to the following diagnoses of cardiovascular disease (CVD) were documented: atrial fibrillation (AF), stroke or TIA (SoT), acute coronary syndrome (ACS), arterial embolism (AE), decompensated heart failure (DHF), syncope (S), ventricular arrhythmia (VA), and cardiac arrest (CA). The objective of this cost analysis was to quantify inpatient treatment costs for each diagnosis in Germany from the perspective of the statutory health insurance. Methods: The analysis was conducted for 2012 using the latest available “G-DRG V2013 Browser 2012 § 21 KHEntgG”. Invoiced diagnosis-related groups (DRGs) including the ICD10-codes corresponding to the considered diagnoses were collected. For available DRGs, the number of cases with normal length of stay and the corresponding cost weights (CWs) were documented. Then, the proportion of an individual DRG within a diagnosis (≥ 1 ICD10-code) was calculated (number of cases for individual DRG / sum of all cases for every DRG within a diagnosis = “DRG weight factor”). Next, the CWs of an individual DRG were multiplied by the “DRG weight factor” to reflect the prevalence of a DRG within an ICD10-code (“weighted CWs”). Finally, the “weighted CWs” were added up to obtain an average CW for a diagnosis, which is based on all DRGs and CWs coded for this single diagnosis. This sum of CWs was multiplied by the state-wide base rate of 2,991.53€ . Results: The following inpatient treatment costs for the respective diagnoses were derived: 2,800€ (AF); 5,000€ (SoT); 5,200€ (ACS); 6,200€ (AE); 3,800€ (DHF); 1,900€ (S); 8,200€ (VA); 15,900€ (CA) (rounded). Conclusions: Inpatient treatment costs for the considered CVD diagnoses vary from 1,900€ to 15,900€ . Generally, hospitalizations due to CVD have a remarkable impact on the budget of German sick funds. The results of this cost study can be used for further health economic analyses in CVD. PCV113 Clinical Pathway And Health Care Resources Utilization Of A Patients Cohort At High Risk Of Cardiovascular Disease Of Local Health Care Unit (Asln°1) Of Milan: A Results Of Intervention On Secondary Prevention Di Matteo S 1, Colombo G L 2, Malnis D 3, Bruno G M 1 e Valutazioni economiche, Milan, Italy, 2University of Pavia, Milan, Italy, 3Dipartimento Governance asl Milano1, Rho, Italy . 1S. A. V. E. Studi Analisi . . . . . Objectives: The objective of the project is to evaluate the impact of health care resources and the related health care costs in relation to outcomes occurred in 10 years on the cohort of 2002 and intervention on secondary prevention cohort 2010 with a follow-up to 3 years. Methods: In 2002, Local HealthCare Unit (ASLn°1) of Milan identified a cohort of 33,977 patients with RCV (Cardiovascular risk). Data analysis was done trough a retrospective claims data study from ASLn°1 of Milan; Physicians who participated were 494. Information has been collected for inclusion in the study regarding some risk factors and treatment. At the end of 2010, all recruited patients were started on a program of secondary prevention, by reducing the effects of smoking and BMI after ad hoc training events for physicians. Results: 12,000 subjects were recruited and followed for 3 years with reduction of spending on the NHS system and reduced incidence of events. A resources utilization analysis has been developed using profiles of treatments and dividing the population into groups with homogeneous treatment (smoking and BMI control vs. standard care), verifying the occurrence of greater outcomes and related health care costs. Data are analyzed with the definition of major cardiovascular events, all-cause mortality and cause-specific, occurrence of diabetes and other chronic greater diseases, verifying the occurrence of greater outcomes and related health care costs. Conclusions: Administrative databases offer low-cost information (since they are already available) regarding more or less all services provided in a health care environment. These sources and their integration are a powerful tool supporting conventional methods used in epidemiological studies and as tools for plan Health care policy. Cardiovascular Disorders – Patient-Reported Outcomes & Patient Preference Studies PCV114 Cardiovascular Risk, Gender And Medication Adherence In Rural Area Of Vietnam Nguyen T P L , CCM S V , Postma M J University of Groningen, Groningen, The Netherlands . . . . . . . Objectives: To examine the relationship between medication compliance, cardiovascular risk and gender in hypertensive patients visiting primary health care centers. Methods: A prospective 1-year study was conducted in rural communes in Vietnam on hypertension management in a population from 35 to 64 years. Data on age, gender, blood pressure and blood test were collected at baseline. Cardiovascular risk was based on the Cardiovascular Risk Prediction Model for populations in Asia. Medication compliance was calculated as the number of days taking the drug divided by the number of days since the first day of the prescription. A threshold of 80% was applied to differentiate between compliance or non-compliance. Taking medication was based on patients’ self-report during each monthly visit. Results: Of total 338 patients met the selection criteria for medication compliance study, 46% was female. Mean age was 53.5 (+/- 6.9) and 77.2 % of patients was < 10% of CVD risk in 8 years. In primary health care settings, medication compliance was 49%. No significant difference in medication compliance in both CVD risk groups (< 10% vs. > 10% risk) was found, also not after controlling for age and gender (adjusted OR was 1.27; 95 % CI: 0.7 – 2.2; p value 0.39). The odds of medication compliance in females was however 0.6 times higher than in males (95% CI: 0.38 to 0.95, p value 0.028). Each 1 year increase, results in patients being 1.04 times more likely to be compliant (95% CI: 1.009 to 1.076, p value 0.01). Conclusions: Medication compliance rate was low among hypertensive patients in Vietnam. CVD risk at the baseline did not significantly differentiate complaint from non-compliant patients. Yet, a major difference in compliance was found for gender. Rather than risk profile, gender should be considered for guiding the choice on who to target for improving medication compliance for hypertensive patients. PCV115 Health Behavior And Medication Adherence Han E 1, Sohn H S 2, Jang S 3 University, Incheon, South Korea, 2Sookmyung Women’s University, Seoul, South Korea, 3Inje University, Gimhae, South Korea . . . . 1Yonsei Objectives: To explore the associations of selected health behaviors with medication adherence in elderly patients aged 65+ years with hypertension, diabetes, or hyperlipidemia. Methods: The Korean National Health Insurance data between January 2010 and June 2011 were used. The study included 662,170 hypertensive, 179,285 diabetic, and 244,702 hyperlipidemic patients. Poor medication adherence was defined as < 80% medication possession ratio from January to June 2011. Health behavior data were from year 2010. Multivariate logistic regression was used. Results: Patients with a waist circumference < 85 (for women) or 90 (for men) centimeters were more likely to adhere to their medications. Current smokers and moderate or heavy drinkers showed poor medication adherence than their counterparts. Mild physical activity was associated with better medication adherence. Conclusions: Public efforts need to focus on improving comprehensive control of both health behaviors and medication adherence. PCV116 Understanding Medication Adherence Using Stated-Preference Data González J M 1, Poulos C 1, Mollon P 2 1RTI Health Solutions, Research Triangle Park, NC, USA, 2Novaris Pharma AG, Basel, Switzerland . . . . Objectives: More than half of people who have experienced a myocardial infarction (MI) are not adherent to their medication regimen, which leads to poorer health outcomes. We used a stated-preference (SP) study to examine factors that could influence patient compliance to prophylactic cardiovascular treatments, and discuss practical issues in using SP methods to explain medication adherence. Methods: Preference data for treatments that lower the risk of cardiovascular events were collected from 464 respondents in the United States with self-reported history of MI using a discrete-choice experiment (DCE). All respondents answered 11 judgment A493 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 questions that presented a pair of virtual patients who were prescribed different treatments defined by: reduction in the risks of nonfatal MI and fatal MI, treatment-related risk of serious infection, mode and frequency of administration, and monthly medication cost. Half of the choice questions asked respondents to select the treatment to which they would most likely be nonadherent. The other half asked respondents to state which of two virtual patients was better off after learning how adherent each was to each medication. Limited dependent-variable models were used to estimate weights indicating the impact of treatment and respondent characteristics on stated-adherence and quantifying the stated impact of nonadherence on respondents’ well-being. Results: Results indicated that reductions in the risk of a nonfatal MI had the largest effect on stated adherence, followed by medication cost, the risk of serious infection, and lastly mode and frequency of administration. Results also show that reductions in compliance had a significant impact on the perceived overall benefits of prophylactic treatments. Conclusions: We find that both clinical and nonclinical factors can impact treatment adherence, suggesting that the flexibility to include a variety of factors with SP models can be useful in understanding patient compliance. PCV117 Patient Adherence Among Adolescents With Arterial Hypertension Nowakowska E , Paczkowska A , Bryl W , Hoffmann K Poznan University of Medical Sciences, Poznan, Poland . . . PCV118 Health State Utilities In Chronic Heart Failure In The Uk Nafees B 1, Cowie M R 2, Patel C 1, Deschaseaux C 3, Brazier J 4, Lloyd A J 5 1ICON plc, Oxford, UK, 2Imperial College London, London, UK, 3Novartis Pharma AG, Basel, Switzerland, 4University of Sheffield, Sheffield, UK, 5Oxford Outcomes, An ICON plc Company, Oxford, UK . . . . . . . Objectives: Previous research has shown the impact of chronic heart failure (CHF) on health-related quality of life (HRQL). Less is known regarding the impact of reduced ejection fraction (HFrEF) on HRQL. The aim of this study was to elicit utility values for CHF with HFrEF or preserved ejection fraction (HFpEF) by New York Heart Association (NYHA) classification system in the UK. In addition, utility values for events such as stroke, myocardial infarction (MI) and chronic kidney disease (CKD) were estimated. Methods: Health states were developed from concept elicitation interviews with CHF patients (N= 10) and cardiologists (N= 5). Draft health states were validated in cognitive debrief interviews with different patients (N= 5) and cardiologists (N= 4) and finalised with scientific input from experts. The resulting health states (n = 10) were piloted with general public to check understanding. General public participants (n=100) completed background questions and rated each state using visual analogue scale and time trade-off (TTO) assessments (with lead time method for states worse than dead). Results: The mean TTO utility for HF-rEF ranged between 0.86 (SD= 0.19) (NYHA class II); 0.60 (SD= 0.23) (NYHA III) to 0.28 (SD= 0.41) (NYHA IV). Equivalent values for HF-pEF were 0.83 (SD=0.24) (NYHA II); 0.55 (SD= 0.28) (NYHA III) to 0.27 (SD= 0.35) (class IV) respectively. Other values were post hospitalisation after stroke (mean= 0.30, SD= 0.43); post MI (mean= 0.45, SD= 0.37) and CKD (mean= 0.78, SD= 0.21). Post-hospitalisation states captured a period of upto three months after the event. Conclusions: This study shows the effect that NYHA class has on HRQL for people with CHF in the UK. The findings showed that HFrEF and HFpEF were very similar. Participants considered events such as recovery from stroke to have significant impact on HRQL. These are important data to consider in evaluating outcomes of treatments and should be reflected in cost effectiveness models in CHF where relevant. PCV119 Acute And Chronic Impact Of Cardiovascular Events On Health State Utilities Matza L S 1, Devine M K 1, Gandra S R 2, Delio P R 3, Fenster B E 4, Davies E 5, Jordan J1, Lothgren M6, Feeny D H 7 1Evidera, Bethesda, MD, USA, 2Amgen, Inc., Thousand Oaks, CA, USA, 3Neurology Associates of Santa Barbara, Santa Barbara, CA, USA, 4National Jewish Health, Denver, CO, USA, 5Evidera, London, UK, 6Amgen (Europe) GmbH, Zug, Switzerland, 7University of Alberta, Portland, OR, USA . . . . . . . . . PCV120 Health State In Patients With Atrial Fibrillation On New Oral Anticoagulants As Assessed With The New Eq-5d-5l Questionnaire At Baseline And 12-Month Follow-Up: Prefer In Af Registry Brüggenjürgen B 1, Schliephacke T 2, Darius H 3, De Caterina R 4, Le Heuzey J Y 5, Reimitz P E 2, Schilling R J 6, Schwertfeger M 2, Zamorano J L 7, Kirchhof P 8 1Steinbeis University Berlin (SHB), Berlin, Germany, 2Daiichi Sankyo Europe GmbH, Munich, Germany, 3Vivantes Hospital Neukölln, Berlin, Germany, 4G. d’Annunzio University, Chieti, Italy, 5Hôpital Européen Georges Pompidou, Université René Descartes, Paris, France, 6Barts and St Thomas Hospital, London, UK, 7University Hospital Ramón y Cajal, Madrid, Spain, 8University of Birmingham Centre for Cardiovascular Sciences and SWBH NHS Trust, Birmingham, UK . Objectives: The aim of the study was to assessment of compliance by adolescents in the field of pharmacological and non-pharmacological methods of hypertension treatment. Methods: The study included 62 patients (20 women, 42 men) diagnosed with hypertension and treated in specialist health care facilities. As a research tool was used questionnaire prepared on the basis of recent literature. Results: The vast majority of respondents (72,7%) declared that regularly taking antihypertensive drugs. The proportion of patients regularly taking antihypertensive drugs was higher in patients treated with monotherapy than polytherapy (48,5% vs 24,2%). Among the methods of non-pharmacological treatment of hypertension the most accepted lifestyle change in the study population was smoking cessation (83,8% of respondents) and reduction of salt consumption (64,5% of respondents), and the least acceptable lifestyle change was to maintain proper body weight by eating a low calorie diet (30,6% of respondents). Conclusions: Adolescents with hypertension in varying degrees adhere to a medical recommendations related to the hypertension treatment. From the available literature data indicate that the current effective way to improve cooperation with the patient’s is education. . recovery; post-event health states represented chronic impact. UK general population respondents valued the health states in time trade-off tasks with time horizons of one year for acute states and ten years for chronic states. Results: A total of 200 participants completed interviews (55% female; mean age = 46.6y). Among acute health states, stroke had the lowest utility (0.33), followed by heart failure (0.60) and ACS (0.67). Utility scores for chronic health states followed the same pattern: stroke (0.52), heart failure (0.57), and ACS (0.82). For stroke and ACS, acute utilities were significantly lower than utilities for chronic post-event (difference = 0.20 and 0.15, respectively; both p < 0.0001). Conclusions: Results add to previously published utilities for cardiovascular events by distinguishing between chronic post-event health states and acute health states that include the event and its immediate impact. Findings suggest that acute and chronic impact should be considered when selecting scores for use in cost-utility models. Thus, the current utilities provide a unique option that may be used to represent the acute and chronic impact of cardiovascular conditions in economic models comparing treatments that may delay or prevent the onset of cardiovascular events. . . . . Objectives: Cost-utility models are frequently conducted to compare treatments intended to prevent or delay cardiovascular events. Most published utilities represent post-event health states without incorporating the disutility of the event or reporting the time between the event and utility assessment. Therefore, the objective of this study was to estimate health state utilities representing cardiovascular conditions while distinguishing between acute impact including the cardiovascular event and the chronic post-event impact. Methods: Health states were drafted and refined based on literature review, clinician interviews, and a pilot study. Three cardiovascular conditions were described: stroke, acute coronary syndrome (ACS), and heart failure. One-year acute health states represented the event and gradual . . . . . . . . . . . . . Objectives: We aimed to understand the short-term impact on quality of life associated with Non VKA Oral AntiCoagulants (NOACs) use in patients with AF. We obtained baseline (BL) and follow-up (FU) data on the health state of AF patients under everyday practice conditions in the PREvention oF thromboembolic events – European Registry in Atrial Fibrillation (PREFER in AF). Methods: PREFER in AF documents AF patients in terms of clinical characteristics, management, quality of life and other outcome parameters. The EuroQol EQ-5D-5L descriptive system and visual analogue scale (VAS) were applied in PREFER in AF at BL and FU to obtain patient-reported generic health-related quality of life information and utility weights. Results: Of the 6390 AF patients at follow-up 1895 (29.7%) had paroxysmal, 1533 (24.0%) persistent, 474 (7.4%) long-standing persistent, and 2488 (38.9%) permanent AF. Comorbidities were highly prevalent. 3344 AF patients (61.1% males, mean age 71.7 ± 9.85 years) provided EQ-5D-5L data both at BL and 12-month FU. On the VAS (range 0-100), the mean score at FU was 68.8 ±18.1 points, with no major differences between patients on NOACs (68.3), VKAs (68.9), AP (70.1), or VKA+AP (71.7), respectively. All scores improved from baseline from 0.36 in patients on VKA to 1.77 for those on AP. The overall utility index at FU was 0.80 ±0.21. At FU the 409 patients on NOACs had a utility score of 0.79 (change from BL, -0.01), the 1789 patients VKAs 0.80 (-0.01), the 237 patients on AP 0.81 (+0.01), the 151 patients on VKA and AP 0.80 (-0.01) and the 749 patients receiving neither VKAs nor AP 0.80 (0.00). Conclusions: Patients with AF present with reduced self-reported quality of life compared to the general population. Patients receiving NOACs had similar quality of life both at baseline and after 12-month FU, when compared to alternative medications. PCV121 Health State In Patients With Venous Thromboembolism On Conventional And Non-Vka Oral Anticoagulants As Assessed With The Eq-5d-5l Questionnaire: Prefer In Vte Registry Cohen A T 1, Bauersachs R 2, Gitt A K 3, Mismetti P 4, Monreal M 5, Willich S N 6, Wolf W P 7, Agnelli G 8 1King’s College, London, UK, 2Max-Ratschow-Klinik für Angiologie, Gefäßzentrum Klinikum Darmstadt GmbH, Darmstadt, Germany, 3Herzzentrum Ludwigshafen, Ludwigshafen, Germany, 4Centre Hospitalier Universitaire Saint-Etienne, Hopital Nord, Saint Etienne, France, 5Hospital Universitari Germans Trias I Pujol, Barcelona, Spain, 6Charité - Universitätsmedizin Berlin, Berlin, Germany, 7Daiichi Sankyo Europe GmbH, Munich, Germany, 8University of Perugia, Santa Maria della Misericordia Hospital, Perugia, Italy . . . . . . . . . . . . Objectives: Non-VKA oral anticoagulants (NOAC), which do not need routine monitoring, have the potential to improve the quality of life (QoL) in patients on long-term treatment for venous thromboembolism (VTE). We aimed to obtain contemporary data on the health state of patients with VTE under daily practice conditions. Methods: PREFER in VTE is a non-interventional study in 7 countries (France, Germany, Austria, Switzerland, Italy, Spain, UK) that prospectively documents patients after an event of acute deep venous thrombosis (DVT) or pulmonary embolism (PE) in terms of clinical characteristics, management, quality of life and other outcome parameters. The EuroQol EQ-5D-5L consists of the 5-dimension descriptive system used to derive utility scores and the visual analogue scale (VAS), measuring self-rated health (scale 0-100). Results: A total of 2790 patients with acute VTE at baseline (BL: 1640 DVT, 1150 PE ±DVT) and 723 patients at an interim analysis at 6 months (443 DVT and 280 PE ±DVT) completed the EQ-5D-5L. On the EQ VAS for current health state, the mean score at 6 months was 73.8 points (change from BL +10.9), with similar values in DVT patients (74.6, change from BL +9.8) compared to PE patients (72.4, change from BL +12.4). Between BL and 6-month follow-up, index values increased in all medication classes (heparin only: 0.66 to 0.75; heparin/VKA: 0.70 to 0.84; NOAC: 0.73 to 0.87). Overall the index value increased from 0.69 to 0.83 (DVT: 0.71 to 0.85, PE: 0.67 to 0.81). Conclusions: Under clinical practice conditions, patients on NOAC and heparin/VKA had larger increases in their health state scores than those on heparin only. This generic QoL tool detected only small differences between treatment options. Six months after the event, patients with DVT had similar self-reported QoL on the VAS compared to patients with PE, and patients with DVT had somewhat higher utility values. A494 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PCV122 The Intermediate Burden Of Diabetes Mellitus In Patients With Cardiovascular Disease (Cvd): A Quality Adjusted Life Year (Qaly) -Analysis Based On Primary Longitudinal Data Laxy M 1, Hunger M 2, Thorand B 1, Meisinger C 1, Kirchberger I 3, Holle R 1 Zentrum München - German Research Center for Environmental Health (GmbH), Neuherberg, Germany, 2Helmholtz Zentrum München, Neuherberg, Germany, 3KORA Myocardial Infarction Registry, Augsburg Hospital, Augsburg, Germany . . . . . . 1Helmholtz Objectives: While the independent influence of metabolic and cardiovascular diseases on either quality of life (QoL) or survival is well studied, the evidence on the combined burden in terms of quality adjusted life years (QALYs) is rather weak. Previous burden of disease studies mostly combined cross-sectional QoL data with mortality statistics from other data sources. However, due to strong model assumptions these studies might be limited in validity and accuracy. The goal of this study was to analyze the intermediate burden of diabetes in patients with previous myocardial infarction in terms of QALYs lost, by introducing a method capable to exploit primary longitudinal data sources of population-based studies. Methods: Data for this analysis were taken from the KORA-Acute Myocardial Infarction (AMI) Registry. QoL (Euro-Qol-5D-index, German tariff) of a subset of patients known to be alive was assessed through postal surveys in 2006 (n= 1022) and subsequently in 2011 (n= 716). Mortality was tracked by the routine surveillance system of the AMI Registry. QALYs gained were calculated assuming a linear QoL change from baseline to follow-up. QoL trajectories of non-responders at follow-up and patients who died before follow-up were approximated by multiple imputation methods using the baseline QoL and covariate structure. Ordinary least square regression models were performed to quantify the QALYs and life years (LYs) lost over the mean observation time. Results: Over a mean observation time of 4 years, patients with diabetes lost 0.14 LYs (3.75 vs. 3.89, p< 0.01) and 0.36 QALYs (2.96 vs. 3.33, p< 0.01) more than patients without diabetes. The QALY-gap was greatest for diabetic patients taking insulin (-0.66, p< 0.01). Sensitivity analyses showed that models were robust concerning model assumptions. Conclusions: The intermediate burden of diabetes for patients with CVD is substantial. Using individual-level data from population-based follow-up of studies is a valuable methodological extension to accurately quantify the burden of chronic conditions. PCV123 Sensitivity Of The Safuca Questionnaire To Detect Differences Between Atrial Fibrillation Patients Treated With Vitamin-K Antagonist Against Those Treated With New Oral Anticoagulants Ruiz M A 1, Anguita M 2, Bertomeu V 2, Cequier Á 2, Muñiz J 2 1Universidad Autónoma de Madrid, Madrid, Spain, 2Sociedad Española de Cardiología, Madrid, Spain . . . . . . Objectives: A secondary analysis was carried out to test if differences existed in reported treatment satisfaction between non-valvular atrial fibrillation patients (NVAF) treated with vitamin K antagonist (VKA) anticoagulants and new oral anticoagulants (NOAC). Methods: A sample of 1318 patients was recruited at random in the FANTASIIA study, between June 2013 and March 2014, from which 77% were using VKA and 23% NOAC at least 6 months before inclusion. The specific treatment satisfaction instrument SAFUCA was used to test differences between groups. The SAFUCA questionnaire is composed by 7 dimensions and overall score, measured in a 0 (least satisfied), 100 (most satisfied) scale. Guyatt’s d was used to estimate effect size. Results: Mean age was 73.8 (SD= 9.4) years, and 42.5% were women. Patients using NOACs attained statistically significant higher values in the overall score and in all SAFUCA dimensions. Overall score effect size (NOAC vs VKA) was medium-small (79.91 vs 73.22, d= 0.43, p< 0.001). Small to large effect sizes were also found by dimension: Effectiveness (77.63 vs 72.70, d= 0.33, p< 0.01), Ease and Convenience (80.99 vs 73.72, d= 0.49, p< 0.001), INR Controls (62.35 vs 57.66, d= 0.32, p< 0.01), Impact on Daily Activities (90.08 vs 81.97, d= 0.55, p< 0.001), Medication Undesired Effects (86.20 vs 79.21, d= 0.47, p< 0.001), Satisfaction with Medical Care (80.30 vs 69.38, d= 0.74, p< 0.001), and Overall Satisfaction (80.30 vs 69.38, d= 0.74, p< 0.001). Conclusions: SAFUCA questionnaire was able to detect satisfaction differences between NVAF patients treated with NOAC and those treated with VKA, presenting medium effect size in most dimensions. This new evidences offer additional support to the questionnaire validity. PCV124 Unmet Needs And Solutions For Heart Failure Admission Steel A , Ansaripour A , Collett S , Wick C, Wagih S , Bakhai A Barnet & Chase Farm Hospitals NHS Trust, London, UK . . . . aggressive therapies such as complex devices or ultrafiltration, as well as enabling an enriched end of life experience for those beyond such therapies. Enabling end stage patients to die in their location of choice would also release considerable resources at the same time. . Objectives: Unplanned patient admissions to hospital with heart failure (HF) are on the increase due to an ageing population and increasing survival post coronary disease. Pre-discharge mortality is high, predictable and often occurs after many days stay for many patients and often without palliative care involvement, not aligned with guidance. Methods: We audited palliative care input to HF patients admitted to our hospital as the predominant condition and subsequently passing away, both before and after an awareness campaign, publishing barriers to end of life dialogues between patient and clinicians, demonstrating the early use of a Get With The Guidelines (GWTG) risk assessment tool to predict mortality and introduction of a daily HF nursing team service. Results: In 2009, amongst 57 HF patients the average time to death was 17.8 days, only 7% received any palliative input before death. The time to death in these patients correlated positively with the (GWTG) predictive score indicating that these deaths could have been anticipated and appropriate palliative involvement triggered. Data from 2010 to 2013 following these interventions (n= 99) showed a marked improvement with 44% of patients having palliative input before death. The average time to death in this group was comparable at 17.22 days. Conclusions: The care of HF patients and their families can be greatly improved with early mortality prediction, sensitive dialogues, routine involvement of HF teams both to enhance survival for patients who will benefit from PCV125 Clinical Psychologists: Closing The Communication Gap Between Physicians And Patients, Leading To Higher Patient Satisfaction And Compliance Djambazov S N 1, Vekov T Y 2, Petrov D 3, Lambeva V 4 clinics Doc Dr Valentina Tsekova, Sofia, Bulgaria, 2Medical University Pleven, Pleven, Bulgaria, 3Bulgarian Medical Union, Sofia, Bulgaria, 4Telelink, Sofia, Bulgaria . . . . . . 1Cancer Objectives: We wanted to compare the patient satisfaction and compliance between a hospital with a trained clinical psychologist, acting as a mediator between physicians and patients and a hospital without one. Methods: The comparison was done between two identical cardiac hospitals, which belong to the same network of cardiac facilities, for a period of six months. At each hospital, 200 patients were included. The patient characteristics, numbers and patient flow were comparable; the facilities were identical as level of comfort and staff training. To measure the patient satisfaction we used a questionnaire. The level of attendance of control visits after discharge we measured with the hospital registry. How many of the 200 patients stick to the discharge therapy after 3 and after 6 months after discharge, we measured with the out-patient centre registry and by telephone interviews. Results: In each hospital 200 patients were included and followed-up. For the hospital without clinical psychologist, patient satisfaction was 79% excellent marks (n= 158), control visits attendance was 42% (n= 84), patient compliance was 72% (n= 144) on the 3rd and 65% (n= 130) on the 6th month after discharge. For the hospital with a clinical psychologist, patient satisfaction was 97% excellent marks (n= 194), control visits attendance was 78% (n= 156), patient compliance was 96% (n= 192) on the 3rd and 89% (n= 178) on the 6th month after discharge. With the help of a trained clinical psychologist, we witnessed the following differences: 18% improvement in patient satisfaction, 36% better attendance to control visits, 24% more compliant patients for both the 3rd and 6th month after discharge. Conclusions: Trained clinical psychologists may play the role of a mediator and close the communication gap between physicians and patients and lead to improved patient satisfaction and compliance. PCV126 Evaluating The Gap Between Physicians’ And Patients’ Understanding Of Patient Needs Djambazov S N 1, Vekov T Y 2, Petrov D 3 clinics Doc Dr Valentina Tsekova, Sofia, Bulgaria, 2Medical University Pleven, Pleven, Bulgaria, 3Bulgarian Medical Union, Sofia, Bulgaria . . . . . 1Cancer Objectives: We wanted to establish the gap between physicians’ and patients’ understanding of patient needs in a hospital setting. Methods: The study was run in the four of the Bulgarian Cardiac Institute clinics. We used questionnaires and within a period of two months, first we asked 30 physicians and 50 patients about what patient needs are according to their understanding. Based on the answers we defined 10 categories for each of the two groups. Then we asked 143 physicians and 500 patients to define which category is most and least important for them by using the Maximum Difference Scaling technique. Results: Courtesy, after discharge recommendations and information about the discharge drugs were the top three most important patient needs according to the patients. Life-saving activities, improvement of quality of life and improve the longevity were the top three most important patient needs according to the physicians. Conclusions: There is a tremendous gap between patients’ and physicians’ perceptions of patient needs. It is very hard for the hospitals to increase patient satisfaction relying on medical services only. Further research is needed to find ways to close that gap. PCV127 Beliefs About Medicines In An Urban Black Hypertension Population Barr J 1, Callender S 2, Tung D 1, Fashote B 1, Lacroix L 3, Young A 2 1Northeastern University, Boston, MA, MA, USA, 2Mattapan Community Health Center, Boston, MA, USA, 3Northeastern University, Boston, MA, USA . . . . . . Objectives: Given ethnic variation in attitudes toward hypertension (Ford 2010; Lewis 2010), study objectives are to determine 1) patient preferences concerning medication use as revealed by responses to Beliefs about Medicines Questionnaire (BMQ, Horne 1999) in an urban black population with hypertension (HTN) and 2) if these preferences were influenced by experiences with parents/friends HTN diagnoses. Methods: After approval from Northeastern University’s Institutional Review Board, coded clinic appointment schedules were used to identify patients scheduled for routine HTN follow-up appointment at a Boston inner-city community health center. Researchers screened for eligibility (English speaking, black ethnicity, taking anti-hypertensive medication), described the study, and requested informed consent. Patients agreeing to participate and completing background questions and the hypertension-specific Beliefs about Medicines questionnaire (BMQ: 18 items, 5 levels, strongly agree-strongly disagree) received a $10 CVSprovided gift card. Results: 189 patients were approached. 94 (49.7%) completed the questionnaire, 11.6% declined, and 38.6% didn’t meet criteria, e.g., didn’t speak English or had previously completed the BMQ. Patients averaged 55.6yo, 67% female. Overall BMQ factor scores were: Specific Necessity (SN), 3.41 ±0.83; Specific Concerns (SC), 2.82 ±0.8; General Overuse (GO), 2.98 ±0.74 and General Harm (GH), 2.54 ±0.72. Patients whose parent had a stroke had higher SN (p= 0.038) and lower GH scores (p= 0.042) than patients without that prior experience. There was no difference in any factors as to whether parents or a friend did/didn’t have hypertension and whether a friend did/didn’t have a stroke. SC were lower in patients with higher educational level (p= 0.002). No statistical associations occurred between A495 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 self-reported adherence results and any of the 4 factor scores. Conclusions: This study provides BMQ factor scores for urban black HTN patients and insight into how experiences of patient’s parents can be used to target increased perception of anti-hypertensive medication necessity and reduce patient-specific medication harm concerns. PCV128 Place Of Residence And Employment Status After Stroke Dewilde S 1, Peeters A 2, Thijs V 3, Annemans L 4, Belgian Stroke Council N P 5 1SHE, Brussels, Belgium, 2Cliniques universitaires Saint-Luc, Bruxelles, Belgium, 3KULeuven, Leuven, Belgium, 4Ghent University & Brussels University, Ghent, Belgium, 5Virga Jesseziekenhuis, Hasselt, Belgium . . . . . . Objectives: To investigate the living and employment situation of patients after stroke, and examine whether this differs by degree of disability. Methods: We conducted an observational retrospective study among 569 post-stroke patients and examined changes in employment and living situation. The investigational period was divided into 3 intervals: < 3,3-6 and > 6 months after stroke. All patients had their mRS (modified Rankin Score, ranging from 0 full health to 5 severely dependent) measured at 3 months after stroke, plus an additional mRS assessment if their stroke was more than 6 months ago. Patients were recruited in 10 regional and university hospitals across Belgium using a convenience sample stratified by mRS. Results: Before their stroke the majority of patients lived at home (99%) despite the fact that 13% had a previous stroke. At 3 months after stroke an association was found between the time spent in an inpatient care facility (hospital, rehabilitation facility, nursing home) and the mRS (p< 0.0001 Weibull survival analysis, average inpatients days were 9.2, 14.1, 27.7, 53.4, 67.0 and 73.4 for mRS 0-5 respectively). Between 3 and 6 months after stroke on average 7% of patients were staying in a rehabilitation facility and 5% in a nursing home; these were mostly patients with mRS > = 3. After six months all patients returned home, except patients with mRS4 (36% home) and mRS5 (38% home). Before stroke 70% of patients were retired; in those working before their stroke, 16-20% patients in mRS categories 0-2 were working again compared to none in mRS category 3-5, of which 14% took early retirement or leave of absence. Conclusions: Experiencing a mild stroke will not affect the patient’s employment and living situation beyond the short term; however suffering a severe stroke is likely to lead to significant changes in place of residence and occupation. PCV129 Uisess-B Oral Health Questionnaire Validity And Reliability In A Mexican Diabetic, Systemic Hypertension And Obese Patients Salcedo-Rocha A L 1, Flores-Larios E A 2, Balderas-Peña L M A 3, Preciado-Serrano M D L 2, García de Alba-García J E 1 1Delegación Jalisco IMSS, Guadalajara, Jalisco, Mexico, 2Universidad de Guadalajara, Guadalajara, Jalisco, Mexico, 3UMAE Hospital de Especialidades Centro Médico Nacional de Occidente IMSS, Guadalajara, Jalisco, Mexico . . . . . . . . . . . . Objectives: Evaluate validity and reliability of the UISESS-B oral health questionnaire in a Mexican diabetic, systemic hypertension and obese patients in the primary health care level. Methods: Were studied 105 subjects with diabetes, systemic hypertension and/or over-weigh-obesity in the context of a diabetic’s control group in the Instituto Mexicano del Seguro Social. Was applied an oral health questionnaire named UISESS-B previously validated. The instrument is divided in three main areas: oral health care habits, oral health and nutrition related habits, and oral health care related symptoms and signs. ResultsConstruct validity was confirmed by the correct compilation of the items included in the two main constructs: habits and perceptions each one including specific multiitem questionnaires (correlation coefficients ranged from 0.30 to 0.80, P>0.001); the reliability was confirmed by the internal consistency (Cronbach alpha: 0.55 for the construct habits and 0.71 for the construct perceptions). Conclusions: The UISESS-B questionnaire shows significant validity and reliability, suggesting its use as an instrument useful to be used as a patient reported outcome in oral health in patients with chronic diseases. PCV130 The “Venous Age”: A New Tool To Sensitize Patients To Their Venous Disease Allaert F A 1, Crebassa V 2 Biotech/CEN Nutriment, Dijon, France, 2clinique du millénaire, Montpellier, France . . . 1CEN Objectives: Many years ago, cardiologists developed on the basis of the Framingham study an “arterial age” which is very useful to sensitize patients to their cardiovascular risk. The purpose of the study was to develop a “venous age” to make people more aware of their venous disease and to better adhere to lifestyle improvement and venous disease treatments. Methods: The score calculation was based on an international epidemiological study conducted in 24 countries in the daily practice or general practitioners. The data base included patients with or without venous disease, whatever the reason for which they were consulting and whatever the level of their venous disease which was systematically described according the elements of the international CEAP classification. Results: The study covers 124 235 patients aged of 52 among whom 69.4% were female. Among them 18.8% had no sign of venous disease (C0), 22.9% had only functional symptoms (C0s), 40.6% had Telangiectasies or reticular veins, 34.8% varicose, 24.9% edema, 14.0% skin changes, 7.3 Healed ulcers and 4.3% active ulcers. The statistical analysis has determined the number of years which must be added to the real age to get the “venous age” by comparison of the age of somebody who has no venous functional symptoms or physical signs. The results provide that number for women and men according the different venous symptoms and sign they present. Conclusions: This first attempt of creating a “venous age” will be certainly improved in the next future, using more complex analysis based on risk factors or other criteria, but it seems already efficient to make people aware of their venous risk and to better adhere to lifestyle improvement and venous disease treatments. PCV131 Preferences Regarding The Attributes Of Oral Anticoagulants In Patients With Atrial Fibrillation Results Of A Discrete Choice Experiment Boettger B 1, Thate-Waschke I M 2, Bauersachs R 3, Kohlmann T 4, Wilke T 5 1IPAM - Institut für Pharmakooekonomie und Arzneimittellogistik, Wismar, Germany, 2Bayer Vital GmbH, Leverkusen, Germany, 3Max-Ratschow-Klinik für Angiologie, Gefäßzentrum Klinikum Darmstadt GmbH, Darmstadt, Germany, 4University Medicine Greifswald, Greifswald, Germany, 5IPAM - Institute for Pharmacoeconomics and Medication Logistics, Wismar, Germany . . . . . . Objectives: Since the introduction of non-vitamin-K-antagonist oral anticoagulants (OACs), an additional option for stroke prevention in patients with atrial fibrillation (AF) compared to vitamin-K-antagonists (VKAs) is available. The objective of this study was to assess patients’ preferences regarding the attributes of these different treatment options. Methods: We conducted a multicenter study among randomly selected physicians who were asked to recruit AF patients. Patients’ preferences were assessed by computer-assisted telephone interviews. We used a Discrete-Choice-Experiment (DCE) with four treatment dependent attributes (need of bridging: yes/no, interactions with food/nutrition: yes/no, need of INR controls/ dose adjustment: yes/no, frequency of intake: once/twice daily) and one comparator attribute (distance to practitioner: < 1km/> 15km). Preferences measured in the interviews were analyzed descriptively and based on a conditional logistic regression model. Results: A total of 140 AF patients (age: 74.0±8.5 years; 57.0% male; mean CHA2DS2-VASc: 6.1±1.1; current medication: 27.1% rivaroxaban, 71.4% VKA, 1.4% other) could be interviewed. Regardless of type of medication, patients significantly preferred the attributes’ level (in order of patients’ importance) “once daily” for “frequency of intake” (binary-coded: once = 1 vs. twice = 0; Coefficient = 0.954; p< 0.001), “no” for “interaction with food/nutrition” (yes vs. no; -0.842; p< 0.001), “no” for “bridging necessary” (yes vs. no; -0.656; p< 0.001) and “≤ 1 km” for “distance to practitioner” (≤ 1 km vs. > 15 km; 0.644; p< 0.001). However, for the attribute “need of INR controls/dose adjustment” (yes vs. no; 0.020; p=0.808) no significant preference in favour of one of the options are shown. Conclusions: In our analyses, “once daily frequency of intake” was the most important attribute for patients’ choice followed by “no interactions with food/nutrition” and “no bridging necessary”. Thus, patients with AF seem to prefer treatment options which are easier to administer. PCV132 Health Related Quality Of Life At One Year Post Discharge In Patients With Heart Failure Escobar A 1, Trancho Z 1, Gonzale-Saenz de Tejada M 1, Quiros R 2, Garcia Perez L 3, Navarro G 4, Bilbao A 5 1Basurto University Hospital (Osakidetza). Red de Investigación en Servicios Sanitarios y Enfermedades Crónicas (REDISSEC)., Bilbao, Spain, 2Hospital Costa del Sol. Marbella, Marbella, Spain, 3D. Servicio Canario de Salud. Red de Investigación en Servicios Sanitarios y Enfermedades Crónicas (REDISSEC), Santa Cruz, Spain, 4Corporació Sanitària Parc Taulí, Sabadell, Spain, 5Basurto University Hospital (Osakidetza). Red de Investigación en Servicios Sanitarios y Enfermedades Crónicas (REDISSEC), Bilbao, Spain . . . . . . . Objectives: There are few studies in Spain about outcomes at 1-year in terms of health-related quality of life (HRQoL) in patients hospitalized by heart failure (HF). The main objective of the present study was to evaluate changes in HRQoL from baseline to 1-year post discharge in patients with HF through two questionnaires, SF-12 and Minnesota Living with Heart Failure questionnaire (MLHFQ). Methods: This has been a prospective study with 502 patients admitted by HF in Basque Country (Spain). Patients completed questionnaires during their hospitalization and at 1-year. The MLHFQ is a specific instrument which has 21 items with an overall scale, physical (8 items) and emotional (5 items) subscales. MLHFQ items are scoring from 0 (best) to 5 (worst). Total score ranges from 0 to 105, physical domain from 0 to 40 and emotional from 0 to 25. SF-12 has two dimensions, Physical Summary Score (PCS) and Mental Summary Score (MSC) which scores range from 0 (worst) to 100 (best). We have used general linear model to study gains in each dimension adjusted by baseline score, age, gender and hospital readmissions in the previous 6 months. Results: Mean age was 76.7 (SD= 10.4), there were a 56.1% of men and 20.3% of readmissions in the previous six months. Regarding both questionnaires and dimensions, baseline status influence in gains, the worse the baseline the more the gains. Likewise men have greater gains and patients readmitted lower in all domains. Age has an influence in MLHFQ physical domain and PCS. Conclusions: Adjusted by baseline score and readmissions, men have greater improvements in all domains of MLHFQ and SF-12. On the other hand, the younger the patients the higher the improvement is, in physical domains however age does not have any influence in psychological domains. PCV133 Literature Review Of Pro Measures Assessing Anticoagulant Therapy Barrett A M 1, Harris N I 2, DeMuro C 2, Kachroo S 3, Phatak H 3 Health Solutions, Manchester, UK, 2RTI Health Solutions, Research Triangle Park, NC, USA, 3Bristol-Myers Squibb Company, Princeton, NJ, USA . . . . . . . 1RTI Objectives: To identify and summarize the key characteristics, strengths, and weaknesses of available patient-reported outcome (PRO) measures assessing treatment satisfaction and health-related quality of life (HRQOL) related to anticoagulation therapy in patients with atrial fibrillation, including evaluating how well the measures meet current regulatory guidance requirements in Europe and the United States. Methods: Publications describing the development, validation, or use of PRO instruments in patients using anticoagulant therapy were identified through a comprehensive literature review. Several sources (PubMed, Patient-Reported Outcome and Quality of Life Instruments Database [PROQOLID], and ClinicalTrials. org) were reviewed to identify instruments. The development process and measurement properties (internal consistency and test-retest reliability, convergent and divergent construct validity, known-groups validity, and responsiveness) of the identified instruments were compared. Results: Five measures assessing anticoagu- A496 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 lation-related PROs were identified: 1) Sawicki questionnaire (assesses treatment satisfaction and HRQOL); 2) Perception of Anticoagulant Treatment Questionnaire (PACT-Q); 3) Duke Anticoagulation Satisfaction Scale (DASS); 4) Anti-Clot Treatment Scale (ACTS) (based on the DASS conceptual model); and 5) Deep Venous Thrombosis Quality of Life (DVTQOL) (assesses HRQOL outcomes related to a primary event of DVT and includes items assessing anticoagulation burden). All of the measures except the Sawicki questionnaire involved anticoagulated patients in the process of item generation and refinement. The ACTS and the DASS exhibited the strongest measurement properties. To a limited degree, the ACTS, DASS, and Sawicki questionnaire have demonstrated responsiveness in clinical trial settings. Conclusions: Although some of the identified measures have shown responsiveness in clinical trials, this review concluded that no existing measure appears likely to support an FDA or EMA PRO label claim in anticoagulation-related treatment satisfaction or HRQOL. However, concepts related to anticoagulation treatment satisfaction and HRQOL are important to patients and should be included in clinical trials, particularly as the burdens and benefits of anticoagulants evolve over time. PCV134 Health Utility Of Acute Coronary Syndrome Patients From An Asian Population Azmi S 1, Anchah L 2, Goh A 1, Fong A 2 Burhani Consulting, Petaling Jaya, Malaysia, 2Sarawak General Hospital Heart Centre, Kuching, Malaysia . . . . 1Azmi Objectives: To compare the health utility of Acute Coronary Syndrome (ACS) patients from an Asian population at baseline admission and 12 months post-ACS. Secondary objectives were to investigate the factors that affect health utility and the impact of using local versus UK tariffs in the analysis. Methods: Primary data was obtained from ACS patients who were admitted to a tertiary-care, general hospital in Malaysia and agreed to participate in the study. The quality of life (QOL) of ACS patients was elicited using validated language versions of the EQ-5D (three severity level) patient reported outcome instrument. QOL data was collected at baseline during initial admission for ACS and at 12-months post-admission. Patient demographic and clinical data were extracted from medical records. Health utility scores were calculated using EQ-5D utility tariffs from Malaysia and the UK population tariff. Results: A total of 112 subjects were recruited into the study of which 104 were used in the primary analysis. Mean age of patients in the analysis dataset was 56.1 years, 88% were male and duration of admission was 6.3 days. Calculated by Malaysian tariff weights, health utility was 0.75 during initial admission, increasing to 0.82 after 12 months (p= 0.012). Among the statistically significant factors associated with lower baseline utility were diagnosis of NSTEMI/unstable angina compared to STEMI (p=0.045), and female sex (p=0.038). Utilities calculated using the Malaysian tariff was consistently higher than those calculated using the UK tariff. ACS utility at baseline was 0.75 and 0.62 (p< 0.001) while utility after 12 months was 0.82 and 0.72 (p< 0.001) respectively. Conclusions: This study investigates several factors that may impact the QOL outcomes of Malaysian ACS patients. It also found significant differences in utility values calculated by Malaysian and UK tariffs, which indicate that the use of local tariffs is more appropriate. PCV135 Health-Related Quality Of Life Impact Of Triple Combinations Of Olmesartan Medoxomil, Amlodipine Besylate And Hydrochlorothiazide In Subjects With Hypertension Haag U 1, Guest J F 2, Soro M 3 GmbH, Schriesheim, Germany, 2Catalyst Health Economics Consultants Ltd., Northwood, Middlesex, UK, 3Daiichi Sankyo Europe, Munich, Germany . . . . 1HaaPACS Objectives: A secondary objective of a phase-III study spanning 54 weeks was to measure changes in the health-related quality of life (HRQoL) of 2,690 patients ≥ 18 years of age with moderate-to-severe hypertension who ended up receiving one of six doses of olmesartan/amlodipine/hydrochlorothiazide (OLM/AML/HCTZ) using the MINICHAL and EQ5D instruments. Methods: Descriptive statistics were used to measure blood pressure and HRQoL scores over the study period. Analysis of covariance (ANCOVA) was used to identify those factors (i.e. age, sex etc.) that could possibly have influenced HRQoL. Linear regression was used to assess the relationship between changes in blood pressure and HRQoL scores. Results: At the study start 90.8% of patients had Grade 2 or 3 hypertension, but at the study end 91.9% had normal/high-normal BP. Patients’ baseline MINICHAL mood and somatic domains scores were 5.5 and 2.6. Over the study period HRQoL improved as both MINICHAL scores decreased by 31-33%. Patients’ baseline EQ5D index and VAS score was 0.9 and 73.4 respectively, increasing by 6% and 12% over the study period. Patients’ QALY gain over the 54 weeks study period was estimated to be 0.029 QALYs. Linear regression was unable to detect any correlation between the changes in blood pressure and HRQoL scores. The ANCOVA model showed that changes in patients’ HRQoL was likely to have been influenced by patients’ grade of hypertension at baseline, the amount of concomitant medication (“pill burden”) and patients’ antihypertensive treatment in the last 26 weeks of the study. Conclusions: OLM/AML/HCTZ reduced blood pressure and significantly increased blood pressure control whilst improving patients’ HRQoL. Reducing patients’ pill burden is likely to increase adherence to treatment and improve blood pressure control. Hence, when prescribing antihypertensive agents physicians should consider the impact that pill burden has on adherence to treatment, achieving blood pressure control and patients’ HRQoL. PCV136 Treatment Patterns And Quality Of Life Of Patients With NonValvular Atrial Fibrillation: An Experience Of A Tertiary Health Care Centers (Treq-Af Study) . . . . . . . . . . . . Objectives: This study aimed to determine anti-coagulant treatment patterns and stroke- and bleeding-related risk factors and to evaluate quality of life (QoL) in nonvalvular atrial fibrillation (NVAF) patients. Methods: This multicenter (12-centers), observational study included ≥ 18 year-old patients (n= 213) diagnosed with NVAF. CHADS2, CHA2DS2-VASc, HAS-BLED scores and EQ-5D scale were used to assess risk factors and QoL and clinical features were recorded at baseline, and 6th and 12th months. Results: The rate of adverse events was 64.7%, major bleeding was 10.8%, stroke was 5.9%, and hospitalization was 25.5% in one-year follow-up. The patients’ treatment patterns were grouped as warfarin, new oral anti-coagulant (NOAC) (dabigatran, rivaroxaban), and antiplatelet agents (AA) (acetylsalicylic acid, clopidogrel). Patient numbers for the groups at baseline, and 6th and 12th months, respectively, were 92, 74, 41 for warfarin, 2, 13, 14 for NOAC, and 39, 29, 26 for AA. The distribution of patients in the warfarin, NOAC, and AA groups regarding CHADS2≥ 2 was 53.8% (n= 49), 50% (n=1), and 63.2% (n=24), respectively; regarding CHA2DS2-VASc≥2 was 86.8% (n=79), 50% (n=1), and 89.5% (n=34), respectively; and regarding HAS-BLED≥3 (high bleeding risk) was 23.1% (n= 21), 0% (n=0), and 18.4% (n=7), respectively. EQ-5D scale scores were 0.85±0.12 and 0.76±0.13 at baseline and 0.67±0.29 and 0.62±0.37 at 12th month for the warfarin and AA groups, respectively; the decrease was significant in the warfarin group (p=0.002) but not in the AA group (p=0.249). The mortality rates of the patients in the warfarin, NOAC, and AA groups in one-year follow-up were 7.6%, 0.0%, and 10.3%, respectively. Conclusions: Our study has demonstrated that a significant number of patients who should be on oral anticoagulants are still treated with AA and the negative effects of warfarin on QoL of NVAF patients as compared to AA. More data is needed with head-to-head comparison of warfarin and NOAC. PCV137 Quality Of Life In Patients With Permanent Cardiac Pacemaker In The Slovak Republic Gerlichova K 1, Simkova E 1, Mastiliakova D 1, Matisakova I 2, Bielik J 1 1Trencin University, Trencin, Slovak Republic, 2Trencin University of Alexander Dubcek, Trencin, Slovak Republic . . . . . Objectives: 2579 permanent cardiac pacemakers (PCPM) were implanted in Slovakia in the year 2012, 532 reimplantations were utilised in 475 resp. 98 per 1000000 inhabitants. Up to now in the Slovak Republic there was not realised the study oriented on OoL in the patients with this treatment. Methods: 100 patients with PCMK were studied, women 58, men 42. The average age was 66.47, duration of illness – 6.93 y. 57 patients were married, 31 divorced or a widower, widow, single 12.39 patients had sick sinus syndrome (SSS). 22 - AV blockade II, 23 - AV blockade III, 16 - brady form of atrial fibrillation. QoL was evaluated by standardised instrument and so Quality of Life Index Cardiac Version – IV). Four domains were examined: health and functional state, mental and spiritual state, family relations. Finally, the total QoL was discovered, too. Kruskall Wallis Anova Test and Mann-Whitney Tests were used to verify the statistical significance. Results: None of rhythm disorders and the duration of illness had impact on QoL. The smallest QoL occurred in health and functional state, the highest QoL occurred in family relations. In the last one (family relations) there was significant difference in comparison to the other domains (p ≤ 0,029), and specifically in the group of married patients. The age had negative significant correlation on QoL (p ≤ 0,027). Women had significant lower QoL in the social and economical fields. Conclusions: PCPM has a certain impact on QoL. The knowledge about the differences in the field of family state and in age gives the incentive to take more attention to older, divorced, widower/widows and single patients and try to realise relevant psychological interventions. PCV138 The Evaluation Of The Health Related Quality Of Life Among Adults With Hypertension Paczkowska A , Nowakowska E , Koligat D , Bryl W , Hoffmann K Poznan University of Medical Sciences, Poznan, Poland . . . . . Objectives: The aim of the study was a subjective evaluation of the quality of life among adults with diagnosed and treated hypertension. The paper also identifies social and clinical factors significantly influencing the quality of life of respondents. Methods: 112 people took part in the study (38 women and 78 men), aged between 19 and 65 years old – in all cases hypertension was diagnosed and treated in a particular health care centre. As a main study tool a questionnaire WHOQOL-BREF in a Polish version was applied. In addition, in order to evaluate the social and clinical factors that influence the quality of life participants were asked to fill anonymous questionnaire prepared specially for this study. Results: The results of the conducted studies indicated that people suffering from hypertension experience remarkably lower quality of life comparing to healthy people. Considerable discrepancies in terms of the quality of life were visible in physical and psychological domains of the WHOQOL-BREF questionnaire. It has been assumed that the quality of life of patients with hypertension is determined by both social (age, gender, education, economic status), and clinical (level of blood pressure, weight, the type of hypertensive therapy, the presence of coexisting diseases). Conclusions: Chronic diseases, including hypertension, distinctively affect the quality of life of patients. The quality of life of patients with hypertension is determined by numerous social and clinical factors. Thus, there is a need to consider the problem of hypertension and its treatment among adult people multidisciplinary – in order to improve their lives. PCV139 Treatment Satisfaction In Patients With Venous Thromboembolism As Measured With Pact-Q2: Prefer In Vte Registry Ozin B 1, Aytemir K 2, Arslan O 3, Ozcan T 4, Kanadasi M 5, Demir M 5, Gokce M 6, Sucu M M 7, Ozdemir M 8, Yigit Z 9, Yavuzkir M F 10, Oto A 2 1Baskent University, Ankara, Turkey, 2Hacettepe University, Ankara, Turkey, 3Dokuz Eylul University, Izmir, Turkey, 4Mersin University, Mersin, Turkey, 5Cukurova University, Adana, Turkey, . 6Karadeniz Technical University, Trabzon, Turkey, 7Gaziantep University, Gaziantep, Turkey, 8Gazi University, Ankara, Turkey, 9Istanbul University, Istanbul, Turkey, 10Elazig University, Elazig, Turkey . Willich S N 1, Bauersachs R 2, Gitt A K 3, Mismetti P 4, Monreal M 5, Wolf W P 6, Agnelli G 7, Cohen A T 8 1Charité - Universitätsmedizin Berlin, Berlin, Germany, 2Max-Ratschow-Klinik für Angiologie, Gefäßzentrum Klinikum Darmstadt GmbH, Darmstadt, Germany, 3Herzzentrum Ludwigshafen, . . . . . . . . . . . . A497 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Ludwigshafen, Germany, 4Centre Hospitalier Universitaire Saint-Etienne, Hopital Nord, Saint Etienne, France, 5Hospital Universitari Germans Trias I Pujol, Barcelona, Spain, 6Daiichi Sankyo Europe GmbH, Munich, Germany, 7University of Perugia, Santa Maria della Misericordia Hospital, Perugia, Italy, 8King’s College, London, UK Objectives: Following an acute deep vein thrombosis (DVT) or pulmonary embolism (PE) event, physicians can choose among heparin, vitamin K antagonists (VKA), and Non-VKA oral anticoagulants (NOAC) for therapy. We assessed patients’ satisfaction with their ongoing anticoagulation treatment. Methods: PREFER in VTE (PREvention oF thromboembolic events – European Registry in Venous Thromboembolism) is an ongoing non-interventional study in France, Germany, Austria, Switzerland, Italy, Spain and UK. The Perception of Anti-Coagulant Treatment Questionnaire (PACT-Q2) is a valid and reliable instrument that allows the assessment of patients’ satisfaction regarding anticoagulant treatment, as well as their opinion about convenience of use. Results: At baseline (BL), a total of 2311 patients with acute VTE (1366 DVT, 945 PE) met the eligibility criterion of current anticoagulation and completed the PACT-Q2 questionnaire. 665 patients were eligible for an interim analysis at 6 months: 7.8% have received heparin only, 33.2% initial heparin/VKA and 9.0% uninterrupted NOAC. In the “convenience” dimension, the score (0-100 range) at BL/6 months was 78.9 ±17.62/ 76.2 ±18.20 points for heparin only, 81.0 ±16.36/ 80.5 ±16.34 for heparin/VKA, and 88.9 ±12.14/ 93.3 ±7.30 for NOACs (all patients: 81.5 ±16.78/ 81.6 ±17.26). Compared to BL, the score improved the most in the NOAC group. In the “anticoagulant treatment satisfaction” dimension, the score at BL/ 6- month was 64.2 ±14.02/64.3 ±16.68 points for heparin only, 65.8 ±14.92/68.7 ±14.46 for heparin/VKA and 68.2 ±16.65/72.9 ±16.70 for NOACs (all patients 65.4 ±15.26/68.5 ±15.76). Conclusions: Overall, patients on current anticoagulation reached relatively high values on the convenience scale, but moderate values on the satisfaction scale. Patients on NOACs rated their convenience and treatment satisfaction substantially higher than patients on heparin/VKA. PCV140 Treatment Satisfaction In Patients With Atrial Fibrillation On New Oral Anticoagulants As Assessed With Pact-Q2 At Baseline And 12Month Follow-Up: Prefer In Af Registry Brüggenjürgen B 1, Schliephacke T 2, Darius H 3, De Caterina R 4, Le Heuzey J Y 5, Reimitz P E 2, Schilling R J 6, Schwertfeger M 2, Zamorano J L 7, Kirchhof P 8 University Berlin (SHB), Berlin, Germany, 2Daiichi Sankyo Europe GmbH, Munich, Germany, 3Vivantes Hospital Neukölln, Berlin, Germany, 4G. d’Annunzio University, Chieti, Italy, 5Hôpital Européen Georges Pompidou, Université René Descartes, Paris, France, 6Barts and St Thomas Hospital, London, UK, 7University Hospital Ramón y Cajal, Madrid, Spain, 8University of Birmingham Centre for Cardiovascular Sciences and SWBH NHS Trust, Birmingham, UK . . . . . . . . . . . . . . 1Steinbeis Objectives: We aimed to understand treatment satisfaction over time under everyday practice conditions after introduction of NOACs (Non vitamin K antagonist oral anticoagulants) in the The PREvention oF thromboembolic events – European Registry in Atrial Fibrillation (PREFER in AF). Methods: PREFER in AF documents clinical characteristics, management, quality of life and other outcome parameters of AF patients. A total of 6412 consecutive patients with ECG-confirmed AF in the previous 12 months were followed up prospectively. The ‘Perception of Anticoagulant Treatment Questionnaire’ is a valid and reliable instrument to assess patients’ expectations (PACT-Q1) and satisfaction regarding anticoagulant treatment, as well as patients’ opinion about treatment convenience of use (PACT-Q2). Results: A total of 2985 patients were willing to fill out the PACT-Q2 questionnaire both at baseline and FU. 1632 of these were on vitamin K antagonists (VKAs), 374 on NOACs, 203 on antiplatelets (AP), 142 on VKA+AP combinations and 634 on neither treatment. In the “convenience” dimension, the overall score (0-100 range) at FU was 84.3 ±16.5. The scores ranged from 81.9 in the no NOAC/VKA/AP group (± 18.43), 84.0 AP (± 16.7), 84.4 NOAC (± 15.9), 85.0 VKA (± 15.9) to 86.8 in the VKA +AP group (± 14.13). In the “anticoagulant treatment satisfaction” dimension of the PACT-Q2, the overall score was 65.1 ±15.7. The scores ranged from 64.5 in no NOAC/VKA/AP (± 14.8), 63.2 in AP group (± 17), 66.0 in NOAC (± 15.05), 64.9 in VKA (± 15.9), to 69.5 in VKA +AP group (± 15.8). Conclusions: Overall, patients on current anticoagulation achieve relatively high values on the convenience scale, but moderate values on the satisfaction scale. While differences in group size and patient characteristics need to be taken into account, patients on NOACs compared to patients on VKAs rate their convenience and treatment comparable to other treatment schemes. Cardiovascular Disorders – Health Care Use & Policy Studies PCV141 Primary Pacemaker Insertion: Gender Differences In Prior Er Utilization Roy D 1, Blanchette C M 2 1University of North Carolina, Charlotte, Charlotte, NC, USA, 2University of North Carolina at Charlotte, Charlotte, NC, USA . . . Objectives: Men’s and Women’s health care experiences differ as they age. These differences lead to disparate health and treatment outcomes between the genders, especially in the area of cardiac health, which has significant disease burden in the US. Women experience symptoms that ‘deviate’ from the ones found in a clinician’s guidebook. The emergency room (ER) also serves as a medical space for initial diagnosis of heart conditions. Thus, looking at the gender based use of ER prior to a cardiac event may help us understand the disparities from a systems perspective. Cardiac pacemakers are used to treat severe and/or symptomatic bradycardia, heart block or a combination of both. This study explored gender differences in characteristics and ER diagnoses associated with pacemaker implants in an inpatient setting. Methods: The study used data from the inpatient and emergency room hospital discharge data from the Florida HealthCare Utilization Project for years 2009 and 2010. Patients with a primary insertion of initial pacemaker and with a diagnosis of bradycardia or heart block or both were included. A unique patient identifier helped map patients across the inpatient and emergency set- tings. Descriptive statistics were used to look at items of interest. Results: There were 1403 discharges meeting the inclusion criterion. These were linked to 1402 ER discharges previous to the surgery. There were no significant differences between the genders in terms of characteristics like race, primary payer, age, etc. However, more men had ER visits than women (52.43% vs 47.57%). The top diagnosis in the ER consisted of conduction disorders (21.03%). Syncope occurred in 10.05% of discharges. Atrial fibrillation was the only ER diagnosis which women experienced more than men. Conclusions: The findings of the study were inconclusive in showing significant differences between the genders in terms of characteristics and ER diagnosis among recipients of primary pacemaker. PCV142 Ldl-C Goal Attainment In Patients With Hyperlipidemia - Estimates From Population-Based Register Data In Sweden Mesterton J 1, Hallberg S 1, Gandra S R 2, Banefelt J 1, Fox K M 3, Johansson G 4, Levin L Å 5, Sobocki P 6 1Quantify Research, Stockholm, Sweden, 2Amgen, Inc., Thousand Oaks, CA, USA, 3Strategic Healthcare Solutions, LLC, Monkton, MD, USA, 4Uppsala University, Uppsala, Sweden, 5Linköping University, Linköping, Sweden, 6IMS Health, Stockholm, Sweden . . . . . . . . . . . Objectives: To estimate low-density lipoprotein cholesterol (LDL-C) goal attainment in patients with hyperlipidemia or prior cardiovascular (CV) events. Methods: Retrospective population-based cohort study conducted using electronic medical records linked to national registers. Patients were included in the study based on a prescription of lipid-lowering treatment between January 1, 2006 and December 31, 2006 or history of CV events (prior to 2006) and followed until December 31, 2012 for estimation of LDL-C goal attainment. Patients were stratified into cohorts based on CV risk level. Propensity score matching was applied to compare patients with new events (myocardial infarction, unstable angina, revascularization, ischemic stroke, transient ischemic attack or heart failure) to patients without new events. Results: Mean LDL-C at the time of the new CV event (index date) was 100.4 mg/dL for patients with CV event history who had a new event (n= 1,101). The mean LDL-C for patients without new events in the same cohort (n= 1,304) was 97.4 mg/dL. The proportion of patients with CV event history and LDL-C < 70 mg/dL was 20.5% and 21.5% for patients with and without new events, respectively. The percentage who attained the goal of 100 mg/dL was 54.1% and 58.1%, respectively. Mean LDL-C for CV risk equivalent (RE) patients with new events (n= 803) was 100.3 mg/dL while patients in the same cohort without new events (n=975) had a mean level of 97.4 mg/ dL. 22.0% of CV RE patients with new events had a LDL-C < 70 mg/dL and 53.8% had a level < 100 mg/dL. The corresponding data for CV RE patients without new events were 19.1% and 57.7%, respectively. Conclusions: The proportion of patients who met LDL-C goals was low, suggesting that current treatment is suboptimal. The proportion of patients who attained the goals was similar in patients with and without new events and across CV risk levels. PCV143 The Clinical Impact Of Rivaroxaban To Chinese At Atrial Fibrillation Patients Results From A Simple Communication Tool Yang L1, Wu JJ2, Zhu G2, Evers T3 University, Beijing, China, 2Bayer Healthcare Company Ltd., Beijing, China, 3Bayer Pharma AG, Wuppertal, Germany 1Peking Objectives: To estimate the clinical impact over three years of switching atrial fibrillation (AF) patients receiving no treatment or aspirin, or Vitamin K antagonist (VKA), to rivaroxaban in China. Methods: A prevalence-based, deterministic budget impact model was developed. The number of AF patients was calculated based on data from literatures. The baseline risk of events and the efficacy and safety data for aspirin and VKA were from meta-analysis. While the efficacy and safety data for rivaroxaban were from ROCKET-AF trial. The current treatment pattern was from a large registry study for Chinese AF patients. 9%, 61% and 30% of AF patients received no treatment, aspirin and VKA, respectively. We just assumed that 5% of the patients who received aspirin currently switch to rivaroxaban per year. Then the model estimates the number of ischemic stroke (IS), systemic embolisms (SE), myocardial infarction (MI), intracranial bleeds (ICH) and major extracranial bleeds (ECH) per year. Results: The number of NVAF patients is 4.5 million. For those patients, there might be 127,469 major IS events, 34,001 SE events, 9,933 MI events, 28,745 ICH events and 56,193 major ECH events under current treatment per year. With 5% patients switching to rivaroxaban from aspirin in the first year, the IS events reduced 4%, SE events reduced 6% and MI events reduced 2% with increased 3% ICH and 1% ECH. To the third year, the IS events continue to reduce 11%, SE events reduced 17% and MI events reduced 6% with increased 10% ICH and 4% ECH compared with the current situation. Conclusions: Rivaroxaban may decrease the clinical burden of AF in China by reducing the incidence of stroke and fatal CV events. Decision-makers can find the exact value of rivaroxaban easily by the simple tool in different situations. PCV144 The Clinical Impact Of Rivaroxaban To Chinese At Deep Vein Thrombosis Patients Results From A Simple Communication Tool Yang L 1, Wu J J 2, Zhu G 2, Evers T 3 University, Beijing, China, 2Bayer Healthcare Company Ltd., Beijing, China, 3Bayer Pharma AG, Wuppertal, Germany . . . . . 1Peking Objectives: To estimate the clinical impact over three years of switching deep vein thrombosis (DVT) patients receiving low molecular weight heparin (LMWH) and vitamin K antagonist (VKA) to rivaroxaban. Methods: A prevalence-based, deterministic budget impact model was developed. The incidence of DVT patient was from literatures. We assumed that 12%, 65% and 23% received treatment for 3months, 6months and 12months, separately. The efficacy and safety data for LMWH+VKA were from meta-analysis. While the efficacy and safety data for rivaroxaban were from EINSTEIN-DVT trial. We just assumed that 20% of the patients who received LMWH + VKA currently switch to rivaroxaban in the first year and A498 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 following with 15% of the rest patients switching to rivaroxaban per year. Then the model estimates the number of DVT, pulmonary embolism (PE), intracranial bleeds (ICH) and major extracranial bleeds (ECH) per year. Results: The new DVT patients are 287,813 per year. For those patients, with 20% patients switching to rivaroxaban from LMWH+VKA in the first year, the recurrent venous thrombus embolism (VTE) events, including DVT and PE, reduced 4.6%, major bleeding events including ECH and ICH reduced 7% with 1% minor bleeding increasing. To the third year, the recurrent VTE events reduced 11.4%, major bleeding events reduced 17% with 2% minor bleeding increasing compared with the current situation. Conclusions: Rivaroxaban may decrease the clinical burden of DVT in China by reducing the incidence of recurrent VTE and fatal bleeding events. Decision-makers can find the exact value of rivaroxaban easily by the simple tool in different situations. PCV145 Cost-Effectiveness Of Disease Management Programs For Cardiovascular Risk And Copd In The Netherlands Tsiachristas A , Burgers L T , Rutten-van Mölken M P M H Erasmus University Rotterdam, Rotterdam, The Netherlands . . . . . . . Objectives: Disease management programs (DMPs) for cardiovascular risk (CVR) and chronic obstructive pulmonary disease (COPD) are increasingly implemented in the Netherlands to improve quality of care and patient’s lifestyle. The aim of the study was to provide evidence about the (cost-) effectiveness of Dutch DMPs as implemented in daily practice. Methods: We compared the 2-year costs and changes in physical activity, smoking behaviour, and utilities between the most and the least comprehensive DMP in four disease categories: primary CVRprevention, secondary CVR-prevention, both types of CVR-prevention, and COPD (total n: 1034). Propensity score matching increased comparability between DMPs. A cost-utility analysis was performed from the health care and societal perspective. Sensitivity analysis was performed to estimate the impact of DMP development and implementation costs on the cost-effectiveness. Results: Patients in the most comprehensive DMPs increased physical activity and had higher smoking cessation probabilities after 2 years in most disease categories. From a health care perspective, the incremental costs were positive in primary CVR-prevention (96% certainty), negative in secondary and both types of CVR prevention (93% and 98% certainty) and indifferent in COPD. The incremental QALYs were positive in all categories (certainty range: 64%-80%). The incremental cost-effectiveness ratio’s ranged from € -114,662 to € 8,849. The results from the societal perspective and the sensitivity analysis were in the same line. Conclusions: The most comprehensive DMPs for CVR and COPD were cost-effective compared to the least comprehensive DMPs. The challenge for Dutch stakeholders is to find the optimal mixture of interventions. PCV146 Comparing Quality Effects Of Patient Care In Integrated And Regular Care For Patients With Hypertension Waehlert L 1, Rex J 2, Engelhard J 2, Altmann V 2, Kostev K 2 1Fresenius University of applied sciences, Idstein, Germany, 2IMS Health, Frankfurt am Main, Germany . . . . . Objectives: This study examines the extent to which Integrated Care Programs lead to an improvement in the quality of patient care. The aim of the study is to carry out a quantitative analysis of differences in quality between patients participating in Disease Management Programs (DMPs) and patients receiving regular care, regardless of health insurance status, program, or region. Methods: The study used data from the representative IMS Disease Analyzer database. It included patients with a confirmed diagnosis of hypertension who started antihypertensive therapy in the period between January 2010 and December 2012. The primary dependent variable of the study was the change in blood pressure after at least six months of antihypertensive treatment. To assess this variable, we determined the proportion of patients with a blood pressure of below 140/90 in the period between day 183 and day 365 after initiation of treatment (index date). In order to eliminate confounding factors, we performed one-to-one matching based on a propensity score. Results: 1,317 patients participating in the integrated care program (ICP) and 1,317 patients not participating in such a program were available for further analyses following the propensity score matching. Patients in both groups were very similar with respect to demographic variables and antihypertensive therapy. The proportion of patients with blood pressure values < 140/90 after one year of treatment was 33.6% in the group of ICP participants and 22.7% in the group of non-ICP patients (p< 0.0001). The chance of reaching the treatment goal was significantly higher in the group of patients participating in an integrated health program (OR: 1.73; 95% CI: 1.45-2.05). Conclusions: It is evident that DMP participants have a significantly better chance of achieving the therapy goal. Thus, it can be established that integrated health care programs have a positive effect on quality. PCV147 Segmentation Is A Key Strategic Tool For Effective Prioritisation And Targeting Of Payers In Highly Competitive Markets; A Client’s Perspective Areteou T Double Helix Development, London, UK . Objectives: The research aimed to develop an attitudinal based, payer segmentation approach to explore payers’ attitudes and behaviours towards the managed entry of novel agents in the anticoagulation area in the health care systems of countries within the EU. The segmentation exercise explored payers’ drivers, motivations, barriers and limitations when assessing, endorsing or restricting new agents. Methods: Qualitative in-depth telephone interviews were conducted to explore payers’ views, along with perceived challenges relating to the entry of novel class of anticoagulation agents. Followed by a quantitative data collection and advanced statistical analysis methodology was employed with regional and local payers in each of the researched markets to define the segmentation accord- ing to attitudes and beliefs relevant to the therapy area. Results: Quantitative segmentation identified key distinct segments of payers displaying unique attitudes and beliefs towards entry of the novel class of anticoagulation agents. The segmentation approach identified key differentiating factors between segments, allowing full profiling of each group. Payers’ underlying values were explored with a view to gain insight to what is important to them as individuals as well as decision makers, what motivates them and what restricts them. Conclusions: Findings from this research were utilized to prioritise targeting of payer segments. In addition, communication and messaging strategies were optimised for these payer groups. Subsequently a post-project feedback workshop with the pharmaceutical client was conducted. The poster will discuss how the research was used by the pharmaceutical client and the benefits of this strategic tool to the brand team. PCV148 Impact Laws And Decrees On Activities: The Ilda Study Citarella A 1, De Liguoro F P 2, Di Martino P 3, Iarrobino A 2, Nava E 4, Ragone P 5, Vercellone A 4, Cammarota S 1 1LinkHealth s. r. l., Naples, Italy, 2MediCoop VESEVO - GPs Association, Torre del Greco, Italy, 3IPPOCRATE - GPs Assosiation, Castellammare di Stabia, Italy, 4Department of Pharmacy, Local health Napoli3 Sud, Castellammare di Stabia, Italy, 5IPPOCRATE - GPs Association, Castellammare di Stabia, Italy . . . . . . . . . Objectives: To assess whether the prescribing pattern of statins changed after reimbursement criteria revision and regional policies in a general practice in southern Italy. Methods: Analysis has been performed on a database of 123 medical practitioners that have managed an average of 190.000 inhabitants in the Campania Region (south of Italy). Prevalence of use and incidence of new treatments were calculated from Jan 2012 to Jun 2013. Statin users were stratified into three groups (Moderate Cardiovascular Risk, MR; High Cardiovascular Risk, HR; Very High Cardiovascular Risk, VHR) according to new criteria for reimbursement for lipid lowering agents revised by the Italian Drug Agency (AIFA) (Nota 13). Results: After the reimbursement criteria revision (November 2012), the prevalence of statin use slightly decreased reaching 6.6% in the second quarter of 2013 (-14% compared to second quarter of 2012). Stratified by level of CV risk, the prevalence of statin use is reduced by 24.9% into MR, 13.1% into HR and 5.9% into VHR, while incidence of new users of 22.4%, 34 5% and 45.8% respectively. In the second quarter of 2013, atorvastatin (+45.3%) was prescribed in 57,5% of patients in MR group (+45.3%), rosuvastatin 5.9% with (-60.1%) and 1.6% with simvastatin + ezetimide (-57.5%); in HR group, 40.7% (+20.8) atorvastatin, 5.3% (-57.0%) rosuvastatin and 2.1% (- 42.3%) simvastatin + ezetimide; in VHR, 56.3% (+15.8) atorvastatin, 12.5% (-37.5%) rosuvastatin and 2.1% (-27.1%) simvastatin + ezetimide. Conclusions: The revision of reimbursement criteria and the regional policies led to significant changes in general practice in southern Italy resulting in a reduction in the statin use, especially in patients who could potentially benefit from it most. The results of the study provide useful information for the general practitioner about areas for improvement prescriptive. PCV149 Assessment Of The Impact Of Legislation On The Utilization Of Statins In Slovakia Minariková D 1, Malovecká I 1, Foltan V 2, Lehocká L 1 University, Bratislava, Slovak Republic, 2Faculty of Pharmacy, Comenius University, Bratislava, Slovak Republic . . . . 1Comenius Objectives: Frequent legislative changes that have brought the Slovak health care reform over the past decade, reflected also on drug policy and especially on the prices of medicines and total consumption of medicines. Generic substitution was introduced to save both health insurance and patient’s finances. Methods: Data on prices and consumption of lipid-lowering agent medicines were collected from Slovak Ministry of Health and Health insurance. Data processing, we used a uniform methodology recommended by the WHO - ATC / DDD classification and basic statistical methods of observing. In case of national data, we reported consumption in units of DDD per 1,000 inhabitants for one day (DID). Results: A class of lipid-lowering agent medicines (C10A) poses in recent years on Slovak market expanding group. Its number significantly increased in direct proportion with the introduction of generic drugs, mainly after 2004, when generic substitution was enacted and later after 2011, when mandatory generic prescribing entry into the force. The proportion of total medicines and generic drugs in the group C10A were x2008total-medicines/%genericdrugs= 55/74,6 types and x2014total-medicines/%generic-drugs= 203/94 types. The number of generic drugs with atorvastatin ranged x2008-2014= 4-18, rosuvastatin x2008-2014= 3-11, simvastatin x2008-2014= 7-7, fluvastatin x2008-2014= 1-1, lovastatin x2008-2014= 1-5. The consumption of medicines with atorvastatin x2008-2014-packages= +Δ 47,7%, x2008-2014value= +Δ 20,1% € , rosuvastatin x2008-2014-packages= +Δ 18,8%, x2008-2014-value= -Δ 77,4 % € , simvastatin x2008-2014-packages= -Δ 48,6%, x2008-2014-value= -Δ 79,6 % € , fluvastatin x20082014-packages= -Δ 50,5%, x2008-2014-value= -Δ 90,4 % € , lovastatin x2008-2014-packages= -Δ 70,1%, x2008-2014-value= -Δ 81,8 % € . In the years 2008-2013 consumption of C10A in Slovakia increased by 77,5 per DDD/1000 inh. /day (DID). OECD statistics from 2011 indicate consumption of lipid-lowering agents amounting to 130 DID, while at the end of 2013 according to our data analysis increased to 142.8 DID. Conclusions: Overall changes in legislation of drug policy brought rising utilization of lipid-lowering agent medicines and in 2011 Slovakia together with Great Britain was on the second place with 130 DID consumption of lipid-lowering agents. PCV150 The Impact Of Pharmaceutical Policies On Pharmaceutical Sales Patterns In Sweden And Japan Imai S 1, Andersson Sundell K 2, Fushimi K 3 Hospital Organization, Tokyo, Japan, 2Sahlgrenska Academy, University of Gothenburg, Gothenburg, Japan, 3Tokyo Medical and Dental University Graduate School of Medicine, bunkyo-ku, Tokyo, Japan . 1National . . A499 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Pharmaceutical expenditure accounted for between 11% to 13% of total health expenditure in European countries including Sweden in 2010 whereas this was 20% in Japan for the same year. We expect that big changes in sales pattern will happen as a result of patent expiry. Other changes have been seen when prescribers get negative/positive information on drugs. In this article we study the impact of patent expiry and negative information relating to pharmaceutical policies on pharmaceutical sales patterns of selected drugs in Sweden and Japan. Methods: we selected angiotensin-converting enzyme inhibitors (ACEs) and Angiotensin Ⅱantagonists (ARBs), since widely used in Sweden and Japan. Seasonal autoregressive integrated moving average (ARIMA) modeling with intervention analysis was used to estimate the change of sales volume. Results: Losartan had a positive change (0.77650, p= 0.0068) in October 2010, Candesartan had a negative change (-0.50760, p= 0.0058) in July 2010. There were no significant differences in the sales volume of Losartan, Telmisartan, except for Candesartan in Japan (0.04868, p=0.7995, -0.38547, p= 0.0880, and -1.21215, p= 0.001, respectively). In this study, we used a publication informed that Candesartan and Telmisartan had not reduced patients’ mortality by a journal July in 2009 as negative information. Conclusions: We found that the sales pattern of selected drugs were changed by negative information and not by the expiry of their patents in Sweden. Whereas in Japan the negative situation for implementing generic substitution like the therapeutic substitution of Trandolapril and the possibility of switching from Candesartan to a combination drug was seen. Further assessment will be needed since factors associated with the changing use of drugs will be infinite. PCV151 The Impact of Modifications of the Formula for Generic Drug Prescription Rate on the Switch To New Brand-Name Drugs With Similar Therapeutic Uses Shimizu S 1, Imai S 2, Ishikawa K B 3, Ikeda S 4, Fushimi K 5 1Institute for Health Economics and Policy, minato-ku, Japan, 2National Hospital Organization, Tokyo, Japan, 3National Cancer Center, Tokyo, Japan, 4International University of Health and Welfare, Ohtawara -City, Japan, 5Tokyo Medical and Dental University Graduate School of Medicine, bunkyo-ku, Tokyo, Japan . . . . . . Objectives: From April 2013, the method of calculating the prescription rate of generic drugs in Japan was changed and protected brand name drugs were excluded from the denominator. In the case of Japan, which does not have a reference pricing system, it is thought that this will lead to a change in prescriptions toward protected brand name drugs rather than drug substitution to generic drugs. The objective of this study is to clarify the trends in relation to the prescription of generic drugs, through the use of administrative data on a nationwide level. Methods: We used survey data from dispensing pharmacies from April 2012 to March 2014. As a comparison, we used the prescription data of 1139 acute care hospitals in which incentive measures for drug substitution to generic drugs had not taken place during the same period. The medical products in question were drugs for diabetes and hypotensive drugs. For data analysis used SQL Server 2008 R2 and R. Results: As the dispensing pharmacy receives additional compensation based on the rate of generic drugs dispensed by that pharmacy in the most recent 3 months, the dispensing rate of generic drugs will have a direct impact on their business. The change in the method of calculating generic drugs has a major impact on the dispensing pharmacies, and in this study we have shown the possibility of dispensing pharmacies shifting more to protected brand name drugs. The dispensing rate of generic drugs by acute care medical facilities has always been low, and thus the impact of the change in calculation of the dispensing ratio is correspondingly low. Conclusions: The results of this study show that when encouraging drug substitution to generic drugs as a policy to reduce drug expenditure, it is necessary to consider measures in relation to the shift to protected brand name drugs. PCV152 Analysis of Cardiac Implants Recalls in the Last Decade: An International Comparison Zhang S X , Kriza C , Schaller S U , Kolominsky-Rabas P L Centre for Health Technology Assessment (HTA) and Public Health (IZPH), Friedrich-AlexanderUniversity Erlangen-Nürnberg, Erlangen, Germany . . . . . . . Objectives: The objectives of this research are to provide an overview of the recalls of cardiac implant medical devices in the last decade according to the different categories of cardiac implant medical devices and analyze the recall reasons. On the basis of this analysis, this research will provide recommendations on how to build a high quality implant registry. Methods: A systematic search was performed focusing on regulatory bodies’ homepages from a range of middle and high-income countries with sufficient information on cardiac implant recalls and the related reasons. Data was extracted for the years 2004-2014 with the following criteria applied: cardiac implant medical device recalls and reasons for recall, i.e. harm or risk to patients excluding labeling or legal problems. Results: 11 countries have been included in the study: United States, Canada, Australia, New Zealand, PR China, China Hong Kong, UK, Germany, Ireland, Switzerland and Saudi Arabia. 104 recall reports have been analyzed in total. The categories of cardiac implants include the following: Implantable Cardioverter Defibrillator (ICD) 40.4%; pacemaker 14.4%; stent 14.4%; as well as Cardiac Resynchronization Therapy (CRT) 13.5%; leads 9.6% and replacement materials 7.7%. Referring to the recall reasons, 32.7% of the reports related to problems with the device battery; 30.7% of devices were recalled due to incorrect therapy delivery; 15.4% devices had software problems; 15.4% devices had connection problems and 5.8% of devices did not deliver correct output data. Conclusions: Due to the high-risk nature of cardiac implants medical devices and their high complication rates associated with considerable associated mortality, the traceability and transparency of safety hazards information are crucial. By analyzing the recall information including recall reasons and cardiac implants categories, important information is gained that can inform a high quality cardiac implant registry for monitoring the safety of cardiac implant patients. PCV153 Dabigatran Users With Non-Valvular Atrial Fibrillation in the Us: A Characterization of Dabigatran Initiators and Switchers Shash D 1, Schnee J 2, Schneider G 3, Schoof N 1, Zint K 1, Clemens A 4, Bartels D B 1 Ingelheim GmbH, Ingelheim, Germany, 2Boehringer Ingelheim Pharmaceuticals, Inc, Ridgefield, CT, USA, 3Evidera, Lexington, MA, USA, 4University Medical Center Mainz, Mainz, Germany . . . . . . . . 1Boehringer Objectives: In routine clinical practice the selection of a particular anticoagulant for treatment of a specific patient may be based on a variety of different factors. The aim of this study was to explore if there were differences in the characteristics of patients with non-valvular atrial fibrillation (NVAF) who started on dabigatran etexilate (DE) and who in the prior year had no oral anticoagulant treatment (initiators) versus those who had previously been treated with warfarin (switchers). Methods: Medco claims data were used to characterize 7,055 NVAF patients from the US with a DE prescription between Feb 2011 and Apr 2012. The first prescription in this period defined the index date. The treatment-groups were stratified by initiators and switchers. Characteristics, comedications, and comorbidities in the 12-monthperiod prior to index date were assessed. All illustrated differences were statistically significant (p < 0.05) (1). Results: Switchers (N = 2,585) had a mean age of 74.0 (±9.6) years, whereas initiators (N= 1,903) were younger (mean 70.0 (±9.6) years). A higher proportion of switchers used comedications compared to initiators, e.g. beta blockers (66% vs. 59%), and gastrointestinal drugs (34% vs. 28%). Switchers were more likely to have congestive heart failure, hypertension, cerebrovascular disease, renal disease and bleeding related hospitalizations when compared to the initiators, and also had a higher mean CHA2DS2-VASc score (4.0 (±1.8) compared to initiators (3.4 (±1.9)). Conclusions: This study shows differences between patients who are initiating DE as first anticoagulant treatment and those who are switched from warfarin to DE revealing that switchers might represent a distinct patient population. Identifying, stratifying or accounting for such differences are necessary in comparisons using real world data. (1) Schoof N et al., Characteristics of patients with non-valvular atrial fibrillation using dabigatran or warfarin in the US. Curr Med Res Opin. 2013 Dec 27. PCV154 Impact of Drug Policy Regulations on the Consumption of Antihypertensive Drugs in Slovakia Psenkova M 1, Foltan V 2, Mackovicova S 1, Marcisova M 1, Minarikova D 3, Tomek D 4 Ltd, Bratislava, Slovak Republic, 2Faculty of Pharmacy, Comenius University, Bratislava, Slovak Republic, 3Comenius University, Bratislava, Slovak Republic, 4Slovak Medical University, Bratislava, Slovak Republic . . . . . . 1Pharm-In Objectives: In 2011, various legislative measures were adopted in Slovakia regarding drug policy. The aim of the submitted work is to evaluate links between the introduction of regulations and the consumption of antihypertensive drugs (AH), expenditures and patient co-payments. Methods: We evaluated data on drugs consumption based on IMS Data. Patient co-payments data were taken from the National Health Information database. When evaluating the average amount of co-payments, we applied a weighted average, which takes into account the level of co-payment and consumption. Results: The consumption of AH (in DOT) increased continually in 2006-2013 (+36%) and the turnover of AH dropped by 4% as a consequence of introducing new regulations. The impact of new regulations was expressed foremost in the consumption of RAS inhibitors, where the decrease in turnover after introduction of clusters in 2012 was 14%. In the evaluated period, the final price of AH was reduced by 35% (from € 8.86 to € 5.78) and the reimbursement was reduced by 52% (from € 7.87 to € 3.76). At the same time, there was an increase in average co-payments by 105% (from € 0.99 to € 2.03). In the evaluated period, the patient co-payment for a fixed AH almost quadrupled (from € 1.05 to € 4.05). The patient paid an average of € 0.93 more for one pack of fixed AH than for a free combination. After the introduction of regulations, the consumption of fixed combinations grew at a slower pace than in the case of monocomponents. Conclusions: The legislative changes in drug policy had a significant impact on the consumption of antihypertensive drugs and on the expense of hypertension treatment. It is necessary to monitor long-term and analyse the impacts of regulations on the prescription of AH and to evaluate factors that can influence the success of hypertension treatment. PCV155 Initiation of Oral Anticoagulant Drugs: Identification of Drivers of Prescribing of New Agents Versus Warfarin Spillane S 1, Bennett K 2, Barry M 1 1HSE Medicines Management Programme, Dublin, Ireland, 2Trinity College Dublin, Dublin, Ireland . . . Objectives: Oral anticoagulants (OACs), used for stroke prevention in atrial fibrillation, include warfarin and the newer drugs (NOACs) dabigatran, rivaroxaban and apixaban. High direct drug costs of the NOACs to the health care payer prompt monitoring of real-world NOAC uptake patterns. This study aimed to identify factors associated with anticoagulation initiation with NOACs versus warfarin. Methods: Analyses were performed using national pharmacy claims data from a means-tested state medical services scheme. First-time initiators of an oral anticoagulant between January 2009 and December 2013 with ≥ 1 year scheme eligibility and ≥ age 50 were identified. Patients whose claims amounted to < 90 days of oral anticoagulation were excluded. Patient characteristics, and number and type of concomitant medications at the time of first oral anticoagulant were recorded and considered as predictors of NOAC initiation using multivariate logistic models (odds ratios, OR and 95% CIs). Results: 34,944 new initiators of oral anticoagulants were included. In 2009, 7.6% of new initiators received a NOAC; this figure rose to 41.2% in 2013. The following were positively associated with NOAC initiation in multivariate analyses: female gender (OR: 1.14,95% CI 1.07-1.21), age < 80 (OR: 1.11, 95% CI 1.04-1.19) and concomitant receipt of NSAID drugs (OR: 4.04, 95% CI 3.72-4.38). Receipt of multiple concomitant medications was negatively associated with NOAC initiation; patients receiving 15+ drug classes had a 42% decreased odds of NOAC receipt (versus ≤ 5 drug classes) (OR: 0.58, 95% CI 0.51-0.66). Specific concomitant drugs negatively A500 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 associated with NOAC use included rate/rhythm control treatments (OR: 0.78, 95% CI 0.73-0.83) and SSRI/SNRI antidepressants (OR: 0.87, 95% CI 0.79-0.96). Regional variation in initiation of NOACs versus warfarin was also observed. Conclusions: Multiple comorbidities may be associated with lower likelihood of NOAC initiation, as recently observed in other jurisdictions. Such uptake patterns have implications for real-world cost-effectiveness and outcomes studies. PCV156 Investment Aspects of Generic Drug Policies in Countries With Severe Resource Constraints Kaló Z 1, Harsányi A 1, Vámossy I 2 Loránd University, Budapest, Hungary, 2Gedeon Richter Plc, Budapest, Hungary . . . 1Eötvös Objectives: The objective of generic drugs policies can be defined as reduction in health care expenditure without compromising health outcomes. This definition is based on the disinvestment aspect of drug policies. However, the objective of generic drug policies can be also defined from an investment perspective, especially in those countries with volume limits for the use of original patented drugs due to economic constraints: increase in population health gain by improved patient access without need for additional health expenditure. Our objective was to compare benefits of generic drugs policies in Germany vs Hungary. Methods: We reviewed the grey literature and IMS database to identify pharmaceutical products with (1) patent expiry in recent years, (2) major therapeutic advancement to previous standard therapies, (3) no direct therapeutic alternative at patent expiry, (4) pharmacy distribution and consequently reliable IMS sales records in different countries. Then we compared aggregated annual volume sales in DOT and ex-factory sales for the selected pharmaceuticals in +/- 3 years before and after first generic entry. Results: In this analysis we present the case of clopidogrel. In Germany the volume sales of clopidogrel products increased by 1.7% with 3 years after first generic entry, in Hungary the increase was 120.5%. The ex-factory sales were reduced after patent expiry in both countries, by 30.1% in Germany and by 59.5% in Hungary. Conclusions: In Germany off-patent clopidogrel generated significant savings without volume increase. In Hungary generic products significantly improved the accessibility of patients to clopidogrel therapy, in addition to reducing pharmaceutical expenditure. Incremental health gain of off-patent medicines should not be underestimated in those countries, where accessibility of patients to patented medicines in restricted. PCV157 The Impact of Drug Policy on the Utilization of Medicines for Treatment of Cardiovascular Diseases in Slovak Republic Gatialova K 1, Foltan V 2, Majtas J 3 1Comenius Univeristy, Bratislava, Slovak Republic, 2Faculty of Pharmacy, Comenius University, Bratislava, Slovak Republic, 3Comenius University, Bratislava, Slovak Republic . . . Objectives: From the total health care costs in Slovak Republic the costs of medicines for treatment of cardiovascular diseases represent about 25%. In the world the proportion is at 8-10%. Accurate data on morbidity from cardiovascular disease in Slovak Republic is not available. National Health Information Center is processing the data on prevalence and incidence of circulatory system diseases, but this includes only those patients who are followed in cardiology in SR. Reportedly, the prevalence of cardiovascular diseases in SR is about 250 000 patients (50.8/ 1000 inhabitants). Methods: The utilization of medicines in period from 2008 to 2013 for treatment of cardiovascular diseases was analysed quantitatively by inderect descriptiv method of evaluating supply of medicines in quantitative units (number of packages), in the number of DDD and in financial indicators reflecting the full value of consumed package. Data were gained from National Health Center and State Institute of Drug Control. Results: The decline of consumption expressed in number of packages was observed in the group of cardiac therapy and by peripheral vasodilatators. The groups of beta blocking agents, agents acting on the renin-angiotensin system and lipid modifying agents showed increase in consumption. In DDD units the consumption deacreased most significantly in the group of peripheral vasodilatators. Rise in DDD units was observed in the group of beta blocking agents, antihypertesives, beta blocking agents, agents acting on the renin-angiotensin system and lipid modifying agents. Atorvastatin was active agent with highest consumption in DDD. The highest average price per package was calculated by lipid modifying agents. After access of generic drugs to the market in 2008 the consumption in financial units declined while consumption in DDD grew in followed period. Conclusions: By using the same amount of health care expenditures there is the possibility to provide treatment to more patients with cardiovascular disease. PCV158 Local Variation in Primary Care Prescribing Behavior in England: Ticagrelor lor example suggests that a positive National level (NICE) recommendation does not necessarily translate into common local prescribing behavior. Local variation (the fifth hurdle of market access) should be considered by pharmaceutical companies when developing market access strategy. PCV159 Drug Utilisation in Cardiovascular Diseases Management in Slovakia: 8 Years Overview Babela R 1, Szydlowski S 2, Rusnak R 1, Fasko M 3 1St. Elizabeth University, BRATISLAVA, Slovak Republic, 2University of Scranton, SCRANTON, Slovak Republic, 3St. Elizabeth University, Bratislava, Slovak Republic . . . . Objectives: Atherosclerotic cardiovascular diseases (CVD) remain the major cause of premature death in Europe. CVD treatment and precribing behavior remains a major challenge for the doctors, payers and regulatory bodies. Key aim of our study was to collect and compare reliable and comparable data on drug utilization for CVD therapy in Slovakia during 8 years (2005-2012). Methods: We utilized review of available costs data sources connected to ATC classification and to cardiovascular diseases (C01-C10). We also looked for Daily Defined Doses (DDD) measurement units. We adopted time frame and data was consequently used for analysis. Costs were used in EUR. Results: Total increase in EUR spent on cardiovascular drugs was more than 60 million EUR (2012 vs 2005). Key growth drivers from selected ATC groups were Vasoprotectives and Calcium channel blockers with 78% and 88% growth respectively (2012 vs. 2005). Amount of drug costs allocated for cardiovascular disease escalated in 2011 with almost 197 million EUR, average price for package reached 5,38 EUR in same year and price per 1 DDD was 0,17 EUR per 1 DDD (2012) compare to 0,16 EUR (2005). Overall unit sales results from 2005 to 2012 show slide growth tendency till 2010 with following slight declined. Growth in units for all ATC group under our scope reached the level of 4,3% (2012 vs 2005). Standardized death rate for CVD decresed from 637,3 in 2005 to 510,4 in 2012 (per 100.000). Conclusions: Increase spending for CVD management translated also into decrease death rate (2012 vs 2005). It is important for regulatory bodies and payers to continue in taking adequate measures that will ensure rational pharmacotherapy alongside with improving prescribing behavior. Combining different measures, such as electronic prescription monitoring and promoting available guidelines for CVD management, could be an effective way. PCV160 Implementation of an Automatic Laboratory Data Checking System To Reduce Deduction of Statins Reimbursement in A Teaching Hospital in Taiwan Lu T H , Chang Y T , Lin Y M Shuang Ho Hospital, Taipei Medical University, New Taipei City, Taiwan . . . . . . Objectives: According to National Cholesterol Education Program Adult Treatment Panel IV, lipid-lowing agent is required if in patient with atherosclerotic risks. The annual cost of statins consumption was the first five human medications in Taiwan. Therefore, disallowed/deduction of reimbursement from Administration of National Health Insurance (NHI) was relatively higher than other drugs. An “Automatic Laboratory data Checking System” was established in order to enhance rational use of statins and to reduce deduction rate of statins reimbursement. This study aims to analyze the economic outcomes after implemented the system. Methods: The major cause of deduction was the lipid profile fragmented in the medical record. To ensure rational use of statins based on NHI regulation, an “Automatic Laboratory data Checking System” in computerized physician order entry (CPOE) system was implemented in a teaching hospital on February 2013. When processing a statin prescription through CPOE system, the prescriber should choose the lipid profile linked with laboratory system in our hospital, or filled in lipid profile performed at outside source. The prescription would be blocked if the inspection date and laboratory data were not adherence to the NHI regulation. Results: After system implementation, the deduction of statins reimbursement was significantly decreased. There were three indicators substantially improved in year 2013 than 2012: The average quarterly deduction was 2.16 million NTD reduced, the average quarterly deduction rate was 14% reduced (18.18% versus 3.95%), and the disallowed reimbursement account for 57.11% medication fee decreased to 7.17%. Conclusions: The present study demonstrated that “Automatic Laboratory data Checking System” lessen the economic burden of statins reimbursement based on NHI regulations. The system was associated with rational use of statins and reducing disallowed reimbursement as well. PCV161 Clinical and Demographics Characteristics of Non-Valvular Atrial Fibrillation Patients Switching From Warfarin To Novel Oral Anticoagulants Sear R D , Jenner H D McKinsey & Co, London, UK Kachroo S 1, Pan X 2, Liu L 3, Kawabata H 4, Phatak H 1 1Bristol-Myers Squibb Company, Princeton, NJ, USA, 2Bristol-Myers Squibb, New Haven, CT, USA, 3Pfizer, New York, NY, USA, 4Bristol-Myers Squibb, Hopewell, NJ, USA Objectives: To understand the level of local variation in community-level prescribing of ticagrelor in England, after national-level recommendation from NICE. Methods: Monthly GP-Practice-level prescribing data was collected for antiplatelet drugs (Chapter 2.9 of British National Formulary [BNF]) in England, between August 2011 and February 2013. Data was obtained from the Health and Social Care Information Centre (HSCIC) and analyzed in Statistical Analysis Software (SAS). The percentage of total antiplatelet spend (net ingredient cost) attributed to ticagrelor was calculated for each GP Practice and Clinical commissioning Group (CCG) cluster. Results: Despite national-level NICE guidance (December 2011) recommending the use of ticagrelor for Acute Coronary Syndrome, uptake of ticagrelor at CCG level varied greatly between August 2011 and February 2013. The proportion of total antiplatelet spend on ticagrelor in February 2013 ranged from 0% to 34∙91%, between CCGs. The highest relative use of ticagrelor was clustered around the Yorkshire region. Conclusions: The ticagre- Objectives: This real-world study evaluated the baseline characteristics of patients with non-valvular atrial fibrillation (NVAF) who had switched from warfarin to novel oral anticoagulants (NOACs). Methods: Retrospective cohort study was conducted using the MarketScan® plus Earlyview data from 10/1/2009 to 12/31/2013. Adult NVAF patients (ICD-9 code 427.31 or 472.32) with one year of baseline period and a history of continuous warfarin use in the baseline period for at least 3 months immediately before the index date (defined as the first NOAC claim) were included. Patients with evidence of valvular heart disease, thyrotoxicosis, pericarditis, mitral stenosis, VTE, cardiac surgery, and endocarditis during the baseline period were excluded. Categorical variables were reported as percentages and frequencies, and continuous variables as means±SD. Categorical variables were compared using Pearson’s chi-squared test while continuous variables were compared using wilcoxon signed-rank test. Results: Among 11,743 eligible patients, 427 (3.64%) switched to apixaban, 8,989 (76.55%) to dabigatran and 2,327 (19.81%) to rivaroxaban. . . . . . . . . . A501 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Apixaban (74.82±11.39 years) patients were older versus those who switched to dabigatran (72.46±10.89 years, p< 0.0001) or rivaroxaban (73.50±11.27 years, p= 0.0193). Apixaban users (45.4%) were more female compared to dabigatran (38.4%, p= 0.0037) and rivaroxaban (40.4%, p=0.0499). The mean CHADS2 score was higher for apixaban users (mean±SD 2.28±1.25) as compared to dabigatran (1.94±1.20, p< 0.0001) and rivaroxaban (2.18±1.25, p= NS) users. Apixaban patients had significantly higher baseline rates for congestive heart failure (p= 0.0111), hypertension (p= 0.0002), renal disease (p= 0.0017) and ischemic stroke/ transient ischemic attack (p= 0.0004) as compared to dabigatran users. Apixaban users (2.34±2.12) also had higher mean charlson comorbidity index scores as compared to dabigatran users (1.98±1.96, p= 0.0002). Conclusions: Patients who switch to apixaban are older and sicker as compared to those switching to dabigatran or rivaroxaban. A detailed evaluation of patient characteristics on the treatment outcomes in NVAF patients switching from warfarin to NOAC is warranted in future. PCV162 Treatment Patterns in Hyperlipidemia Patients With New Cardiovascular Events - Estimates From Population-Based Register Data in Sweden Hallberg S 1, Banefelt J 1, Fox K M 2, Mesterton J 1, Johansson G 3, Levin L Å 4, Sobocki P 5, Gandra S R 6 1Quantify Research, Stockholm, Sweden, 2Strategic Healthcare Solutions, LLC, Monkton, MD, USA, 3Uppsala University, Uppsala, Sweden, 4Linköping University, Linköping, Sweden, 5IMS Health, Stockholm, Sweden, 6Amgen, Inc., Thousand Oaks, CA, USA . . . . . . . . . . . Objectives: To assess treatment patterns of lipid-lowering drugs in patients with hyperlipidemia or prior cardiovascular (CV) events (myocardial infarction, unstable angina, revascularization, ischemic stroke, transient ischemic attack or heart failure) who experience new CV events. Methods: A retrospective populationbased cohort study was conducted using Swedish electronic medical records and national registers. Patients were included in the study based on a prescription of lipid-lowering treatment between January 1, 2006 and December 31, 2006 or history of CV events (prior to 2006) and followed until December 31,2012 for identification of new CV events and assessment of treatment patterns. Patients were stratified into three cohorts based on CV risk level. The index was the date of first new CV event during follow-up. All outcomes were assessed during the year following index date. Adherence was defined as medical possession ratio (MPR) > 0.80. Persistence was defined as no gaps > 60 days in supply of drug used at index date. Results: Of patients with CV event history (n= 6881), 49% were not on treatment at index. Corresponding data for CV risk equivalent and low/unknown CV risk patients were 37% (n=3226) and 38% (n=2497), respectively. Mean MPR for patients on treatment at index was similar across cohorts (0.74–0.75). The proportions of adherent patients (60–63%) and persistent patients (56–57%) were also similar across cohorts. Dose escalation from the dose at index was seen within all cohorts, most notably for patients with low/unknown CV risk as 25% increased the dose after index. 2–3% of patients switched to a different drug after index while 5–6% of patients augmented treatment by adding another lipid-lowering drug. Conclusions: Almost 50% of secondary prevention patients were not on any hyperlipidemia treatment, indicating a potential therapeutic gap. Medication adherence and persistence among patients on hyperlipidemia treatment were suboptimal. PCV163 Determinants of Health Care Utilization in Hypertensive Patients: A Longitudinal Analysis Pinheiro B 1, Fernandes M 2, Antunes M 3 de Investigação Sobre Economia Portuguesa, Lisbon, Portugal, 2Universidade de Lisboa, Lisbon, Portugal, 3Faculdade de Ciências da Universidade de Lisboa, 1749-016, Portugal . . . 1Centro Objectives: DIMATCH-HTA is a cohort study conducted to identify determinants of blood pressure control among Portuguese-speaking African immigrants and Portuguese natives. The aim of this analysis is to study the determinants of health care utilization among hypertensive patients followed in Primary Health Care. Methods: The sample comprises 1243 unbalanced observations from 513 patients. Data were collected through questionnaires administered face-to-face at baseline, 6 and 12 months, and by telephone at 3 and 9 months after enrolment. The variable used to capture health care utilization was the number of visits to the general practitioner related to hypertension, in the three months prior to each interview. The covariates were chosen based on Grossman’s health capital model of demand for health (1972) and Andersen conceptual model (1968). Amongst socioeconomic covariates, gender, age, ethnicity, education, monthly equivalent income, living alone and domestic work were considered. Health status was captured by the presence of diabetes, time since hypertension diagnosis and number of medicines. Self-perception of hypertension and private health insurance were also included. A GLMM model for count data, with a random effect on intercept was estimated. Model adequacy was checked via residual analysis and comparison of observed and predicted values. The analysis was performed in R (version 3.0. 2) using lme4 package. Results: Results of estimated model indicated that total number of medicines (beta= 0.05, p-value= 0.03) and diabetes (beta= 0.40, p-value=0.01) have significant positive impact on health care consumption. Patients whose occupation is domestic work have higher health care utilization (beta= 0.73, p-value= 0.02). On the other hand, those who self-assess their hypertension as controlled (beta= -0.37, p-value= 0.02) and patients with higher income (in log scale, beta= -0.26, p-value= 0.01) have less visits to general practitioner related to hypertension. Conclusions: Based on the analysis it seems that not only variables related to health status, but also socioeconomic determinants impact health care utilization in hypertension. PCV164 The Determinants of Uptake and Diffusion of Innovative Health Technologies. An Empirical Analysis Callea G 1, Armeni P 2, Tarricone R 3, Cavazza M 4, Jommi C 5 . . . . . 1Centre for Research on Health and Social Care Management (CERGAS) and The European Health Technology Institute for Socio-Economic Research (EHTI), Milan, Italy, 2Bocconi University, Milano, Italy, 3Bocconi University, Milan, Italy, 4Centre for Research on Health and Social Care Management (CERGAS), Milan, Italy, 5Università del Piemonte Orientale, Novara, Italy Objectives: The aim of the research is to explore the main determinants driving the diffusion of new medical technologies. Methods: We investigated the diffusion two medical technologies: cardiac ablation for atrial fibrillation (CA-AF) and left atriage appendage closure (LAAC). Our sample consists of all Italian hospitals that adopted the two focal technologies in the period 2009-2012. We classified the hospitals according to ownership, type, teaching status and urban location. In the first regression we investigate the diffusion, i.e. yearly number of procedures performed by each hospital, in the second one the average frequency of use, measured as the average number of days between two subsequent procedures in each hospital. Results: Different types of hospitals show different baseline trends of diffusion. Public hospitals, both teaching and non-teaching, use less CA-AF compared to private non-teaching ones. Among hospital-level variables, the use of DRG-based reimbursement has a positive effect on CA-AT diffusion, not significant for LAAC. The rank in adoption is an important factor only for CA-AF. The impact of the number of other hospitals contemporaneously adopting the technology is negative for both technologies, but significant only for CA-AF. The average time between two subsequent uses is negatively correlated with the diffusion for both technologies. Regional contextual variables, including type of funding and socio-economic variables do not show significant impacts, with the exception of per capita public health expenditure, that enhances the diffusion of both technologies, and the ratio between public health expenditure and GDP in the case of CA-AF. The average frequency of use decreases over time for both technologies. Neither hospital-level variables nor regional-level ones do not show significant effects. The only variable with a significant and negative impact on the frequency of use of CA-AF is the number of competitors. Conclusions: These results are consistent with previous literature. PCV165 Challenges and Opportunities in The Management of Chronic Diseases During The Economic Crisis In Greece: A Qualitative Approach Tsiantou V , Mylona K , Karampli E , Boubouchairopoulou N , Pavi E , Kyriopoulos J National School of Public Health, Athens, Greece . . . . . . Objectives: There is evidence that in Greece, economic crisis has substantially affected chronic patients’ access to health care services. The aim of the study was to depict the current situation and identify the challenges and opportunities regarding the management of chronic diseases, in times of economic crisis and austerity from the aspect of both providers and patients. Methods: Representatives of chronic patients and medical associations were invited to participate in a focus group session. Four diseases (Type 2 Diabetes, Hypertension, COPD, and Alzheimer) were selected based on their epidemiology and socioeconomic impact on the Greek health care system. Fifteen representatives participated and their statements were recorded, analyzed and categorized into 4 categories. Results: Common issues in the management of chronic diseases under study appear to be the low quality of health services, fragmented primary care system and absence of specialized centers for the management of chronic diseases. These problems of the Greek health system were found to be magnified because of the recession. Furthermore, the increased numbers of unemployed, uninsured and patients at risk of poverty, puts additional pressure to the health system and futher undermines the quality of the health services. Economic and geographic barriers in access were reported, strengthening of the primary health system, development of patient registries, patient education regarding self-management and involvement of their associations in decision making was considered critical to the improvement of disease management. Conclusions: The management of chronic diseases was challenging even before the economic crisis in Greece, but the current economic framework poses additional threats for the health care system jeopardizing patients’ health and its sustainability due to an increased risk for future costs. Investment in patients and physicians education regarding chronic diseases management was thought to be the key for improving this situation. PCV166 Catastrophic Health Expenditures and Chronic Condition Patients in Greece Skroumpelos A 1, Pavi E 1, Pasaloglou S 2, Kyriopoulos J 1 1National School of Public Health, Athens, Greece, 2Novartis Hellas, Metamorfosi, Greece . . . . Objectives: Aim of the study was to investigate chronic patients’ out-of-pocket expenditures and the percent of the households subjected to catastrophic health expenditures (CHE) during the economic crisis. Methods: A cross-sectional study was conducted among 1600 chronic patients suffering from diabetes, hypertension, COPD and Alzheimer. Patients were asked to indicate the amount they spent for primary and secondary health care services and for pharmaceuticals. Current household income and income decrease since 2010 were also measured. CHE was defined as any amount spent for health which accounts for more than 20% of the total household income. Results: 1594 patients responded to the survey (99.6%). In 2013, 7.8% of all households with at least one chronic condition patient were subjected to CHE, compared to 3.6% in 2010. The analysis by disease showed that 11.4% in 2013 vs 6.2% in 2010 of the Alzheimer patients faced CHE, while the respective figures was 8.7% vs 2.9% for the COPD patients, 7.1% vs 3.4% for diabetic patients, and 4.2 vs 1.7% for the hypertensive patients. Pharmaceutical expenditures alone were deemed catastrophic for 4.6% of all the above patients in 2013 vs 1.6% before the introduction of the austerity measures and the health care reforms. Of the Alzheimer patients, 6.2% in 2013 vs 2.8% in 2010 faced CHE due to out-of-pocket payments for drugs, while the respective figures were found to be 3.4% vs 1.8% for the diabetic patients, 2.9% vs 0.9% for COPD patients and 1.7% vs 0.9% for the hypertensive patients. Conclusions: After the introduction of austerity measures A502 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 in Greece a substantial increase of the households with at least one chronic condition patient which are subjected to CHE is recorded. There is a need for counter measures or/and an alternative policy context in order to reduce this catastrophic effect of economic crisis. PCV167 Snapshot of Prescribing Practice for Clopidogrel and Esomeprazole Co-Prescription and Cost Evaluation of Guidelines Application Vernaz N , Rollason V , Adlere L , Bonnabry P , Desmeules J University of Geneva Hospitals, Geneva, Switzerland . . . . . Objectives: Through CYP2C19, the antiplatelet clopidogrel and the proton-pump inhibitor esomeprazole demonstrate a pharmacokinetic interaction that could translate into clinical inefficacy of clopidogrel. No medical consensus has been reached to date and therefore different guidelines are available. We aimed to evaluate the prescribing practices in the University Hospitals of Geneva (HUG) by measuring if the co-prescription was staggered as suggested by experts. We also measured the Omeprazole-CLopidogrel-Aspirin (OCLA) study impact on clopidogrel use in our hospital. Methods: Patient’s medical orders and nurse’s drug administration planning’s were analysed from January 2013 to April 2014 and the hospital pharmacy database from January 2000 to April 2014. To measure the “extra costs” of the implementation of different guidelines we built scenarios assuming the clopidogrel or esomeprazole replacement with prasugrel or ticagrelor and pantozole or ranitidine, respectively. Results: Fifty seven percent of patients under clopidogrel had a co-prescription of esomeprazole during the study period. Among them 15% (154/1’000) had a medical order staggering the co-prescription (more than 10 hours apart), 16% a concomitant prescription and 64% no clear information. Five percent had 40 mg esomeprazole twice daily, hindering the possibility of staggering. Surprisingly we found a higher rate of patients having a nurse’s schedule of more than 10 hours (39%, 417/1’071). Switching drugs would lead to increased costs for HUG of € 38’210 for prasugrel, € 34’800 for ticagrelor, € 9’590 for pantoprazole and € 5’205 for ranitidine. A statistical significant decrease in trend of clopidogrel use was observed after the OCLA study publication. Conclusions: The medical order’s information time frame should be mandatory in order to improve the transmission throughout the whole information system and allow a clear staggering of clopidogrel-esomperazole co-prescription avoiding drug-drug interactions when possible. Nurses take the initiative to stagger the co-prescription when these are not clearly defined by medical orders. PCV168 Regional Variation in Hospital Mortality, Length of Stay and Cost of Ischemic Stroke Patients in Alberta Ohinmaa A 1, Zheng Y 2, Jeerakathil T 1, Thanh N X 2, Hakkinen U 3, Kaul P 1, Klarenbach S 1, Friesen D 4, Ariste R 5, Ruseski J 6, Jacobs P 2 of Alberta, Edmonton, AB, Canada, 2Institute of Health Economics, Edmonton, AB, Canada, 3National Institute for Health and Welfare, Helsinki, Finland, 4Alberta Medical Association, Edmonton, AB, Canada, 5Canadian Institute for Health Information, Ottawa, ON, Canada, 6West Virginia University, Morgantown, WV, USA . . . . . . . . . . . . 1University Objectives: This study compares the outcome and health care performance among five health zones in Alberta by evaluating 30 days in-hospital mortality and length of stay (LOS) in patients with acute ischemic stroke, and total hospitalization costs over one year. Methods: Ischemic stroke (ICD-10 code I63) patients (without previous stroke within one year, N= 1,445) hospitalized between April 1, 2007 and March 31, 2008 were followed for one year using hospital Discharge Abstract Database. The severity of the stroke was obtained from the ambulatory care database (NACRs). Median hospital costs by CMG+ group were obtained from Alberta Health. Logistic regression was used to analyse in-hospital mortality; negative binomial regression assessed LOS, and generalized gamma model (log link) for hospital costs. The risk-adjusted outputs were estimated adjusting for sex, all disease-specific co-morbidities, and stroke severity. We calculated observed/ expected results for five zones; South, Calgary, Central, Edmonton, and North Zones. Results: The risk-adjusted 30-days-mortality rates (95% CI) varied from 7.8% (3.4%-12.1%) to 13.5% (9.6%-17.4%) in South and Central zones, respectively. The adjusted mean LOS varied from 16.3 (13.8-19.3) days in South Zone to 26.7 (24.2-29.5) and 29.3 (24.0-35.8) days in Edmonton and North zones, respectively. The results show several statistically significant differences between the first episodes LOS between zones reflecting partly differences in the post-acute care in each location for patients not discharged to home. The one-year-mean hospitalization costs varied from $72,300 ($55,000-$95,100) in North Zone to $25,500 ($20,500-$31,800) in South Zone. Conclusions: The study shows significant variation in outcomes for ischemic stroke between the five health zones. Although the Provincial Stroke Strategy has largely standardized the stroke care between regions, differences in post-acute care arrangements have produced significant LOS and cost differences. More detailed analysis of the reasons for regional variation is needed for improvement of the regional health care outcomes. PCV170 The Association of Hosptial Type and Stroke Centre With Mortality, Length of Stay and Hospital Cost of Ischemic Stroke Patients in Alberta Ohinmaa A 1, Zheng Y 2, Jeerakathil T 1, Thanh N X 2, Hakkinen U 3, Ruseski J 4, Kaul P 1, Klarenbach S 1, Friesen D 5, Ariste R 6, Jacobs P 2 1University of Alberta, Edmonton, AB, Canada, 2Institute of Health Economics, Edmonton, AB, Canada, 3National Institute for Health and Welfare, Helsinki, Finland, 4West Virginia University, Morgantown, WV, USA, 5Alberta Medical Association, Edmonton, AB, Canada, 6Canadian Institute for Health Information, Ottawa, ON, Canada . . . . . . . . . . . . Objectives: This study examines the association of 30 day in-hospital mortality, length of stay (LOS) during the first hospital episode, and hospitalization costs during one year after acute ischemic stroke by type of hospital and by stroke centre status. Methods: New ischemic stroke (ICD-10 code I63) patients (no previous stroke within one year) between April 1, 2006 and March 31, 2009 (N= 4,350) were followed for one year using hospital Discharge Abstract Database. The severity of the stroke was obtained from the ambulatory care database. Median hospital costs by CMG+ group were obtained from Alberta Health. Hospitals were classified as teaching, community large, community medium, and community small hospitals. Hospitals were also classified as comprehensive stroke centre, urban and rural primary stroke centres, and other urban and rural hospitals. The adjusted risk factors in Bayesian Model included sex, age, all disease-specific co-morbidities, and disease severity. The results for four hospital types and five stroke center categories were calculated using the observed/expected approach. Results: The 30 days mortality rates (95% CI) were lowest for teaching hospitals 10.1% (9.0%-11.2%) and large community hospitals (10.0%; 8.3%-11.8%), and the small community hospitals had the highest mortality rates (12.8%; 9.9%-15.8%). The mean LOS (95% CI) varied from 21.7 (20.9-22.6) days in teaching hospitals to 34.2 (28.6-41.0) days in community medium hospitals. The community medium hospitals had significantly higher costs ($62,400; $49,900-$78,000) than the community large hospitals ($32,900; $29,900-$36,200) and teaching hospitals ($37,000; $34,900-$39,200). Both comprehensive stroke and urban stroke centers had lower 30 day mortality rates (95% CI): 9.9% (8.8%-11.1%) and 9.7% (7.3%-12.0%); shorter LOS 21.6 (20.7-22.5) and 25.0 (22.7-27.6) days; and medium levels of costs $39,300 ($36,100-$40,700), compared to other hospitals. Conclusions: The study shows the hospital type and stroke centre had limited effects on the mortality but significant impact on LOS and costs. PCV171 In-Patient Hospital Costs of Stroke: A Focused Literature Review Kritikou P 1, Vemmos K 2, Payne K A 3 BioSource Corporation, London, UK, 2Acute Stroke Unit, Department of Clinical Therapeutics, University of Athens, Athens, Greece,, 3United BioSource Corporation, Dorval, QC, Canada . . . . 1United Objectives: Stroke is the third leading cause of mortality worldwide, with significant associated acute care hospitalization costs. The objective of this literature review was to delineate the costing methodologies employed for the estimation of in-patient hospital costs of stroke. Methods: A PubMed search was performed using the keywords: hospitalization, cost analysis, acute stroke, and cost-effectiveness; limited to publications in English from 2008 onwards. Inclusion criteria were patient-level data collection and detailed description of costing methodology applied. Cost-effectiveness and literature review studies were excluded. Results: In total 22 articles were included in the analysis. Cohort studies comprised 45% of the sample, followed by database analyses (32%), registered-based studies (9%), retrospective chart review studies (9%), and clinical trials (5%). Cost categories measured included direct medical costs (bed and staff, laboratory and imaging investigations, medications, rehabilitation and supportive nursing care), as well as indirect costs for the patients and their caregivers; in 2 studies the economic analysis was performed from a societal perspective. The resource utilization (excluding the database analyses) was identified in the medical records (80%), or from interviews (20%). Unit costs were primarily derived from national listings or hospital accounting files (36% each). The sample sizes (ranging from 100 to over 60,000 patients), as well as the total costs (ranging from US$500 to US$150,000 per patient and from US$70 to US$13,000 per day) varied significantly, as a result of the heterogeneous cost variables described. Conclusions: Methodologies differed in approach, complexity and specific cost variables evaluated. Consequently, the total costs varied significantly across studies which makes direct comparisons of outcomes difficult. A trend towards more sophisticated economic analyses, such as real costs measured versus hospital reimbursement rates, or hospitalization costs before versus after stroke, was observed. A more standardized approach to evaluating in-patient costs of stroke care is warranted. PCV172 Optimizing Process Efficiency Through Implanting Reveal Linq VersUS Reveal Xt/Dx From Three Spanish Hospital Perspective Egea García M 1, Toquero Ramos J 2, García Alberola A 3, Arias Palomares M Á 4, Marti Sánchez B 1 1Medtronic Iberia, Madrid, Spain, 2Hospital Puerta de Hierro, Majadahonda, Spain, 3Hospital universitario Virgen de la Arrixaca, Murcia, Spain, 4Hospital Virgen de la Salud, Toledo, Spain . . . . . . Objectives: Implantable loop recorders (ILR) are devices that continuously monitor heart rhythm in patients with suspicion of cardiac arrhythmias. Reveal LinQ™ is a new insertable holter, an 87% smaller than Reveal® XT/DX that records abnormal heart rhythm up to 3 years. The objective was to develop an economic tool which allows hospitals to quantify their cost savings from the simplified procedure of Reveal LinQ™. The tool was used to compare the costs of implanting Reveal® XT/DX in the cath lab to the costs of inserting Reveal LinQ™ out of the cath lab in three public hospitals of the Spanish National Health Care System. Methods: A cost model was developed to assess the cost per procedure of Reveal LinQ™ and Reveal® XT/DX. The model included data of the personnel needed in the procedure, the hospital setting, the hospitalization previous to the procedure, remote monitoring and post-procedure controls. Results: The total process-related savings of LinQ™ vs. Reveal® XT/DX in Virgen de la Salud, Puerta de Hierro and Virgen de la Arrixaca Hospitals were € 335 (13.3%), € 365 (13.1%) and € 517 (19.2%), respectively. Reveal LinQ™ was associated with a 66% reduction in cardiologist and OR-assistant time in Virgen de la Salud Hospital, a reduction of 15 minutes of cardiologist time in Puerta de Hierro Hospital and a reduction of 3 control visits due to remote monitoring in Virgen de la Arrixaca Hospital. Conclusions: The economic tool showed that the insertion of Reveal LinQ™ is associated with mean savings of € 406 from a hospital perspective compared to previous devices; mainly derived from moving the procedure out of the catheter lab, a reduction of the specialists’ time and in-hospital follow up visits due to remote monitoring. A503 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PCV173 Effect of Oral Nutritional Supplements On Hospital Outcomes in Patients Aged 65+ With Congestive Heart Failure Lakdawalla D 1, Thornton Snider J 2, Perlroth D 3, LaVallee C 2, Linthicum M T 2, Philipson T J 4, Partridge J 5, Wischmeyer P 6 of Southern California, Los Angeles, CA, USA, 2Precision Health Economics, Los Angeles, CA, USA, 3Stanford University, Stanford, CA, USA, 4University of Chicago, Chicago, IL, USA, 5Abbott Nutrition, Columbus, OH, USA, 6University of Colorado School of Medicine, Aurora, CO, USA . . . . . . . . . . 1University Objectives: Hospital admissions for congestive heart failure (CHF) are a major driver of costs for health systems, and CHF is especially prevalent in patients aged 65 and older. This study assessed whether provision of oral nutritional supplements (ONS) in the hospital can reduce these costs, by estimating the effect of ONS use on 30-day readmission rates, length of stay (LOS), and hospitalization episode costs. Methods: Using the 2000-2010 Premier Research database, a large US hospital episode database, we extracted a sample of episodes among patients aged 65 and older with a primary diagnosis of CHF. We excluded episodes involving tube feeding and those ending in death (due to censoring). Using propensity score matching, we created a 1: 1 matched sample of ONS and non-ONS episodes. We applied ordinary least squares (OLS) and instrumental variables (IV) regression analyses to investigate the outcomes of ONS use. The key outcomes studied were 30-day readmission rates, LOS, and episode costs (measured in 2010 US$). Results: Propensity score matching produced a matched sample of 38,418 CHF episodes. Naïve OLS analysis, which did not account for selection bias, suggested that ONS increased costs, LOS, and probability of 30-day readmission. However, using IV regression analysis to control for selection bias revealed that ONS use reduced the probability of readmission within 30 days by 10.1%, from 0.387 to 0.360 (p< 0.01). LOS was reduced through ONS use by 1.28 days (14.2%), from 9.03 to 7.75 (p< 0.01). Episode costs were significantly lower with ONS use, reducing medical expenditures by $1,266 (7.8%), from $16,166 to $14,900 (p< 0.01). Conclusions: In elderly patients hospitalized with CHF, ONS improves 30-day readmission, LOS, and episode cost outcomes. ONS use could provide a low-cost strategy for improving hospitalization outcomes for elderly patients with CHF and reducing burden on health systems from CHF. PCV174 Knowledge Transfer Gap Between Cardiologists and Patients Undergoing Percutaneous Coronary Intervention Regarding Risks Associated With Drug-Eluting Stents: An Asian & European Survey Yan B P 1, Ip A 1, Lee V W 2 Chinese University of Hong Kong, Shatin, Hong Kong, China, 2The Chinese University of Hong Kong, Hong Kong, Hong Kong . . . . . 1The Objectives: The choice of stent used in percutaneous coronary intervention (PCI) is often at the discretion of the interventional cardiologist (IC) without informed shared decision making. We aim to assess the impact of ICs’ awareness of the risk of delayed arterial healing associated with drug-eluting stents (DES) on patient knowledge transfer. Methods: 132 ICs from 11 countries (3 Asian: Malaysia, Hong Kong, Singapore and 8 European: Germany, Italy, UK, The Netherlands, Belgium, Denmark, Russia and Serbia) were invited to complete an online survey using a 4-point scale regarding their (i) familiarity with delayed arterial healing associated with DES; (ii) how concerned they are about delayed arterial healing; (iii) frequency they discuss this risk with their patients and (iv) frequency this risk influence the type of stent they use. Responses from Asian cardiologists were compared with Europeans. Results: 43.2% ICs were Asians and 56.8% were Europeans. Majority of IC were extremely/ very familiar with the risk of delayed arterial healing after DES implantation (63.2% Asian vs. 56.0% European, p=NS). IC who were extremely/very concerned about the risk of delayed healing were more likely to discuss with their patients (Odds Ratio (OR) 2.62, 95% confidence interval (CI) 1.17-5.85, p< 0.01) and influence their stent choice (OR 5.56,95%CI 2.56-12.05, p< 0.01). Although twice as many Asian compared to European ICs were extremely/very concerned about delayed arterial healing with DES (59.6% vs. 32.4%, respectively, p<0.01), there were no significant differences in the frequency this risk was discussed with patients (often/always: 24.6% Asian vs. 26.7% European, P= NS) or influence the type of stent used (often/always: 47.4% vs. 35.7%, P=NS). Conclusions: Many patients are not well informed of the risk associated with DES despite high level of physician awareness and concern of this risk. This knowledge transfer gap exists in both Asia and Europe. PCV175 Acute Ischemic Stroke (Ais) Patient Management in French Stroke Units and Impact Estimation of Thrombolysis On Care Pathways and Associated Costs Schmidt A 1, Bénard S 1, Heroum C 2, Caumette D 3, Delaitre O 3, Le Lay K 3 consultants, Oullins, France, 2CHRU Montpellier, Montpellier, France, 3Boehringer Ingelheim France, Paris, France . . . . . . were to home, 25% to rehabilitative care then home, 2% to rehabilitative care then a nursing home, 7% to long-term care and 6% of stays ended with a patient death. Of a total cost over 1 year of € 610 million (mean cost per patient of € 20,326), 70% concern the post-acute phase. By increasing the proportion of patients thrombolyzed, costs are reduced primarily by a decrease in rehabilitative care, with savings per additional treated patient of € 1,462. By adding improved timing, savings are more than doubled (€ 3,183 per additional patient). Conclusions: By improving thrombolytic management in stroke units, patient journeys through care pathways can be modified, with increased discharges home, a change in post-acute resource consumption and net savings. PCV176 Launching Novel Class Iii Implantable Cardiac Devices for Cardiology in Europe First, Is This Common Commercial Practice Improving Health Care Quality for Europeans Garfield S 1, Armstrong S 2 Market Access, Wayland, MA, USA, 2GfK, Wayland, MA, USA . . 1GfK Objectives: Regulatory hurdles for novel medical devices are lower in Europe than in the US. The costs and evidence requirements to achieve CE marking for class III devices are lower than to achieve FDA-approval via a PMA submission. As such, many companies have chosen to enter European markets with innovative cardiac devices before entering the US market. This study sought to understand the frequency with which that has occurred in the last 10 years, and the correlation between market access timing and cardiac outcomes. Methods: A review of CE mark and FDA approvals for class III implantable cardiac devices was conducted for the period of 2003-2013. Devices were identified and cross referenced to determine which products achieved both CE mark and FDA approval. Those with both, were then compared by the date of approval to determine market access variance in the US versus Germany. Publically available coverage and reimbursement policies were reviewed in each market in combination with relevant disease prevalence rates over the study period. Results: Implantable cardiac devices were routinely available in Germany before the US during the study period. Early use across Europe, in many cases, created additional evidence that was leveraged during the FDA approval and follow-on reimbursement assessment processes in the US. While HTAs for devices was relatively rare in Germany, similar rates of reimbursement and access were achieved in both markets. Rates of cardiovascular associated mortality have dropped dramatically, across both markets though rates initiated their decline earlier in Germany. Conclusions: Germans benefited from earlier access to many innovative implantable cardiac devices during the period of 2003 to 2013, as compared to their American counterparts. The link between early access, clinical outcomes, and cost needs to be further analyzed in future studies. PCV177 Recruiting Cardiologists and Chronic Heart Patients From A Managed Physician Panel To Support Clinical Studies Phase Iii/Iv Or Health Outcome Studies Eichmann F 1, Potthoff P 1, Brown C 2 Health Germany, Munich, Germany, 2All Global, London, UK . . . 1Kantar Objectives: To identify cardiologists willing to participate in Health outcome or Clinical Studies Phase III/IV in internal medicine and to include patients for outcome assessment. Real life and Peri/Post-Approval Studies are becoming more and more important to meet regulatory and market access needs. The objective of the present contribution is to assess the benefits of a managed panel of cardiologists for site and patient recruitment. Methods: In 2012, a representative survey among members of a managed physician panel of cardiologists in US, UK, GER, FR, IT and SP was conducted. 208 cardiologists reported on former experiences with clinical studies phase III/IV and other post-approval studies, their willingness to participate in future studies and their potential adherence to requirements of Good Clinical Practice rules. Results: 87,6% of the cardiologists reported to have formerly participated in clinical studies phase III/IV, and 54,3% in post-approval studies. Over 58% of these cardiologists were willing to participate in future studies (58,2% in clinical studies phase III/IV and 63,1% in other post-approval studies). More than 92,7% of this group was ready to be named as principal investigator to an ethical committee, to report serious events to the sponsor (97,6%), or to participate in a web-based training session (90,3%). Within one month cardiologists see in average 175 patients suffering from chronic heart disease in their practices (Minimum: 20; Maximum more than 500). 93,5% of the cardiologists are willing to ask their patients for informed consent for participation in studies, thereby providing a promising source for recruiting patients. Conclusions: Cardiologists from a managed panel are a time- and cost-effective option for recruiting sites and patients for observational post-approval outcome and safety studies in the big 5 EU countries and the US. Patient incidence estimates are a reliable source for enrollment planning. 1st[è]ve Objectives: This study aims to evaluate the current management and associated costs of acute ischemic stroke (AIS) for patients admitted in stroke units in France and over a 1 year follow-up period as well as to assess the impact of improved thrombolytic management in terms of increasing the proportion of patients receiving thrombolysis and/or treated within 3hrs from symptom onset on functional recovery and care pathways. Methods: A decision model was developed, comprising two components: the first corresponding to the acute hospital management phase of patients with AIS up until hospital discharge and the second corresponding to the post-acute phase. Patient journeys and costs were determined for both phases. Improved thrombolytic management was modeled by increasing the proportion of patients receiving thrombolysis from the current estimated level of 16.7 to 25% as well as subsequently increasing the proportion of patients treated within 3 hours of the onset of symptoms post-stroke from 50 to 100%. The impact on care pathways was derived from clinical data. Results: In 2011, 29,999 stays took place in a stroke unit for AIS in France. 60% of discharges PCV178 Health Care Stakeholders’ Evaluation of A User-Friendly Tool Which Estimates Long-Term Health Gains Following the Reduction of Ldl Levels Laires P Merck Sharp & Dohme, Oeiras, Portugal . Objectives: Demonstration of long-term value of preventive care in terms of health gains and avoided costs associated with disease progression may be of great value for those who need to prioritize health policies. This analysis aimed to evaluate physicians’ and primary care payers’ opinions about a user-friendly tool which estimates long-term health gains following LDL-C reduction. Methods: A user-friendly tool was developed based on a previously published Markov model employed to evaluate health outcomes, including cardiovascular (CV) events and due costs. The model incorporated Framingham risk equations, Portuguese population characteristics, national mortality rates and local costs. Software runs in iOS and the user may simulate for 3, 5 and 10 years the expected CV events drop following a given LDL-C A504 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 reduction. A predefined questionnaire was delivered to physicians’ and primary care payers’ (namely members of the regional health authorities) to survey their opinions about the value of this tool. Results: Overall opinion from 30 physicians and 11 primary care deciders (geographically distributed) was positive, averaging 3.9 in a likert scale from 1 (strongly disagree) to 5 (strongly agree). Physicians averaged 4 while primary care payers’ was 3.7. 71% of respondents ranked 4 to 5 in this overall assessment of the tool. Regarding more specific topics using the same likert scale, global, physicians’ and primary care deciders’ responses averaged as following, respectively: Utility of the tool: 3.6, 3.6 and 3.6; Relevance of the tool: 3.9, 3.9 and 3.9; Value of the tool to understand LDL-C treatment targets: 4.0, 4.2 and 3.6. Conclusions: This approach is useful to understand user’s opinions about a tool which aims primarily to raise awareness on the importance of the LDL-C reduction. Available evidence demonstrates that health care stakeholders in Portugal still need to understand the public health potential of LDL reduction. This tool might be of great value to address this need. PCV179 Treatment Patterns Among Heart Failure Patients Within 30 Days Post Diagnosis: Results From A Us Claims Database Analysis De Camargo Cancela M 1, Hudson E 1, Turner S J 2, Deschaseaux C 3 Ireland Ltd, Dublin, Ireland, 2Novartis Pharmaceuticals, East Hanover, NJ, USA, 3Novartis Pharma AG, Basel, Switzerland . . . . . 1Novartis Objectives: Clinical guidelines recommend ACEIs (angiotensin converting enzyme inhibitors), ARBs (angiotensin receptor II blockers) for patients intolerant to ACEI, beta blockers (BBs), aldosterone antagonists (AAs) and diuretics as the pharmacological treatment for heart failure (HF). This study assesses the treatments prescribed within 30 days post diagnosis among HF patients in a real world setting based on an administrative claims database in the US. Methods: This was a retrospective cohort study conducted using MarketScan database. Adult patients having ≥2 HF-related medical claims or 1 hospitalisation with primary HF diagnosis between April 2009 and March 2012, and with a minimum of 12 months pre- and post-index continuous medical and pharmacy eligibility were included. Index date was defined as the first HF-related medical claim between April 2009 and March 2012. Patients with HF diagnosis in the 12 months pre-index period were excluded. Demographics, clinical characteristics and index treatment (defined as a 30 days window period after HF diagnosis) were analysed. Results: Among 121,904 patients included in the analysis, 48.3% were >75 years of age and 35.0% were 18-64 years of age. Diabetes (27.0%), CV related conditions (26%) and COPD (16.3%) were the most prevalent comorbidities. Overall, 37.6% patients were not prescribed any treatment related to HF within the first 30 days after HF diagnosis (no HF severity available). Among those prescribed treatment, the following treatment breakdown was observed: ACEIs, 29.3% of patients; ARBs, 8.5%; BBs, 46.6%; AAs, 7.4%; and diuretics, 45.7%. Prescription of fixed dose combination of ACEIs with diuretics or ARBs with diuretics was very limited. Conclusions: Findings suggest that a substantial proportion of patients do not receive HF specific medications following the month of diagnosis. Further research is necessary to explore the reasons for lower than expected HF medication claims in these patients considering their high risk for morbidity and mortality. PCV180 Pattern of Benzodiazepines Utilization In Outpatients With Hypertension in Serbia Tomas A 1, Horvat O 1, Tomic Z 1, Ban M 1, Sabo A 2 1Faculty of Medicine, University of Novi Sad, Novi Sad, Serbia and Montenegro,, 2Faculty of Medicine, University of Belgrade, Belgrade, Serbia and Montenegro . . . . . Objectives: Benzodiazepines (BZD) are often administered to patients with arterial hypertension in addition to antihypertensive agents. The aim of this study was to analyze patterns of BSD usage among hypertensive outpatients in Serbia according to sex and age. Methods: Data on BZD issued on prescription for the International classification of diseases (ICD) code I10 (essential arterial hypertension) were collected from all state-owned pharmacies in Novi Sad (population 350,000) from September 2011 to February 2012. Consumption was calculated using the ATC/ DDD methodology. Results: BZD accounted for 2% of all drugs issued in treatment of hypertension. The total amount of BZD issued was 2.27 DDD/1000inh/day. Bromazepam (51.38 %) and diazepam (37.56 %) were the most commonly used BSD and accounted for almost 90% of all BSD utilized. Lorazepam accounted for 8.36% of total consumption. Other agents (alprazolam, nitrazepam, midazolam, klonazepam) accounted for less than 3% altogether. The BZD consumption increased with patient’s age – 91% being prescribed to the patients over the age of 50. Consumption was highest in age group 70-80 years. BZD were more often prescribed to female (68.63%) than to male outpatients (31.37%). Conclusions: This study confirmed consumption of BZD among outpatients with hypertension. It is most common in hypertensive female and elderly population. However, since the benefits of BZD administration in hypertension control remain unclear, the adequacy of this practice is questionable. This work was supported by the Ministry of Science and Technological Development, Republic of Serbia, project No. 41012. PCV181 Logistics of Monitoring of Vitamin K Antagonists in Western European Countries: Prefer in Vte Registry Monreal M 1, Bauersachs R 2, Gitt A K 3, Laeis P 4, Mismetti P 5, Willich S N 6, Cohen A T 7, Agnelli G 8 1Hospital Universitari Germans Trias I Pujol, Barcelona, Spain, 2Max-Ratschow-Klinik für Angiologie, Gefäßzentrum Klinikum Darmstadt GmbH, Darmstadt, Germany, 3Herzzentrum Ludwigshafen, Ludwigshafen, Germany, 4Daiichi Sankyo Europe GmbH, Munich, Germany, 5Centre Hospitalier Universitaire Saint-Etienne, Hopital Nord, Saint Etienne, France, 6Charité Universitätsmedizin Berlin, Berlin, Germany, 7King’s College, London, UK, 8University of Perugia, Santa Maria della Misericordia Hospital, Perugia, Italy . . . . . . . . . . . Objectives: In patients with acute deep venous thrombosis (DVT) or pulmonary embolism (PE) there is a lack of comparative data on the logistics of international normalized ratio (INR) monitoring in patients on vitamin K antagonists (VKA) on maintenance anticoagulation. Methods: This interim analysis is based on data of 2863 patients enrolled in the PREFER in VTE registry in 7 Western European countries (France, Germany, Italy, Spain, UK, Austria and Switzerland). 1689 had DVT (only) and 1174 had PE (±DVT). Results: At 6-month follow-up, 51.5% of the patients were treated with VKA. INR measurements at 6-month follow-up (UK: 3 month) were performed in the hospital setting for 13.1% of these patients (most frequently in Spain 42.9%, UK 32.7%,, Italy 18.0%), in the anticoagulation center in 23.6% (in Spain 40.0%, Italy 37.0%, UK 16.4%), in the physician’s office in 16.8% (Germany 90.7%, UK 41.8%), by the patient self-measured in 4.9% (Spain 8.6%), or at biology labs or other institutions in 36.9% (France 92.4%). The patient’s mean travel distance to the INR site was 5.8 ±9.16 km overall, the range was 0-90 km (mean varied across countries from 2.4 to 7.7 km). Mean travel time was 14.3 ±18.4 min (across countries 5.7 to 18.2 min). Patients most frequently used their private car/motorbike (50.8 %) or walked (16.4%). The mean number of INR measurements over the 6-month period was 16.8, the number of INR measurements per month was 2.8 ±1.2 overall (across countries 1.8 to 3.2). Conclusions: In the various countries, different institutions are responsible for routine INR measurements. While biology labs are almost exclusively used in France, there is no equivalent for such institutions in other countries. INR self-measurement plays a minor role. Patients usually have the INR sites in their vicinity, and the average number of measurements shows little variation between countries. PCV182 Treatment Patterns and Health Resource Utilization Among Atrial Fibrillation Patients in United Arab Emirates and Saudi Arabia Johnston K 1, Osenenko K M 1, Donato B M K 2, Qatami L 3, Alawi A A 4, Binbrek A S 5, Hersi A S 6, Mould J F 7, Levy A R 1 1ICON Epidemiology, Vancouver, BC, Canada, 2Bristol-Myers Squibb Company, Wallingford, CT, USA, 3Bristol-Myers Squibb Company, Dubai, United Arab Emirates, 4Sheikh Khalifa Medical City, Abu Dhabi, United Arab Emirates, 5Rashid Hospital, Dubai, United Arab Emirates, 6King Saud University Medical City, Riyadh, Saudi Arabia, 7Pfizer, New York, NY, USA . . . . . . . . . . . . . . . . . Objectives: Atrial fibrillation (AF) is the most common cardiac arrhythmia, and accounts for one-third of hospitalizations for cardiac disturbances. The majority of descriptive data on management of AF patients are from western countries, with limited information available from the Middle East region. The objective of this study was to characterize treatment patterns and health resource utilization among AF patients in the Kingdom of Saudi Arabia (KSA) and United Arab Emirates (UAE). Methods: A retrospective chart review was undertaken at three hospitals in UAE and three in KSA, to identify AF patients diagnosed between January 2005 and June 2010. Patient charts were sampled consecutively backwards by diagnosis date, from June 2010 until the target sample was reached. AF was identified based on ICD-9 code (427.31), from a sample of patients defined by any history of anticoagulant use. Data on demographic and disease-related characteristics, treatment patterns, health resource utilization, and international normalized ratio (INR) control were abstracted from diagnosis until June 2012. Results: Among eligible AF patients (UAE, n= 157, KSA, n= 152), the majority were diagnosed with chronic AF (80.9% in UAE, 63.7% in KSA) as opposed to paroxysmal AF. Treatments prescribed to AF patients differed between countries: warfarin monotherapy was widely used in UAE (59.9%), while a variety of warfarin- and aspirin-based combination therapies were used in KSA, with no single dominant regimen. Warfarin + bisoprolol (12.5%) and aspirin + bisoprolol (10.5%) combination therapies were the most common regimens in KSA. Patterns of health care utilization also varied, with hospitalization and emergency room visits more common in KSA, and outpatient visits more common in UAE. Conclusions: Treatments and health resources used by AF patients varied between KSA and UAE. While some differences may result from differences in patient and disease characteristics, they likely also reflect variation in management strategies across the regions. PCV183 Ldl-C Lowering Efficacy of Evolocumab (Amg 145) Could Reduce Apheresis in Patients At High Risk for Cardiovascular Events in Germany Villa G 1, Schmid T 2, Lothgren M 1, Michailov G 2 (Europe) GmbH, Zug, Switzerland, 2Amgen GmbH, München, Germany . . . . 1Amgen Objectives: Individuals at high risk for cardiovascular events who fail to achieve treatment goal (LDL-C<100 mg/dL) despite being on maximal lipid lowering treatment qualify for apheresis, a procedure that costs approximately 40,000€ per patient per year in Germany. The current model aims to assess the ability of evolocumab (AMG 145), an investigational medication that is being evaluated in clinical trials, to reduce the proportion of patients requiring apheresis in Germany. Methods: Data on secondary prevention patients eligible for apheresis, excluding homozygous familial hypercholesterolemia patients, were extracted from the German IMS Disease Analyzer 2011-2013 database (n= 8,262) and included in the analysis. The calculated mean LDL-C reductions observed in the DESCARTES and LAPLACE-2 evolocumab trials, ranging from 59.3% (95% CI [54.9%, 63.8%]) to 72.3% (95% CI [69.1%, 75.4%]), were applied to baseline LDL-C levels of the identified patient-profiles, following a probabilistic approach. The goal for such patients was LDL-C< 100 mg/dL, as defined in the German lipid association (DGFF) guideline. Results: The mean ± standard deviation LDL-C levels of the sample decreased from 150.2 ± 32.9 mg/dL at baseline to 61.1 ± 13.8 mg/dL (DESCARTES) and to 41.7 ± 9.5 mg/dL (LAPLACE-2) after evolocumab treatment. From an initial proportion of 100% of patients eligible for apheresis at baseline, evolocumab treatment led to a proportion of 1% (DESCARTES) and 0% (LAPLACE-2) of patients requiring apheresis. Thus, in the analyzed population, apheresis-related costs could be largely reduced. Conclusions: The use of evolocumab in the treatment algorithm of high-risk patients not at LDL-C goal could allow reducing the invasive, time-consuming, burdensome and costly weekly apheresis treatments. As a result, significant savings of apheresis-related costs could be achieved by the German Health Care System. A505 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 INDIVIDUAL’S HEALTH – Clinical Outcomes Studies PIH1 Predictive Validity of Inappropriate Prescribing Criteria for Adverse Drug Events, Hospitalizations, and Emergency Department Visits: A Time-To-Event Comparison of the Beers and Stopp Criteria Brown J , Li C , Painter J , Hutchison L C , Martin B University of Arkansas for Medical Sciences, Little Rock, AR, USA . . . . . . Objectives: Adverse drug event (ADEs), hospitalizations, and emergency department (ED) visits are important sequelae of inappropriate prescribing. Explicit measures are useful methods to detect inappropriate prescribing in the elderly. This study compares the predictive validity of the 2002 Beers, 2012 Beers, and the STOPP criteria for these outcomes. Methods: A retrospective cohort design was implemented using commercial claims data between 2006 and 2009. Subjects included those 65 years and older continuously eligible for medical and pharmacy benefits for at least 6 months to assess baseline comorbidities and 3 months of follow-up. Time varying cox proportional hazard models were estimated using a monthly time interval where indicators of inappropriate prescribing exposure in month (i-1) where related to outcomes in month (i) to minimize temporal ambiguity between exposure and outcome. Measures of model discrimination (c-index) and hazard ratios (HR) were calculated to compare un-adjusted and adjusted models with the 2002 Beers, 2012 Beers, and STOPP exposures models. Results: The final cohort included 174,275 contributing 361,621 person years of follow up. The prevalence of inappropriate prescribing was 34.1%, 32.2%, and 27.6% for the 2012 Beers, 2002 Beers, and the STOPP criteria and the rates of ADEs, ED, and hospital visits were 21.0, 140.3, 67.2 per 1000 person years. All inappropriate prescribing criteria modestly discriminated ADEs in unadjusted analyses: STOPP (HR= 2.89 [2.68-3.12]; C-index= 60.7%), 2012 Beers (HR=2.51 [2.33-2.70]; C-index=60.3%), 2002 Beers (HR=2.65 [2.46-2.85]; C-index=60.5%). Similar results for model discrimination measures were observed for ED visits (STOPP [C-index= 59.0%], 2012 Beers [C-index= 58.5%], 2002 Beers [C-index= 58.5%]) and hospitalizations (STOPP [C-index=59.8%], 2012 Beers [C-index=59.0%], 2002 Beers [C-index=58.8%]). Conclusions: All three of the criteria were modestly prognostic for ADEs, EDs, and hospitalizations with the STOPP criteria slightly outperforming the Beers criteria. Inappropriate prescribing is common in the elderly and significantly increases the risk of subsequent hospitalizations, ED visits, and ADEs. PIH2 Prevalence of Chronic Diseases Among Older Patients (> 65 Years) in German General Practitioner Practices Kostev K 1, Rex J 1, Engelhard J 1, Altmann V 1, Jockwig A 2 1IMS Health, Frankfurt am Main, Germany, 2Fresenius University of Applied Sciences, Idstein, Germany . . . . . Objectives: In Germany, the share of persons aged >65 years is 21%. Germany therefore has the highest share of older inhabitants compared to other european countries. The aim of the present study was to evaluate the prevalence of defined chronic diseases among older patients (> 65 years) in German general practitioner (GP) practices. Methods: Data from Disease Analyzer database including 1,100 GPs were used. 2,722,706 patients with at least one consultation in the time between April 2013 and March 2014 were included and analyzed. Results: Overall, 774,361 (28.4%) of patients by GPs were over 65 years. In patients aged >65 years 66% were diagnosed with hypertension, 41% with lipid metabolism, 33% with diabetes mellitus, 29% with coronary heart disease, 20% with rheumatoid arthritis, 19% with cancer, 15% with chronic obstructive pulmonary disease and 14% with heart insufficiency. Conclusions: Patients older than 65 years make up 28% and therefore comprise the main client group of GPs. The prevalence of chronic diseases is in this age very high. On account of their multimorbidity, the care of these patients is challenging. It is therefore important to place particular emphasis on geriatric patient care as a part of the study of medicine as well as in physicians assistant degree programs. PIH3 Current Situation of Paediatric and Adult Patients With Fragile X Syndrome: Preliminary Data From the Explain Fxs Registry Haessler F 1, Gaese F 2, Pittrow D 3, Huss M 4, Peters H 4, Kretschmar C 5, Brinkman M 6, Elstner S 7, Colla M 8 1Klinik für Psychiatrie, Neurologie, Psychosomatik und Psychotherapie im Kindes- und Jugendalter, Universität Rostock, Rostock, Germany, 2Isar-Amper-Klinikumr gGmbH, Klinikum München-Ost, Haar, Germany, 3Technical University Carl Gustav Carus, Dresden, Germany, 4RheinhessenFachklinik Mainz, Mainz, Germany, 5Städt. Krankenhaus Dresden-Neustadt, Dresden, Germany, 6Novartis Pharma GmbH, Nuremberg, Germany, 7Evangelisches Krankenhaus Königin Elisabeth Herzberge gGmbH, Berlin, Germany, 8Charité/ ECRC, Berlin, Germany . . . . . . . . . Objectives: The Fragile X syndrome (FXS), caused by a CGG repeat expansion in the FMR1 gene on the X chromosome, is the most common inherited form of mental retardation and autism. Representative data on the characteristics and management of FXS patients in Germany are lacking. Methods: EXPLAIN FXS is an ongoing prospective longitudinal observational study which evaluates characteristics and management of patients with a genetically confirmed diagnosis of FXS in ambulatory care. The registry uses a non-probability sampling approach to collect data on the characteristics, therapeutic drug and non-drug interventions, psychosocial parameters, quality of life, caregiver burden, and health economic parameters such as hospitalisation days. Results: After an average 9-month recruitment period in 10 centres, 33 patients have been included (86% males, mean age 22 years, age range 2- 82 years). Only 1 and 4 patient attended regular school, or regular school with integration measures, whereas the others were in special schools for the mentally impaired. Siblings were affected by FXS in 59% (10% unknown). On the 6-item FXS disease score by Giangreco (on 25 patients), the mean value was 6.7 ±2.4 points (range 2.4 to 11.0). Autistic behaviour was noted in 89%, hyperactivity in 50%, impaired intelligence in 96%. An FXS phenotype was often met, as 63% of patients were described to have a long, narrow face, and 89% large or prominent ears. Comorbidities were frequent, in particular anxiety disorders in 65%, ADHD in 61%, impaired social behaviour in 85%, depression in 23%, and seizures in 4%. Medical therapy for comorbidities was often prescribed. Conclusions: The registry is expected to provide much-needed data on the characteristics and management situation of FXS patients in Germany, to allow comparisons with other countries, and to enable gap analyses based on current guidelines for management of these patients. ClinTrials. gov identifier is NCT01711606. PIH4 Feasibility of A Network Meta-Analysis in Endometriosis van Nooten F E 1, Novak A 2, Langham J 3 1Astellas, Leiden, The Netherlands, 2Anovak-Services, Apeldoorn, The Netherlands, 3PHMR Associates, London, UK . . . . Objectives: To conduct a systematic review of randomised controlled trials comparing effectiveness of GnRHa in the treatment of endometriosis in order to assess the feasibility of conducting a meta-analysis and indirect treatment comparison. Methods: Inclusion criteria were all published randomized controlled trials comparing GnRHa versus any other pharmacological treatment or placebo in the treatment of laparoscopically confirmed moderate to severe endometriosis. Electronic databases were searched up to February 2014 (Embase, Medline, Cochrane, Clintrials. gov and clinical conferences). Search results were screened, eligible studies were assessed for risk of bias and data extracted. Studies were evaluated for the feasibility of inclusion into a network meta-analysis. Results: A total of 1,670 titles and abstracts were screened. Sixty-nine papers relating to 62 studies were identified as eligible for inclusion. Fourteen of the 62 were excluded due to high risk of bias, a further 21 did not present sufficient data. The remaining 27 studies showed clinical heterogeneity in the measurement of disease severity and efficacy outcomes and quality of reporting. Efficacy outcomes were measured using various methods and at different time points (3, 6 to 12 months). All but one measured pain, 12 using modified versions of the Biberoglu and Behrman (B&B), 7 using a similar method and 4 used a visual analogue scale. Quality of life was measured in 7 studies, most commonly by a generic HRQoL instrument, the SF-36. Adverse event recording varied. Conclusions: A core set of studies were identified that could be incorporated into a quantitative analysis assessing effectiveness of GnRHa for treatment of moderate to severe endometriosis over 6 months. A network meta-analysis would potentially be feasible using more commonly reported pain outcomes, (modified) B&B; HRQoL and reported adverse events. PIH5 The Efficacy of Oximes in Acute Human Organophosphorus Poisoning; An Updated Meta-Analysis Abdollahi M 1, Nikfar S 1, Mirfazaelian H 2 University of Medical Sciences, Tehran, Iran, 2Tehran University of Medical Sviences, Tehran, Iran . . . 1Tehran Objectives: The present study iims to meta-analyze clinical trials evaluating the efficacy of oximes on organophosphorus (OP) poisoning treatment. Methods: PubMed, Scopus, Google Scholar, and clinicaltials. gov were searched for studies investigating the effects of oximes in the treatment of OP poisoning. Mortality, intermediate syndrome, intensive care unit (ICU) admission rate, hospital stay duration, and intubation rate were the key outcomes of interest. Data were searched in the time period of 1966 through December 2013. Results: Thirteen studies (eleven clinical trials and two historical cohorts) that met our criteria were included in the analysis. Pooling of data showed that relative risk (RR) of need for intubation in OP poisoning for eight included trials comparing oximes with placebo was 1.27 with 95% CI= 0.73 to 2.23 (P= 0.4). RR of only one observational study was 1.57 (95% CI= 0.79 to 3.2, P> 0.05). The summary of RR for mortality rate in 9 studies was 0.38 (95% CI= 0.65 to 2.97, P= 0.41) and for one observational study was 1.33 (95% CI= 0.54 to 3.29, P> 0.05). The RR for ICU admission rate in OP poisoning for three trials comparing oximes to placebo was2.12 with 95% CI= 0.89 to 5.03 (P= 0.09). For only one observational study, RR was 0.81 (95% CI= 0.49 to 1.25, P> 0.05). For intermediate syndrome, while the RR of only trial comparing oximes with placebo was 1.89 (95% CI= 1.27 to 2.91, P< 0.05), for only one observational study, it was 1.43 (95% CI= 0.7 to 2.96, P> 0.05). For hospital stay duration, the RR of four studies was 0.75 with 95% CI = -0.51 to 1.99. Conclusions: According to these data, oximes beneficence in OP poisoning is unclear and there is possible increase in incidence of intermediate syndrome. PIH6 labour induction with prostaglandins: what works best? A systematic review, network meta-analysis and cost-effectiveness analysis Keeney E 1, Alfirevic Z 2, Caldwell D M 1, Dowswell T 2, Dias S 1, Jones L 2, Navaratnam K 2, Welton N J 1 1University of Bristol, Bristol, UK, 2University of Liverpool, Liverpool, UK . . . . . . . . . . Objectives: The number of women whose labours are induced is increasing with average rates in England in excess of 20%. A broad range of methods for induction are available, including pharmacological, mechanical and alternative interventions with considerable variation in effectiveness, safety and cost. Of the pharmacological interventions, the most commonly used are prostaglandins. However, even within this class, there are a variety of different prostaglandins with a variety of modes of administration. The aim of this piece of work is to identify the most costeffective and safe prostaglandin for cervical ripening and labour induction in the NHS setting. Methods: A systematic review and network meta-analysis (NMA) of all relevant randomised trials comparing all prostaglandins for labour induction, and a cost-effectiveness analysis represented in the form of a decision tree model incorporating cost and maternal and neonatal health outcomes. Results: The NMA of randomised controlled trials comparing 11 different types of prostaglandins for labour induction demonstrated that the best treatment for achieving a vaginal birth within 24 hours was vaginal misoprostol tablet ≥ 50μ g with a probability of being A506 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 best of 53%. The safest treatment in terms of caesarean section risk was titrated (low dose) oral misoprostol solution with a probability of being the best of 63%. Giving equal weight to both outcomes, titrated low dose oral misoprostol solution had the best overall outcomes (average rank of 1.5) followed by vaginal misoprostol ≥ 50μ g (average rank of 1.6). Cost and utility data are now being gathered to inform the cost-effectiveness analysis. Conclusions: The NMA and cost-effectiveness analysis will be of value to clinicians, pregnant women, guideline developers and policy makers within the NHS. We will discuss how the model can be extended to include other types of intervention for induction of labour. PIH7 Examination of the Efficiency of Electrical Stimulation in Case of Stress and Urge Incontinence Oláh A 1, Turnerné Hilmer H 1, Müller Á 1, Knisz J 1, Fullér N 1, Boncz I 2, Pakai A 3, Stromájer-Rácz T 1 1University of Pécs, Pécs, Hungary, 2Faculty of Health Sciences, University of Pécs, Pécs, Hungary, 3University of Pécs, Zalaegerszeg, Hungary . . . . . . care (15.6%), others (12.3%), indigestion (2.9%), energy drinks (2.7%), constipation (1.9%), antihistamine (0.4%), skin ointments for allergy (0.4%), and skin ointments except for allergy (0.2%), respectively. Only 5 people (1%) reported side effects after taking OTC drugs and dietary supplements. The most common reason for buying OTC medicines and dietary supplements was to choose products they prefer (33.8%); some participants had taken OTC drugs and dietary supplements for improving overall health status (19.8%). Also, the use of OTC drugs and dietary supplements was associated with chronic disease (p=0.034), education level (p<0.001), and household income (p< 0.001), Conclusions: The results demonstrate the prevalence of OTC use in the elders for more than 3 months is unsubstantial and multivitamins and dietary supplements are commonly taken by the elderly. In addition, the elders with chronic diseases, higher income, and/or lower education level are inclined to take OTC medicines. . . Objectives: Electrical stimulation is one kind of conservative treatment in urinary incontinence. Electrical stimulation can trigger the contraction of pelvic floor muscles, the functioning of these muscles can be improved, and the strength of muscle contraction can be increased as well. The aim of this research is to examine the efficiency of electrical stimulation treatment in case of stress and urge incontinence. Methods: In the course of the examination non-random, purposive sampling was applied. Self-made questionnaires were given to the participants before and after treatment. Incontinence Impact Questionnaire (IIQ-7) and Urogenital Distress Inventory (UDI-6) international valid questionnaires were applied to measure life quality. The target group was woman clients between the age of 30-65, and above 65 years. The examination took place at Kaposi Mór Teaching Hospital, Somogy County, from 15 February, 2013 to 15 February, 2014. The analysis of results was performed with MS Excel 2007 program. Results: The research was conducted with 100 participants. In the course of the examination 45% of the participants reported improvement after treatment regarding stress incontinence while in case of urge incontinence only 15% experienced positive change. Due to the treatment the number of the regular liner users was decreased. The number of users was 75 (75%) before treatments, which reduced to 44 (44%) after treatment. Electrical stimulation proved to be more efficient in case of stress incontinence than in urge incontinence. Conclusions: Follow-up of the participants for several months would give information about the long-term benefits of electrical stimulation. In the future the improvement of the knowledge of patients on incontinence including the methods of conservative treatment is important. PIH11 Identifying Potentially Inappropriate Medication (Pim) and Major Risk Factors for Hospitalization for Elderly Patients Admitting To Teaching Hospital: Study From Indian Perspective Shah C S SJM college of pharmacy, chitradurga, India . . Objectives: Inappropriate polypharmacy is a particular concern in older people and is associated with negative health outcomes. Objective of this study was to identify potentially inappropriate medication (PIM) and major risk factors for hospitalization in elderly patients. Methods: It was an observational prospective study in semi government teaching hospital situated in south India (Karnataka). This study was performed for 3 month and data were collected from elderly patients having age above 60 year who were admitted to medical ward. Results: Current analysis involving data of 150 (64 % male) elderly in patients with mean (SD) age of 67.9 (8.13) years. Majority of patients (30 %) were diagnosed with asthma followed by cancer (16 %). Mean (SD) hospital stay was 15 (5.23) days for cancer followed by 3 (0.77) days for asthma patients. Total 24 prescriptions were found with PIM as per beer’s criteria 2012. Out of which 50 % of the inappropriate medication found exclusively in asthma patients. Insulin was reported to be the single most PIM in 25 % of the prescriptions. Conclusions: Asthma was found to be the single most common disease leads to hospitalization among elder patients. Cancer was found to be the single common disease which increases hospital stay. Group of patients with asthma were receiving highest numbers of inappropriate medication compare to other patients. INDIVIDUAL’S HEALTH – Cost Studies PIH8 Effects of A Multidisciplinary Home-Based Medication Review Program On Hospital Admissions In Older Adult Singaporeans PIH13 Potential Efficency and Cost Savings To the English Nhs By A Laparoscopic Aproach To Total Abdominal Hysterectomy Cheen H H M , Ong K Y , Lim S H , Chng S G J , Chen L L , Lim P S , Ng S M , Chang W T Singapore General Hospital, Singapore Farrugia M East Kent University Hospitals NHS Foundation Trust, Margate, England . . . . . . . . . . . . . . . . . . Objectives: The study aimed to (i) evaluate the effectiveness of a multidisciplinary home-based medication review (HBMR) program in reducing hospital admissions and (ii) determine the prevalence of drug-related problems (DRP) in older adult Singaporeans. Methods: A retrospective observational study was conducted at an academic medical centre in Singapore. Patients referred to the HBMR program between March 2011 and December 2012 were included. Home visits were conducted by a team comprising a pharmacist and care coordinators. Frequency of hospital admissions was determined 6 months before and after HBMR. DRP identified were categorised and their outcomes (resolved/unresolved) recorded by the pharmacist. Summary statistics were used to report patient characteristics and prevalence of DRP. Incidence of hospital admissions was compared using Poisson regression, while paired t-test was used to compare associated costs (in Singapore dollars) and length of stay (LOS) before and after HBMR. Results: The analysis included 107 patients with mean (SD) age of 75.6 (7.6) years. There were 52 (48.6%) males, and 89 (83.2%) were Chinese. The implementation of the HBMR program resulted in a 41% reduction in risk for hospital admissions (IRR: 0.59; 95% CI: 0.47 – 0.73, p< 0.001). Mean (SD) costs of hospital admissions reduced from $16,957.77 ($16,118.35) before HBMR to $7,488.76 ($12,773.40) after HBMR (p< 0.001). In a subgroup analysis of 62 patients who had hospital admissions before and after HBMR, mean (SD) LOS decreased from 26.5 (22.4) days to 17.6 (17.8) days (p= 0.010). The team identified 525 DRP from 1,353 medications reviewed. Of these, 34 (6.7%) and 174 (34.1%) DRPs were resolved with and without physician involvement respectively. The most common DRP identified were failure to receive drug (n= 163, 31.0%) and untreated indication (n= 140,26.7%). Conclusions: The multidisciplinary HBMR program was effective in reducing hospital admissions in older adult Singaporeans with multiple DRP. PIH10 Over the Counter Medication and Dietary Supplements Use Among Older Adults Lee H 1, Choi S E 1, Lim J 2, Kang D 1, Lim E A 1, Cheong C 3 1Korea University, Yeongi-gun, South Korea, 2Korea University, Sejong, South Korea, 3Seoul National University, Seoul, South Korea . . . . . . . . Objectives: The aim of this study was to examine the prevalence and patterns of over-the-counter (OTC) medicine and dietary supplements use among older adults in South Korea for more than 3 months period. Methods: The data were drawn from the 2009 Korea Health Panel database by selecting 2,910 Koreans aged 65 years or older. Data on self-reported OTC drugs and dietary supplements use longer than 3 months and demographic characteristics were collected from participants using a questionnaire and multiple logistic analysis was performed to estimate the factors associated with OTC and dietary supplements use. Results: 424 of 2,910 (14.6%) older adults took at least one OTC medication or dietary supplement. Most frequently used categories of OTC drugs and dietary supplements were vitamins and dietary supplements (46.9%) followed by analgesics (16.7%), gum . Objectives: Hysterectomy is the most common procedure in gynaecology and carries a very high patient satisfaction rating by patients. As a result, a number of economic benefits arise, including reduction in average length of stay (LOS), as well as readmission and surgical site infection rates using a minimally invasive route. The cost savings and efficiency gains accrued to the English NHS are presented in the form of a budget impact model. Methods: A budget impact model (BIM) was developed in order to compare the costs of two treatment strategies – total abdominal hysterectomy (TAH) and total laparoscopic hysterectomy (TLH). Micro-costing data was obtained from EKHUFT for a subset of TAH and TLH procedures. The latter was extrapolated at the National level using OPCS-level Hospital Episode Statistic (HES) in order to calculate cost and efficiency savings vis-à-vis average LOS and readmissions. This calculation relies on a conservative increase in laparoscopic adoption from 22 to 30%. Results: In 2013, 25,548 TAH procedures were carried out compared to 5,611 TLH. The LOS was 3.6 days for TAH compared to 1.3 days for TLH. The readmission rates are significantly higher for TAH at 1.64/1000 procedures compared to 0.18/1000 for TLH. The estimated savings on increasing the TLH rate nationally by 8% will free 4000 patient days. On taking into account the lower readmission rate, lower LOS and potential income from additional procedures carried out from additional bed availability, the micro-costing of the procedure shows an overall financial benefit to the NHS. Conclusions: A increase in the laparoscopic approach to hysterectomy by 8% will result is significant economic benefits for the English health service. This study confirmed the reduced LOS, with a lower readmission rate following the laparoscopic approach. The budget impact of the calculated increase in costs of TLH can be absorbed within tariff. PIH14 Economic Assessment of Preeclampsia: Screening, Diagnosis, Treatment Options, and Long Term Outcomes, A Systematic Review Zakiyah N 1, van Asselt A 1, Baker P 2, Postma M J 1 1University of Groningen, Groningen, The Netherlands, 2Gravida: National Centre for Growth and Development, The University of Auckland, Auckland, New Zealand . . . . . Objectives: To provide a comprehensive overview of the existing evidence on the health economics of screening, diagnosis, and treatment options in preeclampsia. Methods: A systematic literature search was undertaken using three electronic databases (MEDLINE, Embase, Cochrane) to identify all English language full papers published between 1994 and 2014 containing economic evaluation or budget impact analysis (BIA) studies concerning the screening, diagnosis, treatment and prevention of preeclampsia. In addition, the included studies were appraised by means of the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement and the guidelines on reporting format for budget impact analysis (ISPOR Principles of Good practice for Budget Impact Analysis). Results: From an initial total of 119 citations, six papers fulfilled the eligibility criteria, five of which were economic evaluations and one was a BIA. All studies were published A507 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 between 2001 and 2012. One study was on early screening using maternal markers, and concluded that this may be cost-effective from a payer perspective. Three other studies focused on treatment and management, and concluded that induction of delivery is the most cost-effective option in term preeclampsia. Also, magnesium sulphate was found to be a cost-effective treatment, but more so in low income countries. One study was a comprehensive HTA report on combination strategies of screening and prevention, which found that providing calcium to all pregnant women without prior testing was the most cost-effective strategy. The BIA, on introducing a novel diagnostic test, concluded that the savings associated with this test would offset its initial cost. Conclusions: Novel biomarkers in screening for and diagnosing preeclampsia show promise as a cost-effective approach, from a payer perspective. Moreover, magnesium sulphate seemed to be a cost-effective option to manage seizures in preeclamptic women while delivery remains the ultimate efficient treatment, especially for women with term preeclampsia. PIH15 Price Variation in Obstetrical Services in A Rural State Erten M Z , Phillips J , Jones C , Kappel S , Meyer M University of Vermont - College of Medicine, Burlington, VT, USA . . . . . . Objectives: To determine and quantify the level of price variation in inpatient care for obstetrical services across hospitals in the State of Vermont, USA. Methods: We used data from Vermont Healthcare Claims Uniform Reporting and Evaluation System (VHCURES) for the calendar year 2012. We generated single-line summaries for all inpatient obstetrical services with an admission date in 2012. We excluded records for Medicaid, secondary payments either to Medicare or another commercial insurance, denied claims, non-VT resident claims and adjustments to existing claims. This produced a total of 4,019 records for the evaluated services from 13 Vermont hospitals. We calculated the allowed amount for each Diagnosis Related Group (DRG) (total amount a hospital received from a payer, including any prepaid amounts related to the service, plus the amount due from a patient through copayment, co-insurance and deductible). The outcome of interest is the average price of a specific DRG at each hospital as a percent of group average price in Vermont Hospitals. To reflect patient severity we adjusted prices by case-mix. Results were generated for hospitals only when there were at least 10 discharges in a specific hospital. Results: The average prices for a vaginal delivery with complicating diagnoses varied between 66% to 156% of the group average, vaginal delivery without complicating diagnoses varied between 74% to 182%, neonate with other significant problems varied between 60% to 206% and normal newborn varied between 77% to 180%. Conclusions: There is wide variation in payments for obstetrical services across Vermont hospitals. Further analysis is necessary to understand the role of various factors that contribute to this variation. During this transition period to a single-payer system in the rural state of Vermont, policy makers will need evidencebased information to ensure undiminished access to care in obstetrical services. PIH16 Malnutrition in Institutionalized and Community-Dwelling Older Adults in Spain: Estimates of Its Costs To the National Health System Rodríguez-Mañas L 1, Abizanda P 2, Barcons N 3, Lizán L 4 1Getafe University Hospital, Madrid, Spain, 2Complejo Hospitalario Universitario de Albacete, Albacete, Spain, 3Nestlé Health Science, Barcelona, Spain, 4Outcomes 10, Castellon, Spain . . . . Objectives: To estimate the economic impact of malnutrition on annual direct cost in Spanish institutionalized or community-dwelling older adults. Methods: A systematic review was carried out on the economic burden and use of medical resources associated with malnutrition in institutionalized or communitydwelling older adults. National and international databases until December 2013 were searched. Based on results from the literature review, an Excel-based tool was developed to estimate the annual cost difference between malnourished and well-nourished institutionalized and community-dwelling older adults in Spain. Malnutrition prevalence, annual frequency of general practitioner (GP) visits and hospitalizations were the main inputs considered in the model. Unit costs were derived from health care cost databases available in Spain (Euros, 2014). One-way sensitivity analysis (OWSA) was performed. Results: The results of the systematic review showed that malnutrition implies higher medical costs in the study population, particularly due to the length of hospital stay, number of hospital admissions and GP visits. Economic results showed that the annual cost associated with the use of resources of a malnourished patient (€ 5,000.66) was 3.5 times higher than that of a well-nourished (€ 1,433.78). OWSA showed that prevalence of malnutrition was the variable with the greatest impact on results. Decreasing the prevalence of malnutrition according to minimum values identified in the literature (from 15.6% and 4.4%, to 2% and 3% in institutionalized and community-dwelling patients respectively), caused the annual cost per patient with malnutrition to drop from € 5,000.66 to € 2,511.83, while increasing it to the maximum values (62% and 23%, respectively), implied a rise from € 5,000.66 to € 24,291.93 in the annual cost per patient. Conclusions: Adopting measures to reduce the prevalence of malnutrition in institutionalized or community-dwelling older adults would reduce by up to 50% the annual cost per patient associated to GP visits and hospitalizations. PIH17 The Prevalence and Cost of Illness in Women With Endometriosis in Ukraine Piniazhko O 1, Zalis’ka O 1, Vernikovskyy I 2 Halytsky Lviv National Medical University, Lviv, Ukraine, 2Lviv Regional Perinatal Center, Lviv, Ukraine . . . establishing now in Ukraine. Methods: We performed a retrospective analysis of the real data from 450 medical records of patients who were treated in a Lviv Regional Perinatal Center in 2013. The results of our previous study of treatment cost of endometriosis in Ukraine were used (Zalis’ka O., Piniazhko O. et al. 2014). We used the prices from Ukrainian database of Morion company (Kiev) on 01.06.2014 (1 EUR = 16,04 UAH). Results: According to the analysis 70% of women with endometriosis choose treatment hormones with the duration of six month usually and 30% of patients undergo surgical treatment. Annual direct cost of endometriosis per patient varies from € 224 to € 487 depending on the chosen scheme of pharmacotherapy, which differences 2 times. Accordingly, costs correlated with clinical severity and patient’s financial position and ability to pay. Based on the data of State Statistics Service of Ukraine, we estimated that the number of women of reproductive age in Ukraine was 14,7 million in 2013. Therefore when to extrapolate data the number of women with endometriosis could amount approximately 1,47 million and 1,03 million women would take pharmacotherapy instead of surgery. The annual cost of treatment hormones of endometriosis could amount from €230,5 million to €501,6 million. Conclusions: The economic burden of endometriosis in Ukraine demonstrates a significant need in the optimization of the management of this disease within the introduction of health insurance system and health technology assessment of the applicable hormones. PIH18 Burden and Cost of Multiple Chronic Diseases in A Large Cohort of Elderly in Italy Rossi E , Cinconze E , De Rosa M CINECA Interuniversity Consortium, Casalecchio di Reno, Italy . . . Objectives: Life expectancy in Italy is constantly increasing (+2,4 years for men and +1,7 years for women from 2001 to 2011) with a consequent effect on health needs and expenditure. Furthermore, on overall population, 21% has one chronic disease, 9,3% two and 8% more than three. The aim of this study is to evaluate the burden of multiple chronic diseases and associated costs in a large community setting of Italian elderly population, starting from Cineca ARNO Observatory. Methods: ARNO Observatory is a comprehensive population-based database which integrates since 1987 administrative data from Local Health Units to monitor health data in the real world. On a population of 10,5 millions of inhabitants we selected a cohort of 2.166.000 (21%) subjects over 65 years. Outpatient drug prescriptions, inpatient hospital discharges, diagnostic, lab tests prescriptions were analyzed during 2012 to evaluate chronic diseases, multiple chronic conditions and illness direct costs. Results: From selected cohort, 82,5% has at least one chronic disease and this proportion rises up to 89,6% in over 80s. Hypertension is the most common disease (67%), followed by dyslipidemia (30%) and diabetes (18%). Over 50% of elderly has multiple chronic diseases but the risk of developing concomitant chronic illnesses rises with ages (60% of people aged > = 85 have comorbidities). Integration of administrative data led to the evaluation of cost of illness which varies from 2.528€ for hypertension to 13.613€ for Acute Coronary Syndrome; the average cost of patients with 2 or more comorbidities is twice than patients with only one chronic disease. In general, more than a half of cost is due to hospitalization (57%), 27% to drugs and 16% to diagnostic exams (16%). Conclusions: Analyses of real world data represent an effective tool for the evaluation of elderly population diseases and can be a valid instrument to support clinical governance. PIH19 Epidemiological and Financial Burden of Preterm Labor Hospitalizations – An Analysis of German Claims Data Wetzka S 1, Fleßa S 1, Lange A 2, Zygmunt M 3 1Department of Health Care Management, Faculty of Law and Economics, Ernst-Moritz-ArndtUniversity Greifswald, Greifswald, Germany, 2Department of Neonatology and Pediatric Intensive Care, Hospital for Pediatrics, University of Greifswald, Greifswald, Germany, 3Department of Obstetrics and Gynecology, University of Greifswald, Greifswald, Germany . . . . Objectives: Preterm Labor (PTL) in pregnant women is a common obstetrics inpatient diagnosis. It is important to differentiate between threatened PTL associated with potentially avoidable hospitalization and PTL resulting in a preterm birth (PTB) with respect to their burden to the health care system. The prevention and diagnostic-driven reduction of such hospital admissions for observational reasons are supposed to reduce costs. The aim is to analyze the epidemiological and financial impact of programs to prevent avoidable hospital admissions due to threatened PTL. Methods: A cohort of singletons at the University hospital Greifswald with PTL-related hospital admissions prior to birth was analyzed with respect to resource use and obstetrics outcomes. This bottom-up analysis was conducted from the perspective of the statutory health care system based on accounting information claims from the hospital`s database for the period 2007 to 2012. ICD-10 codes were used to identify the study population. All costs were inflated to 2012 using the German GDP index. Results: Out of 4,408 singleton births, potentially avoidable hospital admissions were found in 248 cases (5.6 %). In this subgroup, 17 infants were delivered preterm (avg. 34.8 weeks), resulting in a PTB rate of 6.9 % compared to a general rate of 8.7 % in Germany. These hospitalizations led to average costs of about 1,600 EUR per case (min. 300 EUR; max. 5,600 EUR). Conclusions: Although threatened PTL affects one singleton pregnancy in 20, the fraction of costs resulting from observational admissions associated with threatened PTL is less than 0.7 percent compared to the total department costs. This finding conflicts with the cost-saving potential of prevention programs (e. g. pH self-assessment) suggested in the literature. 1Danylo Objectives: Endometriosis affects 10% of women of reproductive age and WERF EndoCost study has recently shown that it imposes substantial economic burden on society through the reduced economic and personal productivity. The aim of the study was to estimate the prevalence of endometriosis and the cost of illness from the payer’s perspective in Ukraine as the model of health insurance is PIH20 Incidence and Long-Term Cost of Oral Steroid-Related Adverse Events in Chronic Diseases in Poland Garbacka M 1, Zapalska A 1, Borowiack E 1, Tronczyñska D 2, Wepsiec K 2, Dziurda D 2, Krzyzanowska A 2 1NUEVO HTA CLP, Cracow, Poland, 2GlaxoSmithKline, Warsaw, Poland . . . . . . . A508 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Although the relationship between use of glucocorticoids (GC) and occurrence of adverse events (AE) is widely acknowledged, the estimation of risk size of specific AE is still imprecise. The aim of study was to quantify the incidence and economic cost of selected steroid-related AE in GC-users regardless of baseline chronic disease. Methods: Review of the available data about the consequences of long-term use of oral GC (depending on prednisone or its equivalent dose and period of use) was conducted. From 162 full-text publications (10024 abstracts), four with mixed population and > 5 years median follow–up were valuable. Hip fracture (HFr), cataract (CAT) and diabetes mellitus (DM) were chosen as the common and most cost generating AE connected with oral GC treatment. A Markov model with a lifetime horizon (30 years) was developed to forecast incidence and health care cost of three regimen (non-GC, low dose GC < 2.5 mg, high dose GC > 7.5 mg). Direct medical costs were included in the analysis. Results: For a lifetime horizon the incidence of HFr, CAT and DM increased from 0.77% to 5.49%, 23.48% to 91.04% and 12.34% to 17.02% (7.1, 3.9, 1.4 fold increase) respectively for comparison non-GC versus high dose GC. For selected cohort of 1,000 you need to treat 34, 2, 22 patients respectively (low dose GC instead of high dose GC) to prevent one additional case of HFr, CAT, DM. Shorter duration of steroid therapy (5 years) provide two Quality-Adjusted Life Months gained (per one patient) and leads to 2,230, 35,460, 6,920 avoided cases of HFr, CAT, DM (per 100,000 cohort). The use of low-dose or non-GC is cost-effective strategy (total cost per patient 2,958 PLN, 1,301 PLN, respectively) compared with high-dose GC (10,823 PLN). Conclusions: Oral GC treatment can lead to dose-dependent increase in the risk of selected AE. PIH21 Potentially Inappropriate Medication in the Elderly – Relevance and Economics of the 30 Top-Selling Priscus Agents in Germany Pohl-Dernick K 1, Meier F 1, Maas R 2, Schöffski O 1, Emmert M 1 1Friedrich-Alexander University Erlangen-Nürnberg, Nuremberg, Germany, 2Friedrich-Alexander University Erlangen-Nürnberg, Erlangen, Germany . . . . . Objectives: Some drugs increase risk for adverse effects in elderly patients. Accordingly, lists of potentially inappropriate medication (PIM) that should be avoided in elderly patients have been proposed. In 2010 an expert panel published a PIM list adapted to the German drug market (PRISCUS-list) which lists 83 inappropriate agents and their recommended surrogates. This study calculates the amount of drug reimbursement of PIM and the potential saving using appropriate surrogates recommended by the PRISCUS list from the perspective of statutory health insurance (SHI). Methods: Data was provided by AOK Research Institute (WIdO). Study material consists of a register extraction of the top 30 drugs (by sales) on PRISCUSlist in 2009 for patients ≥ 65 years of the entire SHI-population. We calculated the percentage of sales and defined daily doses (DDD) for patients ≥ 65 compared to the total SHI-population. Costs for the recommended substitution were estimated by different scenarios. Results: In 2009, the proportion of the top 30 drugs on the PRISCUS-list that were prescribed to patients ≥ 65 was 58.2%. Sotalol was the drug with the largest proportion of DDD prescribed to patients ≥ 65 (92.9%). Drug reimbursement for the top 30 PIM medications prescribed to patients ≥ 65 were € 305.7 million (54.3% of total reimbursement). Reimbursement for Solifenacin was highest with € 32.5 million. Prescription of the surrogates would lead to increasing costs for the German health care system. Those were calculated to range between € 325.9 million and € 810.0 million. Conclusions: This is the first study assessing the economic burden of PIM according to PRISCUS-list in Germany. The results show that a more appropriate medication for the elderly comes along with additional costs. For a final evaluation of relevance and economics of PIM, costs of adverse drug events caused by PIM and clinical feasibility of substitution have to be taken into consideration. PIH22 Cost-Effectiveness Analysis of Use of Dydrogesterone in Premenstrual Syndrome Kolbin A , Vilum I , Kurylev A , Balykina Y , Proskurin M Saint Petersburg State University, Saint Petersburg, Russia . . . . . Objectives: The primary objective of the study was the analysis of pharmacoeconomic expediency of administration of dydrogesterone (Duphaston®) for premenstrual syndrome (PMS) treatment in comparison with micronized progesterone (Utrogestan®). Methods: The mathematical modeling with dydrogesterone or micronized progesterone was applied in the study. For the calculation of the efficacy data of clinical trials were used. Costs were calculated on the basis of Russian prices (grls. rosminzdrav. ru). The model was constructed as following: for each branch of the decision tree, cost and efficacy for a group of 100 patients (female aged 18 – 45 years) and per patient were analyzed. Modeling duration was 3 months (therapy during three cycles). The cost-effectiveness ratio (CER) and incremental cost-effectiveness ratio (ICERs) were calculated. Results were evaluated to cost-effectiveness threshold. Efficiency was estimated on the basis of clinical trials (effectiveness). Calculation of cost included: the cost of drugs administration course; costs of consultative and diagnostic appointments of the gynecologist; the cost of inefficient therapy – costs of additional diagnostic examination. The comprehensive sensitivity analysis was performed. Results: The cost of the total course of therapy with dydrogesterone was more expensive in comparison with micronized progesterone – 84,1 EUR against 82,6 EUR. Strategy of administration of dydrogesterone showed more efficiency in comparison with micronized progesteron (8% increase of effectiveness). CER for dydrogesterone and micronized progesterone were 115.20 and 127.07 respectively. The ICER was 18.75 Eur per patient that is much lower than a cost-effectiveness threshold in Russia (27922.8 EUR). The sensitivity analysis confirmed conclusions of the main scenario. Conclusions: The strategy of administration of dydrogesterone in PMS is economically expedient from the point of view of cost-effectiveness ratio. In addition, increase of effectiveness was noted for use of dydrogesterone. The study was conducted at Abbott support. PIH23 Cost-Benefit Model of Varying Nexplanon and Other Long-Acting Reversible Contraceptive (Larc) Methods: Uptake Compared to the Oral Contraceptive Pill: UK Perspective Praet C , D’Oca K Merck Sharp and Dohme Ltd, Hoddesdon, UK . . Objectives: Cost is considered one of the major barriers to greater use of LARC (Long-Acting Reversible Contraceptive) methods, especially cost of treatment initiation. However, when considering their contraceptive efficacy alongside cost of pregnancy, LARC methods are deemed by NICE to be more cost-effective than combined oral contraceptive pills even at one year of use. (NICE LARC CG30 2005). Methods: A 3 year time-horizon cost-benefit model was developed to assess budgetary impact of increasing LARC uptake (implant, IUD, IUS and injectable) compared to the oral contraceptive pill, in UK women aged 16-49 who currently use the following contraceptives of interest: non-LARC method (defined as contraceptive pill only) or LARC methods (IUD, IUS, injectable, implant). A weighted-average price based on current market shares was calculated, for all contraceptive pills currently available in the UK. Increased uptake of any LARC method was offset against a reduction in contraceptive pill usage. Unintended pregnancies, based on typical failure rate, occurring with all treatments considered was taken into account. Results: Of approximately 14,750,000 women aged 16-49 in the UK, official statistics confirm 37% use contraceptive methods of interest to our model. This proportion formed our cohort of approximately 5,500,000 UK women aged 16-49, which was followed over a 3 year time horizon. A 100% increase in uptake of each LARC method would lead to a 49% decrease in oral contraceptive pill uptake. Over a three year period this would save 374,794 unintended pregnancies, and elicit financial savings of £630,831,022, on which £54,098,847 is attributable to treatment costs (ingredient, consultations, removal/insertion costs) and £576,732,175 to the cost of unintended pregnancies (live birth, miscarriage, abortion, ectopic pregnancy). Conclusions: The model projects that increasing LARC uptake will result in a significant reduction in the number of unintended pregnancies, with consequent savings to the NHS across the UK. PIH24 Misoprostol Vaginal Insert Pharmacoeconomic Model for 5 European Countries Walczak J 1, Bierut A 2, Dowgiallo J 2, Pacocha K 1, Pieniazek I 1, Stelmachowski J 1, Opala T 3, Sobkowski M 3, Baev O 4 1Arcana Institute, Krakow, Poland, 2Ferring Pharmaceuticals Poland Sp. z o. o., Warsaw, Poland, 3Ginekologiczno-Polozniczy Szpital Kliniczny Uniwersytetu Medycznego im. K. Marcinkowskiego w Poznaniu, Poznan, Poland, 4Federal State Budget Institution “Research Center for Obstetrics, Gynecology and Perinatology” Ministry of Healthcare and Social Development of the Russian Federation, Moscow, Russia . . . . . . . . . Objectives: Our aim was to assess the costs and consequences of labour induction using misoprostol vaginal insert (MVI) compared with currently used technologies using a specifically developed user-friendly decision model developed for Austria, Poland, Romania, Russia and Slovakia. Methods: The model was developed in Microsoft Office Excel and compares clinical and safety aspects like time to vaginal delivery, time to active labour, occurrence of cesarean delivery and adverse events of MVI with selected comparators. Efficacy and safety data were retrieved from targeted literature review, conducted in the main medical databases. Country-specific information about costs and resource use was incorporated into the model. Local data were collected for each country via a specifically developed questionnaire. The model considered the hospital and public payer perspectives. The model generated results as an incremental difference between the total costs related to labour induction with MVI or a comparator. The threshold price of MVI was also calculated. Results: Local Key Opinion Leaders recommended the following comparators: dinoprostone vaginal insert (DVI; Austria), dinoprostone vaginal tablets (Dtab; Austria, Slovakia), dinoprostone cervical gel (Dgel; Poland, Russia, Slovakia) and oxytocin (Austria, Poland, Romania, Russia). The hospital perspective was chosen as default (additionally the public payer perspective was adopted for 2 countries). The use of MVI in most scenarios is related to a reduction in time consumption of hospital staff and in the length of patients’ stay in hospital wards. MVI was less costly or marginally more expensive in 80% of cases. Conclusions: Induction of labour with the use of MVI using a hospital perspective, brought savings in most countries and scenarios in comparison to other prostaglandins (DVI, Dtab, Dgel). PIH25 Cost Effectiveness Analysis of A Vaccination Programme for the Prevention of Herpes Zoster and Post-Herpetic Neuralgia in Adults Aged 65 and Over in Norway Suseg P 1, Olsen D 2, Préaud E 3, Uhart M 3 1LINK Medical Research AS, Oslo, Norway, 2SPMSD, Drammen, Norway, 3Sanofi Pasteur MSD, Lyon, France . . . . Objectives: Herpes Zoster is a very painful and debilitating disease for which no satisfactory treatment exist. A vaccine is licensed in Europe for the prevention of Herpes Zoster (HZ) and postherpetic neuralgia (PHN) in adults aged ≥ 50 years and is recommended in France and UK. The objective of this study was to assess the cost-effectiveness of vaccination programs in people aged 65 years and over in Norway. Methods: An existing European Markov cohort Model was adapted to the Norwegian health care setting. Health states considered are healthy, HZ, PHN, healthy post-HZ and death. HZ and PHN states are further split by pain severity (mild, moderate or severe). A vaccine efficacy durability model based on the pivotal trial data was included to simulate waning in the efficacy. The cost-effectiveness outcomes were assessed from both the third party payer and the societal perspective. First, analysis comparing a HZ vaccination policy for adults aged ≥ 65 years with a no vaccination policy was done. Then, analysis comparing vaccination policies of 5-years age class cohorts (from 65 to 100 years old) to a no vaccination policy were conducted. Input data were obtained from Norwegian sources whenever avail- A509 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 able. Results: The strategy of vaccinating people over 65 yo showed incremental cost-effectiveness ratios of 41,467€ /QALY gained from a payer perspective and 40,733€ /QALY from a societal perspective. The analysis by age group showed that the ICER is age-dependent, the lowest ICER (30,780€ /QALY in both perspectives) having been found in the cohort 70-74 years old. In sensitivity analyses data on the duration of PHN, utilities and vaccine efficacy duration showed a major impact on the results. Conclusions: Our cost-effectiveness analysis shows that a HZ vaccination policy for adults aged ≥ 65 years in Norway could be cost-effective and provide substantial public health benefits in the Norwegian health care system. PIH26 Cost-Effectiveness Analysis of Surgical Management of Stress Urinary Incontinence With Single-Incision Mini-Sling Versus Tension-Free Vaginal Obturator in Spain Castañeda E 1, Sanz-Granda Á 2, Hidalgo A 2, Meza D F 1, Marqueta J M 1, Carreras M 1 de Palamós, Palamós, Girona, Spain, 2Weber, Economía y Salud, Majadahonda, Madrid, Spain . . . . . . . . 1Hospital Objectives: To analyze the cost effectiveness of surgical management of stress urinary incontinence (SUI) in women with single-incision mini-sling (SIMS) compared with tension-free vaginal obturator (TVT-O) Methods: A cost-effectiveness analysis based on the results of interventions performed with TVT-O (2005-2008) and SIMS (2008-2011) in women with a diagnosis of SUI was performed. The clinical effectiveness was defined as an objective cure at 12 months (pad-test < 1 g/h). A perspective of the hospital payer was adopted; therefore, only direct health care costs (diagnostic and surgical procedures, medical devices, medications, hospital stay times and staff) were included. CI95% of total cost was estimated by bootstrapping; later, different sensitivity analyses were conducted Results: Procedures were carried out in 81 women (44 in the SIMS group and 37 in the TVT-O). A small difference (6.7%) in clinical effectiveness was observed in favour of SIMS, however, it was not statistically significant (SIMS: 93.2% and TVT-O: 86.5%). The total annual cost per patient with SIS was lower (2,059€ ; CI95%: 1,914-2,285) than with TVT-O (2,821€ ; CI95%: 2,661-2,997), showing a statistically significant cost saving of 762€ (CI95%: 516-987). In the base case, the probability of SIMS is less costly than TVT-O was 100%. The sensitivity analysis showed that the cost determinant was the length of the hospital stay, observing that an equivalent cost was only achieved if there was no cause for any hospital stay with TVT-O. Conclusions: The use of singleincision mini-sling is associated with an comparable clinical effectiveness but with a 762€ per patient reduction of the average annual cost, compared to tension-free vaginal obturator. Therefore, the results suggest that, over a post-operative period of twelve months, tension-free tape single-incision mini-sling is a dominant alternative to tension-free vaginal obturator because of a lower cost and a comparable effectiveness. PIH27 A Cost-Effectiveness Analysis of Different Types of Labor for Singleton Pregnancy – Real Life Data Lakic D 1, Tadic I 1, Odalovic M 1, Petrovic B 2, Petrova G 3 1University of Belgrade Faculty of Pharmacy, Belgrade, Serbia and Montenegro, 2University Hospital for Gynecology and Obstetrics “Narodni Front”, Belgrade, Serbia and Montenegro, 3Medical University Sofia, Faculty of Pharmacy, Sofia, Bulgaria . . . . . Objectives: To assess cost, clinical outcomes and cost-effectiveness of different types of labor in singleton pregnancies. Methods: A decision model was used to compare vaginal labor, induced labor and planned cesarean section. All data were taken from the Book of Labor from the University Hospital for Gynecology and Obstetrics “Narodni Front” in Serbia, for labors conducted during one month period in 2011. Successful delivery, (i. e. labor that began up to 42 gestation weeks, without maternal mortality and the newborn Apgar scores greater than or equal to seven in the first and fifth minute of life) was considered as the outcome of the cost effectiveness analysis. To test the robustness of this definition probabilistic sensitivity analysis was performed. Results: From a total of 667 births, vaginal labor was conducted in 98 cases, induced vaginal in 442, while planned caesarean section was performed 127 times. Emergency caesarean section as a complication was much higher in the vaginal labor cohort compared to the induced vaginal cohort (OR = 17.374, 95% CI: 8.522 to 35.418, p < 0.001). The least costly type of labor was induced vaginal labor: average cost 461 euro, with an effectiveness of 98.17%. Both, vaginal and planned cesarean labor, were dominated by the induced labor. The results were robust. Conclusions: Elective induction of labor was associated with the lowest cost compared to other types of labor, with favorable maternal and neonatal outcomes. PIH28 Cost Effectiveness of Pentavalent Rotavirus Vaccine (Rv5) in Slovenia Gebremeskel B G 1, Pokorn M 2, Zakotnik B 2, O’Brien M A 3 1Merck & Co., Inc/Rutgers University, West Point, PA, USA, 2University Medical Centre Ljubljana, Ljubljana, Slovenia, 3Merck & Co, Inc., West Point, PA, USA . . . . . . Objectives: To assess the potential impact of universal vaccination with RV5 on health care burden and costs associated with rotavirus gastroenteritis (RGE) among a hypothetical birth cohort of 21,938 Slovenian children in their first five years of life Methods: A Markov model was used to evaluate the cost per quality-adjustedlife-year (QALY) and public health impact of vaccination with RV5 from the health care payer and societal perspectives. The base case assumes three dose coverage rate of 94% at 2, 4 and 6 months among the vaccinated, the remaining receiving 1 or 2 doses. In the absence of universal vaccination 819 hospitalizations and 3,276 office visits were projected to occur in the first 5 years of life. RGE associated costs include direct and indirect costs associated with parental work loss. The efficacy of RV5 in reducing health care resource utilization is based on the results of the Rotavirus Efficacy and Safety Trial (REST). Results: A universal RV5 vaccination program is projected to reduce hospitalizations, office visits and parental work loss by 94, 76 and 87% respectively. The cost per case avoided would be 165 Euros, and cost per hospitalization avoided would be 1,639 Euros with implementation of universal vaccination. The cost per QALY saved would be 29,452 Euros and 20,453 Euros from the health care and societal perspectives respectively. Conclusions: RV5 is projected to avert substantial number of RGE hospitalizations and office visits in Slovenia and would be considered a cost effective intervention. KEYWORDS: rotavirus vaccine, cost effectiveness, QALY. PIH29 Health Economic Model On the Costs and Effects of Rotavirus Vaccination in Romania Preda A L 1, Moise M 1, Standaert B 2 1GlaxoSmithKline, Bucharest, Romania, 2GlaxoSmithKline Vaccines, Wavre, Belgium . . . . Objectives: Rotavirus gastroenteritis (RVGE) is one of the most frequent diseases in children less than 5 years old. Today, no recommendation for general rotavirus vaccination exists in Romania, which leads to a vaccination coverage rate of < 8%. We model whether the strategy of “Universal Mass Vaccination” should be cost-effective from the Social Health Insurance Fund (SHIF) perspective. Methods: We adapted an already published simple model to estimate the cost-effectiveness of rotavirus vaccination in Romania. It helps identifying the drivers of the economic analysis in a birth cohort of 194,411 children (Romania 2012) followed over a period of 5 years. In the model, vaccine efficacy results from international clinical trials are combined with Romanian epidemiology and cost data from the SHIF perspective for 2012. The model assumes a theoretical vaccination rate of 100% and uses discount rates of 3% on costs and effects. Results are tested on their robustness using univariate sensitivity analysis. Results: The model predicts that a two-dose rotavirus vaccine could avoid around 82,581 mild, 51,328 moderate and 3,075 severe RVGE cases. Moderate and severe events are associated with physician visits and hospital stays respectively. These also lead to a total cost saving of € 2.4 million for indirect cost estimates. The main factors responsible for the savings are the reduction in hospital stays, in productivity losses of parents and in medical visits. Sensitivity analysis indicates the importance of good epidemiological data and cost numbers on productivity loss. Conclusions: General vaccination against rotavirus could avoid many of the severe diarrhea events in children less than 5 years old in Romania. Depending on the price per course, the vaccination strategy will lead to societal cost gain. PIH30 Cost-Effectiveness Analysis of Coffee Consumption for Prevention of All-Cause Mortality in Germany Drees L 1, Theidel U 1, Campbell C M 2, Popelar B V 2, O’Day K 2, Mittendorf T 1 GmbH, Hannover, Germany, 2Xcenda, LLC, Palm Harbor, FL, USA . . . . . . . . 1Herescon Objectives: Coffee contains over 1,000 distinct molecular compounds and is one of the most widely consumed beverages worldwide. Epidemiologic studies have shown an inverse relationship between coffee consumption and all-cause mortality. This analysis aims to assess the cost-effectiveness of coffee from a blended German consumer and payer perspective. Methods: An existing decision-analytic model was adapted with German data. A cohort life-table analysis was developed to model life-years (LYs) of German coffee consumers vs. non-consumers over a lifetime horizon. Age- and gender-specific mortality rates were used to model survival outcomes. Relative risks of death by average coffee intake (cups/day) were obtained from a recent large, prospective cohort study. Cost were considered for cost per cup (home prepared and from a national sample of low and high-cost vendors) and for health care. Incremental analyses were conducted by cost, sex, and level of daily coffee consumption. Deterministic and probabilistic sensitivity analysis was performed. Results: Coffee increased undiscounted LYs in 1, 2-3, 4-5, and 6+ cup/day male (0.65, 1.10, 1.33 and 1.10) and female (0.45, 1.21, 1.51, 1.41) consumers, respectively, versus non-consumers. ICERs per undiscounted LY gained were € 3,938/ € 7,047/ € 16,271 for males and € 5,514/ € 5,066/ € 13,537 for females, respectively, for 1, 2-3, and 4-5 cups/day consumption of home-prepared coffee; ≥ 6 cups/day was strictly dominated. Consumption of 4-5 cups per day purchased from high-cost vendors was not cost-effective (male: € 168,780/ female: € 135,636). Results were consistent throughout the sensitivity analyses, whereas coffee effectiveness in preventing death and coffee acquisition cost has the largest impact on ICERs. Conclusions: In this analysis, coffee consumption was associated with increased LYs and was shown to be potentially cost-effective, especially if homeprepared or purchased from low-cost vendors. Given the observational nature of the study data, further research is warranted to validate these findings. PIH31 Cost-Effectiveness Analysis of Screening Syphilis Among Pregnant Women Chuluunbaatar E National Yang Ming University, Ulaanbaatar, Mongolia . Objectives: Maternal and congenital syphilis prevalence rates are currently rapidly increasing in Mongolia. On-Site screening and same-day treatment for syphilis in pregnancy prevents greater numbers of congenital syphilis and its complications. The Ministry of Health has been implementing on-site rapid screening test (RT) intervention and same day treatment approach for maternal syphilis with the contribution of the World Health Organization. Objective of the study was to understand the cost- effectiveness (CE) of screening antenatal syphilis using the RT strategy, to compare this intervention with RPR testing strategy. And in order to estimate CE we found out maternal syphilis prevalence. Methods: Ingredientsbased cost data and epidemiological data were collected retrospectively from the pregnancy medical records. Decision analysis was used to estimate the incremental CE of on-site PT compared to the current practice, off-site PRP/TPHA. Descriptive analysis has done for prevalence of syphilis, and economic analysis has done to calculate costs. Stability of cost-effectiveness ratios were evaluated by the univariate sensitivity analysis. Results: With antenatal syphilis prevalence of 3.0% (1.46% in urban and 5.7% in sub-urban area), the cost effectiveness was US$ 14.60/DALY A510 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 for RT intervention and US$ 17.88/DALY for RPR intervention. Cost-effectiveness ratios (CERs) were more sensitive to the prevalence rate, sensitivity of tests, and DALY discount rate. Conclusions: Using the on-site antenatal rapid testing, same day treatment for positive results, and confirmed by RPR testing approach is costeffective in Mongolia. PIH32 Cost Effectiveness of Calcium Supplement in Reducing Preeclampsia-Related Maternal Mortality Chicaiza-Becerra L A 1, Garcia-Molina M 1, Oviedo S 1, Urrego J 2, Rincon C J 3, Gomez P I 1, Rubio-Romero J A 1 1Universidad Nacional de Colombia, Bogotá, Colombia, 2Facultad de Ciencias Económicas, Universidad Nacional de Colombia, Bogotá, Colombia, 3Universidad Nacional de Colombia, Bogota, Colombia . . . . . . . . . . . Objectives: To estimate the cost-effectiveness of the supply of calcium of 1200mg per day from the week 14 of pregnancy to all pregnant women compared to not supplying it to reduce the incidence of preeclampsia. Methods: A decision tree was built in TreeAge® with outcome measured in life years gained (LYG) associated to the reduction in maternal deaths. The costs were included from the perspective of the Health System in Colombia. Pharmaceutical costs were obtained from 2008 SISMED (1) and the value of the procedures was calculated by adjusting the values of Tariff Manual ISS 2001 + 30% (2), these values were compared with information of costs supplied by three EPS. All costs are expressed in Colombian pesos of 2010. The discount rate was 0%. It was performed sensitivity univariate and probabilistic analyzes for costs and effectiveness. Results: Compared to no intervention, calcium supplement is a dominant alternative. If the incidence of preeclampsia is lower than 51.7 per 1000 pregnant women or the cost per tablet of calcium of 600 mg is greater than $454, calcium supplement is no longer a cost-effective alternative in Colombia for a threshold of 3 times the GDP per capita in Colombia of 2010 by LYG, equal to $36,143,550. Conclusions: Supply of calcium to all pregnant women from week 14 of gestation is a dominant alternative compared to no intervention, which saves 200 LYG, while it decreases costs in the order of $5,304 million pesos per 100.000 pregnant women. PIH33 Economic Evaluation of Ulipristal Acetate for the Treatment of Patients With Moderate and Severe Symptoms of Uterine Fibroids in Romania Lorenzovici L 1, Székely A 1, Ágh T 2, Vámossy I 3, Kelemen L J 4, Finta H 5, Kaló Z 6 Research Romania Ltd, Tirgu Mures, Romania, 2Syreon Research Institute, Budapest, Hungary, 3Gedeon Richter Plc, Budapest, Hungary, 4Gedeon Richter Romania SA, Corunca, Romania, 5University of Medicine and Pharmacy of Tirgu Mures, Tirgu Mures, Romania, 6Eötvös Loránd University, Budapest, Hungary . . . . . . . ing an additional unintended pregnancy with ulipristal acetate as compared to levonorgestrel is estimated to be 418€ . Ulipristal acetate is most cost-effective in the subgroup of intake within 24 hours, where it is more efficacious at a lower cost compared to levonorgestrel. Conclusions: Ulipristal acetate is a cost-effective alternative to levonorgestrel, given that the cost of avoiding an additional pregnancy with ulipristal acetate is less than the average cost of said pregnancies. Therefore, French minors should have free access to ulipristal acetate directly in a pharmacy. Ulipristal acetate should be used rapidly after unprotected intercourse (within 24hours) to benefit from its cost-saving potential compared to levonorgestrel use. . 1Syreon Objectives: Ulipristal acetate is a selective progesterone receptor modulator that has been demonstrated to be an effective 3-month pre-operative treatment for moderate to severe symptoms of uterine fibroids in adult women of reproductive age. The aim of this analysis was to assess the cost-effectiveness of 5 mg ulipristal acetate as an add-on therapy to standard pre-surgical observation and treatment or immediate hysterectomy in Romania. Methods: A Markov model was developed using a 10-year time horizon. Ulipristal acetate was compared with pre-surgical observation and immediate hysterectomy. The model comprised the following mutually exclusive health states: mild, moderate, severe, or persistent severe excessive bleeding disorder; myomectomy; post-myomectomy with mildly to moderately excessive bleeding disorder; post-myomectomy with severely excessive bleeding disorder; hysterectomy; post-hysterectomy; post-menopause; and death. Transition probabilities and utility values were obtained from clinical trials and the scientific literature. Resource utilisation and unit costs were derived from the consensus panel of clinical experts and the Romanian National Insurance House tariffs. Cost vectors in RON were converted to EUR by using 2013 Romanian National Bank average exchange rate (1 EUR = 4.419 RON). Results: Adding a 3-month course of ulipristal acetate to pre-operative observation was predicted to achieve an additional 0.021 quality-adjusted life years (QALYs) at an estimated incremental cost of 367 € , which would result in an incremental cost of 17,749 € /QALY. When 3 months of ulipristal acetate therapy was compared with immediate hysterectomy, the incremental cost-effectiveness ratio was reduced to 2,300 € /QALY. The results were most sensitive to the utility value of the post-hysterectomy health state but responsive to changes in other model parameters. Conclusions: The results of this analysis suggest that adding ulipristal acetate treatment to standard pre-surgical therapy represents a good value for money in Romania. The inclusion of societal benefits may considerably reduce the cost-effectiveness ratio. PIH34 The Cost-Effectiveness of Emergency Hormonal Contraception With Ulipristal Acetate Versus Levonorgestrel for Minors in France Schmid R HRA Pharma, Paris, France . Objectives: To compare the cost and effectiveness of two emergency contraceptive methods in minors in France and to support the payer’s analysis if it is worth to deliver ulipristal acetate for free to minors. Methods: Based on a decisionanalytical model, the cost-effectiveness of two emergency contraceptive methods is compared. Pregnancy rates, outcome of unintended pregnancy in minors and resource utilization are derived from literature. Resources and their costs are considered until termination or a few days after delivery. Costs are taken from a collective perspective. Sensitivity analyses are performed on the most sensitive input parameters. Results: Using emergency contraception is superior to no method. The cost of an unintended pregnancy in a French minor is estimated to be 1630€ (1330€ - 1803€ ). Almost 4 million€ (3.1-13.7million€ ) could have been saved by using ulipristal acetate instead of levonorgestrel in 2010. The incremental cost of avoid- PIH35 Cervical Assessment With Progesterone in the Prevention of Preterm Birth: A Strategy Based On Cost-Effectiveness Fonseca E B 1, Nishikawa A M 2, Paladini L 2, Clark O A C 2 1Universidade Federal da Paraíba, João Pessoa, Brazil, 2Evidencias, Campinas, Brazil . . . . . . . . INTRODUCTION: Preterm birth (PTB) complications are estimated to be the second most common cause of death in under-five children and responsible for 3.1million neonatal deaths. According to a worldwide analysis, Brazil is one of the top ten countries with the highest number of PTB. Considering its long-term costs, strategies that reduce incidence may be cost-effective. Treatment with progesterone is one of the interventions recommended for PTB prevention due to the evidence supporting its efficacy in women with short cervix and prior history of preterm delivery. Objectives: Determine whether treatment with progesterone for pregnant women with a short cervical length < 25mm identified in routine measurement of second-trimester transvaginal cervical length by ultrasound in low-risk singleton pregnancies is a cost-effective strategy under the Brazilian Healthcare System perspective. Methods: A cohort model was developed according to the disease and resources use. Epidemiology of pregnancies at risk of PTB eligible for progesterone treatment were obtained from published data. To obtain national clinical data, births were categorized by gestational week age at delivery specialist opinion. Progesterone effectiveness data were obtained from systematic reviews, meta-analysis and specialist opinion. Costs included screening test, prenatal consultation, progesterone and neonatal hospitalization. Exchange rate was 1USD= 2.30BRL. Results were presented in cost/PTB avoided. Results: Considering 278.100 PTB, the inclusion of screening test to identify pregnant women with short cervix and its prophylaxis with progesterone shows significant economic savings of USD74 million. Although the expenditure on drug, screening test and prenatal consultation increment the total costs, the reduced number of PTB (263,052 vs 278.100) and neonatal UTI hospitalization length (4,098,543 days vs 4,518,056 days) resulted in a total economic saving. Conclusions: Prevention of PTB is dominant in women with short cervix as compared to a no-prophylactic strategy scenario resulting in economic savings to the Brazilian health care system. PIH36 Cost-Effectiveness of Palivizumab Use in High Risk Children From Brazilian Health System Perspective Fernandes R R A UERJ - IMS, Rio de Janeiro, Brazil . . . Objectives: This study aimed to investigate the cost-effectiveness of palivizumab to different combinations of risk groups, such as premature children born with gestational age (GA) ≤ 28 weeks, GA ≤ 32 weeks, children with congenital heart disease and bronchopulmonary dysplasia. Methods: Literature review was performed to search effectiveness data. One Markov model (base case), and one decision tree (alternative scenario) were built with a cohort simulation along a 18 years-time horizon for the base case and a 1 year time horizon for the decision tree. Base case consider sequelae after infection, and alternative scenario not. The Health System perspective was used, with a discount rate of 5%. A probabilistic sensitivity analysis (Monte Carlo simulation) with probability distributions fitted to the variables, was performed under the different structural assumptions, as well as a deterministic analysis, using Tornado diagram, to verify the variable modifications able to alter the responses of the model. A threshold analysis was used to estimate the price that palivizumab would fit under an acceptability threshold proposed for the health system. Results: The option of using the prophylaxis just in preterm children born with GA ≤ 32 weeks dominated all others. The incremental effectiveness of base case analysis compared with no prophylaxis (base line) was 0.19 QALY. However, this strategy was not cost-effective, presenting an incremental cost-effectiveness ratio (ICER) of R$ 81,627.31/QALY, value above of World Health Organization (WHO) proposed threshold of three times GDP per capita (R$ 63,756.00/QALY). The ICER of GA ≤ 32 weeks in alternative scenario was 2.023.045,72, showing the importance of considering sequelae in analysis. Sensitivity analysis showed that some variables when altered were able to change model final answers. Conclusions: Threshold analyses demonstrated that palivizumab price must be reduced in at least 22% to be incorporated to all populations use, based on WHO threshold. PIH37 Cost-Effectiveness Analysis of the New Biomarkers for Diagnosis of Acute Kidney Injury in Children After Cardiac Surgery Bogavac Stanojevic N 1, Petrovic S 1, Lakic D 2, Peco Antic A 3, Vulicevic I 4, Ivanisevic I 5, Kotur Stevuljevic J 1, Jelic-Ivanovic Z 1 1University of Belgrade, Faculty of Pharmacy, Belgrade, Serbia and Montenegro, 2University of Belgrade Faculty of Pharmacy, Belgrade, Serbia and Montenegro, 3School of Medicine, University of Belgrade, Belgrade, Serbia, Belgrade, Serbia and Montenegro, 4University Children’s Hospital, Belgrade, Serbia and Montenegro, 5University Children’s Hospital, Belgrade, Serbia and Montenegro . . . . . . . . Objectives: Children undergoing cardiac surgery for congenital heart disease are more likely to experience development of acute kidney injury (AKI) in the immediate postoperative period. In current clinical practice, AKI diagnosis is based on a rise in serum creatinine (sCr) levels, which occurs 2-3 days after the initiating renal insult. Many new biomarkers offer promise for earlier AKI diagnosis. The objective A511 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 was to assess the incremental cost effectiveness of using serum CysC (sCysC), urine NGAL (uNGAL) and urine L-FABP (uL-FABP) for the diagnosis of AKI in children after cardiac surgery compared with current diagnostic method (monitoring of sCr level). Methods: We developed a decision analytical model to estimate qualityadjusted life years (QALY), lifetime costs and incremental cost-effectiveness of different biomarker-based diagnostic strategies which can be used in clinical practice compared to current strategy. This model simulates detection of AKI, its progression to chronic kidney disease (CKD) and CKD treatment in cohort of patients younger than 18 years. Results: The cost-effectiveness ratios were between $1485/QALY for sCr and $3579/QALY for uNGAL. uNGAL and sCys C strategies yielded higher costs and lower effectiveness (ie. dominated) compared to uL-FABP strategy. uL-FABP added 1.43 QALY compared to current diagnostic method at an additional cost of $8521.87. ICER for uL-FABP compared to sCr was $5959.35/QALY. Probabilistic sensitivity analyses indicated that the uL-FABP strategy was cost-effective for all 10.000 patient simulations at specified $50000/QALY threshold. Conclusions: Our results suggest that the use of uL-FABP is likely to represent an economically advantageous strategy for early AKI diagnosis in children after cardiac surgery. However, we need rapid screening uL-FABP test to ensure timely and efficient AKI treatment. mize selection bias. Patients in both groups were required to be at least age 18 years, with continuous medical and pharmacy benefits 1 year before, and 1 year after the index date. One-to-one propensity score matching (PSM) was used to compare health care costs and utilizations during the follow-up period between the menopausal and the comparison groups, and were adjusted for baseline demographic and clinical characteristics. Results: After risk adjustment by PSM, a total of 67,740 patients in each cohort were matched. More menopausal patients had inpatient admissions (15.18% vs. 11.89%, p< 0.0001) and other service (99.98% vs. 90.29%, p<0.0001) and pharmacy visits (86.55% vs. 71.23%, p<0.0001) compared to those without menopause. Menopausal patients also incurred significantly higher other service visit ($11,215 vs. $8,812, p<0.0001) and pharmacy costs ($2,448 vs. $1,878, p< 0.0001) than comparison patients. Conclusions: In the U.S. Medicaid program, menopausal patients had higher health care utilization and incurred higher costs than those without menopause, highlighting the economic burden of the disease. PIH38 Cost-Effectiveness Analysis of the Therapy of Endometriosis 1Temas Tuletova A , Dochshanova A , Skakova R Medical University Astana, Astana, Kazakhstan Objectives: Infantile Hemangioma (IH) is one of the most common childhood benign tumours. Recent studies have demonstrated the success of propranolol for involution of IH and the higher clinically effective and safe compared with corticosteroids. The purpose of this study is to estimate the cost-utility of propranolol, a new medicinal product authorized for this specific paediatric indication (3.75 mg/ mL, oral solution) versus corticosteroids (5.00 mg, tablets), used in clinical practice in absence of other authorised therapies of proliferating IH requiring systemic treatment. Methods: A life-time (30 years) mixed decision tree and Markov model has been developed to describe the pathway of infants with IH and to assess costs and outcomes (Quality-Adjusted Life Years – QALYs – gained) from the perspective of the Italian National Health Service (INHS). Clinical inputs derive from the manufacturer’s pivotal trial and literature review, validated by key clinicians in Italy. The economic evaluation considers direct medical costs associated with IH (drug acquisition, hospital admissions and outpatient visits) derived from public sources. The atopic dermatitis as a proxy for IH utilities, the Infants Dermatitis Quality of Life Index and the Children’s Dermatology Life Quality Index were used to estimate quality of life. Probabilistic sensitivity analyses (PSA) were performed to investigate model parameter uncertainties. Costs and health benefits have been discounted at an annual rate of 3.00%. Results: The cumulative costs are €2,399.32 and € 1,859.68 while cumulative QALYs are 19.11 and 18.95 for propranolol and for corticosteroids respectively, corresponding to an ICUR of € 3,372.75/QALY. PSA results suggest that 94.60% of the 1000 iterations fall within a € 30,000 cost-effectiveness threshold considered acceptable for a marginal unit of effectiveness. Conclusions: The propranolol (3.75 mg/mL, oral solution for paediatric use) for the treatment of proliferating IH can be considered cost-effective compared to corstisteroids (5.00 mg, tablets) in the INHS perspective. PIH41 Cost-Utility Analysis Comparing Propranolol With Corticosteroids in the Treatment of Proliferating Infantile Hemangioma in Italy Paragò V 1, El Hachem M 2, Bonamonte D 3, Diociaiuti A 2, Teruzzi C 1 - A Quintiles Company, Cassina de’ Pecchi, Italy, 2Bambino Gesù Children’s Hospital, Rome, Italy, 3University of Bari, Bari, Italy . . . . Objectives: Endometriosis is a disease with social and economic impact. We analyses the clinical and pharmaco-economical efficacy of the treatment of genital endometriosis. Methods: This open-label prospective comparative research was performed in 2012-2013 on Center of Endometriosis of Astana. Results of treatment taken from 180 female patients with endometriosis aged 19-35 years old were included into the study. All of them were divided on three groups depending on the methods of therapy. 1st group have a endo-surgery treatment (60 patient), 2nd group (60 patient) - only hormonal therapy (Dienogest 2 mg per day for 6 months) and 3rd group - combined therapy (after endo-surgery taken the Dienogest 2 mg per day for 6 months). For the calculation of cost/effectiveness index we included all of the direct medical expenses. Results: In the first group efficacy index is 66,7%, in the second group – 70,0% and in the third group – 91,7%. Index cost/effectiveness for endo-surgery is 143 298 KT (1$= 182KT), cost/effectiveness for hormonal therapy is 92 428 KT and cost/effectiveness for combined methods is 115 718 KT. Surgical treatment has the low efficiency of the high costs compared with other therapies. The indicator of cost/effectiveness of the combined therapy is higher than in the hormonal treatment: higher efficiency cost in additional costs. For the objectification of this fact calculated Incremental Index per patient: 186 500 - 107 833 / 92.0 - 70.0 = 182 KT. So, the increase of one percent of effectiveness combined therapy compare to hormonal therapy reflects additional cost, not more than 182 KT per patient. Conclusions: Hormonal therapy resulted as the cost-saving therapy of genital endometriosis in young women. The higher efficiency of treatment of endometriosis are needed additional expenses. PIH39 Cost-Utility Analysis of Preventive Home Visits in Older Adults Brettschneider C 1, Luck T 2, Fleischer S 3, Roling G 3, Beutner K 3, Sesselmann Y 3, Luppa M 4, Behrens J 3, Riedel-Heller S 4, König H H 1 1University Medical Center Hamburg, Hamburg, Germany, 2University Leipzig, Leipzig, Germany, 3University Medical Center Halle (Saale), Halle (Saale), Germany, 4University of Leipzig, Leipzig, Germany . . . . . . . . . . . Objectives: Most elderly prefer to grow old in the community within a familiar environment, instead of moving to a nursing home. Preventive home visits based on multidimensional geriatric assessment can be one strategy to support this preference and might additionally reduce health care costs, due to the avoidance of costly nursing home admissions. The purpose of this study was to analyse the cost-effectiveness of preventive home visits from a societal perspective in Germany. Methods: This study is part of a multi-centre, non-blinded, randomised controlled trial. Participants were older than 80 years and living at home. Three home visits were conducted to identify self-care deficits and risk factors, to present recommendations and to implement solutions. The control group received usual care. A cost-utility analysis using QALY based on the EQ-5D was performed. A cost-effectiveness acceptability curve controlled for confounding variables was constructed. A sensitivity analysis to control for the influence of the mode of QALY calculation was performed. Results: 278 individuals were included in the analysis. Mean total cost (+874 EUR) and number of QALY (+0.0014) were higher in the usual care group, but differences were not significant. The probability for costeffectiveness of preventive home visits increased from 6% (Willingness-to-pay: 0 € / QALY) 30% (Willingness-to-pay: 250.000 € / QALY). The results were robust to the mode of QALY calculation. The probability of cost-effectiveness did not exceed 30%. Conclusions: We found convincing evidence that the evaluated preventive home visits programme is not cost-effective. PIH40 Evaluation of the Economic Burden of Menopausal Women in the U. S. Medicaid Program Li L 1, Shrestha S 1, Baser O 2, Wang L 1 1STATinMED Research, Plano, TX, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . Objectives: To evaluate the economic burden of menopausal women enrolled in the U.S. Medicaid program. Methods: Menopausal women (International Classification of Disease, 9thRevision, Clinical Modification [ICD-9-CM] diagnosis code 627) were identified using U.S. Medicaid data from 01JAN2008 through 31DEC2010. The initial diagnosis date was designated as the index date. A separate group of patients without a menopause diagnosis but of the same age, race, and gender was identified, matched and designated as the comparison group. A random index date was chosen to mini- . . . . PIH42 Cost Effectiveness Analysis of A Vaccine To Prevent Herpes Zoster and Postherpetic Neuralgia in Italy Coretti S , Ruggeri M , Codella P Università Cattolica del Sacro Cuore, Rome, Italy . . . Objectives: The aim of this study was to assess the cost-effectiveness of HZ vaccination compared to no vaccination strategy which only involves the treatment of patients affected by HZ, within the Italian context. Methods: The natural history of HZ and PHN was mapped through a Markov model with lifetime horizon and cycles lasting one month. Both third party payer (the Italian National Health Service) and societal perspectives were adopted. Costs and Effectiveness data was derived from literature and discounted by 3.5%. Model results are expressed in terms of incremental cost-effectiveness ratio (ICER). Both deterministic and probabilistic sensitivity analyses were performed to appraise the effect of parameters’ variation on model results. Results: A population of patients with HZ aged between 60 and 79 years was hypothesized. The ICER of the vaccination equaled € 12,155 per QALY under the NHS perspective and € 11,118 per QALY under the societal perspective. Moreover, under NHS perspective the cost per HZ-episode avoided and the cost per PHN-episode avoided amounted to € 1,098 and € 8,742 respectively. Considering a cost-effectiveness threshold of € 30,000/QALY, the probabilistic sensitivity analysis showed that vaccination is cost-effective regardless of the perspective adopted, in 99% of simulations. Conclusions: Results showed that a vaccination program against herpes zoster and post-herpetic neuralgia is cost-effective in Italian patients aged between 60 and 79 years. PIH43 Cost-Utility Analysis of A Medication Review With Follow-Up for Older People With Polypharmacy in Community Pharmacies in Spain: Consigue Program Jódar-Sánchez F 1, Malet-Larrea A 2, Martín J 3, Garcia L 4, López del Amo M P 3, MartínezMartínez F 3, Gastelurrutia-Garralda M A 5, Garcia-Cárdenas M V 6, Sabater-Hernández D 6, Benrimoj S I 6 1Virgen del Rocío University Hospital, Seville, Spain, 2University of the Basque Country, Vitoria, Spain, 3University of Granada, Granada, Spain, 4Andalusia School of Public Health, Granada, Spain, 5University of Granada, San Sebastián, Spain, 6University of Technology Sydney, Sydney, Australia . . . . . . . . . . . . . . Objectives: The objective of this study was to estimate the incremental costeffectiveness ratio (ICER) of a medication review with follow-up (MRF) service for older people with polypharmacy in community pharmacies against the alternative of receiving usual dispensing. Methods: The study was designed as a longitudinal cluster randomized trial carried out over six months of follow-up. The target A512 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 population consisted of older people (≥ 65 years) with polypharmacy (≥ 5 drugs) and the study was conducted in 178 community pharmacies in Spain. A total of 1,403 patients were enrolled, 688 in the intervention group (IG) and 715 in the control group (CG). The analysis adopted the perspective of the National Health Service (NHS). In order to analyze the uncertainty of ICER results, we performed a nonparametric bootstrapping with 5,000 replications. Results: Both groups reduced the average number of prescribed medications, although this reduction was greater in the IG (0.28 drugs; p< 0.001) than in the CG (0.07 drugs; p= 0.063). Patients in the IG showed an improvement in their quality of life by 0.0528 in the utility score (p< 0.001). By contrast, patients in the CG showed no differences in their quality of life by 0.0022 in the utility score (p= 0.815). We obtained an ICER of € 8,542/QALY and € 6,777/QALY for the first and second scenario respectively, and a MRF as dominant strategy for the third, fourth and fifth scenario. For a willingness to pay of € 30,000/ QALY, the probability of the MRF being cost-effective, compared to usual dispensing, is in a range between 98.2% and 100% for the five scenarios. Conclusions: MRF is an effective intervention for optimizing prescribed medication and improving the quality of life of older people with polypharmacy in community pharmacies. The results from the cost-utility analysis suggest that MRF is cost-effective. PIH44 Retrospective Analysis of the Economic Burden of Long-Term Care Facility Residents Diagnosed With Alzheimer’s Disease in the United States Huang A 1, Shrestha S 1, Baser O 2, Wang L 1 1STATinMED Research, Plano, TX, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . Objectives: To evaluate health care resource utilization and costs of patients diagnosed with Alzheimer’s disease (AD) and residing in long-term care facilities. Methods: A retrospective database analysis was performed using the Minimum Data set (MDS) linked to 5% Medicare data from 01JAN2008 through 31DEC2011. AD patients were identified using International Classification of Disease, 9thRevision, Clinical Modification (ICD-9-CM) diagnosis code 331.0. The first diagnosis date was designated as the index date. A comparison cohort was created for patients without an AD diagnosis, using 1: 1 propensity score matching (PSM) to control for baseline characteristics (age, region, gender, index year, baseline Charlson Comorbidity Index [CCI] score). For the comparison cohort, the index date was randomly chosen to reduce selection bias. Patients in both cohorts were required to be age ≥ 65 years, with at least two consecutive quarterly assessments in MDS data in the 6 months pre-index date and 1-year continuous medical and pharmacy benefits enrollment pre- and post-index date. Study outcomes (health care costs and utilizations) were compared between the cohorts. Results: After 1: 1 matching, a total of 2,158 patients were identified for the disease and comparison cohorts, and baseline characteristics were balanced. The AD cohort had a higher percentage of inpatient stays (33.73% vs. 24.93%, p< 0.0001), outpatient visits (92.22% vs. 89.99%, p= 0.01) and skilled nursing facility (SNF) (32.53% vs. 28.41%, p< 0.01) and hospice admissions (11.03% vs. 7.14%, p< 0.0001) than the comparison cohort. The AD cohort also incurred higher inpatient ($5,442 vs. $4,001, p< 0.001), SNF ($5,679 vs. $4,523, p< 0.01) and hospice stay costs ($3,164 vs. $2,047, p< 0.001) as well as carrier claim ($2,907 vs. $2,686, p= 0.03) and pharmacy costs ($5,043 vs. $4,722, p= 0.01), compared to the comparison cohort. Conclusions: AD was associated with higher health care resource utilization and a significantly higher economic burden. INDIVIDUAL’S HEALTH – Patient-Reported Outcomes & Patient Preference Studies PIH45 Validation of the Adherence Barriers Questionnaire (Abq) – An Instrument for Identifying Potential Risk Factors Associated With Medication-Related Non-Adherence Müller S , Wilke T IPAM, University Wismar, Wismar, Germany . . Objectives: Medication-related non-adherence is a major challenge in the reallife treatment of patients. To meet this challenge successfully, adherence interventions with a tailored approach towards patient-specific adherence barriers are needed. Therefore, a reliable and practicable questionnaire for identification of those adherence barriers in specific patients is needed. The aim of this investigation is to develop and validate such a questionnaire. Methods: The “Adherence Barriers Questionnaire (ABQ)” was developed and tested in 432 patients with atrial fibrillation in a multicenter observational cohort study. Evaluation of the questionnaire included an assessment of internal consistency as well as factor analysis. Criterionrelated external validity was appraised by comparing the ABQ score with the score of a self-report adherence measure and with a clinical parameter (time in therapeutic range (TTR) regarding INR values in the VKA-based stroke prophylaxis treatment of patients). Results: The final 14-item ABQ scale demonstrated high internal consistency (Cronbach’s alpha= 0.820). Factor analysis identified a three-factor solution, representing intentional adherence barriers with 5 items (31.9% of the variance), medication- or health care system-related adherence barriers with 5 items (13.3% of the variance) and unintentional adherence barriers with 4 items (7.7% of the variance). The ABQ correlated significantly with self-reported non-adherence (Spearman’s rho=0.438, P< 0.001) as well as TTR (Spearman’s rho=-0.161, P< 0.01). Patients with above-average ABQ scores (increased number of existing adherence barriers) were significantly (p< 0.005, Pearson Chi-Square) more likely to have a poor anticoagulation quality (TTR<60%) than patients with a lower ABQ score (44.6% versus 27.3%). Conclusions: The ABQ is a practicable, reliable and valid instrument for identifying specific barriers to medication-related adherence. Future research is required to examine the ability of the ABQ to identify patient perception/behavior changes over time which may be important for the measurement of success of adherence interventions. PIH46 Stress Load Factors in the Scope of Students Szunomár S 1, Nagy J 1, Pakai A 2, Fullér N 1, Stromájer-Rácz T 1, Boncz I 3, Oláh A 1 of Pécs, Pécs, Hungary, 2University of Pécs, Zalaegerszeg, Hungary, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . . 1University Objectives: The examination of the effects of stress in university students during the exam period, compared with demographic data. Methods: Prospective research was made in the exam period. Altogether 181 university students participated in the study, in the course of which online questionnaire were applied. In the first part of the questionnaire demographic questions were listed, while Student Nursing Stress Index (SNSI) questionnaire was applied to measure stress level. Emotions perceived in a given moment could be evaluated by the Brunel Mood Scale. In the last part of the questionnaire Marlowe- Crowne Senior Short Form Social Desirability Scale (MCSDS) was applied to evaluate behaviour desired by the society. The analysis of results was performed with SPSS 20.00 and MO Excel 2007 programs. For data analysis descriptive statistics, χ 2-test, t-test, variance analysis and regression analysis besides the significance level p< 0,05. Results: The division of genders was 28 (15,47%) male and 153 (84,53%) female participants in the research (n= 181). The average age was 21,62±3,07 years. The measurement of stress showed that sleeping time of students can be significantly affected by stress (p< 0,001). Those whose parents are divorced reached higher points in the value of stress (p=0,038). Stress load caused by the exams did not show significant difference between specialities (p> 0,05). In the course of the research we found that senior students experienced significantly more stress in the exam period than freshmen (p= 0,013). Conclusions: As the result of the measurements it can be said that stress is continuously present in students. Besides the requirements of the university students have to cope with several other problems. PIH47 Failure To Obtain the First Prescribed Refill (Early Medication NonPersistence): A Meta-Analysis of Rates and Causes of Variation in Rates By Chronic Disease Class and Analytic Methods Atkinson M J 1, Trivedi B 1, McHorney C A 2 Market Access Services, Inc., San Diego, CA, USA, 2Covance Market Access Services, Inc., Gaithersburg, MD, USA . . . . . 1Covance Objectives: Medication non-adherence is often classified by the timing of nonadherence. Primary non-adherence is the failure to fill a newly-prescribed medication (Rx). Rx non-persistence is the failure to continue therapy after the initial fill. A recent classification − early non-persistence (ENP) − is the failure to obtain the first prescribed refill of a new Rx (single dispensation only). In this meta-analytic review, we compare the rates of ENP across studies by four moderator dimensions: (1) chronic disease class; (2) symptomatic vs. asymptomatic chronic condition; (3) length of the baseline treatment-free interval; and (4) whether ENP estimates were adjusted for Rx switches. Methods: Fifty-eight studies that contained data on ENP were identified using a PubMed literature search and searches of each article’s reference list. ENP rates were recorded for 91 distinct samples. ENP was defined as the failure to obtain the first prescribed refill within 30 days of the first-refill date. ENP rates were weighted by sample size and combined to provide pooled fixed effect size estimates for the moderator dimensions. Results: Across all samples, the overall weighted ENP rate was 23.6%. Observed difference between ENP rates by disease class were largely explained by whether the treatment focused on symptom control or not: symptomatic ENP rate= 39.5% vs. asymptomatic ENP rate= 17.7%. ENP rates were affected by two aspects of methodology: (1) length of baseline treatment-free interval (shorter, favoring treatment-experienced=17.1% vs. longer, favoring treatment-naïve=27.2%); and (2) Rx switches accounted for in the ENP estimates (accounting for switch= 18.8% vs. not accounting for switch=26.0%). Conclusions: ENP is as high, and can be higher than primary non-adherence. Most extant studies simply document the rate of ENP. Given that ~ 24% of patients are “one-and-done,” it is imperative to: (1) understand the drivers of ENP and (2) develop patient-centered interventions to stem the epidemic of ENP. PIH48 Patients’ Acceptance of Their Medication: Results From a French Multi-Diseases Study With Patient Online Community Using the Acceptance By the Patients of Their Treatment (Accept©) Questionnaire Gilet H 1, Chekroun M 2, Arnould B 1 . . . 1Mapi, Lyon, France, 2carenity.com, Paris, France Objectives: Lack of adherence and persistence can be major barriers to treatment efficiency in real world, for many chronic diseases. Measuring patients’ acceptance of their medication is thus gaining importance as it is likely to help better understand and predict patients’ behavior towards treatment. The generic ACCEptance by the Patients of their Treatment (ACCEPT) questionnaire was developed to measure patients’ acceptance of their medication. The objective of this study was to evaluate for a variety of diseases the level of acceptance in real life using a patient online community. Methods: This study was observational, cross-sectional, conducted through the French Carenity platform. All patients connected were invited to complete an online questionnaire including demographics, chronic disease and treatment, and the 25 ACCEPT items. ACCEPT includes 6 multi-item acceptance dimensions (Medication Inconvenience, Long-term Treatment, Regimen Constraints, Side Effects, Effectiveness and General; range 0-100; higher score= greater acceptance) and one single-item acceptance dimension (Numerous Medications; range 1-3). Patients included in the analysis were suffering from any chronic diseases with at least 50 respondents and currently receiving a treatment for this disease. Results: Responding patients had breast cancer (n= 57), type 1 diabetes (n= 101), type 2 diabetes (n= 213), fibromyalgia (n= 135), rheumatoid arthritis (n= 98), multiple sclerosis (n=260), ankylosing spondylitis (n=134) or bipolar disorder (n=65). Most respondents were female (49% to 100%), with mean age 44 to 61. Mean (SD) ACCEPT General score was: 36 (33) for breast cancer, 64 (31) for type 1 diabetes, 54 (32) for type 2 diabetes, 35 (31) for fibromyalgia, 39 (31) for rheumatoid arthritis, 50 A513 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 (31) for multiple sclerosis, 34 (33) for ankylosing spondylitis and 45 (33) for bipolar disorder. Conclusions: This study enabled ACCEPT data to be collected in real life for a variety of chronic diseases. These data can be of major interest to help evaluating and interpreting level of acceptance in future studies. PIH49 Attitude Change Among 18-19 Years Old Boys After School-Drug Prevention Program Danku N 1, Csákvári T 2, Vajda R 1, Boncz I 3 1University of Pécs, Pécs, Hungary, 2University of Pécs, Zalaegerszeg, Hungary, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . Objectives: The aim of our study was to explore the efficiency of “FÜGE” school-drug prevention program based on the students attitude change. Methods: The research program was made at the Zipernowsky Károly Secondary Technical School with a standard questioner. Pre-test was made 10 days before and post-test 10 days after the school-drug prevention program. 94 people participated in the research. We processed our data with the help of MS excel 2007 and we prepared a T-test with it. Results: The results showed that the rate of those, who know everything about drugs have grown from 52,1% to 59,6% and those, who did not know anything decreased from 3,2% to 2,1% (p= 0,181) before and after the drug-prevention program. The effect of the “FÜGE” program also changed the student’s sense of danger regarding drugs significantly. There was a significant change (p= 0,008) in the trying of marihuana and hashish. Based on the given answer’s T-tests there were also significant differences in the test of hallucinogenic drugs (p= 0,012) and amphetamine, speed (p= 0,046). There was an almost significant (p=0,071) correlation regarding the occasional usage of amphetamine, speed. Furthermore, after the program was made, significantly more students believes that trying (p= 0,001), using occasionally (p= 0,050) and using regularly (p=0,003) herbal drugs is dangerous. However, less than 7% percent of the students agreed the statements: “I have learned a lot from the occasions.” and “I received answers for a lot of question that I was interested in earlier.” Conclusions: “FÜGE” program is considered successful, because the participants’ knowledge increased, their sense of danger changed, but only half of them had a positive opinion about the program’s information-amount. We have to state that the changes in the knowledge about drug usage not necessarily lead to changes in behavior. PIH50 Patient Perspective: Pro Compliance and Effective Reminder Strategies Ross J , Holzbaur E , Wade M , Rothrock T Almac Clinical Technologies, Souderton, PA, USA . . . . Objectives: Survey data results providing patient perspectives on compliance and reminder-use in studies including Patient Report Outcomes (PROs) is shared. This presentation looks at patients’ preferred reminder modes, identifies what may impact diary compliance, and evaluates reminder strategies. Methods: The survey was conducted in 2013, including patients globally (N=405) who participated in at least one clinical trial in the past two years with patient diaries. Patients were asked about their most recent diary experience (including compliance), future trial participation (including preferences for receiving reminders-content/modality/timing), and personal technology behaviors. Results: Only 53.6% of patients reported always being compliant with completing diary entries. Factors associated with non-compliance were age, dissatisfaction, and hard to remember diaries. Reasons for non-compliance included: “They Forgot” (51.4%), “Too Busy” (41.1%), “Access” (27.6%), and “Other” (2.7%). Patients (77.2%) provided high attractiveness ratings for reminders in future trials. Preferred reminder modes include: text messages (67.2%), hand-held alerts (34.3%), phone calls (34.1%), calendar alerts (32.6%), and email (6.2%). The majority of patients want to receive reminders for: diaries (97.3%), appointments (95.8%), and medication (95.0%). Most patients indicated checking text messages and emails daily. Significantly more patients check text messages immediately compared to email suggesting that text messages may be more effective for reminding patients. Majority reported owning some mode of technology; therefore, reminder strategies should be tailored to patient mode availability. Conclusions: As self-reported by patients, forgetting is the top reason for non-compliance which can be highly prevalent in trials. Results show patients feel reminders are helpful. Patients would like to be reminded of various events. Reminder strategies should account for patient preferences and mode availability to keep patient satisfaction high. Effective reminder strategies can positively affect compliance, satisfaction and patient experiences in clinical trials. Reminder practices can also be rolled out to benefit compliance in health care. PIH51 Using the Analytic Hierarchy Process To Derive Health State Utilities From Ordinal Preference Data Reddy B 1, Kind P 2, Adams R C 1, Walsh C 3, Barry M 4 1National Centre for Pharmacoeconomics, Dublin, Ireland, 2University of Leeds, Leeds, UK, 3Trinity College Dublin, Dublin, Ireland, 4HSE Medicines Management Programme, Dublin, Ireland . . . . . . Objectives: EQ-5D is a standardised instrument for use as a measure of health outcome. There are well-documented problems regarding how best to measure worse than dead states using the Time Trade Off (TTO) approach. We attempted to generate utilities from ordinal relationships between health states (HSs) to overcome these issues, using the Analytic Hierarchy Process (AHP) approach. AHP is a multiple criteria decision analysis technique based upon pairwise comparisons, useful for structuring complex prioritisation problems. It has been widely used in health settings. Methods: The technique outlined was applied to the Measurement and Valuation of Health (MVH) study dataset. The number of occasions that each HS was preferred to each of the others was measured and represented by a 45*45 matrix. This was subsequently transformed into a scale indicating the significance of the difference between each pair of HSs. We describe 5 approaches to structure pairwise comparisons of HS preference (2 concave, 2 convex, 1 linear). A score for each HS was derived from this matrix’s principal eigenvector, and the matrix’s consistency calculated. Results: All approaches predicted the rankings of HSs found in the MVH report well. However, the utilities subsequently derived followed an unconventional, bunched shape compared to the original study. By optimizing the parameters in order to minimize the sum of squared errors between approaches, a more suitable approach (“Beta rank fit”) was identified. Utilities could in principle therefore be derived using this method alone, without recourse to TTO models. Conclusions: This paper outlines an approach that may be suitable for converting ordinal preference data into cardinal utilities, and offering a number of advantages over previously described approaches. Ranking exercises for participants are considerably easier to carry out than full TTO studies, so the approach may be suitable for resource limited settings or for underrepresented subpopulations. PIH52 A Utility Algorithm forthe Pressure Ulcer Quality of Life – Utility Instrument (Puqol-Ui) Czoski Murray C J 1, Meads D M 1, McCabe C 2, Edlin R 3, Rutherford C 4, Hulme C T 1, Nixon J 1 of Leeds, Leeds, UK, 2University of Alberta, Edmonton, AB, Canada, 3University of Auckland, Auckland, New Zealand, 4University of Sydney, Sydney, Australia . . . . . . . . . . 1University Objectives: Pressure Ulcers are an important health care problem, recognized as ‘Never Events’ by the US Government. To date, there is no instrument to capture their health utility impact, or the value of treatments and prevention strategies. The Pressure Ulcer Quality of Life Utility Instrument (PUQol-UI) is a condition specific preference-based measure designed to capture the impact of having a pressure ulcer (PU) on an individual’s health related quality of life and will allow calculation of QALYs necessary for cost-effectiveness analyses. PUQol-UI consists of 7 domains (Pain, Mobility, Activities of Daily Living, Energy, Depression, Burden and Social Function). Each domain has three possible response levels: ‘No Bother’, ‘Little Bother’, and ‘A lot of Bother’. Methods: A valuation exercise obtained Time TradeOff values for 51 PUQol-UI health states in 200 interviews with the UK General Population. OLS, Random Effects and Fixed Effects linear regression models were fitted and evaluated using tests of standard goodness of fit and estimation and validation sample predictive performance. Results: The Random Effects model was superior in fit and predictive performance, with 83% of states predicted to within 0.1 of the observed mean. Preliminary analysis of the psychometric properties of the PUQoL-UI indicates adequate levels of validity and may offer measurement advantages over the generic EQ-5D measure. Conclusions: The PUQol-UI is a useful addition to the portfolio of condition specific utility measures available to researchers interested in economic evaluation of technologies for the management of pressure ulcers, and health care decision makers responsible for funding such technologies. PIH53 Time-Trade-off Modelling of Health Utility Values for Menopausal Symptoms and Their Treatment Pisa G 1, Kurz P 2, Potthoff P 1, Eichmann F 1 1Kantar Health Germany, Munich, Germany, 2TNS-Infratest, Munich, Germany . . . . Objectives: Impaired Health-Related Quality of Life (HRQoL) of women due to various symptoms of menopause impairments has been given increased importance in the past years. The objective of the present study is to estimate utility values for symptoms relevant for menopause-specific disturbances and to convert them into women’s willingness to give away months of life (time-trade-off) for relief of those symptoms. Methods: A time-trade-off (TTO) model was applied to estimate the utilities of 7 symptoms caused by menopause impairments. A German version of the QualiPause Inventory (QPI) was used for assessing the severity of the symptoms. A quota sample of 478 women, aged 45 to 60 years, was interviewed. Health states were presented to participating women on a mobile computer screen, and they were asked to specify the willingness to give away months/years for the relief of the symptoms, using time-scaled graphic slide controls visible on the computer screen. A total of 45 health states were valued out of a potential of 2,187 defined by the classification system. Logistic regression and Bayes methods were used to estimate the utility values. Utility values were converted into trade-off willingness for life months. Both methods led to almost identical results. Results: Willingness to trade-off life months for relief of symptoms ranges between Zero and 132 months of life with a median of 12 months. 25% of the women were willing to trade-off more than 45 months, 5% more than 100 months and 1% even 132 months or more. Among the more severely rated symptoms are dryness of the vagina, bleedings, and anxiety. Conclusions: Time-Trade-Off techniques can be used to estimate preference values for health states affected by menopausal symptoms and their relief. These values can be used to estimate differential outcomes of hormone replacement therapy. PIH54 Geographical Variations of Health Perception in the Us, Using Brfss Data 2012 Duccini F D M 1, Roïz J 2 . . . . 1Ensai, BRUZ, France, 2Creativ-Ceutical, London, UK Objectives: To determine whether and how the location influences the way people perceive their health in the US using BRFSS data 2012. The explained variable, general health, has five modalities (excellent (1), very good (2), good (3), fair (4) and poor (5)). Methods: Using BRFSS data 2012, a descriptive study and chi-square test have been conducted crossing the general health variable with the location variable. This variable has been combined in ten modalities (New England, MidAtlantic, East North Central, West North Central, South Atlantic, East South Central, West South Central, Mountain, Pacific and Guam/Puerto Rico). It has been followed by an ordered logit model to explain general health variable by the location using stepwise selection. Results: The study has been carried on 474,124 weighted individuals from BFRSS data 2012. The chi-square value is 962,244 and the p-value was lower than 0.001. Some regions such as Pacific, mid Atlantic, mountain and new England were found in a higher proportion among the excellent and very good health groups than in all the population (P: 11.7% and 10.6% vs 10.5%; MI: 11.9%, A514 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 11.6% vs 11.2%; M: 9.1% and 8.8% vs 8.4%; NE: 5% and 4.7% vs 4.3%). When others such as the East South Central and the West South central regions were found in a higher proportion among the good, fair and poor health groups (ESC: 5.8%, 6.7% and 9.6% against 5.6% WSC: 13.6%, 13.4% and 15.6% vs 12.5%). The coefficients found in the ordered logit model were all significant and have confirmed the descriptive study. Conclusions: The location influences the way people perceive their health in the US. Next step would be to look at other socio-demographics variables such as people’s revenue, race or education. PIH55 A Comparison of Value for Health States Worse Than Dead Between Japan and Uk Noto S 1, Shimozuma K 2, Saito S 3, Shiroiwa T 4, Fukuda T 4, Moriwaki K 5, Izumi R 1 University of Health and Welfare, Niigata, Japan, 2Ritsumeikan University, Kusatsu, Japan, 3Okayama University, Okayama, Japan, 4National Institute of Public Health, Saitama, Japan, 5Kobe Pharmaceutical University, Kobe, Japan . . . . . . . 1Niigata Objectives: To clarify the difference of value for health states worse than dead (WTD) between Japan and UK. Methods: A web survey was conducted asking respondents whether each health states is WTD before discrete choice experiment (DCE) tasks (DCE results not described). Health states were described using the EQ-5D-3L descriptive system. The 48 health states were blocked into 24 sets for DCE tasks. All respondents were asked 12 sets tasks randomly. We compared value for WTD between Japan and UK, UK’s value referred the article (Bansback et al., 2011). Results: A sample of 1242 members of the market research panel was invited by email to participate in the survey. Of these, 1085 (87%) completed all tasks. The mean age of participants were 49.5 years (SD= 16.6). High numbers of value of WTD were confirmed. 50.5% of respondents judged health state 33333 to be value of WTD, but 77.0% judged in UK. Similarly, 45.7% and 41.0% of them judged to be value of WTD for health state 33332 and 33323, respectively (72.0% and 60.0% in UK). Conclusions: Our findings suggest that Japanese value of worse health states not to be low in comparison with UK’s. On the other hand, it was thought that participants might not understand the tasks. PIH56 Patient-Reported Fall Related Health Care Services in Elderly Women Tamulaityte-Morozoviene I , Tamulaitiene M , Stukas R , Surkiene G , Dadoniene J , Alekna V Vilnius University, Vilnius, Lithuania . . . . . . Objectives: Although the falls in elderly people lead to serious health consequences, the economic burden is underestimated. The aim of this study was to examine the fall-related out-patient medical care in community-dwelling elderly women. Methods: Women aged 65 years and older who visited National Osteoporosis Center for diagnostic or treatment procedures, were interviewed by phone recording the consequences and health care procedures related to every fall sustained during the previous 12 months. Results: The study population consisted of 310 women who reported one or more fall, one in three of them had fallen twice or more. Of all women who fell, 280 (90.3%) reported their fall resulted in an injury, and 77 (15.3%) falls led to bone fractures. Fall related medical care was provided to 135 women: to 43.5% of those who fell and 48.2% of those who sustained injuries from falling received medical attention. Among these, the highest percentage reported using of out-patient medical services. The number of out-patient visits reported (535 visits in total) ranged from 1 to 13, and in 70 cases (51.9%) – from 2 to 4 visits. Different types of out-patient health care were used by 43.5% of women who fell. The majority of specialists visited were orthopaedist, surgeon, and radiologist. An ambulance was used by 11.9%, and family doctor was visited by 19.4% of fallers. The mean number of health care procedures was higher in women who sustained a fracture, as compared to those who did not: 4.9 (95% CI 4.4–5.4) and 0.67 (95% CI 0.29–0.76), respectively; p< 0.0001. Conclusions: From all self-reported falls registered in women over 65 years, 90.3% resulted in any injuries. The mean number of out-patient visits per faller was 1.73 (95 % CI 1.36–2.1). Objectives: The EuroQol-5D (EQ-5D) is a generic patient-reported outcome measure (PROM) allowing comparisons to be made across different diseases and conditions. The instrument has been used in the UK’s National Health Service (NHS) since 2009 to collect data from patients to assess the effectiveness of a number of surgical interventions. The aim of this study was to investigate whether the EQ-5D domains behave similarly across patient samples. Methods: The data were derived from published Hospital Episode Statistics (HES) for April 2013 to March 2013. The EQ-5D had been completed by patients undergoing four surgical procedures: groin hernia repair (N= 21831), hip (N= 37800) and knee replacement (N= 40429) and varicose vein repair (N= 4681). The partial credit model (Masters, 1982) was applied to the data. Uniform differential item functioning (DIF) and non-uniform DIF (criterion difference > 0.5 logits) was assessed across the four interventions, gender, age group and the interactions. Results: There was significant uniform DIF between the 4 interventions with 50% of all possible contrasts demonstrating DIF. The only domain not affected by DIF was Discomfort/Pain. There was DIF present in 2/3 of the contrasts for Anxiety/Depression, Mobility and Self-care and in 50% of the Usual Activities domain. DIF was also demonstrated across age groups for the Mobility and Anxiety/Depression domains. No DIF was found for gender. Finally, non-uniform DIF was demonstrated for age group by intervention. The Mobility domain showed the greatest degree of non-uniform DIF (20/24, 83% of the contrasts). Conclusions: The finding that the EQ-5D performs differentially depending on the patient group is an important one and means that the instrument should be used cautiously when comparisons across different surgical interventions are being made. This has potentially major ramifications for the use of the instrument as a measure of efficacy in the NHS. PIH59 Antenatal Depression and Its Risk Factors Among Women in Chengdu of China Results From A Hospital Based Survey Zhao Z M 1, Pan X F 1, Qi X R 2, Li S Q 1, Zhao Y 1, Chang H 1, Xue Q P 1, Wen Y 1, Liu X 2, Yang C X 1 University, Chengdu, China, 2West China Women’s and Children’s Hospital, Sichuan University, Chengdu, China . . . . . . . . . . . . . . . . 1Sichuan Objectives: Mental health of pregnant women is essential for maternal and neonatal health. However, there is lack of statistics of antenatal depression in China. The study aimed to investigate the prevalence of antenatal depression and explore its risk factors among pregnant women in Chengdu of China. Methods: Women at third trimester of pregnancy were screened for symptoms of depression at antenatal clinics of West China Second Hospital between 28 October 2013 to 28 February 2014 based on the Chinese version of the Edinburgh Postnatal Depression Scale (EPDS) and a psychosocial risk factors checklist. Results: A total of 2243 pregnant women aged 30.0±4.0 years participated in the survey. The mean EPDS score was 8.43 (standard deviation: 3.97). With a threshold score of 13, 14.2% were screened as having symptoms of depression. Age (P=0.007), education level (P<0.001), occupation (P=0.001), number of children (including the fetus) (P=0.018), number of miscarriage/ abortion (P=0.048), and age of first pregnancy (P=0.001) were associated with antenatal depression in univariable analysis but not multivariable analysis (P> 0.05 for all). Women who were dissatisfied with living conditions (OR= 1.81; 95% CI: 1.38-2.38), had a poor relationship with mother-in-law relationship (OR= 2.20; 95% CI: 1.652.92), and had unplanned pregnancy (OR= 1.34, 95% CI: 1.02-1.76) were more likely to show antenatal depression symptoms. Conclusions: Our study shows antenatal depression might be prevalent among Chinese women in Chengdu. Early detection and intervention for antenatal depression may be necessitated to improve maternal and neonatal health after more systematic studies and reliable data are available. PIH60 Disutility Associated With Erectile Dysfunction in the Middle-Aged Or Older Males Väätäinen S , Martikainen J A University of Eastern Finland, Kuopio, Finland . . . Objectives: To assess the long-term grading of health-related quality of life (HRQL) of care-needed elderly who had received the occupational therapy in Japan. Methods: We carried out a 2-year follow-up study of multicenter trial. The subjects were recruited from 26 nursing homes in Japan. The proxy of the subjects completed a questionnaire of the Health Utilities Index Mark3 (HUI3). We tested the long-term effect of occupational therapy and aged natural grading of care-needed elderly. Results: 55 male and 85 female subjects remained at final follow-up. The mean age of subjects was 76.9 years. The global score of HUI3 of baseline was 0.377 (SD= 0.270). Their score was improved for three month (mean score=0.418, SD=0.284), but had deteriorated to 0.328 (SD=0.324) 2-year later. Speech, Ambulation, Emotion and Cognition had deteriorated significantly among 8 attributes of HUI3. In regression analysis, higher care level significantly increased risk of deteriorating HRQL of care-needed elderly. Conclusions: Our findings suggest that occupational therapy have short-term effect for care-needed elderly. However, we cannot affirm that that occupational therapy has long-term effect. The aging may deteriorate their HRQL of care-needed elderly naturally. Objectives: Erectile dysfunction (ED) affects millions of males world-wide. While it is obvious that ED affects individuals Quality of Life, the quantifiable data on disutility associated with ED is still lacking. Moreover, the health utility – impact has not been studied using multiple health utility instruments previously. Our aim was to quantify the disutility caused by different levels of ED using two preference-based health utility indices simultaneously. Methods: A total of362 middle-aged or older (52-75 year old) males responded to the five-item International Index of Erectile Function (IIEF-5) in the cross-sectional sample of Savitaipale Study in 2007-2008. The lower score in IIEF-5 (range 1-25) corresponds to more severe ED. Health utility was assessed with two separate validated preference-based instruments, 15D and SF-6D. Data were adjusted for age, number of morbidities and marital status. Minimally important differences (0.02-0.03 for 15D and 0.04 for SF-6D) were used to guide the clinical interpration of the results. Results: Both 15D and SF-6D were significantly correlated with IIEF-5 (p<0.001). When examining the ED categorically, the adjusted marginal disutility (0.023 in 15D and 0.038 in SF-6D) was statistically significant (p< 0.05) and clinically noticiable even at the mild ED (IIEF-5 score of 22-25). The marginal disutility progressively increased with increasing level of dysfunction, and was highest among the males who had not had sexual activity in past 6 months (0.060 in 15D and 0.093 in SF-6D, p<0.001 on both). On average, a one point decrease in IIEF-5 corresponded to a 0.003 decrease in 15D (p<0.001) and 0.004 in SF-6D (p<0.001). Conclusions: Erectile dysfunction can cause a substantial disutility on males. While this condition may not be life treathening or is not considered a major public health problem societally, the marginal disutility associated with severe ED is comparable or even greater than disutility associated with many chronic morbidities. PIH58 Differential Item Functioning and the Eq-5d: Evidence From the Uk Hospital Episode Statistics PIH61 Reference Eq-5d-3l and Eq-5d-5l Data From the Italian General Population Smith A B University of York, York, UK 1University PIH57 Long-Term Grading of Health-Related Quality of Life of CareNeeded Elderly: A 2-Yr Follow-Up Study Noto S 1, Izumi R 1, Moriwaki K 2 University of Health and Welfare, Niigata, Japan, 2Kobe Pharmaceutical University, Kobe, Japan . . . 1Niigata . . Scalone L 1, Cortesi P A 1, Mantovani L G 2, Ciampichini R 1, Cesana G 1 of Milano - Bicocca, Monza, Italy, 2Federico II University of Naples, Naples, Italy . . . . . . . A515 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: No recent Italian norm EQ-5D data were available. Furthermore, norm data from the new descriptive system with 5 levels were completely missing. The main objective of the present study was to assess an Italian general population reference data using both the standard EQ-5D-3L version and the recently introduced EQ-5D-5L. Methods: Large-scale telephone survey was conducted in November 2013 on 6,800 subjects from the general population of the Lombardy region, with 9.8 million residents. They were recruited to be representative of the Lombardy general adult population as regards age (from 18 years), gender and geographical distribution. Each participant underwent a telephone interview including the Italian version of the 5L and 3L descriptive system, then, to minimize memory effects, between the two descriptive systems the participants were asked to report their socio-demographic data, and finally they answered the question on the visual analogue scale (VAS). The data collected with the 3L and 5L version descriptive system were converted into utilities. Results: Participants were 48% male with a mean (SE) age of 51.9 (0.21). Around half (51.3%) of the participants specified they have a paid or unpaid work, 15.8% were housewives, 6.2% students were, 5.3% idles and 26.5% retired. Overall no problems were reported by 86.5% (3L) and 84.2% (5L) with mobility, by 96.1% (3L) and 94.2% (5L) with self-care, by 88.0% (3L) and 84.9% (5L) with usual activities, by 58.4% (3L) and 52.8% (5L) with pain/discomfort, and by 66.5% (3L) and 61.7% (5L) with anxiety/depression. The mean (standard error) and median VAS was 78.2 (0.2) and 80. Mean (SE) utility index obtained from both the 3L and the 5L versions was 0.915 (0.001). Conclusions: Reference EQ-5D-3L and EQ-5D-5L data on the Italian general adult population are now available. Although these data were collected in the Lombardy region we can consider our results a good proxy of the full Country. PIH62 Patient Preferences: Pro Mixed Modes – Epro Versus Paper Ross J , Holzbaur E , Wade M , Rothrock T Almac Clinical Technologies, Souderton, PA, USA . . . . Objectives: This presentation expands on a previous ISPOR presentation on patient acceptance of the use of Mixed Modes for collecting PROs in trials. The ISPOR PRO Mixed Modes task force recommends when mixing modes to avoid mixing paper with ePRO. However, interest in using paper for PRO collection still exists. This presentation will investigate survey data to examine if patients prefer ePRO over paper. Methods: The research (conducted in 2013) includes patients globally (N= 405) who participated in at least one clinical trial requiring patient diaries in the past two years. Patients were asked about previous diary experiences and future trial participation. The previous presentation showed most patients are in favor of mixed modes–mainly due to being able to choose their preferred mode. This presentation focuses on patients with prior experiences with both paper and ePRO (N=167). Results: Of paper/ePRO experienced patients, 77.3% preferred ePRO; 76.1% had high agreement that ePRO makes dairy participation easier; 73.1% had high agreement that ePRO-use makes them more willing to participate in future diaries. Of patients who prefer ePRO, those who used ePRO in their most recent trial had significantly higher satisfaction ratings (87.0%) than those who used paper (55.2%), p< 0.001. Low agreement ratings were associated with dissatisfaction and longer times per diary entry. Conclusions: These findings show most patients prefer ePRO and satisfaction rates are higher when patients use their preferred ePRO mode. Sponsors should consider using ePRO due to patient preference, as higher satisfaction is associated with optimal compliance and data quality when implemented appropriately. As lower agreement was associated with dissatisfaction and longer times per entry, this indicates there may have been issues with ePRO implementation or instrument selection. Proper implementation planning should include appropriate ePRO mode/instrument selection, ensuring ease of use while keeping patient burden low and satisfaction high. PIH63 Implementation of An Ambulatory Pharmacist-Managed Anticoagulation Clinic In Qatar: Development of A New Service and A Pilot On Patients’ Satisfaction and Quality of Life Awaisu A , Kheir N , Mohamed Ibrahim M I , Al-Taweel H M , Elmubark A E Qatar University, Doha, Qatar . . . . . . . . Objectives: Pharmacist-managed anticoagulation clinics have been shown to improve the quality of life (QoL) of patients receiving anti-clot treatment. The first pharmacist-managed anticoagulation clinic in Qatar was established at Al-Wakrah Hospital in March 2013. This study aims to report the development of a new pharmacist-managed service and to determine the patients’ satisfaction with the new service and their overall QoL using a validated instrument called Duke Anticoagulation Satisfaction Scale (DASS). Methods: A new pharmacistmanaged anticoagulation clinic was successfully develop through agreements with physicians on the scope of the service. A prospective cross-sectional study using 25-item DASS QoL instrument was conducted at the Anticoagulation Clinic of Al-Wakra Hospital. An Arabic-translated version of the tool that was conceptually equivalent to the original English version was developed through linguistic validation and cultural adaptation processes. Each item was assessed using a 7-item Likert-type scale with lower values indicating a better QoL and greater satisfaction. The primary outcome measures were QoL and satisfaction. Results: Of the 50 patients attending the anticoagulation clinic, 25 consented to participate in the study. The mean total QoL score of the population was 66±24 (range 34-118), indicating modest QoL. Male patients reported a better QoL than female patients (61.7 ± 19.5 vs. 73.3 ± 30.7; p= 0.255). Furthermore, participants who were naïve to anticoagulation treatment showed better QoL compared to non-naïve participants (61.3±22.3 vs. 80.3±26.0; p= 0.093). However, these differences did not reach statistical significance. Conclusions: Patients receiving anticoagulation service managed by pharmacists in Qatar have expressed satisfaction with the service and a modest QoL that was comparable to what has been reported in the literature. Additional studies with larger samples are required to further document the value of the new service. PIH64 Family Preferences in the Volume VerSUS Outcome Debate: Implications for the Delivery of Complex Pediatric Care O’Leary G 1, Lockhart A 1, Mullenger R 2, Warren A 2, Hancock Friesen C 2, Levy A 2, Molinari M 2, O’Blenes S 2 1IWK Health Centre, Halifax, NS, Canada, 2Dalhousie University, Halifax, NS, Canada . . . . . . . . Objectives: A Relationship between volume and outcome for complex medical procedures has been used as an argument for regionalization; however, this must be balanced against preferences to have care delivered close to home. The objective of our study was to determine how families trade-off variations in risk against the ability to have complex pediatric care delivered locally. Methods: Twenty parents of children without serious medical problems seen in an outpatient clinic participated in a probability trade-off experiment involving two scenarios in which they were asked to imagine their child required a complex medical procedure (‘lowrisk’= 5% mortality, ‘high-risk’= 30% mortality) available locally or at an alternate large center 2.5 hours away by air. Numeric and graphic representations of mortality risk were reduced in a stepwise fashion for procedures performed at the alternate center. Thresholds at which participants chose to travel were identified. Participant’s decisions were then challenged by increasing the costs incurred by travelling to the alternate center. Results: In the low-risk scenario, participants chose not to travel until absolute risk was reduced by 2±0.2% (relative risk reduction of 39±3%). In the high-risk scenario, a larger absolute risk reduction (5.1±0.8%, p= 0.0001) but smaller relative risk reduction (17±3%, p= 0.0001) triggered a decision to travel. In the low-risk scenario, only 2 of 8 participants with household income > $100,000/ yr changed their decision to travel when faced with additional costs; however 8 of 12 with lower income changed their decision (p= 0.07). In the high-risk scenario, 1 of 8 in the high income group changed their decision compared to 7 of 12 in the low income group (p= 0.04). Conclusions: Many families would trade substantially higher risk to have complex pediatric care delivered locally. These results have implications for policy development related to delivery of complex care at smaller children’s hospitals located far from large urban centers. PIH65 Evaluating Prevalence of Self-Medication in Bahawalpur Masood I 1, Ahmad M 1, Khan M S 2, Minhas M U 1 Islamia University of Bahawalpur, Bahawalpur, Pakistan, 2The Islamia University of Bahawalpur, Punjab-Pakistan, Bahawalpur, Pakistan . . . . . . 1The Objectives: Aim of this study was to determine the prevalence and pattern of selfmedication among different classes in Bahawalpur community Methods: It was a cross-sectional descriptive study targeting residents of Bahawalpur including almost every class and gender. Sample size was calculated and 10% was added to encounter non response, respondents were selected through convenience sampling method. The data was collected using a pre-tested self-administered questionnaire. The data collection tool was tested and restructured after a pilot study on a small number (10% of the calculated sample) of population was tested and re-structured. The data was analyzed using SPSS version 15 and the results were tabulated Results: A total 420 of the participants responded including literate 280 (66%) illiterate 140 (33%). Most of the respondents were motivated towards self-medication due to high cost of prescription medicines (n=312; 74.3%), weak trust on physicians (n= 404; 96.2%) and drug sellers (n= 217; 51.7%). Significantly high percentage of medical professionals (n=111,77.6%; p=008) had opinion that self-medication gives desired results as compare to respondents with no-medical background (n= 180; 65%). Conclusions: It was concluded that self-medication is common among the residents of Bahawalpur and prevails more among literate and medical health care professionals as compare to illiterate and those not with medical background. PIH66 A Systematic Review To Identify the Use of Preference Elicitation Methods in Health Care Decision Making Janus S I M 1, Weernink M G M 1, van Til J A 1, Raisch D W 2, van Manen J G 1, IJzerman M J 3 1University of Twente, Enschede, The Netherlands, 2University of New Mexico College of Pharmacy, Albuquerque, NM, USA, 3University of Twente and MIRA institute for Biomedical Technology & Technical Medicine, Enschede, The Netherlands . . . . . . . . . . . . . . Objectives: Preference elicitation methods (PEMs) offer the potential to increase patient-centered medical decision-making (MDM), by offering a measure of benefit along with a measure of value. Preferences can be applied in decisions on: reimbursement, including health technology assessment (HTA); market access, including benefit-risk assessment (BRA), and clinical care. The three decision contexts have different requirements for use and elicitation of preferences. The aim of this systematic review was to identify studies that used PEMs to represent the patient view and identify the types of health care decisions addressed by PEMs. Additionally, PEMs were described by methodological and practical characteristics within the three contexts’ requirements. Methods: Search terms included those related to MDM and patient preferences. Only articles with original data from quantitative PEMs were included. Results: Articles (n=322) selected included 379 PEMs, comprising matching methods (MM) (n= 71, 18.7%), discrete choice experiments (DCE) (n= 96, 25.3%), multi-criteria decision analysis (n= 12, 3.2%), and other methods (i. e. rating scales), which provide estimates inconsistent with utility theory (n= 200, 52.8%). Most publications of PEMs had an intended use for clinical decisions (n= 134, 40%), HTA (n= 68, 20%), or BRA (n= 12, 4%). However, many did not specify an intended use (n= 156, 41.1%). In clinical decisions, rating, ranking, visual analogue scales and direct choice are used most often. In HTA, DCEs and MM are both used frequently, and the elicitation of preferences in BRA was limited to DCEs. Conclusions: Relatively simple preference methods are often adequate in clinical decisions, because they are easy to administer, give fast results, place low cognitive burden on the patient, and low analytical burden on the provider. MM and DCE fulfill the requirements of HTA and BRA but are more complex for the respondents. There were no PEMs that had low cognitive burden, and strong methodological underpinnings which could deliver adequate information to inform HTA and BRA decisions. A516 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PIH67 Comparison of Equity Weights of Life Year Gains: A Discrete Choice Experiment for Japanese and Korean General Public Goto R 1, Mori T 2 1Kyoto University, Kyoto, Japan, 2Konan University, Kobe, Japan . . Objectives: Setting priorities with limited public resources has gained heated interests worldwide. Weighting health gains differently for different groups in the population is another manner to consider equity in cost-effectiveness analysis. However, there is only a few empirical analysis eliciting general public preference. This research is to compare equity weights of Japanese and Korean. Methods: We conducted a web-based survey in Mar 2013 including a discrete choice experiment (DCE) to elicit general publics’ equity weight for life gains of those from different groups. We selected attributes and designed this experiment following manners used in Norman (2013). Thus, we analyzed weights according to the difference of gender, smoking status, life style, caring status, income and age. Results: 1,280 Japanese and 580 Koreans completed questionnaires and were eligible for analysis. Japanese put higher weight on male (p< 0.001), non-smokers (p< 0.001), those with lower income (p< 0.001), carer (p< 0.001) and those with an expected age of death less than 45 years (p< 0.001). Korean have the same patterns of preference according to income (p< 0.001), caring (p< 0.001) and smoking status (p= 0.026). However, they equally consider groups from different gender (p= 0.331) and age groups. For both countries, respondents tend to prefer groups with same characteristics as them. Conclusions: People from two Asian developed countries with universal health insurance shows different equity weights. These may reflect the variations of cultural backgrounds and coverage of health care services. PIH68 How Iranian People Think About Generic Substitution? Mehralian G 1, nourmohammadI S 2, Yousefi N 1, Peiravian F 1 1Shahid Beheshti University of Medical Sciences, School of Pharmacy, Tehran, Iran, 2Azad University, Tehran, Iran . . . . Objectives: The growth of pharmaceutical expenditure and patients’ out of pocket has motivated researchers to explore underlying factors affecting on generic substitution both in developed and developing countries. The purpose of this study is therefore to explore how Iranian people think about generic medicines and what underlying factors should be taken into account by policy makers to promote the culture of generic substitution. Methods: A cross-sectional descriptive study inviting Iranian people was performed using a self-administrated anonymous questionnaire. Besides the demographic section, 34 items of developed questionnaire were categorized to 7 main factors including: experience of patients, efficiency of medicines, cost of medicines, physician’s role, pharmacist’s role, negetive perception of patients and government interventions. Results: After analyzing 1310 completed questionnaires, results showed the among the aforementioned factors the government interventions has the first priority to encourage patients to use of generics medicines instead of brand medicines and followed by physician’s role, pharmacist’s role, efficiency of medicines, cost of medicines, experience of patients and negetive perception of patients. Conclusions: In conclusion, the trust of Iranian’s society to the government, physicians and pharmacists would be a worthy opportunity to reduce health care expenditure as well as patients’ out of pocket by taking evidencebased decisions toward promotion of generic substitution. PIH69 Assessment of Health States and Erectile Dysfunction-Associated Quality of Life Among Adult Males and Females With Male Partners in Germany, the United Kingdom and the United States Menegatou E 1, Claeys C 2, Wasserman D 2, Tomaszewski K 2 1KJT Group, Inc., Amsterdam, The Netherlands, 2KJT Group, Inc., Honeoye Falls, NY, USA . . . . Objectives: This study assessed the quality of life associated with various states of erectile dysfunction (ED) among adults: both diagnosed ED sufferers and nonsufferers in Germany, the United Kingdom and the United States. Methods: A multi-national, cross-sectional online survey was conducted among a representative sample of 2,000 adults (n= 500 Germany, n= 500 UK, n= 1,000 US) with an equal number of men and women in a heterosexual relationship. Respondents rated either their or their partner’s erectile function, and health state utilities were measured using standard gamble (SG) and visual analog scale (VAS) in counterbalanced order. Utilities were estimated for one level from the Erection Hardness Scale: penis gets hard but not hard enough for penetration. Differences were examined by measure, country and respondent demographics. Results: The SG and VAS measures yield similar, but not identical mean estimates for the ED health state. Significant variation exists by measure as well as by country. In comparing the utility assessments between the countries using SG, German respondents have significantly higher average utility (0.49) for ED than do respondents from the UK (0.40), and US (0.41). When examining differences by gender, females have greater utility for the ED health state when compared to males. Men with ED report a lower utility for the ED health state when compared to men without ED. By contrast, females whose partners have ED report a higher utility for the ED health state when compared to females with partners without ED. Although utility is consistent across most ages, an increase in utility exists for adults aged 75 and over. Conclusions: Notable differences in ED utility emerged by country: Germans have higher utility for ED when compared to UK and US respondents. Significant differences in ED utility between males and females suggests there is an important relationship disconnect between men with ED and their female partners. PIH70 Does Price Matter? The Impact of Cost Information On Patient Decision Making Chan J Y 1, Butt T 2 School of Economics, London, UK, 2University College London, London, UK . 1London . . Objectives: In publicly funded health systems such as the United Kingdom (UK) National Health Service (NHS), patients do not normally face the full economic cost of treatment decisions, nor are they aware of the potential cost to the system. We investigated whether patient awareness of treatment costs, either to the system or to themselves, would affect treatment choices. Methods: 344 representative members of the UK public were recruited via an online survey panel. Respondents were required to make treatment decisions in three different health conditions (sore throat, psoriasis and sciatica). Respondents were presented with condition-specific patient decision aids (Option Grids™), each supported by: 1) no cost information, 2) cost to the NHS (drug/ procedure tariff), 3) cost to patient (drug/ procedure tariff), and 4) access cost to patient (flat cost for all options). Differences in treatment choices were explored using ANOVA. Significant differences within each health condition were subsequently explored using t-tests. Results: A significant number of respondents switched choice to the cheapest intervention when tariff costs to either the system (p< 0.05) or themselves (p< 0.01) were considered versus no cost information when choosing between treatments for psoriasis. For all three health conditions, presenting flat access costs increased the likelihood (p< 0.01) of respondents choosing the treatment option known to have the highest tariff price. Conclusions: Cost information influences treatment decisions. We observed that awareness of cost to the system or to oneself encouraged the choice of lower priced treatment options, whereas flat access charges encouraged the choice of treatment known to be more expensive. Provision of cost information may therefore be important for informed decision making, and could also be a policy tool to generate cost savings for the health system. PIH71 Health Literacy and Self-Reported Health Status Using the Eq5d-5l: An Exploratory Analysis Rey-Ares L 1, Augustovski F 2, Irazola V 3, Garay O U 4, Gianneo O 5, Fernández G 5, Morales M 5 for Clinical Effectiveness and Health Policy (IECS), CABA, Argentina, 2Economic Evaluations & HTA Department; Institute for Clinical Effectiveness and Health Policy (IECS) and Professor of Public Health, University of Buenos Aires, Buenos Aires, Argentina, 3Institute for Clinical Effectiveness and Health Policy, Buenos Aires, Argentina, 4IECS Institute for Clinical Effectiveness and Health Policy, Buenos Aires, Argentina, 5Fondo Nacional de Recursos, Montevideo, Uruguay . . . . . . . . 1Institute Objectives: To describe health literacy (HL) in Uruguayan general population and its relation with self-reported health status. Methods: As part of an ongoing Uruguayan EQ-5D-5L valuation study, we included the Short Assessment of Health Literacy-Spanish questionnaire (SAHL-S), a previously validated instrument that evaluates HL through 18 items combining word recognition and comprehension. Low HL is defined by identifying ≤ 14 correct items. We included participants with valid SALH-S responses, complete sociodemographic characteristics, self-reported health status with the EQ-5D-5L, and report of previous experience with illness. This preliminary analysis describes sociodemographic characteristics, HL and selfreported health status and explores the independent association between EQ visual analogue scale (VAS) score and HL using standard linear regression. Results: Of 773 participants 60.2% were women (mean age 42.02 years; SD: 15.51). VAS mean was 79.34 (DS: 16.39). 52.9% participants had at least one limitation in any of the EQ-5D domains, 75.9% had experience with illness and 51% in caring others. Educational attainment (EA) distribution was 17.2% up to primary, 52.3% up to secondary and 30.5% up to tertiary or higher education. Low HL was present in 39.8% of the population. In bivariate analysis aging and low HL were associated with poorer VAS scores (coef -0.276; p= 0.000; coef -3.028; p= 0.012). Higher VAS scores were observed with higher EA (coef 2.832; p= 0.001). Multiple regression shows HL is related to VAS independently of age, but this association loses its statistical significance -becoming borderline- after adjusting for EA and experience in caring others (coef -1.93; p= 0.098). Conclusions: HL is a recently developed construct that combines formal education and acquired knowledge related to health. This is the first study that describes HL in Uruguay, and shows that is associated with self-reported health. Further studies are needed to explore the potential value added to standard educational level measurement. PIH72 Assessing the Translatability of the Term “Frustrated” McKown S 1, Angun C 2, Talbert M 3, Brandt B A 4, Gawlicki M C 4 1Corporate Translations, Inc., Chicago, IL, USA, 2Corporate Translations Inc., Odunpazarı, Eskişehir, Turkey, 3Corporate Translations Inc., Chicago, IL, USA, 4Corporate Translations, Inc., East Hartford, CT, USA . . . . . . . Objectives: The objective of this study was to assess the translatability of “frustrated,” a term commonly used to describe a range of emotions in Clinical Outcomes Assessments (COA). “Frustrated” includes many constructs, such as “discouragement,” “anger” and “upset.” Previous studies have shown that terms including multiple constructs in English, such as “bother,” are not sufficiently translatable across all languages. Methods: Back-translations of questionnaires containing the word “frustrated” were analyzed to assess the translatability of the term. The following related constructs were also included in analysis: “discouraged,” “angered,” “disappointed” and “upset.” Data collection forms resulting from cognitive debriefing were also analyzed to determine subjects’ interpretation of “frustrated” as translated in other languages. Results: “Frustrated” proved to be very problematic for 13 out of the 24 languages in this study, most notably for Eastern European, Indian and Asian languages. For example, “frustrated” was translated as “indignant” in Greek, “disillusioned” in Hungarian, “irritated” in Japanese and “discouraged” in Korean. Out of 245 subjects, 13% took issue with “frustrated,” indicating that it was not understood or not appropriate for their languages. Analyses of related constructs showed that “discouraged” and “angered” were best suited for use, as these terms were translated with no issues in all 12 languages available for analysis. “Upset” was found to be equally problematic, and thus rejected as a recommended construct. Conclusions: “Frustrated” is not recommended for use in COAs intended for international data pooling. Similar to the findings of previous studies, more spe- A517 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 cific terms, such as “discouraged” and “angered,” translate with greater conceptual equivalency. Therefore, when seeking to measure the various concepts associated with the term “frustrated,” measuring more specific constructs independently using separate questionnaire items is recommended. PIH73 Mobile Phone Use in Patient Reported Outcomes– An Updated Literature Search O’Gorman H Exco InTouch, Nottingham, UK had “symptoms” as a primary endpoint; a drug for rheumatoid arthritis (RA) had “functioning” as its lead secondary endpoint; the remaining six drugs (for pulmonary arterial hypertension (PAH), Crohn’s Disease, smoking cessation, Myasthenia Gravis, asthma, and overactive bladder) had “HRQOL”, “symptoms, and “functioning” as minor secondary endpoints. Three drugs -indicated for PAH, seizure, and RA- had PRO claims in their labels. Conclusions: Although not yet prominent in Japan, PROs are used in drug clinical trials and label claims. Symptoms, Quality of Life, and Functioning are the most common PROs used. . Objectives: To demonstrate the increasing use of mobile phones to collect patient reported outcomes in research as a valid method of data collection. Methods: A literature search was conducted looking at articles published between 2009 and 2014 that referenced electronic diaries of some description. Articles were pulled out that specifically referenced mobile or cellular phones. Results: 39 of out of 191 articles found specifically referenced mobile. The studies referenced were carried out on populations with an age range of 8 years up to 80 (mean 35.4; SD 16.6) and were split into 15 therapy areas including metabolic and genetic disorders, pain, weight management, sexual activity, respiratory, multiple sclerosis and gastroesophageal reflux disease. Population size ranged from 12 to 994 (mean 208.3; SD 269.2), and subjects reported for a minimum of 7 days (up to 6 reports per day) to a maximum of 2 years (mean 154.3 days; SD 170.6). Notably, 18 out of the 39 studies allowed the subjects to use their own mobile phone for the reporting and 19 articles referenced smartphones specifically. Conclusions: All concluded that mobile phones were suited to collect data from subjects. It was noted that the use of mobiles was acceptable as they are used them in everyday life and found to be convenient; the technology was also inexpensive to implement. The fact that 46.2% of the studies allowed the subjects to use their own mobile phones for the reporting emphasises the practicality of using mobile phones in patient reported outcomes. Although the mean age of all the studies was relatively low, the age range was very wide and researchers can be confident that older populations could use mobile phones to collect these data. The technical evolution of mobile technologies and ubiquitous nature show that this technology is a valid means to collect patient reported outcomes. PIH74 Regulatory Issues in Pro Advertising: A Review of the Ddmac/Opdp Letters From 1998 To 2013 To Identify Pro Claims Violations and Examine Their Evolution Over Time Acquadro C 1, Regnault A 2, Arnould B 2 Research Trust, Lyon, France, 2Mapi, Lyon, France . . . PIH76 Comparing the Equivalence of Eq-5d-5l Across Different Modes of Administration O’Gorman H 1, Mulhern B 2, Brazier J 2, Rotherham N 1 InTouch, Nottingham, UK, 2University of Sheffield, Sheffield, UK . . . . 1Exco Objectives: Interest in delivering Patient Reported Outcome Measures (PROMs) using mobile devices (e-PROMs) has increased in recent years. However there is debate about the level of equivalence between the traditional pencil and paper and electronic modes of administration. The aim of this study is to compare the equivalence of delivering a widely used generic PROM (EQ-5D-5L) pencil and paper and mobile phone administration modes. Methods: A mobile version of the EQ-5D-5L was developed with guidance from the EuroQol Group. Two hundred respondents from a research cohort of people in South Yorkshire were identified, and randomly allocated to one of the administration modes based on stratifications for age and gender (and across a range of self-reported health issues). The EQ-5D-5L was completed either using a mobile device or the standard paper version which were sent out to the respondent. Follow up usability questions were also included. EQ-5D equivalence was compared at the dimension and utility and VAS score level using ANOVA. Results: Response rates were comparable across the arms, with the majority of respondents owning a smartphone. The mean EQ-5D-5L utility and VAS scores and the frequency of respondents endorsing the individual EQ-5D-5L categories across each of the dimensions does not differ across the administration modes. The majority of the mobile phone completion sample agreed that the mobile version of EQ-5D-5L was easy to complete, and that the phone was easy to use, and that they would complete e-PROMs again. Conclusions: Completing e-PROMs using mobile phones produces equivalent results and response rates to pencil and paper methods, and respondents are positive towards completing questionnaires using these methods. This provides evidence that e-PROMs are valid for use to collect data in a range of settings including clinical trials, routine care, and as, for example, health diaries. 1Mapi Objectives: According to the Federal, Food, Drug and Cosmetic Act (FD&C Act), prescription drug promotion must not be false or misleading, have fair balance, be consistent with the approved product labeling, and only include claims substantiated by adequate and well-controlled clinical studies. The Office of Prescription Drug Promotion (OPDP), formerly the Division of Drug Marketing, Advertising and Communications (DDMAC), was set up to protect the public health by assuring prescription drug information is truthful, balanced and accurately communicated. The objective of this study was to review the DDMAC/OPDP warning and notice of violations letters to find out 1) how many violations were in relation to PRO and HRQL claims and 2) how those evolved after the publication of the FDA PRO draft guidance in 2006. Methods: DDMAC letters were identified on the “Enforcement Activities by FDA” webpage. Letters from 1998 to 2013 were all reviewed manually to identify violations in relation to PRO and HRQL claims during the periods before and after the publication of the guidance (1998-2005 vs. 2006-2013). Results: 763 letters were reviewed. Each letter included information about one or more violations of the FD&C Act, such as “Omission of Risk Information”, “Overstatement of Efficacy”, “Unsubstantiated Superiority Claims”, etc. The review showed a letter volume on the decline (n= 524 for 1998-2005, n= 239 for 2006-2013), with an increase in PRO violations: 19.50% of all letters (1998-2005) vs. 30.5% (2006-2013). HRQL violations were rarer after 2006 and were more often detected as implicit: 20 false HRQL claims, of which two were considered implicit (1998-2005) vs. seven false HRQL claims, of which four were considered implicit (2006-2013). Examples will be presented. Conclusions: The FDA guidance on PRO measure seems to have had an influence on HRQL information: less ads with explicit violations and a OPDP’s tendency to argue over implicit claims. PIH75 The Use of Patient Reported Outcomes (Pros) By the Pharmaceutical Industry in Japan – A Brief Review of Pmda Data in Comparison With Fda and Ema-Approved Label Claims PIH77 Are Patient Reported Outcomes Relevant To Patients? Learnings From A Patient Advocate Survey Holtorf A P 1, Palacios D 2, Brixner D 3 1Health Outcomes Strategies, Basel, Switzerland, 2Novartis Pharma AG, Basel, Switzerland, 3University of Utah, Salt Lake City, UT, USA . . . . Objectives: Increasingly, patients become active participants in making decisions on their therapy. A survey was conducted to understand the experience and expectations of patient organizations (POs) with patient reported outcomes (PRO) as they are measured today. Methods: An online survey was conducted in English language throughout May 2014 among 40 participants at a global cross disease patient forum to prepare a discussion of the relevance and usefulness of patient reported outcomes from the patient perspective. The participants represented a broad range of disease specific and disease independent patient organizations from various countries including USA, European countries, Asia, Latina America, Middle East and Australia. Results: Current PROs were perceived as useful but not optimal for informing patients in making their own therapy decisions. All of 9 typical PRO domains were considered important (between 3.9 and 4.7 on a 5 point scale) with the most important being symptoms (4.6±0.89), Physical Function (4.65±0.59) and psychological well-being (4.7±0.47). The participants thought that PROs should be part of all studies throughout the entire life cycle of products including evidence for clinical research, reimbursement decisions, listing decisions, health technology assessment (HTA) or comparative effectiveness (CER) studies (all between 4.25 and 4.6 on a 5-point scale). Increasingly, POs develop their own instruments to elicit PROs from the patient perspective and as patient based evidence. Conclusions: The concept of patient reported outcomes is good in principle but more is needed for integrating additional aspects which are relevant for the patients themselves to understand the full impact and consequences of the therapy. Patient reported outcomes are a key endpoints from the patient perspective and should be elicited throughout the entire development and marketing cycle of products. Ledesma D A 1, Tanaka E 1, Adachi K 1, Rossi B 2 . . . . . 1Bayer Yakuhin, Ltd., Tokyo, Japan, 2Bayer Yakuhin, Ltd., Osaka, Japan Objectives: The use of patient-reported outcomes (PROs) in label claims in the US and Europe is regulated by the US FDA and the EMA, respectively. Japan’s Pharmaceuticals and Medical Devices Agency (PMDA) does not have such regulations. This study was done to determine whether Japan-based pharmaceutical companies utilize PRO endpoints at all and in what way, by investigating their inclusion of PROs in pharmaceutical clinical trials and drug information materials. Methods: We searched the websites of ClinicalTrials. gov and the PMDA for information on 14 drugs which had received PRO claim approvals from both the US FDA and EMA from 2006-2010. Search terms were the generic and/or brand names of the selected drugs (in English and Japanese, as appropriate). PROs were classified as “symptoms”, “functioning”, and “HRQOL” based on the PRO scale used. A table comparing PRO type, endpoint positioning, and US and Europe-approved label claims versus the PRO information reported in Japan for the same drug was created. Results: Of the above fourteen drugs, four are not yet available in Japan. One drug with an FDA and EMA-approved “symptoms” claim did not have such in its Japan clinical trial. Of the nine remaining drugs, the PRO endpoints were as follows: two drugs, indicated for epileptic seizure and for benign prostatic hyperplasia, PIH78 The Endometriosis Health Profile (Ehp) – A Case Study of Successful Epro Collaboration Two R 1, Wilkins G 1, Cox A1, Jenkinson C 2, McEvoy K 3, Churchman D 4, Walzer A 5, Wichmann K 5 Ltd, Banbury, UK, 2University of Oxford, Oxford, UK, 3CRF Health, London, UK, 4Isis Outcomes, Oxford, UK, 5Bayer Pharma AG, Berlin, Germany . . . . . . . 1PharmaQuest Objectives: To migrate the UK English Endometriosis Health Profile (EHP) from paper to ePRO format for completion by respondents on a touchscreen tablet device. Following migration, to produce translations of the UK English ePRO version in 25 languages. Methods: The draft ePRO version of the EHP was reviewed by the questionnaire developer, the translation project manager and the sponsor. During the initial review the questionnaire was assessed for linguistic equivalence with the paper version and for usability in relation to the target patient group. A number of factors were considered including layout, response input method and forced completion. Decisions were made based on the recommendations of the developer, translation vendor and ePRO vendor according to the specialism of each party, taking into consideration the capabilities of the software and the requirements of the A518 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 patient group. Following the initial review the tablet-based ePRO version was pilot tested with 10 endometriosis patients who were native speakers of UK English. The translated versions of the EHP were adapted for ePRO administration and the resulting screenshots proofread for accuracy. Results: Feedback from patients indicated that some amendments to the formatting and ordering of instructions would be beneficial. However, all respondents indicated that the ePRO version of the EHP was easy to use and preferable to a paper-based questionnaire. The ePRO format posed some difficulties for specific languages which required an adjustment to the layout or wording structure. Conclusions: Cognitive debriefing and usability testing confirmed that the ePRO version of the questionnaire was an accurate representation of the original paper version. This was achieved via cooperative input at the initial review stage to ensure that all aspects were considered and close collaboration throughout the project to find appropriate solutions to the challenges posed by the ePRO administration of the EHP. PIH79 Evaluating the Translatability of Physical Assessment Clinical Outcomes Assessment (Coa) Items McKown S 1, Talbert M 2, Brandt B A 3, Gawlicki M C 3 . . . . . . 1Corporate Translations, Inc., Chicago, IL, USA, 2Corporate Translations 3Corporate Translations, Inc., East Inc., Chicago, IL, USA, Hartford, CT, USA Objectives: The objective of this study is to determine which physical assessment Clinical Outcomes Assessment (COA) questionnaire items are most translatable. Methods: Eighteen (18) physical assessment items were analyzed, using back-translations to determine conceptual equivalency with the source text. Previous studies have regarded 80% source conceptual equivalency as a translatability benchmark. Translator feedback regarding cultural appropriateness also influenced whether certain physical assessment items were optimal for use in questionnaires, as some items might be translatable, but not appropriate for the target country. Results: Physical assessment items were grouped into three categories: highly translatable, moderately translatable, and problematic. Highly translatable examples, “running errands,” “getting around town” and “washing yourself” were translated with conceptual equivalency 100% of the time, and were considered culturally appropriate for all languages. Moderately translatable items, such as “taking a trip” may be misconstrued, as the distance implied may differ across languages. “Bathe yourself” was translated with 87% conceptual equivalency, and may be misunderstood as “taking a bath,” when the intention is to clean oneself. Problematic items, such as “going out,” translated with conceptual equivalency only 76% of the time, requiring a destination or activity to be comprehensible in other languages. “Walking a block,” although achieving 84% conceptual equivalency, is not applicable nor understood in many languages. When measuring distance walked, it is recommended to be specific, using examples such as 100 meters or a kilometer. Conclusions: Specific concepts appear to be more translatable and universal, such as “getting around town” and “walking 100 meters. ” Items left open for interpretation, such as “bathing yourself” or “taking a trip” can cause moderate translation difficulties. Additionally, many languages cannot coherently translate less specific items, such as “going out,” or translate culturally specific items. Although “walking a block” is translatable, such an example is unlikely to be understood in many languages. PIH80 Theoretical and Practical Possibilities of the Measurement of Postoperative Pain in Obstetric Intensive Ward Oláh A 1, Toldyné Beck M 1, Müller Á 1, Knisz J 1, Gelencsér E 2, Szunomár S 1, Boncz I 3, Fullér N 1 1University of Pécs, Pécs, Hungary, 2University of Pécs, Kaposvár, Hungary, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . . . Objectives: The measurement of the degree of pain and the exploration of influencing factors to compare data about the evaluation of postoperative pain given patients and nurses on a numeric scale. The connection between BMI index, age and verification of need for opiate associated with BMI was evaluated. Our aim was to measure whether nurses underestimate the level of pain in patients as well to show the advantages of multimodal analgesia and the additional opiate needs. Methods: The survey was conducted between the 1st November and 15th of December in 2013 among 40 patients who had surgery and 10 nurses dealing with them afterwards. Data was collected at the sub intensive ward of the Department of Obstetrics and Gynaecology, University of Pécs. The analysis of results was performed with MO Excel 2007 program. For data analysis absolute and relative frequency, Chi-square test, two-sample t-test, correlation and linear regression analysis besides the significance level p< 0,05. Results were presented with main confidence interval. Results: Nurses underestimated the patients’ pain value (p= 0,011). Tight and continuous connection with the patient’s vital parameters had not been proved (p>0,05). Connection was not found between the age and BMI index of the patients (p= 0,134), however in case of the mostly overweight sample group the need for opiate wasn’t increased (p= 0,62). It is shown by the survey that open surgeries aren’t result in higher pain value which was not significant after adequate analgesia (p> 0,05). The benefit of multimodal therapy in analgesia had been proved, although it had not resulted in decreased need for opiate (p=0,807). Conclusions: It is essential to conduct a survey on the knowledge of health care workers in the future and to organize possible workshops and create protocols. Moreover, pain should be monitored minimum three times a day (Shugarman LR., 2010, Canada). PIH81 Health Related Quality of Life in Patients Receiving Home Enteral Nutrition in Spain Assessed By A Specific Questionnaire: Nutriqol® Apezetxea A 1, Cuerda C 2, Virgili N 3, Irles J A 4, Cuesta F 5, Casanueva F 6, Carrillo L 7, Layola M 8, Lizán L 9 1Hospital Basurto, Bilbao, Spain, 2Hospital Universitario Gregorio Marañón, Madrid, Spain, 3Hospital Bellvitge, Barcelona, Spain, 4Hospital Universitario Nuestra Señora de Valme, Sevilla, . . . . . . . . . . Spain, 5Hospital San Carlos, Madrid, Spain, 6Hospital Universitario Santiago de Compostela, Santiago de Compostela, Spain, 7Centro de Salud Victoria de Acentejo, Santa Cruz de Tenerife, Spain, 8Nestle health science, Barcelona, Spain, 9Outcomes’10, Castellon, Spain Objectives: To assess Health Related Quality of Life (HRQoL) in patients receiving Home Enteral Nutrition (HEN) using NutriQoL® questionnaire in Spain. Methods: NutriQoL®, a specific questionnaire, developed and validated in Spain, for the assessment of HRQoL of patients receiving HEN regardless of the underlying condition was administered to a prospective cohort from 9 Spanish hospitals. It includes 17 pairs of items of HEN-related HRQoL grouped in two dimensions: 1) physical functioning and activities of daily living; 2) social life aspects, scoring from -51 (worst HRQoL) to 51 (best HRQoL). Cluster analysis using k-means identified groups of patients with similar HRQoL. Results: A total of 140 subjects (61.4% men; mean (SD) age: 62.7 (15.41) participated. NutriQoL® mean total score was 14.98 (14.86). Dimension 1 and 2 scored 13.55 (11.71) and 1.40 (4.74). Cancer patients presented lower HRQoL compared to neurological and malabsorption patients (12.76 vs. 18.11 vs. 17.37; p= 0.098). Patients receiving oral HEN as a supplement referred higher HRQoL than those receiving HEN by gastrostomy or nasogastric tube (19.54 vs. 14.00 vs. 7.02; p< 0.001) as their only nutrition route (19.33 vs. 8.18; p< 0.001). Up to 71.4% of patients referred HRQoL improvements since the introduction of HEN. Cluster analysis resulted in 4 groups according to NutriQoL® score. Cluster 1 [32.23 (5.83)]: Neurologic patients receiving oral HEN as a supplement. Cluster 2 [18.19 (3.94)]: oncologic disease receiving HEN by gastrostomy as a supplement. Cluster 3 and 4: [3.9 (4.67) and 12.21 (5.95)]: oncologic patients receiving oral HEN as a supplement, with differences in terms of severity (Charlson index 2.45 (2.65) vs. 3.14 (2.57). Conclusions: NutriQoL® results demonstrated a sample with a fairly good HRQoL, where the introduction of HEN had improved their HRQoL. In patients receiving HEN, physical functioning and activities of daily living were better predictors than social life domain. PIH82 Quality of Life in Pregnant Women Attending Anti-Natal Clinics in Rural and Urban Areas of Delta State Arute J E 1, Eniojukan J F 2, Odili V O 1 STATE UNIVERSITY, ABRAKA, Nigeria, 2Niger Delta University, Wilberforce Island, Nigeria . . . . . . 1DELTA Objectives: Preventing problems for mothers and babies depends on an operational continuum of care with accessible, high quality care before and during pregnancy, childbirth, and the postnatal period. The objective of this study is to evaluate the quality of life of pregnant women attending antenatal clinics in rural and urban areas of Delta State, Nigeria. Methods: A descriptive cross sectional study design was used. Six hundred and ninety nine pregnant women attending antenatal clinics in selected hospitals were interviewed using a 31 item pretested, structured questionnaire developed using the World Health Organization Quality of Life (WHOQOL) on Pregnancy assessment brief template. Data assessed include socio-demographics and questions relating to physical, psychological, social and environmental health were used to assess the health related quality of life values in each of the study participants. Data collected were analyzed using the Statistical Package for Social Sciences (SPSS) software version 16.0. The level of statistical significance was set at P< 0.0001. Results: The mean age of the respondent was 27±5.04. The HRQOL mean scores were highest for the environmental domain (26.39±5.34) and lowest for the social relationship domain (11.43±1.81). The overall QOL mean scores in the other two domains were: physical health (24.84±3.74), psychological health (21.84±3.03). Significant differences were observed in all domains except social relationship. Conclusions: The health related of life (HRQOL) in pregnant women was found to be lower in those living in the rural areas than their counterparts in the urban areas in all domains except social relationships. INDIVIDUAL’S HEALTH – Health Care Use & Policy Studies PIH83 Hospital Drg Costing and Health Services Use of Very Pre-Term Infants From the Proprems Neuro Study Across 10 Hospitals in Australia and New Zealand Sia K L 1, Gold L 1, Jacobs S 2, Cheong J 2, Opie G 3, Garland S 2, Donath S 4, Hickey L 2, Boland R 4, Webster C 5 1Deakin Health Economics, Melbourne, Australia, 2Royal Women’s Hospital, Melbourne, Australia, 3Mercy Hospital for Women, Melbourne, Australia, 4Murdoch Children’s Research Institute, Melbourne, Australia, 5Northern Hospital, Melbourne, Australia . . . . . . . . . . . Objectives: Mortality and morbidity of very preterm (born < 32 weeks’ gestation) and very-low-birth-weight (VLBW, < 1500g) infants impose substantially on finite health resources. This study estimated costs of hospital and non-hospital services for a cohort enrolled in ProPrems Neuro, of very preterm/VLBW infants from birth to 24 months’ age corrected for prematurity. We also tested the sensitivity of results to the costing approach used. Methods: ProPrems Neuro study assesses the 2-year outcomes of very preterm/VLBW infants from Australia and New Zealand from 2007-2011 in a prospective multicentre, double-blinded randomised controlled trial of probiotic administration. Infants’ health resource use was collected from medical assessment records at birth hospitals, parent report and, with parental consent, from Medicare Australia (Government database) for resource use up to 24 months. Hospital costs were calculated separately by the Victorian (State) Casemix funding approach and the newly implemented national activity-based funding (ABF) algorithm. AR-DRG diagnostic/procedural codes were used to classify inpatient episodes by prematurity/birth weight, complications, length of stay (LoS), hospital and patient characteristics. Costs were measured in 2013 Australian dollars. Results: 1099 preterm infants across 10 hospitals were included. Average costs were highest for infants with birth weight < 750g: $224,158 with mean LoS 105 days. Cost comparison between Casemix and ABF systems showed significantly lower costs using the national algorithm. Mean cost difference for the largest participating hospital was $9132 (95%CI 5998, 12267; p< 0.001). Final results of infant health service use to 2 A519 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 years corrected age will be presented in November 2014 with follow-up completion and data release from Government database by August 2014. Conclusions: Preterm infants showed high services use including hospitalisation, with associated high costs. Costs vary by patient characteristics and costing approach. This study’s results should inform effective resource planning of neonatal health services and development of future prevention interventions aimed at preterm birth. PIH84 Daily Dose and Costs Associated With Maintenance Therapy of Topical Testosterone Agents Among Hypogonadal Men . . . . . . . . Objectives: Topical testosterone agents (TTAs) are commonly used to raise low serum levels of testosterone in men. After initiating at a recommended starting dose (RSD), patients may undergo dose titration to achieve an appropriate maintenance dose. The objective of this study was to compare daily maintenance doses and costs of treatment with TTAs from US payer perspective in adult men diagnosed with Hypogonadism (HG). Methods: Adult men with a HG-associated diagnoses initiated at the RSD with Axiron® (Lilly USA, LLC; RSD 60mg per day; N= 209), AndroGel® 1% (AbbVie Inc.; 50mg per day; N= 614), AndroGel® 1.62% (AbbVie Inc.; 40.5mg per day; N= 235), or Testim® (Auxilium Pharmaceuticals, Inc.; 50mg per day; N= 558) between January 1, 2011 and March 31, 2012 were identified in a database of commercially insured beneficiaries. Patients were required to have continuous eligibility and no claims of the index therapy in 12 months prior to and at least 1 month of continuous eligibility following initiation. Baseline demographic characteristics, Charlson Comorbidity Index (CCI), comorbidities, and testosterone use were compared using chi-squared test for categorical variables and Wilcoxon ranksum test for continuous variables. Mean dose was estimated per-person per-day (PPPD). Risk-adjusted dose and payer costs PPPD were estimated with a generalized linear model. Results: Maintenance dose was attained at month 4. Mean dose PPPD in month 4 was 68.45mg, 56.68 mg, 51.18mg, and 59.24mg and risk-adjusted dose PPPD was 113.3%, 113.5%, 126.3%, and 118.7% of RSD for Axiron, AndroGel 1% (non-significant vs. Axiron), AndroGel 1.62% (P< 0.001 vs. Axiron), and Testim (P= 0.047 vs. Axiron), respectively. Risk-adjusted third-party payer costs PPPD were $7.49, $9.49, $10.39, and $9.59 (all P< 0.001 vs. Axiron), respectively. Conclusions: Maintenance dose as a proportion of RSD was the least among Axiron and AndroGel 1% patients, while third-party payer costs for the maintenance dose were lowest among Axiron patients. PIH85 The Efficiency Evaluation of the Rule From Drug Market Regulation Chamber (Cmed) Proposed To Public Medicines Acquisitions in Antitrust Market Conditions SantAnna F A ANVISA, brasilia, Brazil . . Objectives: The aim of this work is measure the efficiency of public medicine purchase considering competitive markets applied by the rule that establishes PMVG. Methods: The competitive market was defined as more than 5 (five) pharmaceutical manufactories. Two lists were built, one with PMVG (PMVG databank), established by CMED, and the other list contains the public prices acquisitions before and after the enacted CAP legislation (Health Price Bank – BPS databank). The lists were separated in two groups. The first contains the prices of PMVG databank and BPS databank between 2004 and 2006 (group 1), and the second complains the prices for the period between 2007 and 2012 (group 2). The comparison of two groups, before CAP (group 1) and after CAP (group 2), was conducted by Student t-test one-tailed of the mean medicine prices (significance level of 5%, assuming unequal variances). Results: The list resulted in a selection of 29 medicines. Six (6) medicines had an average BPS price before CAP less than or equal to the average BPS price after CAP, however, only 1 was statistically significant (α ≤ 0.05). In contrast, 23 medicines had an average BPS price after CAP significantly lower or equal to the average BPS price before CAP, but only 5 (five) no statistically significant (α ≤ 0.05). Conclusions: For competitive market we have observed efficiency loss by application of CMED legislation. PIH86 The Efficiency Evaluation of the Rule From Drug Market Regulation Chamber (Cmed) Proposed To Public Medicines Acquisitions Araujo M A M ANVISA, Brasilia, Brazil . . PIH87 Implementation of A Collaborative Pharmacy Practice Model in Nursing Homes of A Swiss Canton: Drug Cost Monitoring Between 2009 and 2012 Zeukeng M J 1, Niquille A 2, Locca J F 2, Perraudin C 3, Bugnon O 1 of pharmaceutical sciences, University of Geneva, Lausanne, Switzerland, 2University of Lausanne, Lausanne, Switzerland, 3School of pharmaceutical sciences, University of Geneva, University of Lausanne, Lausanne, Switzerland . Kaltenboeck A 1, Boytsov N 2, Hayes-Larson E 1, San Roman A 1, Ivanova J 1, Birnbaum H 3, Foster S 2, Vazquez J 2, Muram D2, Swindle R 2 1Analysis Group, Inc., New York, NY, USA, 2Eli Lilly and Company, Indianapolis, IN, USA, 3Analysis Group, Inc., Boston, MA, USA . cally significant (α ≤ 0.05). In contrast, 47 medicines had an average BPS price after CAP significantly lower or equal to the average BPS price before CAP (α ≤ 0.05). Conclusions: It is concluded that this CMED legislation is economic and social efficient. . Objectives: The aim of this work is measure the efficiency of this rule that established PMVG for public medicine purchase. Methods: Two lists were built, one with PMVG (PMVG databank), established by CMED, and the other list contains the public prices acquisitions before and after the enacted CAP legislation (Health Price Bank – BPS databank). The lists were separated in two groups. The first contains the prices of PMVG databank and BPS databank between 2004 and 2006 (group 1), and the second complains the prices for the period between 2007 and 2012 (group 2). The comparison of two groups, before CAP (group 1) and after CAP (group 2), was conducted by Student t-test one-tailed of the mean medicine prices (significance level of 5%, assuming unequal variances). Results: CMED implemented The Price Adequacy Coefficient (CAP) by publishing the Resolution CMED nº 04 on 18thof December of 2006. It is a mandatory minimum discount to be applied by pharmaceutical companies and distributors in the Factory Price of medicines (PF) authorized, resulting on a public price cap (PMVG). There is no evidence of the efficiency, or the real need for this regulatory activity performed by CMED geared to the drug public acquisitions. The list resulted in a selection of 68 medicines. Fourteen (14) medicines had an average BPS price before CAP less than or equal to the average BPS price after CAP, however, only 4 were statisti- . . . . . . 1School Objectives: The aging of the population and the increase of chronic disease patients represent the current medical and socio-economic challenges. In 2008, facing considerable increase of medicines cost in nursing homes (NH) in the canton of Vaud (Switzerland), professionals around the resident were invited to develop a collaborative pharmacy practice model derived from the successful experimentation in the canton of Fribourg. The intervention called Quality Circle in NH is an evidence-based program with a direct influence on the prescriber, that promotes team based care and clinical guidelines. The model has showed sustainable evidence of the cost containment of medicines without affecting the quality of care. The objective of this study was to assess the first results of economic impact of the program in NH engaged between 2009 and 2012. Methods: Individual data by NH resident was derived from community pharmacists’ invoice and from resident admission data. The evolution of the daily mean drug cost per resident was compared to the evolution of the daily mean drug cost of the population over 65 years old in primary care from cantonal pharmacy invoice (Swiss Health Observatory, OBSAN). Results: Between 2009 and 2012,13 NH were entered into the program (601/6’492 beds, 9.3%), between 9 and 11 NH were monitored. In 2012, the mean age of residents was 88 (SD 7.3) years old; 78% were women. From 2009 to 2012 the daily mean drug cost per resident of the program decreased to 10.3% from 6.8 to 6.1 EUR, representing a clear reduction compared to cantonal invoice that record an increase of 0.59%. Conclusions: The monitoring of the program shows first positive results due to the reduction of drug costs. Nevertheless, this model has an operating cost. The next step will be to determine the break even point of the intervention. PIH88 Fda Cdx Category Medication Use During Pregnancy in the United States Raval A , Pan X , Sambamoorthi U West Virginia University, Morgantown, WV, USA . . . Objectives: To assess the patterns and factors associated with prescription of FDA classified C, D and X category drugs (unsafe medications) during pregnancy. Methods: Cross-sectional analysis was conducted of pregnant women aged 20 and 49 years using data from the Medical Expenditure Panel Survey (MEPS) for the years 2009 and 2011. Difference in the demographic variables, socioeconomic status, access to health care, and chronic conditions were tested for the use of safe medications. All analyses accounted for the complex survey design of MEPS. Results: The study sample consisted of 1603 pregnant women (whites [595 (58.8%)], married [924 (66.3%)], employed [918 (64.5%)], no chronic illness [1103 (65%)]). A total of 413 (28.3%) women utilized medication of FDA category CDX. The use of the category C medications was highest [337 (23.4%)] drugs followed by category D [101 (6.9 %)], and category X [64 (4.4)]. Women with chronic illness [adjusted Odds Ratio (aOR) = 1.61; 95% CI = 1.03, 2.52, P < 0.001], having fair or poor perceived health status (aOR = 1.95; 95% CI= 1.15, 3.31, P < 0.05), doing regular physical activity (aOR = 1.50; 95% CI= 1.12, 2.01, P < 0.01) were more likely to receive unsafe medication, while pregnant women with less than higher secondary education (OR = 0.30; 95% CI = 0.11,0. 81, P < 0.05), with near poor (aOR: 0.53, 95% CI: 0.31, 0.91, P< 0.05) or middle income (aOR: 0.62, 95% CI: 0.40, 0.95, P < 0.05), were less likely to receive unsafe medications. There were no differences rates of unsafe medications use by age, race, marital status, employment status, mental health, and smoking status. Conclusions: Nearly one-third of all pregnant women used FDA category CDX drugs. Interventions should be carried out to reduce utilization of FDA CDX especially among the women with chronic conditions and poor health status. PIH89 Evaluation of Patient and Financial Outcomes Associated With Advanced Infertility Treatment Options Locklear J 1, Bozkaya D 2, Migliaccio K 2, Bramley T 3, Phillips A L 1, Mahony M 1 1EMD Serono, Inc., Rockland, MA, USA, 2Xcenda, L.L.C., Palm Harbor, FL, USA, 3Xcenda, Palm Harbor, FL, USA . . . . . . . Objectives: A model was designed to evaluate infertility treatment protocols and examines the medical and pharmacy budget impact from a patient and health plan perspective. Methods: An Excel-based platform (Microsoft Excel 2010) was designed to model the cost and expected number of pregnancies associated with the following advanced infertility treatments: natural cycle intrauterine insemination (IUI), IUI with clomiphene citrate (IUI-C), IUI with gonadotropins (IUI-G), in vitro fertilization (IVF) and IVF with intracytoplasmic injection (IVF+ICSI). The model is designed to customize inputs within key parameters including population, treatment strategies and health plan coverage. Standard fertility treatment strategy can choose to enter one of five standard first line treatment pathways including IUI, IUI-C, IUI-G, IVF or IVF+ICSI. Fast track fertility treatment strategy does not include the use of IUI-G. Health plan coverage allows end user to input drug coverage (yes/ no, copayment per cycle and patient coinsurance) and procedure coverage (yes/no, number of cycles covered, copayment, number of covered doctor visits per cycle and patient coinsurance). Results: Default model population of women initiating fertility treatment is 1,555. The total cost to the health plan per year (health plan A520 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 perspective) for the standard fertility treatment strategy is about 50M USD compared to the fast track fertility treatment strategy at about 41M USD for a net budget impact of 9M USD. The average cost per patient per year (patient perspective) for the standard fertility treatment strategy is approximately 4,800 USD compared to the fast track fertility treatment strategy at an estimated 4,200 USD for a net budget impact of approximately 600 USD. Results vary upon user inputs. Conclusions: An Excel-based model was developed to assist managed care organizations and employers with the development of an optimal fertility benefit design. The model serves as an educational tool to evaluate various fertility benefit designs in terms of patient and financial outcomes. PIH90 Potentially Inappropriate Medicines and Potential Prescribing Omissions in Older People and Their Association With Health Care Utilization: A Retrospective Cohort Study Moriarty F 1, Cahir C 2, Fahey T 1, Bennett K 3 College of Surgeons in Ireland, Dublin, Ireland, 2Trinity College Dublin, Dublin 8, Ireland, 3Trinity College Dublin, Dublin, Ireland . . . . mary profile we used the data of frequency by age and gender for this specialised profile obtaining this from the minimum joint data base of the hospital discharge. The profile of the pharmaceutical expenditure by age and gender was obtained from the pharmacy invoice, once the public contribution was deducted. The estimated population in 10 years was obtained from the National Statistic Institute. (INE). Results: aging results in a cumulative annual rate for the period 2011-2021 for public health expenditure growth of 5.34%. The aging effect implies a cumulative annual rate in the period 2011-2021 of 6.43% and a decrease of -0.69% due to the range effect (decrease in population). Of all the segments, the largest increase is in the pharmaceutical costs with an accumulated increase of 8.81%, of which 5,65% is in primary attention and 4,78% in specialised attention. Conclusions: According to our results, population growth or aging are determining aspects in public health expenditure increase. Using data directly from each region will explain the differences. In the case of Extremadura, the age factor is very important when increasing the pressure of the public health cost, having a special influence in the field of pharmaceutical expenditure and primary attention. 1Royal Objectives: Older people are vulnerable to medicine-related adverse effects. In response to these concerns, prescribing indicators have been developed addressing: Potentially Inappropriate Medicines (PIMs), medicines with unfavourable riskbenefit ratios and Potential Prescribing Omissions (PPOs), omission of indicated medicines with a clear benefit. Little is known about the impact of PIMs and PPOs on health care utilization. This study aims to determine the association between PIMs and PPOs and health care utilization. Methods: This is a retrospective cohort study of 2,051 community-dwelling participants in The Irish Longitudinal Study on Ageing (TILDA) aged ≥ 65 years with linked medication dispensing history from a national pharmacy claims database. PIM and PPO exposure in the 12 months prior to participants’ TILDA interviews was determined using validated prescribing indicators: Screening Tool for Older Persons’ Prescriptions (STOPP), the Screening Tool to Alert doctors to Right Treatment (START), Beers’ criteria and Assessing Care of Vulnerable Elders (ACOVE) indicators. Outcome measures used were self-reported number of hospital visits (emergency department or inpatient admissions) and general practitioner (GP) visits in the previous year. Poisson regression models were used to determine the associations between PIMs and PPOs and these outcomes, adjusting for age, sex, education, number of medications, chronic conditions, and health insurance status. Results: Overall PIM prevalence was 19.8-52.7% and PPO prevalence was 43.6-44.8% depending on the screening tool applied. Independent of screening tool used, PIMs and PPOs were significantly associated with hospital visits. For example, the adjusted Incident Rate Ratio (IRR) for each additional STOPP PIM was 1.24 (95%CI= 1.15-1.35). With the exception of START PPOs, PIM and PPO exposure were also significantly associated with GP visits (adjusted IRR= 1.10 (95%CI= 1.06-1.15) for each additional Beers’ PIM for example). Conclusions: PIM/ PPO exposure is independently associated with increased health care utilization, supporting application of PIM/PPO indicators as robust measures of health care quality and patient safety in relation to prescribed medications. PIH91 Impact of Assisted Reproductive Therapy (Art) On Infant Health and Health Care Cost Outcomes Tabano D C , Schroeder A , Sullivan K , Vaidya N University of Colorado School of Pharmacy, Aurora, CO, USA . . . . . Objectives: Assisted reproductive therapy (ART) has increased dramatically in the US over the past several decades, nearly doubling from 1999 to 2008. Prior research has evaluated multiple outcomes from ART including newborn survival and birth weight as well as cost analyses measuring cost per live birth; despite robust information on ART as a whole there is no Colorado-specific data on neonatal intensive care unit (NICU) outcomes following ART and its economic implications. Methods: Using data from the Colorado Department of Public Health and Environment (CDPHE) - Colorado Birth Certificate Database from 2007-2012, we use multivariable logistic regression to determine if ART births are associated with a higher risk of NICU admission compared to non-ART (no fertility treatment) births. We compare the risk of NICU admission among the full birth cohort and a singleton-only cohort controlling for plural births and birth order. We use Colorado state Health Care Policy and Financing (HCPF) fiscal year 2007-08, Colorado Centre for Reproductive Medicine costs and Colorado State Medicaid 2012 fee schedule data to estimate average NICU admission rates, total ART procedural costs and average costs of delivery, respectively. Results: 190,795 live births in 2007-2012 were included into the birth cohort for analysis (12,666 ART births; 178,129 non-ART births). ART births resulted in a 52% increased risk of being admitted to the NICU compared to non-ART births (OR 1.52 [95% CI 1.38,1. 69]); singleton-only ART births had a 39% greater risk of being admitted to the NICU compared with singleton non-ART births (OR 1.39 [95% CI 1.18,1. 65]). Average NICU admission costs were estimated at $ 6,165.78 per ART birth and $ 331.85 per non-ART birth. Conclusions: ART births in Colorado have a higher risk of NICU admission compared to non-ART births. The economic impact of NICU admissions is 18.6 times greater among ART births compared to non-ART births. PIH92 Aging Impact Over the National Health Cost in Extremadura Public Health Expenditure of Extremadura in the Period 2011-21 Hidalgo A 1, Alonsog S 2, Vizcaino S 2, Montaño L 2, Alvarez M2, Maiilo M A 2, Muñoz J I 2, Sanchez-Chorro J L 2 1Instituto Max Weber, Majadahonda, Spain, 2Servicio Extremeño de Salud, Merida, Spain . . . . . . . . . PIH93 Kazakhstan VerSUS Uzbekistan: A Review of the Drug Provision Systems Rainova K , Izmirlieva M , Ando G IHS, London, UK . . . Objectives: As is the case with many of the Commonwealth of Independent Countries (CIS), since the collapse of the Soviet Union, both Kazakhstan and Uzbekistan have been re-building health care provision, and improving access to medications for their population. This study compares the two separate paths the countries have taken, and aims to establish the outcomes achieved by the two systems as well as the direction of future reforms. Methods: Secondary research focused on analysing the systems in place in the two countries, focusing on drug provision. The study assessed the mechanisms in place, drawing comparisons between the two systems, with a particular focus on the outcomes achieved. Results: Uzbekistan provides medications free of cost for certain categories of patients, while in Kazakhstan, the government has approved a list of drugs that are provided for free as part of the guaranteed volume of free medical care. In both the cases these fall under an outpatient setting. Although the procurement of drugs is carried out mostly via tenders, Kazakhstan’s system involves establishing price ceilings. In Uzbekistan, retail and wholesale margins are controlled. Between 2002-2012, public health expenditure as a percentage of total health expenditure rose from 54% to 58% in Kazakhstan and from 45% to 53% in Uzbekistan. Life expectancy however, increased from 65.9 to 69.6 years in Kazakhstan and from 67.1 to 68.1 years in Uzbekistan. Conclusions: With growing government health expenditure, reflecting the expansion of the health care systems, the countries are likely to increasingly look into containing costs. Given that some pricing mechanism is already in place in Kazakhstan, it may consider implementing tighter pricing regulations, moving closer to those seen in Europe. In Uzbekistan, the government may potentially consider expanding the beneficiary categories while ensuring competiveness within the tendering process. PIH94 Psychometric Properties of the 16-Item Sort form Version of the Menopause Cervantes Health-Related-Quality-of-Life Scale: the Cervantes-Sf Coronado P 1, Sánchez Borrego R 2, Ruiz Díaz M A 3, Rejas Gutiérrez J 4 Universitario San Carlos, Madrid, Spain, 2Clínica DIATROS, Barcelona, Spain, Spain, 3Universidad Autónoma de Madrid, Madrid, Spain, 4Pfizer S.L.U., Alcobendas/Madrid, Spain . . . . . 1Hospital Objectives: The Cervantes scale is a specific health-related-quality-of-life (HRQoL) questionnaire developed in Spanish women through and beyond menopause. The original 31-item scale was reduced to a less time consuming 16-item sort form: The Cervantes-SF. The aim of this work was to assess the psychometric properties of the Cervantes-SF in a routine clinical sample of perimenopause and postmenopause women. Methods: Peri and postmenopause adult women were recruited in twelve outpatient clinics of Gynecology. All of the patients completed both scales, the 31-items form and the abridged version, however, order of administration was balanced equally to avoid administration bias. A sub-sample of 31 women answered the sort form within 1-2 weeks later (rete-test). Correlation between forms and testretest reliability were used to test measurement stability. Item analysis, internal consistency reliability, item-total and item-domain correlations and item correlation with the generic Spanish version of the EQ-5D-3L questionnaire were also studied. Results: A sample of 215 women [mean age 55 years old (SD= 5.3)] was enrolled. Internal consistency was good (Cronbach’s a=0.829) but slightly lower than that of the original scale (a= 0.895). Dimension reliabilities ranged between a= 0.636 (Health) and a= 0.923 (Vasomotor). Correlations between extended and reduced subscales was high and significant in all cases (p< 0.001), ranging from r= 0.790 for Health to r= 0.872 for Vasomotor. Correlation between total scores was also high (r= 0.885) and no differences were found between mean scores (Effect size= 0.353). Short-form total score correlation with EQ-5D utility score was negative and significant (r= -0.487) and also with EQ-5D Health VAS (r= -0.432). Test-retest correlation was high (r= 0.886). Completion of Cervantes-SF required half of the time than the original scale. Conclusions: The abridged 16-item Cervantes scale (Cervantes-SF) maintained the original psychometric properties. This version extends 51% of the original length, being faster to apply and making it specially suitable for routine medical practice. . Objectives: To estimate the effect of demographic component in the evolution of public health expenditure of Extremadura in the period 2011-21 Methods: we estimated health expenditure profiles by age and gender in 2011. Then, we used population projections to calculate future health spending. To obtain those profiles we used data obtained from the information systems of Extremadura. For the pri- PIH95 Predictions for Medical Subsidy Enrollment Among Young Children From High-Risk Families in Taipei Lussier E C 1, Ting T T 2, Lin C Y 3, Hung N 1, Chen C Y 1 University, Taipei, Taiwan, 2Academia Sinica, Taipei, Taiwan, 3Taipei Tzuchi General Hospital, Taipei, Taiwan . . . 1National Yang-Ming . . . . . . A521 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: The current study looks to explore young children from high-risk families in the Taipei City setting. High-risk social welfare intervention was investigated. First, differences between enrollees and non-enrollees for medical subsidy program among high-risk family whose cases were started in 2009 or 2010 will be looked at. Second, the study will try to determine the social welfare intervention’s effectiveness in increasing application for medical subsidy program and finding any predictors which may have helped or harmed application for enrollment. Methods: The study sample included under 6 year old children high-risk families (n=199). Highrisk family database and medical subsidy database were linked. Differences between high-risk subsidy enrollees (n= 87) and non-enrollees (n= 112) and effectiveness of a social welfare intervention in increasing subsidy application were investigated in a pre-post analysis of high-risk social welfare intervention. Individual level as well as relative residential level characteristics were explored. Results: Medical subsidy enrollment was correlated with younger age at time of a high-risk intervention and relative district level variables. Pre-post comparison suggests high-risk interventions significantly increased subsidy application by 7.4%. Logistic regression indicates older age at time of intervention was associated with 40% less chance of application. Conclusions: The study provides empirical evidence for potential effects of a high risk social welfare intervention on the accessibility to health care. Findings also show where policy makers can improve intervention in order to address the needs children at-risk, especially for different age groups. PIH96 What Factors Are Associated With Vaccination Programme Success? van Oorschot D A M Radboud University, Nijmegen, The Netherlands . . . Objectives: When the WHO launched the Expanded Programme on Immunisation (EPI) in 1974, < 5% of the world’s children were vaccinated against polio, measles, diphtheria, tetanus, pertussis and tuberculosis. Nowadays coverage rates are increasing and more vaccines have been added to the programme. This project aims to identify factors associated with successful programmes based on EPI coverage. Methods: The relationship between multiple socio-demographic and economic factors and EPI coverage (primarily obtained from the WHO and World Bank) was investigated using simple linear regression, Principal Component Analysis (PCA) to identify explanatory variables, and finally multiple linear regression analysis. 132 countries with data on self-funded health care programmes were included in the analysis. These were ranked according to Gross National Income (GNI) per capita and lower (L), lower-middle (LM), upper-middle (UM) and upper (U) quartiles were identified. Results: Income (GNI/capita) was not significantly associated with achieving high EPI coverage rates. Within the income groups the factors trending with improved EPI coverage included: 5-yr mortality and corruption index in L, sanitary facilities in LM, 5-yr mortality, sanitary facilities, birth-rate and life-expectancy in UM. In U all countries achieved > 90% coverage. Conclusions: Identifying simple predictive variables of successful vaccination programmes is complex because of multicollinearity. However, by exploring within homogeneous income groups, it was possible to identify underlying factors related to vaccination programme success. As the core EPI vaccines were introduced 40yrs ago one would expect introduction to have been fully implemented thus reducing the likelihood of a relationship between country income and coverage today, however it could have been expected some time ago. To further explore the relationship between country income and vaccination programme success, one could expand the analysis to include the newer vaccines as soon as coverage information is available for a majority of the countries. PIH97 Patient-Reported Outcomes (Pro) in Go/No-Go Decision Making in Drug Development Ingelgard A P 1, Nokela M 2, Cole J C 3, Berger A K 3 1Covance Market Access, Solna, Sweden, 2Covance Market Access Services, Solna, Sweden, 3Covance Market Access Services, Inc., San Diego, CA, USA . . . . . . . Objectives: Despite existing examples of the great impact successful integration of PRO science and know-how may have on improving market access, PROs are typically not part of drug development decision making such as go/no-go decisions. The objective of this study is to identify decision making instances in drug development when important go/no-go decision making also could have included PROs but typically weren’t. Additionally, search for any strategies and activities that might better enhance the likelihood for senior decision makers also to integrate PRO science and knowhow in drug development decision making. Methods: A literature search was carried out including terms such as PRO, strategy, decision-making, go/no-go decisions, drug development and phases I to IV. Results: Typical go/no-go decisions mentioned in the literature include decisions to move a drug candidate from one development phase to the next stage, i. e. from target-to hit to launch decisions. There were very few references found where go/no-go decision-making in drug development also included PROs. Many references also states the necessity to include the patient (e. g. listen to the patient) early on in drug development. However, no concrete suggestions on how to carry out these ambitions in practice were found. Conclusions: There is little research to be found in the literature on go/no-go decision making in drug development where PRO science and knowhow are taken into account. The lack of clear practical guidance and examples on when and how to start inclusion of PROs science and knowhow in go/no-go decision making may be one impediment to their successful inclusion. Perhaps, a best practice rule could be as simple as including a PRO opportunity assessment in phase I-II go/no-go decisions where the feasibility, pros and cons for potentially including PRO in the clinical development program are summarized. At least, a first overview of PROs potential is done. PIH98 Estimation of Serum Calcium Level in Peri and Postmenopausal Women: A Comparative Study Paras P 1, Shobha Rani R 1, Shobha G 2 1Al-Ameen College of Pharmacy, Bangalore, India, 2St. philomena’s hospital, Bangalore, India . Objectives: To estimate the calcium levels in peri menopausal and postmenopausal women and to evaluate the need for calcium supplementation among them. Methods: A prospective study was conducted at gynecology department of a tertiary care hospital for the period of six months to estimate the serum calcium levels among them. Study populations were divided in to perimenopausal and post menopausal group. All the eligible patients were enrolled after obtaining informed consent. Study subjects from both the group were estimated for serum calcium levels to identify if they need calcium supplements. Unpaired T test was performed to find out any significant difference between both groups and pearson’s correlation cofficient (r) was applied to assess the relation between age and calcium levels. Results: During the study period, 53 of 100 patients enrolled were postmenopausal with the mean age of 60.8±10.47 years and 47 of 100 patients enrolled were perimenopausal with mean age of 44.6±3.54 years. The mean calcium level of 47 perimenopausal women was found to be 9.32±0.55 (reference level: 8.0-11.0 mg/dl) and 8.56±0.54 for 53 postmenopausal women. In post menopausal women there was highly significant drop observed in serum calcium levels with increasing age, compared to peri menopausal women. (CI: 95%, p< 0.0001, r: -0.81). Conclusions: The serum concentrations of calcium in majority of our study population were within the normal range. There was a good source of dietary intake of calcium in most of the patients. The levels of calcium were lower in postmenopausal women compared to perimenopausal women. Since there is an negative effect of calcium on the bone mineral density in postmenopausal women, it can be recommended that calcium supplementation can be given as prophylaxis to prevent the long term bone loss and to decrease the risk of fracture and osteoporosis. . . PIH99 Patient Characteristics and Medication Treatment Patterns Among Men With Erectile Dysfunction (Ed), Lower Urinary Tract Symptoms Secondary To Benign Prostatic Hyperplasia (Bph-Luts), Or CoOccurring Ed and Bph-Luts In the Uk Primary Care Setting Ilo D 1, Raluy-Callado M 2, Graham-Clarke P 3, Donaldson R 2, Birt J 4, Sadasivan R 2, Zhu Y 4, Neasham D 2 1Lilly UK, Hampshire, UK, 2Evidera, London, UK, 3Eli Lilly Australia Pty Ltd., NSW, Australia, 4Eli Lilly and Company, Inc., Indianapolis, IN, USA . . . . . . . . Objectives: Describe patient characteristics and medication treatment patterns among newly diagnosed cases of BPH-LUTS, ED, and co-occurring ED and BPH-LUTS. Methods: Retrospective cohort study using UK CPRD data on incident BPH-LUTS and incident ED patients indexed between June 2010 and May 2011. Patient records were analysed from 12 months pre-index and up to 24 months post-index. Results: The cohort included 8912 men with BPH-LUTS-only, 2589 with an ED diagnosis followed by BPH-LUTS, 8093 with ED-only and 1641 with a BPH-LUTS diagnosis followed by ED, all aged ≥ 40 years. The majority of BPH-LUTS patients (~90%) were diagnosed and managed within GP practices. Men were diagnosed with BPH-LUTS alone at an older average age (68±11.9 years, IQR= 59–77) compared to men in the ED/BPH-LUTS group (67±9.5 years, IQR= 61–74, p= 0.002). Men were diagnosed with ED at an older average age (65±9.2 years, IQR= 59–72) in the BPH-LUTS/ED group compared with ED-only patients (57±9.1 years, IQR= 50–64, p< 0.001). Time between diagnoses was longer for ED/BPH-LUTS patients (6.8±4.76 years) versus BPH-LUTS/ED patients (5.8±5.10 years). BPH-LUTS and ED treatment patterns were similar for patients with and without co-occurring conditions. Most patients were initially prescribed alpha-blockers (62.9% BPHonly, 65.5% ED/BPH-LUTS) or anticholinergics (14.9% BPH-only, 14.0% ED/BPHLUTS). For ED, most patients were initially prescribed sildenafil (51.6% ED-only, 49.6% BPH-LUTS/ED) or tadalafil (24.3% ED-only, 26.0% BPH-LUTS/ED). At six months post-diagnosis, ~47% incident BPH-LUTS patients and ~78% ED patients were not on any BPH-LUTS or ED treatment, respectively. Conclusions: Study data suggests > 80% of patients are managed as either BPH-LUTS- or ED-only. Average age of BPH-LUTS/ED patients at ED diagnosis suggests patients may suffer from ED years before seeking medical attention (p= 0.002). Presence of the co-occurring condition does not appear to impact treatment choice, however, a lower proportion of ED patients initiate treatment after diagnosis, compared with BPH-LUTS patients. SYSTEMIC DISORDERS/CONDITIONS – Clinical Outcomes Studies PSY1 Pain Incidence and Analgesic Consumption During Haemodialysis Sessions: Impact on Health-Related Quality Of Life Rebollo P 1, Arenas M D 2, Castejón N 1, Reichert García J 2, Delgado Conde P 2, Gutierrez Rivas P 2, Gil González M T 2 1LASER ANALYTICA, Oviedo, Spain, 2Hospital Vithas Perpetuo Socorro, Alicante, Spain . . . . . . . . . Objectives: To analyse the incidence of pain and the need of analgesics during haemodialysis sessions, and its impact on Health-Related Quality of Life (HRQoL). Methods: Data about the number of sessions in which 172 patients showed pain and needed analgesics were collected in 2 haemodialysis units in Spain during 3 months. Age, sex, comorbidities, (diabetes and cancer history), time on haemodialysis, pain complaints during haemodialysis sessions, intake of analgesics, opioids and antidepressant drugs, were collected. Generic HRQoL was assessed by means of the computer adaptive test CAT-Health, previously validated, through an iPAD. A negative score means that the HRQoL is worse than that of general population and a positive score, indicates that it is better. Results: Mean age (S. D) was 66.87 (13.32), being 44 patients (25.6%) aged over 75 years. 55.8% were male, 34.3% diabetic and 11.6% had cancer history. The median time undergoing haemodialysis was 51.50 months (27.75-84.50). 81 patients (47.4%) had pain during some session. The mean number of haemodialysis sessions with pain was 4.78 (range between 1 and 21 sessions), which represents 12.7% of the total number of sessions. 67 patients (39%) were usually taking analgesics, 37 (21.5%) opioids and 29 (16.9%) antidepressant drugs. Patients taking analgesics showed worse HRQoL: mean (S. A522 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 D) CAT-Health scores of -1.2412 (1.6065) vs -0.0119 (1.3495) (p< 0.001). Patients taking opioids also showed worse scores: -1.1009 (1.9349) vs -0.3288 (1.4169) (p= 0.037). Differences on HRQoL according to antidepressant drugs were not statistically significant. Episodes of pain during dialysis were concentrated in 24 patients who had 6 or more painful sessions. These patients were taking analgesics more frequently (86.4% vs 36.9%; p< 0.001) and showed worse CAT-Health score: -1.5391 (1.6348) vs -0.3243 (1.4981) (p= 0.005). Conclusions: Pain during haemodialysis sessions is very common and requires the frequent use of analgesics, having a negative impact on patients’ HRQoL. PSY2 Clinical Utility of the Collect Scale To Assess Comorbidities In Patients With Chronic Lymphocytic Leukemia De La Serna J 1, Carbonell F 2, Giraldo P 3, Lopez A 4, González I 5, García C 5, Orofino J 5, Roset M 6, Perulero N 6, Rios E7 1Hospital Universitario 12 de Octubre, Madrid, Spain, 2Consorci Hospital General Universitari Valencia, Valencia, Spain, 3Hospital Universitario Miguel Servet, Zaragoza, Spain, 4Hospital Universitari Valle d’Hebron, Barcelona, Spain, 5Roche Pharma, Madrid, Spain, 6IMS Health, Barcelona, Spain, 7Hospital Virgen de Valme, Sevilla, Spain . . . . . . . . . Objectives: COLLECT scale assesses comorbidities in patients with Chronic Lymphocytic Leukemia (CLL). Validation of COLLECT was a secondary objective of the MABERYC non-interventional study. The aim is to assess the clinical utility and the validity of COLLECT to guide treatment regimen prescribed to CLL patients. Methods: MABERYC study included patients with CLL, being or not previously treated, initiating treatment with Rituximab+chemotherapy. COLLECT was administered at baseline and 12 months following treatment finalization. Treatment response (TR) and safety was also assessed in the last visit. COLLECT categorizes comorbidity in low (0-3), moderate (4-7) and high (> 7). Changes in COLLECT were categorized in improvement (reduction≥ 2), without changes (variation< 2) and worsening (increase≥ 2). Results: MABERYC included 218 patients, 179 completed COLLECT at baseline. Patients had a mean age of 67.5 years, 73% were male, 53% were naïve, 37% had moderate comorbidity and 27% high comorbidity. At baseline, 42% of patients initiated treatment with Rituximab-Fludarabine-Ciclofosfamide (RFC), 30% with Rituximab-Bendamustine (RB), 18% with Rituximab-Clorambucile (RC) and 10% other patterns. Mean COLLECT score was higher in older patient, higher ECOG, previously treated (5.2 vs 4.2) and those receiving less aggressive treatments (5.8 RC, 5.4 RB, 3.6 RFC). Changes in COLLECT were analysed in 134 patients. CR could be associated to improvement in comorbidity. Complete remission was reached by 53% of patients with COLLECT improvement, 47% without changes, and 32% worsening. Number of adverse events (AEs) treatment related tend to be higher in patient with lower comorbidity (1.6 vs 0.9), using more aggressive treatments. Total number of AEs (related or not to treatment) tend to be higher in patients with higher comorbidity (7.8 vs 5.8). Conclusions: COLLECT scale assess comorbidity which is related with patients’ profile, treatment regiment prescribed, CR and AEs. COLLECT could assist decision-making on the intensity of the chemotherapy regimen to prescribe. PSY3 Impact of Biologics Use on Depression and Anxiety Frequency and Health Care Resource Utilization In Psoriasis: An Analysis Using the Quebec Provincial Drug Reimbursement Program Database Lachaine J 1, Lambert-Obry V 1, Gaumond S 2, Desjardins O 3 1University of Montreal, Montreal, QC, Canada, 2AbbVie Canada, Saint-Laurent, QC, Canada, 3AbbVie Canada, St-Laurent, QC, Canada . . . . Objectives: Psoriasis is a chronic inflammatory disease of the skin that cannot be cured. For patients with active moderate to severe psoriasis, biologics use is associated with an improvement in patients’ quality of life especially by reducing prevalence of psychological disorders. The objective of this study was to assess the impact of biologics use on depression and anxiety frequency and the number of medical visits. Methods: A retrospective study of the Quebec provincial drug reimbursement program (RAMQ) database was conducted using a randomly selected group of patients who have received at least one diagnosis of psoriasis between January 1st, 2007 and June 30th, 2012. To assess the impact of biologics use, time series analyses were performed. Time series analyses evaluate changes in the slope of a trend pre- and post-intervention, herein defined as biologics initiation. Trends in depression and anxiety frequency and medical visits frequency were compared for each year for a 5-year period before and after biologics initiation to assess the differences in slopes. Results: A total of 43,400 patients with psoriasis were included in the study (mean age= 54.6 [SD= 21.9] years, 53.7% females), of which 1,108 (2.6%) used a biologic agent. For patients who needed to be treated with biologics, the rates of change in the depression and anxiety prevalence increased by 3.4% and by 4.2% per year prior to biologics initiation respectively. After biologics initiation, the trends were still increasing, but at a statistically lower rate of 2.5% (p= 0.028) and of 2.4% (p=0.012) per year. Medical visits per patient increased during the 5-year period before biologics initiation. Visits frequency has reduced during the 5-year period after biologics initiation with a trend decreasing annually (p=0.002). Conclusions: The present analysis illustrates that biologics use reduces the increase in depression and anxiety frequency and decreases the number of medical visits. PSY4 A Real-World Characterization of Patients With “Moderate-ToSevere” Systemic Lupus Erythematosus Strand V 1, Johnson J 2, Vandeloo C 3, Galateanu C 3, Lobosco S 2 1Biopharmaceutical Consultant, Portola Valley, CA, USA, 2Adelphi Real World Ltd, Macclesfield, UK, 3UCB Pharma, Brussels, Belgium . . . . . Objectives: To characterize the patient (pt) group classified by physicians as having “moderate-to-severe” systemic lupus erythematosus (SLE) disease severity, and assess disease burden. Methods: Data were extracted from the Adelphi 2013 Lupus Disease-Specific Program, a multinational survey of clinical practice. Physicians com- pleted Patient Record Forms (PRFs); pts self-reported data including EQ-5D and the Work Productivity and Activity Impairment Index for SLE (WPAI-Lupus) in Patient Self-Completion Records (PSCs). Pt eligibility was determined by physicians; disease activity and severity were based on physician assessment. Data across countries were pooled. Results: Data were collected from rheumatologists in the USA (n=97), France (n= 37) and Germany (n= 35), including PRFs (550/200/207, respectively) and PSCs (303/109/149, respectively). Physician assessment of disease severity was predominantly based on affected organs and symptoms (45% and 35% of rheumatologists, respectively); 15% based severity on test results/clinical assessments. No disease activity index was widely used, 58% used their own assessment. Physician assessment of severity was imperfectly correlated with control of disease activity (activity controlled in 56.1% of “moderate-to-severe” pts, uncontrolled in 6.1% of “mild” pts). Pts with “moderate-to-severe” severity presented with greater severity and organ involvement, and a higher proportion experienced flares per 12-month period than “mild” pts (78.4% vs 52.6%). “Moderate-to-severe” severity was associated with a greater impact on HRQoL (EQ-5D: 0.72 vs 0.86; WPAI: 35.0 vs 16.3) than “mild” disease. Fewer “moderate-to-severe” pts were employed (full-time employment: 35.8% vs 48.8%), and a higher proportion required a care provider (6.6% vs 3.3%). Conclusions: SLE severity is not consistently assessed or defined in clinical practice: measures used in clinical trials are not routinely adopted in daily practice, whilst organ involvement and symptoms are central to physician assessment of severity. Correlation between severity and control of disease activity is imperfect. “Moderate-to-severe” severity is associated with a greater burden than “mild” disease. PSY5 Effectiveness of Heavy-Light Chain Quantitative Test: A Systematic Review Kim J H , Shin H W , Lee M National evidence-based healthcare collaborating agency, Seoul, South Korea . . . . . Objectives: Heavy-light chain (HLC) quantitative test can identity and quantify the heavy and light chain of each immunoglobulin class. The purpose of this study was to evaluate the effectiveness of HLC quantitative test. Methods: To evaluate the effectiveness of HLC quantitative test, systemic literature review using Ovid-MEDLINE, EMBASE, Cochrane library and eight domestic databases including KoreaMED had performed until October 10th, 2013. We included five cohort studies and one diagnostic evaluation study in the final evaluation. Two reviewers independently assessed the quality of included studies and extracted data on study. The qualities of these studies were assessed according to Scottish Intercollegiate Guidelines Network (SIGN) tool. Results: The correlation between HLC quantitative test with previous tests was evaluated in one study which patients with increased monoclonal IgA were enrolled, and the correlation coefficient with comparator tests was reported as 0.94 in that study. Clinical significance of quantitative HLC test for predicting prognosis was also reported in five cohort studies. Survival rate in patients with higher HLC ratio was significantly lower, and the increased IgA κ /λ ratio or IgM κ /λ ratio was significantly correlated with higher survival rate in patients with monoclonal gammaglobulinemia. The body of evidence as a whole suggests a Grade C for HLC quantitative test. Conclusions: HLC quantitative test is safe and effective test that can quantitatively measure the identified immunoglobulin type and predict the prognosis of patients with monoclonal gammopathy. PSY6 Disappearance of B-Symptoms In Comorbid Patients Receiving FirstLine Obinutuzumab (Ga101) -Chlorambucil (G-Clb) Or RituximabChlorambucil (R-Clb) for Chronic Lymphocytic Leukemia (Cll) Wenk Andres S 1, Becker U 2, Klawitter S 1, Wiesner C 3, Bernhardt A 2 Pharma AG, Grenzach-Wyhlen, Germany, 2F. Hoffmann-La Roche Ltd., Basel, Switzerland, 3Genentech Inc., South San Francisco, CA, USA . . . . . 1Roche Objectives: The CLL11 study (stage 2, NCT02053610; sponsored by F. Hoffmann-La Roche, Basel, Switzerland), compared G-Clb with R-Clb in 663 patients with previously untreated CLL and comorbidities. We compared the presence, disappearance and duration of absence of B-symptoms (fever, night sweats and weight loss) for G-Clb with R-Clb in CLL11. Methods: Patients were randomised to receive six 28-day (D) cycles (C) of G-Clb (N= 333) or R-Clb (N= 330; G: 1000 mg D1, 8 and 15 C1, D1C2–6; R: 375 mg/m2 D1C1, 500 mg/m2 D1C2–6; Clb: 0.5 mg/kg D1 and 15 each cycle). B-symptoms were assessed by a physician at baseline, D1C1, D8C1, D1 of each subsequent cycle, 28 days after last study drug and every 3–6 months during follow-up for progression-free survival. The data cut-off for this analysis was 9thMay 2013. Results: At baseline, 260 patients had B-symptoms (n= 126, G-Clb; n= 134, R-Clb). During the treatment period, B-symptoms disappeared in 107 (85%) and 124 (93%) patients who received G-Clb and R-Clb, respectively (relative risk [RR]: 0.92, 95% confidence interval [CI]: 0.84–1.00; p= 0.0554). Median time to first disappearance of all B-symptoms was 32.0 days for G-Clb and 35.0 days for R-Clb (hazard ratio [HR]: 1.24,95% CI: 0.96–1.61; p= 0.103). At end of treatment, 96 (76%) and 108 (81%) patients who received G-Clb and R-Clb, respectively, were B-symptom free (RR: 0.95, 95% CI: 0.83–1.07; p= 0.3899). In patients whose B-symptoms disappeared, the 25% quartile for duration of absence of B-symptoms (median not reached) was 16.4 months for G-Clb and 10.4 months for R-Clb (HR: 0.59, 95% CI: 0.36–0.97; p=0.0387). Conclusions: There was no clinically meaningful difference in B-symptom freeness at end of treatment, however the absence of B-symptoms was prolonged by 6 months for G-Clb versus R-Clb. PSY7 Comparison of Disease Status and Outcomes of Patients With Ankylosing Spondylitis (As) Receiving Adalimumab Or Etanercept Monotherapy In Europe Narayanan S 1, Lu Y 2, Hutchings R 2, Baynton E 2 1Ipsos Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, London, UK . . . . Objectives: To compare the disease status and outcomes of patients with AS receiving adalimumab and etanercept monotherapy in Europe. Methods: A multi- A523 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 country multi-center medical chart-review study of AS patients was conducted among rheumatologists in UK/France/Germany/Italy/Spain to collect de-identified data on patients who were recently treated with a biologic as part of usual care. Physicians were screened for duration of practice (3-30yrs) and patient volume (incl. > 5AS biologic patients/month) and recruited from a large panel to be geographically representative in each country. Eligible patient charts (≥ 3) were randomly selected from a sample of prospective patients visiting each center/practice during the screening period. Physicians abstracted patient diagnosis, treatment patterns/dynamics and patient symptomatology/disease status/outcomes. Patients on adalimumab/etanercept monotherapy were analyzed. Results: 329 eligible AS patient charts were abstracted; 141 on adalimumab (male: 89%, age: 41.9yrs, average months on adalimumab: 29.5, 95% on first biologic) and 102 on etanercept (male: 78%, age: 43.2yrs, average months on etanercept: 32.2, 98% on first biologic). Top-3 comorbidites (adalimumab vs. etanercept) were dyslipidemia: 6% (range: 0% (Germany) -11% (Italy)) vs. 8% (range: 0% (UK/Germany) -15% (Spain)), obesity: 4% (range: 0% (Italy) -7% (Germany)) vs. 6% (range: 0% (UK) -11% (France)), and depression: 6% (range: 0% (Italy/Spain) -15% (UK)) vs. 5% (range: 0% (Germany/Spain) -7% (Italy/France)). Among patients with available data, latest lab measures documented were (adalimumab vs. etanercept): ESR: 15.1mm/h (range: 11.6 (Germany) - 23.6 (Italy)) vs. 13.6mm/h (range: 8.0 (Germany) -19.2 (Italy)), CRP: 6.7mg/dl (range: 2.6 (Spain) -8.9 (France)) vs. 7.3mg/dl (range: 1.9 (Germany) -10.0 (Italy)), rheumatoid factor-positive: 6% (range: 0% (Spain) -10% (Germany/Italy)) vs. 10% (range: 0% (UK) -19% (Italy)). Latest disease severity measures documented were (adalimumab vs. etanercept): Swollen Joint Counts: 0.8 (range: 0.1 (Germany) -1.4 (UK)) vs. 0.5 (range: 0.3 (France) -0.9 (Italy)), Tender Joint Counts: 1.2 (range: 0.5 (Germany) -1.9 (Italy)) vs. 1.1 (range: 0.4 (France) -2.1 (Italy)), HAQ: 1.1 (range: 0.7 (Spain) -1.5 (UK)) vs. 1.3 (range: 1.0 (France) -3.0 (Germany)). Conclusions: Among AS patients receiving adalimumab or etanercept monotherapy, disease severity differed within the EU5, with patients on adalimumab, and patients in Italy, having marginally higher burden and poorer outcomes. Factors influencing the observed patterns and the impact of specific biologic treatments on these observations warrant further scrutiny to optimize therapeutic interventions and improve outcomes. PSY8 A Double-Blind Controlled Study of the Efficacy and Safety of Long-Acting Amfepramone Treatment In Mexican Obese Patients H 1, Pizarro M 2, Rosado J 3, Rizzoli Cordoba A 2, Fernández Soto Molina Castellanos Perez del Valle C 4, Reyes García J G 4 1HS Estudios Farmacoeconómicos, Mexico City, Mexico, 2Hospital Infantil Federico Gomez, Distrito Federal, Mexico, 3FES Zaragoza, Mexico City, Mexico, 4Instituto Politécnico Nacional, Mexico city, Mexico . . . . . . . erase 45U/kg). Non significative differences implies overlapped confidence intervals for the three therapies in all the efficacy endpoints. Conclusions: This analysis suggests a comparable effect of the velaglucerase and taliglucerase on platelet count increase, as well as a comparable effect of the three enzymes approved for Gaucher disease on Hemoglobin concentration increase, Liver and Spleen volume reduction. PSY10 The Oncology Pain Treatment Clinical Studies Quality Assessment Giermaziak W 1, Faluta T 2, Bondaryk Z 1, Markowska A 1, Deszcz M 3 Medical Library, Warsaw, Poland, 2Military Institite of Medicine, Warsaw, Poland, 3Medical University of Warsaw, Warsaw, Poland . . . . . 1Main Objectives: Opioids have been extensively used in the treatment of oncological pain for many years. Due to various therapeutic doses and imprecise measurements of the reduction of pain, it is difficult to make meta-analysis. There are only a few precise systematic reviews assessing efficacy of pain medicines (especially meta-analysis). The study goal was to assess the quality of published data of oncological pain treatment and create a meta-analysis when it is possible. It has been led in the context of a limited amount of meta-analysis for the opioid preparations’ use. Methods: Opioids have been extensively used in the treatment of oncological pain for many years. Due to various therapeutic doses and imprecise measurements of the reduction of pain, it is difficult to make metaanalysis. There are only a few precise systematic reviews assessing efficacy of pain medicines (especially meta-analysis). The study goal was to assess the quality of published data of oncological pain treatment and create a meta-analysis when it is possible. It has been led in the context of a limited amount of meta-analysis for the opioid preparations’ use. Results: The primary research provided data of high quality. The average Jadad scale rating for all primary publications was equal to 3.43. The individual studies presented the effectiveness of individual drugs. However, their variation with respect to the scale assessing either pain intensity or diversity of a population makes the presentation of meta-analysis impossible to be presented. Conclusions: Oncology pain treatment has been assessed in high number of published studies. However due to their heterogeneity the meta-analisis methodology is limited to be used. PSY11 A Pilot Study of the Effectiveness of Treatment Patients With Hemophilia In Ukraine Zalis’ka O 1, Mudrak Y 2, Mudrak I 2 1Danylo Halytsky Lviv National Medical University, Lviv, Ukraine, 2Vinnyca National Medical University after M. Pyrogov, Vinnyca, Ukraine . . . Objectives: Amfepramone is an anorectic drug used for the short-term treatment of obesity; however, its efficacy and safety in the long-term has been scarcely studied. To determine the efficacy and safety of long-acting amfepramone treatment in Mexican adult obese patients by a double blind, randomized and placebocontrolled clinical trial study. Methods: One hundred and fifty six volunteers with a body mass index (BMI) greater than 30 kg/m2 and less than 45 kg/m2 were randomized to receive a tablet of 75 mg amfepramone or placebo daily during 6 months. Primary outcomes were the absolute body weight loss and the percentage of patients who achieved at least 5% or 10% weight loss at 3 and 6 months, whereas secondary outcomes were the improvement of anthropometric and metabolic parameters. Results: Amfepramone treatment showed a superior efficacy to decrease the body weight than placebo at 3 (-4.9 ± 0.25 versus 0.7 ± 0.32 Kg) and 6 (7.7 ± 0.52 versus 1.1 ± 0.7 Kg) months. In addition, Sixty-four and thirty-four patients achieved at least 5% or 10% weight loss, respectively, with amfepramone at 6 months, compared with eight and zero patients of placebo. Amfepramone also improved BMI and waist circumference, but not waist-hip index (WHI), glucose, total cholesterol, low-density lipoproteins (LDL), high-density lipoproteins (HDL) and triglycerides at 3 and 6 months. Amfepramone produced only mild adverse events and they were presented in a greater number than placebo only at 3 months, being the main adverse event dry mouth. Conclusions: Data suggest that amfepramone is effective and well tolerated in the long-term treatment of Mexican obese patients. Objectives: The incidence is 87% of hemophilia and often leads to disability (90% of patients). WHO recommends annual demand of 1 patient of 30 000 IU of factor VIII, but in Ukraine with a national program funded an average 5786 IU per year (20% of the need). In Ukraine, began production of the domestic drug factor VIII “BioKlot A” from 2012. The aim was to determine the effectiveness of substitution treatment “BioKlot A” native medicine in the treatment of patients with hemophilia A. Methods: In a pilot clinical study examined 168 patients diagnosed with hemophilia A severe form of male, aged 18-52 years, who were hospitalized. Experimental group received “BioKlot A” control group – “Oktanat” in equal doses of 40 IU / kg / day during the first day, 20 IU / kg / day - the next day. All patients were determined by the intensity of joint pain, joint swelling index, joint circumference, blood count, and others criteria. Results: Within 24 hours after “BioKlot A” (average dose rate of 10 000 IU) joint pain significantly decreased after 48 hours - the pain disappeared and the joint volume decreased by 14% in 72 hours - the amount of diminished joint by 22% and increased range of motion by 28%. Time coagulation by Lee-White 24 hours decreased from 30 ± 0,2 minute to 19 ± 0,2 minute. In the control group, the average dose rate amounted to 10750 IU “Oktanat” and dynamics of clinical and hematological data were not significantly different from the experimental group. Conclusions: The established similar effectiveness of factor VIII “BioKlot A” (Ukraine) and “Oktanat” (Austria) with clinical and laboratory criteria. Production of factor VIII preparation in Ukraine has significant socio-economic impact, providing improved quality of life of patients. PSY9 Comparative Effectiveness Study of Enzymatic Replacement Therapies In the Treatment of Gaucherxs Disease On Adults PSY12 How To Improve Health Outcomes In the Treatment of Chronic Myeloid Leukemia Muciño-Ortega E 1, Mendoza C F 1, Rubio M E 2 1Pfizer S.A. de C.V., Ciudad de México, Mexico, 2Pfizer S.A. de C.V., Mexico City, Mexico 1Novartis, Porto Objectives: This study aims to perform a comparative effectiveness study of licensed Enzymatic Replacement Therapies (ERT´, taliglucerase, imiglucerase and velaglucerase) on the treatment of Gaucher´s disease on adult patients. Methods: A systematic review of published clinical trials was performed on three Medical databases. Inclusion criteria were randomized, double-blind and phase III clinical trials measuring the efficacy of ERT on adult patients. Using meta-analysis methodology, the difference in the clinical outputs due the use of a specific ERT in four endpoints was assessed: Hemoglobin concentration increase, platelet count increase, Spleen and Liver volume reduction. Since no placebo group is reported in clinical trials of orphan drugs, outcomes previous to ERT were considered as control. In this study we report the p-value of test for subgroup differences to compare among the three alternatives. Results: For the higher dose (60U/kg for the three enzymes), the data reported allowed a taliglucerase, velaglucerase and Imiglucerase comparison in efficacy measures (except the analysis of imiglucerase in platelet count endpoint). Hemoglobin concentration increased a mean of 1.79 g/dL [95% CI 1.08, 2.51; p-value for therapy differences= 0.82]. Spleen volume decreased a mean of -5.17 Multiples of Normal (MN) [-10.09, -0.25; p-value for therapy differences= 0.90]. ERT showed a trend favouring liver volume reduction, -0.28 MN [-0.57, 0.01; p-value for therapy differences= 0.80] and increased platelet count, 33.11x109/L [-1.67, 67.89; p-value for therapy differences= 0.74]. However, for data limitations, we were unable to perform the comparison for the lower dose (taliglucerase and imiglucerase 30U/kg, velagluc- Objectives: Imatinib and nilotinib are two tyrosine kinase inhibitors (TKI) used in the treatment of chronic myeloid leukemia (CML). Recent data reveals that patients who achieve deep molecular responses may be able to stop TKI, which has important clinical and economic implications. The efficacy of nilotinib was compared to imatinib in newly diagnosed patients and the proportion of patients achieving deep molecular response MR4.5, after 5 years, was 31% for imatinib and 54% for nilotinib. The aim of this analysis was to estimate the number of patients remaining in TFR with nilotinib compared with imatinib. Methods: According to STIM trial, 39% of patients remained in complete molecular response after discontinuation of imatinib. In the Stop 2G-TKI study, 62.2% remained in undetectable molecular residual disease after nilotinib discontinuation. Based on this data, we developed an outcomes model to estimate the number of patients remaining in TFR long term after consolidated treatment with nilotinib versus imatinib. It was assumed that after 5 years of treatment with imatinib or nilotinib, patients with consistent MR4.5 are eligible to interrupt treatment. After 12 months, patients who maintain sustained undetectable BCR-ABL will remain in TFR. Results: Considering an annual CML incidence of 1.33 per 100,000 patients, of which 8% are intolerant to treatment, we estimated 128 de novo CML patients eligible for TKI treatment in Portugal per year. For this cohort of patients, treatment with nilotinib would result in an additional 28 patients in TFR when compared with imatinib (43 versus 15). Given the number of patients remaining in TFR, overall treatment costs with nilotinib, after . . . . . Viriato D 1, Marques A 1, Fonseca M 1, Almeida A 2 Salvo, Portugal, 2Instituto Português de Oncologia de Lisboa, Lisboa, Portugal . . . . A524 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 6 years, would be 34% lower. Conclusions: Treatment with nilotinib is expected to result in better health outcomes, with more patients achieving TFR. This initial TKI investment should reflect itself in long term economic benefits. PSY13 Pre-Symptomatic Genetic Testing In Familial Amyloid Polyneuropathy: The Reproductive Options Inês M1, Rodrigues F 2, Saraiva J 3, Costa J 4 1Instituto de Medicina Molecular, Lisboa, Portugal, 2Centro Hospitalar e Universitário de Coimbra, Coimbra, Portugal, 3Centro Hospitalar e Universitário de Coimbra, Medical Genetics Unit, Hospital Pediátrico, Coimbra, Portugal, 4Institute of Molecular Medicine, Lisbon, Portugal . . . Objectives: Familial Amyloid Polyneuropathy (FAP) is a rare, rapidly progressive, debilitating and life-threatening neurodegenerative disease. Pre-symptomatic genetic testing (PST) can contribute to reduce FAP prevalence, informing carriers about risk of transmission to offspring’s and available reproductive options such as Pre Natal Diagnosis (PND) or Pre Implantation Genetic Diagnosis (PGD). This study aims to describe FAP carrier’s reproductive options after PST and to analyse its socio-demographic determinants. Methods: Data from a cohort of 145 FAP carriers that underwent PST at Medical Genetics Unit (HP/CHUC – Portugal, 2000-2012) was used to analyse reproductive options of FAP carrier’s. An econometric logistic model was specified to identify determinants for natural reproduction option. Results: The subjects were mainly women (55%) with mean age of 35 years (SD= 15) at PST entry. Most participants were in a relationship 92/145 (63%) and mean number years of education was 9 years (SD= 14). 75/145 (52%) of subjects were already parents, reporting a total of 144 children. 21/145 (14%) subjects decided not to have offspring following their positive test result. No child adoption was reported. 27/145 (19%) of subjects report offspring posteriori to PST: 19/27 (70%) subjects undergo natural reproduction and 8/27 (30%) reproduction with PND or PGD support. A total of 9 offspring non-FAP carriers (with PND/PGD) and 24 offspring from natural reproduction were observed until data collection. The logistic regression for decide on natural reproduction confirms statistical significance (p-value<0.05) for only two variables: being in relationship and having previous children. No statistical significance was observed for gender, age at PST or number of education years. Conclusions: There is evidence of a high proportion of carriers with offspring’s when initiate PST and high proportion of natural reproduction within FAP carriers after PST. An ethical physiological case-by-case approach is essential to learn more about the determinants for preventing FAP offspring transmission. PSY14 Trends In Prescription Opiate Use Among Patients With Commercial Or Government Sponsored Health Insurance In the Us From 2010-2013 Bonafede M M 1, Palmer L A 2 Health Analytics, Cambridge, MA, USA, 2Truven Health Analytics, Bethesda, MD, USA . . . prostate cancer) (C60-63), 10.7% had malignant neoplasms of digestive organs (C1526) and 8,4% showed malignant neoplasms of ill-defined, secondary and unspecified sites (C76-80). The outpatient diagnosis rate for CLL was 94.9%, inpatient rate 0.6% and in-and outpatient rate 4.6%. Overall, 266 of 1,405 pts (18.9%) (175 men [65.8%], 91 women [34.2%]) received chemotherapy in 2012 (ATC Code L01* 74.1%, PZN 9999092 23.0%, OPS 854* 2.9%). Most patients received outpatient treatment (94.0%), with 5.3% of patients received both out- and inpatient treatment and 0.7% inpatient treatment. The most commonly used treatments were rituximab (26.7%), bendamustine (20.2%), chlorambucil (11.1%), cyclophosphamide (7.7%), fludarabine (6.2%) and other treatments (28.1%). Conclusions: The majority of patients being diagnosed with CLL did not require treatment within a time period of a year. Approximately 1/3 of patients had a second malignancy, predominantly skin cancer. Treatment was primarily composed of chemotherapy or chemoimmunotherapy. PSY16 Acute Myeloid Leukemia and Myelodisplasic Syndrome Treated With Intensive Chemotherapy In France Based On National Hospital Databases (Pmsi) Thiebaut A 1, Lafuma A 2, Bureau I 2, Boyaval G 3, Bensoussan C 3, Godard C 3 1Department of Hematology, Grenoble University Hospital, La Tronche, France, 2Cemka-Eval, Bourg la Reine, France, 3MSD France, COURBEVOIE, France . . . . . . Objectives: To estimate the annual number of patients in France with Acute Myeloid Leukemia (AML) and MyeloDysplasic Syndrom (MDS) and treated with intensive chemotherapy susceptible to induce neutropenia. Methods: French hospital databases named PMSI record medical information about all the hospitalizations performed annually in France. From 2006 to 2012, databases allow linking the stays over time of a given patient with an anonymous number. In this study, PMSI databases were used to identify patient, aged more than 15, with diagnosis of AML or MDS and who were alive during the year 2012. Then, patients who underwent hospital stays for chemotherapy during more than 5 days were identified and considered as neutropenic according to experts opinions. Results: Since 2006, 51,386 patients with at least one diagnosis of AML or MDS and aged more than 15 were identified, from which only, 16,006 had at least one hospital stay in 2012 and 3,468 were hospitalized more than 5 days for chemotherapy. Among those patients, 55.2% were male, mean age was 60.4 years, 30% died during a hospital stay and 19% (664) were bone marrow grafted during the year 2012. These patients had 1.8 stays for chemotherapy per year with average chemotherapy duration of 27 days. Two third of these patients (34%) were diagnosed in 2012 and 23% in 2011. Conclusions: Among the 16,006 patients diagnosed AML or SMD and hospitalized in France in 2012, 3, 468 (21%) received intensive chemotherapy inducing neutropenia, putting them at high risk of invasive fungal infection. . 1Truven Objectives: To describe the prevalence of outpatient prescription opiate use among patients with Commercial, Medicare Supplemental, or Medicaid insurance in the US after 2010. Methods: Patients were identified using the Truven Health Commercial and Medicare Supplemental databases from 2010-2013 and Medicaid Multi-State databases from 2010-2012. We estimated the proportion of patients with at least one outpatient opiate prescription for each calendar year among patients with continuous medical and pharmacy enrollment for the entire calendar year, stratified by insurance type and age (< 18, 18-64, 65+). Average opiate days supply per calendar year was also described. Results: On average, a total of 32.9 million (m) patients were included in each calendar year analysis with an average of 6.7m Commercially-insured children, 20.2m Commercially-insured non-elderly adults, 2.5m Medicare Supplemental adults (age 65+), 2.6m Medicaid-insured children and 0.8m Medicaid-insured adults per year. Among commercially-insured non-elderly adults, the prevalence of opiate prescriptions decreased from 25.1% to 23.4% from 2010 to 2013 (p< 0.001) with a similar decrease among Medicare Supplemental patients (29.5% to 27.6%, p< 0.001). Among Medicaid adults, opiate prescription prevalence was lower prevalence in 2012 (38.9%) than in 2010 (44.1%) or 2011 (45.2%). Among children (age <18), opiate prescription prevalence use was slightly higher for Medicaid than commercially-insured children (9.9% versus 7.4%, p<0.001); opiate prescription prevalence dropped by 15.3% among commercially-insured patients from 2010 to 2013 (p<0.001) and by 12.1% among Medicaid children from 2010 to 2012 (p< 0.001). Average annual days supply increased over time among Medicaid adults (92.8 to 100.5 days), commercially-insured non-elderly adults (41.7 to 44.5 days) and Medicare Supplemental adults (65.9 to 68.6 days); days supply per year was steady among children (calendar year range: 7.5-9.1). Conclusions: There were modest reductions in the prevalence of prescription opiate use from 2010-2013. Further and more detailed monitoring is needed to parallel efforts to curtail inappropriate use. PSY17 Allogeneic Stem Cell Transplant With Graft Versus Host Disease In France In 2012 Based On National Hospital Databases (Pmsi) Thiebaut A 1, Lafuma A 2, Bureau I 2, Boyaval G 3, Bensoussan C 3, Godard C 3 1Department of Hematology, Grenoble University Hospital, La Tronche, France, 2Cemka-Eval, Bourg la Reine, France, 3MSD France, COURBEVOIE, France . . . . . . Objectives: To estimate the number of patients receiving Allogeneic Stem Cell Transplant (ASCT) and presenting with graft versus host disease (GVHD) in 2012 in France. Methods: French hospital databases named PMSI record medical information about all the hospitalizations performed annually in France. From 2006 to 2012, databases allow linking the stays of a given patient over time with an anonymous number. In this study, PMSI databases were used to identify patient, aged more than 15, with allogeneic stem cell transplant through adequate Diagnosis Related Group and who were alive, hospitalized during the year 2012. Patients with diagnosis of GVHD or typical diagnoses of GVHD symptoms (sclerodermia, diarrhea, mucitis and keratitis) were identified. Results: Since 2006 to 2012, 9,855 patients received allogeneic stem cell transplant in France of which 3,469 died during a hospitalization over the period, with an annual death rate stable around 20%. In 2012, 3,842 ASCT patients aged more than 15 were hospitalized at least once, and the proportion with GVHD was estimated to 40% (1,574). Acute GVHD diagnosis or typical GVHD symptoms occurring less than 100 days after grafting appeared in 274 patients and chronic GVHD was diagnosed in 1300 patients. Mean age was 47.8, 61.4% were male and 22% of these patients died during a stay in 2012. Among the ASCT patients hospitalized in 2012, only 5% were grafted in 2006 and this figure regularly increased with the year of the procedure to reach 35% grafted in 2012, corresponding to a mean annual mortality rate of around 20%. Conclusions: Among the 3,842 ASCT patients hospitalized in France in 2012, 1,574 (40%) presented GVHD, putting them at higher risk of immunosuppression and invasive fungal infection, but also at higher risk of death. PSY15 Prevalence and Treatment of Chronic Lymphocytic Leukaemia (Cll) In Germany: An Analysis of Sickness Funds PSY18 The Presence of Anxiety Among Elementary Schools Obese Children 1Janssen-Cilag Tomeczkowski J 1, Leisten M K 1, Metin H 1, Khuen C 1, Fleischmann J 2, Tapprich C 1 GmbH, Neuss, Germany, 2Janssen-Cilag Germany, Neuss, Germany 1University Objectives: No national registries for chronic lymphocytic leukaemia (CLL) exist in Germany. The objective of this analysis was to examine the number of patients with CLL diagnosed (with or without other diagnoses of cancer) and to characterize the types of treatment being utilized and care settings using sickness funds claim data. Methods: This analysis evaluates data from 1,771,225 beneficiaries in 2012 from different statutory sickness funds. Patients with CLL were identified by ICD-10 C91.1*, oncological co-diagnoses by ICD-10 C00-79; D37-48 and chemotherapy by Anatomical Therapeutic Chemical (ATC) Code L01*, pharmacy number (PZN) 9999092 and/or operating and procedure code (OPS) 854*. Results: In total, 1,405 patients with a diagnosis C91.1* (CLL) were identified (prevalence 79/100,000) with 60.2% were male and 39.8% female. Overall, 32.2% of patients with CLL had a co-diagnosis with another cancer; 25.1% melanoma (C43-44), 21.3% had neoplasms with unknown behaviour (D37-48), 11.8% had malignant male GU neoplasms (90% Objectives: From among the consequences of the obesity very considerable the emotional and distressful disturbances with which it grew fat children day, than day they fight. Assessing a what kind of measure the sport influences it on was our aim the nutritional and the mood life, furthermore, that the obesity influences the learning and the mood life. Methods: We made a questionnaire survey in the primary schools of Szekszárd, 13-14 year ones among students. Inclusion it was a criterion that only 7-8. are class children may have taken a part in the examination, furthermore without parents consent, and sufferers may not have taken a part in the examination only in a serious illness. All element 149 heads. At the compilation of the questionnaire used Helena (Healthy Lifestile in Europe by Nutritrion in Adolescence) name from a project for us essential questions, and we complemented it their edited our questionnaire implied altogether 150 questions with questions, . . . . . . . Germán Z 1, Botár A 2, Sziládiné Fusz K 1, Pakai A 3, Boncz I 4, Fullér N 1, Oláh A 1 of Pécs, Pécs, Hungary, 2Balassa János Hospital, Szekszárd, Hungary, 3University of Pécs, Zalaegerszeg, Hungary, 4Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . . A525 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 so. Results: Our 1. hypothesis, that the children with normal weight move much, than their obese companions, not proven true (p= 0,778). It was the next supposition that less are truth with a subject for the students with normal weight they have difficulties, than for the obese students. According to the examination, the normal one and the obese child with a subject truth in the look of difficulty, we found a significant difference (p= 0,015). The third hypothesis, that, in the obese group taller the anxious children’s proportion, than the did not grow fat in a group, not proven true. The overweight children (n= 46, medium= 109, 76) anxiety with a significantly bigger degree was not showed based on the scale the normal one tápláltsági state at students (n= 95, medium= 104, 56 dot, p= 0,515). Conclusions: The enrolled students showed no significant difference between the observed parameters of the overweight and sporting habits. SYSTEMIC DISORDERS/CONDITIONS – Cost Studies PSY20 Clinical and Cost-Effectiveness and Budget Impact of Routine Use of Bispectral Index Monitors In Theatres Vincent O 1, Sabatelli L 2, Craig J 1, Taylor M 1 1York Health Economics Consortium, York, UK, 2GLOB MOD, Barcelona, Spain . . . . Objectives: Bispectral Index (BIS) monitoring systems monitor spontaneous electroencephalography, track sedative drug effects and help guide anaesthetic administration. Evidence shows such monitoring enables savings on anaesthesia and reduces adverse events compared with monitoring clinical signs only. However, the cost-effectiveness of monitors is uncertain. Methods: A cost-effectiveness model and budget impact analysis were developed to compare the outcomes of monitoring the depth of anaesthesia with BIS monitors compared with standard clinical monitoring. Over a five-year period, the model estimated the incremental cost per quality-adjusted life year (QALY) and incremental number of adverse events avoided in theatres with anaesthetists using BIS monitors compared with standard care. Data values were obtained from peer-reviewed literature. Subgroup analysis was conducted for an elderly patient group. Sensitivity analyses explored uncertainty in the model. A budget impact model examined the financial impact of adopting BIS monitors across the UK. Results: The modelled results showed that using BIS monitors dominated clinical observation of signs and use of electrocardiograph and other devices, being cost saving by £82 per operation in the adult population, whilst improving QALYs by 0.016 per patient. The budget impact analysis showed a cumulative saving of over £136 million if theatres in the UK adopted a phased increase in monitor use such that 1.35 million surgical procedures were conducted using these monitors in five years time. Conclusions: Adopting BIS monitors is cost-effective and results in substantial cost savings compared with observing clinical signs plus conventional devices only. . . . . . Objectives: to estimate the costs of drug supply for registered rare diseases (RD) patients in Russia in 2013. Methods: The study included two steps: 1) the analysis of the current situation with financing of drugs for RD patients in Russia in 2013 and 2) estimating the unmet needs - necessary weighted average costs (nWAC) for a pathogenetic pharmacotherapy of registered patients with RD in the same year. Number of registered patients with 24 RD listed in the Government Regulation of Russian Federation № 403 from April 26, 2012 was estimated for the studied year. Number of registered patients and actual financing has been received from the database of regional public health services. The nWAC for expensive pathogenetic pharmacotherapy of patients with 24 RD were calculated on the basis of standards of therapy and experts’ survey results. The current situation with financing of RD drugs in Russia in 2013 was compared with unmet needs (nWAC). Results: 4.4 billion rubles ($125.7 million) were spent from regional budgets in Russia for 11,173 patients (0.008% from all population of Russia) with 24 RD in 2013. The largest patient group with 3,460 patients was a classic phenylketonuria group. The prognostic annual nWAC for pharmacotherapy of the most expensive patient with type VI mucopolysaccharidosis were 40.9 million rubles ($1.2 million) per child and 71.5 million rubles ($2.0 million) per adult. The nWAC for pharmacotherapy of registered patients with 24 RD in 2013 were 15.7 billion rubles ($448.6 million). The annual nWAC for pharmacotherapy of the most expensive disease (idiopathic thrombocytopenic purpura (Evans syndrome) was estimated as 2.9 billion rubles ($82.8 million) per group of 2,700 patients. Conclusions: The state financing of pharmacotherapy for patients with 24 RD in Russia should be increased approximately in 3.5 times to fulfill the unmet needs. PSY22 Budget Impact Analysis of Drugs for Ultra-Rare Non-Oncological Diseases In Europe Schlander M 1, Gandjour A 2 1University of Heidelberg, Wiesbaden, Germany, 2Frankfurt School of Finance & Management, Frankfurt, Germany . Kanters T A , Steenhoek A , Hakkaart L Erasmus University Rotterdam, Rotterdam, The Netherlands . . . . Objectives: The relatively low budget impact of orphan drugs is often used as an argument in reimbursement decisions. However, overall, the budget impact of orphan drugs can still be substantial. In this study, we assess the uptake and budget impact of orphan drugs in the Netherlands. Methods: We examined the number of orphan drugs, the number of patients and budget impact of orphan drugs in the Netherlands in the period 2006 to 2012, both for inpatient and outpatient orphan drugs. Budget impact was provided in absolute numbers and relative to total pharmaceutical spending. Results: The number of orphan drugs and patients treated increased substantially over the period studied. Overall, budget impact increased substantially over a period of six years, both in absolute terms (426% increase) as well as relative to total pharmaceutical spending (378% increase). Growth rates decreased over time. In 2012, 17% of available drugs had an individual budget impact of more than € 10 million per year. Conclusions: Individual budget impact of orphan drugs is often limited, although exceptions exist. However, in total, the budget impact of orphan drugs is considerable and has grown substantially over the years. This could potentially influence reimbursement decisions for orphan drugs in the future. PSY25 Budget Impact Analysis of Introducing Biosimilar Infliximab for the Treatment of Auto Immune Disorders In Five European Countries . Sura M 1, Gerasimova K 1, Omelyanovsky V V 1, Avxentyeva M 1, Tatarinov A 2, Fedyaev D 2 1The Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia, 2Financial Scientific Research Institute of the Ministry of Finance of Russia, Moscow, Russia . PSY24 Budget Impact of Orphan Drugs In the Netherlands In The Period 2006-2012 Jha A 1, Upton A 2, Dunlop W 1 1Mundipharma International Limited, Cambridge, UK, 2Abacus International, Bicester, UK PSY21 Estimating the Costs of Drug Supply for Rare Diseases Patients In Russia . estimate was based on prevalence data for URDs for which patented drugs are currently available and for which drugs are in clinical development and hence may be expected to be launched in the foreseeable future. A power function was used to estimate the relation between (decreasing) prevalence and (increasing) cost per patient. For drugs in development, we applied phase duration data and attrition rates from the Tufts Center for the Study of Drug Development database. Results: A total of 18 drugs under patent protection for non-oncological URDs were identified. Furthermore, 29 drugs for non-oncological URDs under development that have the potential of reaching the market by 2021 were found. Total budget impact over 10 years was estimated to be € 14,112 and € 4,965 million for approved and pipeline URD drugs, respectively (total: € 19,077 million). Relative to total pharmaceutical expenditures in Europe, spending on drugs for URDs is estimated to rise from 0.7% at present to 1.6% in 2021. Univariate sensitivity analyses and extreme scenario analyses suggesting robustness of this projection will be presented. Conclusions: Our analysis does not support concerns regarding an uncontrolled growth in expenditures for drugs for URDs. Nevertheless, continuous monitoring of the budget impact as an input to rational policy making is recommended. . Objectives: Ultra‐rare disorders (URDs) have been defined by a prevalence of less than 1 per 50,000 persons. On a per patient basis, the annual acquisition costs of drugs for URDs can be very high, and there have been concerns that expenditures for these products might escalate in the future. The goal of this study was therefore to provide a budget impact analysis (BIA) of drugs for ultra-rare non-oncological diseases in Europe. Methods: The BIA had a time horizon of 10 years (from 2012 to 2021) and adopted the perspective of all European payers in combination. The . . Objectives: Biosimilar infliximab has been approved by EMA for the management of inflammatory autoimmune disorders including rheumatoid arthritis (RA), ankylosing spondylitis (AS), Crohn’s disease, ulcerative colitis (UC), psoriatic arthritis (PsA), and psoriasis based on quality, safety and efficacy profiles comparable to infliximab. The aim of this study was to evaluate the five-year budget impact of introducing biosimilar infliximab in the management of RA, AS, Crohn’s disease, UC, PsA, and psoriasis from the health care system perspective. Methods: An Excel-based budget impact model was developed. The numbers of patients eligible for infliximab were calculated based on disease prevalence rates in Germany, Italy, Belgium, the Netherlands and the United Kingdom. The price of biosimilar infliximab is not yet known; therefore three discount scenarios versus infliximab (10%, 20%, and 30%) were applied. Market share was assumed to be 25% in the first year in all scenarios. Annual market share growth was varied in each of the scenarios at 20%, 30% and 40%, respectively. Results: The combined net budget savings for Germany, Italy, Belgium, the Netherlands and the United Kingdom in the first year were € 17.8, € 35.5 and € 53.3 million for the 10%, 20% and 30% price discount scenarios, respectively. Over a 5 year period the net budget savings were € 132.8, € 322.8 and € 532.8 million for the 10%, 20% and 30% price discount scenarios, respectively. Conclusions: The introduction of biosimilar infliximab as a treatment option for patients with RA, AS, Crohn’s disease, UC, PsA, and psoriasis could achieve substantial cost savings for health care systems. In the price discount scenarios tested, the total combined savings across Germany, Italy, Belgium, the Netherlands and the United Kingdom over a 5 year period ranged from € 132.8 million to € 532.8 million. The net budget impact was highly sensitive to market uptake rates and the price discount applied. PSY26 Budget Impact Analysis of Belimumab In the Treatment of Patients With Systemic Lupus Erythematosus In Russian Federation Kulikov A , Komarov I , Pochuprina A I.M. Sechenov First Moscow State Medical University, Moscow, Russia . . . Objectives: To estimate budget impact analysis (BIA) of belimumab plus standard of care (SoC) comparing the SoC alone in the treatment of patients suffering from systemic lupus erythematosus (SLE). Methods: BIA was conducted of the belimumab treatment plus SoC vs SoC alone. Costs of treatments in both groups contained following direct medical costs: costs of drugs and administration, costs of diagnostic laboratory and instrumental procedure, costs of inpatient and outpatient visits, costs of SLE complications and also adverse event costs. A five-year time horizon was used. All costs in both groups were estimated to their present value using a 5 % discount rate. Results: Costs of the course of belimumab therapy were 2,118,449 RUB/ 45,581 EUR for 5 years but the difference in the required budget funds between belimumab treatment groups and SoC alone treatment group amounted to 1,876,965 RUB/ 40,385 EUR for 5 years with a 5 % discount rate. The decrease of the difference in the required budget funds between these two groups was due to a lower A526 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 frequency of inpatient visits and pulmonary, cardiovascular, renal and skin complications of SLE in group treated with belimumab. Therefore the use of belimumab led to a reduced difference in the required budget funds from 2,118,449 RUB/ 45,581 EUR to 1,876,965 RUB/ 40,385 EUR and the reduction ran as high as 241,484 RUB/ 5,196 EUR for 5 years. Conclusions: The use of belimumab in the treatment of patients with SLE resulted in budget spending. However, a good safety profile and efficacy of belimumab help reduce the costs for 241,484 RUB/ 5,196 EUR for 5 years in belimumab treatment group. PSY27 Public Expenditure On Authorised Orphan Drugs In the Czech Republic Between 2008 and 2013 Vocelka M 1, Spurna M 1, Fuksa L 2, Hambalek J 1 Institute for Drug Control, Prague, Czech Republic, 2Charles University, Faculty of Pharmacy, Hradec Kralove, Czech Republic . . . . 1State Objectives: The aim of the study was to determine public expenditure on medicines for rare diseases from the public health sector in the Czech Republic (CZ). Methods: We identified orphan medicinal products (OMPs) registered by the European Medicines Agency (EMA) until December 2013 in the public database of EMA. In the base-case scenario, medicinal products available in the CZ were considered OMPs only within the interval of marketing authorisation date and December 2013 or OMP designation withdrawal date or date of withdrawal of use. Reports on consumption and real expenditures of these OMPs came from all health insurance companies in the CZ. Exchange rate of 25.4 CZK = 1 EUR was used. Results: Overall, 86 OMPs were authorised within the European Union (EU) between 2008 and 2013. Of these, 50 OMPs (58.0%) were covered from the Czech public health insurance at some point within this period. The number of registered OMPs increased from 54 in 2008 to 84 in 2013, while the number of covered OMPs doubled from 24 to 41, respectively. The annual public expenditure on OMPs rose simultaneously from 43 to 72 million EUR, while the total expenditure on drugs increased from 1.7 to 2.0 billion EUR. The OMP share of total pharmaceutical sales grew steadily from 2.5% in 2008, reaching 3.4% in 2011 and plateaued at 3.6% in 2013. Twenty-one oncological OMPs (51.2%) generated up to 72.6% of the overall orphan costs in 2013. Conclusions: Compared to other EU countries, the Czech public expenditure on OMPs of 3.6% of all drugs seems to be relatively high. The major part of OMP costs is in oncological diagnoses. One of the limitations is the exclusion of designated OMPs not authorised by EMA since these OMPs still might be present on the market and reimbursed. This aspect needs further investigation. PSY28 Qutenza® Estimated Costs Per Patient In Primary Versus Secondary Care. A Comparison Between Qutenza®, Pregabalin and Lidocaine for the Treatment of Peripheral Neuropathic Pain Darba J 1, Kaskens L 2, Villa G 2 1Universitat de Barcelona, Barcelona, Spain, 2BCN Health Economics & Outcomes Research S.L., Barcelona, Spain . . . Objectives: The objective of this analysis was to estimate and compare the annual cost per patient of the administration of capsaicin 8% patch, Qutenza®, in Primary and Secondary Care for the treatment of adult patients with peripheral neuropathic pain (PNP). The annual cost per patient treated with Qutenza® was also compared with Pregabalin and Lidocaine in Primary Care. Methods: The costs per patient for each treatment were estimated by analysing the health care resources associated with the use of pregabalin, lidocaine and Qutenza®for the treatment of PNP from the perspective of the Spanish National Health System. Healthcare resources associated with the three pharmacological treatments considered were administration time and health care personnel, complementary non-pharmacological treatment, hospitalisation, adverse effects and concomitant medication, EUR 2013. Total costs per patient were estimated and compared both including and excluding concomitant medication to estimate the effect on the treatment costs. Results: The annual treatment cost per patient when Qutenza® is administrated in the hospital was estimated at € 5,387.32 and € 5,259.19 when administered in Primary Care. Qutenza® applied in Primary Care for the treatment of PNP showed annual savings of € 128.14 per patient compared with the administration of Qutenza® in Secondary Care. Overall, the annual treatment costs per patient with pregabalin or lidocaine were estimated at € 14,074.14 and € 5,998.09, respectively, resulting in savings of € 8,814.95 and € 738.9 respectively when compared with the annual cost per patient when treated with Qutenza® in Primary Care. Conclusions: The introduction of Qutenza® into Primary Care for the treatment of PNP is expected to result in annual savings due to lower administration costs associated with the use of Qutenza® and savings in health care personnel in Primary Care. Qutenza® also showed to be the less expensive option in comparison with pregabalin or lidocaine, independent of the area of administration. PSY29 Economic Impact Linked To the Reduction of Exacerbations When A Treatment Regime With Inhaled Antibiotics Is Switched To Aztreonam Lysine In Patients With Cystic Fibrosis and Chronic Pulmonary Infection Caused By Pseudomonas Aeruginosa Sole A 1, Poveda J L 1, Giron R M 2, Prados M C 3, De Gracia J 4, Casado M Á 5 Universitario La Fe, Valencia, Spain, 2Hospital La Princesa, Madrid, Spain, 3Hospital Universitario La Paz, Madrid, Spain, 4Hospital Universitari Vall d’Hebron, Barcelona, Spain, 5Pharmacoeconomics & Outcomes Research Iberia, Madrid, Spain . . . . . . . . . . 1Hospital Objectives: Estimate, from the Spanish National Healthcare System (SNS) perspective, the required reduction of exacerbation episodes in different patient profiles to maintain constant the overall treatment cost if a chronic treatment with inhaled antibiotics is switched to aztreonam lysine in patients with cystic fibrosis (CF) and chronic pulmonary infection by Pseudomonas aeruginosa (PA). Methods: A cost comparison model was developed considering two criteria to define patient profiles: current drug treatment (12 options linked to tobramycin and/or colistimethate sodium) and current exacerbation episodes (1-6/year). For each patient profile the overall treatment cost was calculated. Considering a switch to aztreonam lysine, the reduction of exacerbation was estimated in order to maintain the overall treatment cost. For the base case, ex-factory prices were considered; for sensitivity analysis, discounts ranged from 15 to 50%. Results: Switching to aztreonam lysine would not increase overall treatment costs in 8/12 options; even maintaining current exacerbation episodes could lead to SNS savings. In the remaining options, the required reduction of exacerbations varies depending on the patient’s current situation, ranging from 110% (one episode/ year) to 18% (six episodes/year). In the sensitivity analysis, there are 3 options in which switching to aztreonam lysine would not imply higher treatment costs. For the remaining alternatives, the required reduction of exacerbations does not change significantly from the base case: from 116% (one episode/year) to 19% (six episodes/ year). Conclusions: In CF patients with chronic pulmonary infection by PA, hospitalizations costs due to exacerbations have a greater economic impact than drug treatment costs. Using aztreonam lysine, instead of other treatments, would not imply a budget impact in a wide range of patient profiles if exacerbation episodes can be reduced. As the number of current exacerbations increases, a lesser reduction of these events is required by a switch to aztreonam lysine to avoid incremental treatment costs. PSY30 Impact of Generic Substitution On the Prescribing of MeprobamateContaining Combination Analgesics In South Africa Truter I Nelson Mandela Metropolitan University, Port Elizabeth, South Africa . Objectives: Meprobamate is a constituent of various combination analgesics in South Africa. Due to its abuse potential, the scheduling status of meprobamate is currently under review with the possibility of introducing stricter controls on its availability. The primary aim was to determine the impact of generic substitution on meprobamate-containing combination analgesic prescribing. Methods: A retrospective, cross-sectional drug utilisation study was conducted on prescription data of a medical insurance scheme administrator in South Africa for 2011. Results: A total of 97 491 analgesics were dispensed to 31 854 patients during the year. Within ATC category N02B, 62.10% of prescriptions were for analgesic combinations, of which 20 326 prescriptions were for meprobamate-containing analgesics at a cost of R282 930. A total of 10 404 patients received meprobamate-containing analgesics. Overall, 20.85% of all analgesics contained meprobamate. Patients who received meprobamate-containing analgesics were slightly older (39.52 years) compared to patients who received analgesics in general (33.61 years). Mandatory generic substitution exists in South Africa, unless otherwise indicated by the prescriber or if the patient refuses. Twenty-two trade names of meprobamate-containing analgesics were dispensed, of which 17 (the originator plus 16 branded generics) contained exactly the same dosages of active ingredients (320 mg paracetamol, 8 mg codeine phosphate, 32 mg caffeine and 150 mg meprobamate). In previous studies, the originator product was the most often prescribed, yet in this study the originator product only constituted 3.72% of products prescribed (average cost: R30.42) compared to 70.63% for the most popular branded generic (average cost: R11.65). Conclusions: The price difference between generic equivalents and the originator product, together with mandatory generic substitution, were responsible for the change in prescribing patterns. PSY31 The Economic Burden of Treating Thalassemia In Greece Geitona M 1, Karagianni V 2, Kattamis A 3, Voskaridou E 4, Drosou M 5, Vini D 5, Kalogeropoulou M 6 1University of Peloponnese, Corinth, Greece, 2Technological educational institute of Athens, Egaleo, Greece, 3Ag. Sofia Children Hospital, Goudi, Greece, 4Laiko General Hospital of Athens, Ambelokipi, Greece, 5Agios Panteleimon General Hospital, Nikaia, Greece, 6Center for Health Services Management and Evaluation, National and Kapodistrian University of Athens, Athens, Greece . . . . . . . Objectives: The objective was to estimate the economic burden of B-thalassemia and to compare the cost of alternative iron chelation strategies in patients with transfusion (TDT) and non-transfusion dependent thalassemias (NTDT) in Greece. Methods: A multicenter, non interventional, retrospective, observational, phase IV study was conducted in the 3 out of 8 thalassemia units in Attica. Patients’ data and resource use, including medication and related consumables, blood transfusions, lab tests, diagnostic examinations and hospital visits were recorded from January 2009 to December 2011, and classified per diagnosis and iron chelation treatment strategies. Greek NHS perspective was considered. Results: 331 thalassemic patients were recorded whereas the majority (87.6%) was diagnosed with Betathalassemia major, 31% were treated with deferasirox (DFX), 23% with deferoxamine and deferiprone, (DFO+DFP), 9.7% with deferiprone (DFP) and 5.1% with DFO. Patient mean annual cost per patient was € 32,064, including all treatment strategies, for the period 2009-2011, ranging from € 30,997 in 2009 to € 32,564 in 2011. Medication (45.9%), blood transfusions (38.1%) and consumables (14.9%) were the most important cost drivers. Mean cost per patient treated with DFX was estimated at € 35,928, with DFO+DFP at € 34,035, with DFO at € 31,637 and with DFP at € 17,208 for the same period. DFX significantly reduced (p< 0.001) the number of blood transfusions in relation to DFO+DFP with 19 and 25 transfusions per year, respectively. Similarly, DFX significantly reduced the volume of blood transfusions (p< 0.001) in relation to DFO with 9,693ml and 10.312ml per year, respectively. Conclusions: Rational health care decision making process should be based on the holistic perspective of analytical resource use and disease management data rather than focusing on the price per product per se. Especially in Greece, a country where Beta-thalassemia incidence in general population is ~8 %, the consideration of all cost components should be taken into account in health resource allocation. PSY32 Economic Considerations On the Use of Mifamurtide In the Treatment of Osteosarcoma In Spain Brosa M 1, García del Muro X 2, Mora J 3, Villacampa A 1, Pozo T 4, Adán C 4, Grande M 4, García E 4, Cubells L 4 . . . . . . . . . A527 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1Oblikue Consulting, Barcelona, Spain, 2Institut Català d’Oncologia, L’Hosp. de Llobregat, Spain, Sant Joan de Déu, Esplugues de Llobregat, Spain, 4Medical Department. Takeda Farmacéutica España, Madrid, Spain 3Hospital Mepact® (mifamurtide) is the first drug approved for treatment of high-grade resectable non-metastatic osteosarcoma in patients aged 2-30 in the last 20 years. It follows a randomised clinical trial showing a statistically-significant and clinicallyrelevant decrease in the risk of death without compromising safety. Objectives: This study assessed the cost-effectiveness and budget impact of mifamurtide added to the standard chemotherapy of cisplatin, doxorubicin and methotrexate with this same standard therapy (ST) without mifamurtide. Methods: A Markov model was built to combine trial-based outcomes with Spanish resource use and unit cost of compared options. The analysis has been carried out from the perspective of the Spanish National Health Service, with a time horizon of up to 60 years in the base analysis. A probabilistic sensitivity analysis was carried out to assess the influence of the uncertainty of the variables introduced into the model. All the costs are expressed in euros for the year 2011 and, beyond the first year, both costs and effects (quality-adjusted survival) have been discounted with an annual rate of 3%, following local recommendations. Results: The observed greater efficacy of mifamurtide in the trial translates into a gain of 3.03 (undiscounted) and 1.33 (discounted) QALYs and an additional cost of € 102,000. The estimated budgetary impact of using mifamurtide in 10% to 100% of the potential population would cost € 671,000 and € 6.7 million respectively. Conclusions: The iCER of mifamurtide in Spain is in the low band (< € 100,000) of the iCERs obtained by other orphan drugs and would have a limited, predictable and affordable cost in Spain. PSY33 Cost Analysis In the Treatment of Patients With Systemic Lupus Erythematosus In Russian Federation Kulikov A , Komarov I , Pochuprina A I.M. Sechenov First Moscow State Medical University, Moscow, Russia . . . Objectives: To evaluate and compare costs of belimumab treatment plus standard of care (SoC) vs SoC alone in patients with systemic lupus erythematosus (SLE). Methods: The cost analysis was conducted of the belimumab treatment plus SoC vs SoC alone for one year. Based on the results of clinical trials to assess safety and efficacy of belimumab (BLISS-52 and BLISS-76) the cost structure of treatment was determined. Results: The costs were evaluated in the following groups: drug costs, costs of inpatient and outpatient visits, costs for correction of adverse events and costs for treating of SLE complications. The costs of inpatient visits comprised 91,830 RUB/ 1,976 EUR for SoC alone and 52,991 RUB/ 1,140 EUR for patients treated with belimumab plus SoC. The correction of adverse events came to 7,763 RUB/ 167 EUR for SoC alone, while for belimumab plus SoC it was 7,846 RUB/ 169 EUR. The costs of treating SLE complications totaled 384,976 RUB/ 8,283 EUR and 370,229 RUB/ 7,966 EUR for SoC alone and belimumab plus SoC groups of patients, respectively. Conclusions: The costs of inpatient visits in belimumab treatment group were lower than those of the SoC group resulting from a reduction in the frequency of SLE flare. The costs for correction of adverse events were similar in both patients groups due to a good safety profile of belimumab. The level of costs for treating SLE complications was lower in belimumab as compared to that in SoC group as a consequence of lower frequency of cardiovascular, pulmonary, renal, and skin complications in the belimumab treatment group. A good safety profile and efficacy of belimumab, which had been demonstrated in clinical trials, led to a decrease in costs of inpatient visits and treatment of SLE complication in patients treated with belimumab. PSY34 Opioid Prescribing and the Impact of Branded Generics Truter I Nelson Mandela Metropolitan University, Port Elizabeth, South Africa . Objectives: Opioids are the mainstay of therapy for patients with moderate to severe pain. Mandatory generic substitution exists in South Africa, unless otherwise indicated by the prescriber or if the patient refuses. Few studies have been conducted in South Africa analysing the prescribing patterns and cost of opioid analgesics. Methods: A retrospective, cross-sectional drug utilisation study was conducted on prescription data of a medical insurance scheme administrator in South Africa for 2011. The database contained 2 298 312 records for medicine, medical devices and procedures. Results: A total of 97 491 analgesics were dispensed to 31 854 patients during the year. Within ATC category N02, opioids (NO2A) accounted for 26.55% of analgesic prescriptions at a cost of R1 071 230.14. A total of 9 793 patients were prescribed 25 888 opioid analgesics. The average age of patients was 41.50 (SD= 16.61) years. Female patients were slightly younger (average age: 40.69 (SD= 17.05) years) than male patients (average age: 42.32 (SD= 16.09) years). Nine different active ingredients and two combination products were prescribed. Tramadol, an atypical opioid, was the most often prescribed (68.11%), followed by pethidine (14.39%) and morphine (8.38%). The average cost per prescription was R41.38 (SD=R69.29). Fentanyl had the highest average cost per prescription (R454.43), and pethidine the lowest (R8.53). Overall, the average cost for an originator product was R56.25 and for a generic product R28.95. Tramadol accounted for 68.11% of prescriptions and 80.40% of cost. Tramadol had the most most branded generic equivalents on the market (7), yet only had 42.89% generic prescribing. Most prescriptions were issued by private hospitals (62.89%), followed by pharmacies (24.41%) and general medical practices (12.16%). Only 35.48% of the products were prescribed on a chronic basis. Conclusions: Tramadol dominated opioid prescribing. The study confirmed price differences for opioid analgesics between branded generics and originator products. PSY35 Pill Burden, Health Care Resource Utilization and Costs Among Subpopulations of Immediate Release Hydrocodone Users BenJoseph R 1, Yang S 2, Yang E 2, Holly P 1, Boulanger L 2 Pharma L.P., Stamford, CT, USA, 2Evidera, Lexington, MA, USA . 1Purdue . . . . Objectives: To assess pill burden, health care resource utilization (HRU), and costs among patients with long-term immediate release (IR) hydrocodone use. Methods: We performed a retrospective analysis of health care claims from 2011-2012 Truven MarketScan® Commercial, Medicare supplemental, and Medicaid Multistate databases. Patients with IR hydrocodone prescription for ≥ 90 days during 6 month baseline period (July 2011- December 2011) with continuous enrollment during baseline and 12 month follow-up periods were selected. The final population was sub-categorized by prescribed coverage days (PCD) of IR hydrocodone during baseline into 90-119, 120-179, and ≥180 days. Chi-square or ANOVA analyses were used to test pill burden, HRU and costs (standardized to 2013 US dollars) during baseline and follow-up periods across subpopulations. Results: A total of 36,174 commercial, 32,699 Medicaid, and 8,873 Medicare IR hydrocodone users were selected. In the baseline period, subgroups with longer PCD had significantly more average hydrocodone pills per month yet fewer HRU and medical costs (all p< 0.05). However, during the follow-up period, groups with longer PCD had greater increase in number of inpatient hospitalizations and other types of HRU (length of stay, outpatient hospital visits, office visits, and emergency room visits). The subgroup of patients with PCD < 120 days had lower annual all-cause medical costs during follow-up compared with baseline (decreasing $2,624, $2,955, $4,209 per patient per year in Medicaid, Medicare and commercial patients, respectively), while patients with longer PCD during baseline had increased costs (p< 0.05). For example, Medicaid patients with 120-179 PCD had an increase of $1,874 and those with ≥ 180 PCD had an increase of $4,348. These trends were similar for all insurance types. Conclusions: Extended length of PCD, particularly after 120 days, corresponds with higher patient burden including elevated pill burden and rising HRU and costs in both commercial and public insurance patients with long-term IR hydrocodone use. PSY36 Prevalence-Based Measurement of the Economic Burden of Rare Diseases: Case Review To Determine the Annual Cost of Acromegaly In France Ponet O 1, Tasdemir E 2, Magestro M 3, Griner B P 4, Cummins G 5, Van Engen A 6, Kreeftmeijer J 6, Niemira J 4, Tao C 7 1Novartis France, Rueil Malmaison, France, 2Novartis Oncology, Origgio / VA, Italy, 3Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA, 4Quintiles, Cambridge, MA, USA, 5Quintiles, Durham, NC, USA, 6Quintiles Consulting, Hoofddorp, The Netherlands, 7Quintiles Consulting, Cambridge, MA, USA . . . . . . . . . . Objectives: Although acromegaly is acknowledged as requiring resource-intensive treatment, its ultimate economic burden is unclear. As an extension of work presented at ISPOR 2013 International Conference (New Orleans, US), the objective of this research is to measure the annual economic burden of acromegaly in France using a case-review methodology with a prevalence-based sample of patients diagnosed with acromegaly. Methods: A case-review method was used with a sample of 22 endocrinologists reviewing 88 patient cases (4 cases per physician) diagnosed with acromegaly. The patient case histories included: resource utilization including office visits and hospitalization, diagnostic procedures and labs, medications prescribed, medical procedures preformed, and an estimate of lost productivity. A micro-costing analysis was conducted to obtain costs in the prior 12 months for each patient case reviewed using published literature, medical fee schedules, and pharmaceutical cost databases to assign costs to treatments and medical procedures identified in the survey data. Annual costs were examined across a broad range of patients of different ages, gender and time from diagnosis. Two biomarkers were used to categorize acromegaly patients as Controlled vs. Uncontrolled: Insulin Growth Factor-1 (IGF-1) and Growth Hormone (GH). Several patient characteristics were used as control factors when comparing annual economic costs: age, sex, and time from diagnosis. Statistical tests and confidence intervals were calculated to determine the significance of patient characteristic effects on economic burden. Results: Three patient subgroups were used to classify uncontrolled acromegaly patients: IGF-1, GH and both IGF-1 and GH. The annual per-patient economic burden of disease costs ranges from € 29,000 to € 79,400 across these groups. These cost ranges are benchmarked to other studies to provide context and validity. Conclusions: The total economic burden of acromegaly in France is significant. Understanding the factors impacting burden of illness will inform future improvements in treatment practice. PSY37 Rates of Diagnosed Opioid Abuse Or Dependence and Incremental Direct Health Care Costs Among Patients With Long-Term Use of Immediate Release Hydrocodone BenJoseph R 1, Yang E 2, Huse S 2, Bhagnani T D 2, Holly P 1, Kansal A 3 Pharma L.P., Stamford, CT, USA, 2Evidera, Lexington, MA, USA, 3Evidera, Bethesda, MD, USA . . . . . . . 1Purdue Objectives: To estimate rates of diagnosed abuse and incremental health care costs among long-term immediate release (IR) hydrocodone users. Methods: We performed a retrospective analysis of health care claims from 2011-2012 Truven MarketScan® Commercial and Medicaid Multistate databases. Patients with IR hydrocodone prescription ≥ 90 days during 6 month baseline period (July 2011December 2011) with continuous enrollment during baseline and 12 month follow up periods were selected. Opioid abuse was defined as a patient having at least one medical claim indicating opioid abuse/dependency (ICD-9-CM diagnosis codes: 304.0x, 304.7x, 305.5x, 965.00,965.02 or 965.09) during the study period. Rates of opioid abuse during baseline and follow up periods were estimated by plan type, along with total health care costs, standardized to 2013 US dollars. Generalized linear model regressions were used to estimate incremental costs in the post-index period. Results: A total of 32,699 Medicaid and 36,174 commercial IR hydrocodone users were selected in the study. Rates of abuse were 96 and 36 per 10,000 patients, respectively. Abusers had higher unadjusted annual total health care costs than non-abusers during both baseline and follow up period (p< 0.05). After controlling for baseline characteristics of age, gender, Charlson Comorbidity Index, pill burden A528 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 and costs (all p< 0.05), the adjusted annual incremental costs in abusers versus nonabusers were $28,882 (95% Confidence Interval [CI]: $28,455-$29,311) and $15,523 (95% CI: $15,389-$15,657) per patient among Medicaid and commercially insured patients, respectively, during the post-index period. The main cost driver was inpatient hospitalization which comprised 88% of unadjusted incremental costs during follow up in Medicaid insured population and 53% in commercial insured population. Conclusions: Diagnosed opioid abusers among long-term IR hydrocodone users impose significantly higher financial burden in both Medicaid and commercially insured populations in terms of all-cause direct health care expenditures, ranging from $28,882 to $15,523 per abuser per year. PSY38 Prevalence-Based Measurement of the Economic Burden of Rare Diseases: Case Review To Determine the Annual Cost of Acromegaly In Italy Premoli E 1, Tasdemir E 2, Magestro M 3, Griner B P 4, Cummins G 5, Van Engen A 6, Kreeftmeijer J 6, Niemira J 4, Tao C 7 1Novartis Pharma - BU Oncology, Origgio, Italy, 2Novartis Oncology, Origgio / VA, Italy, 3Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA, 4Quintiles, Cambridge, MA, USA, 5Quintiles, Durham, NC, USA, 6Quintiles Consulting, Hoofddorp, The Netherlands, 7Quintiles Consulting, Cambridge, MA, USA . . . . . . . . . . Direct medical costs were estimated using tariffs for laboratory tests, diagnostic exams, visits, and prices for drugs. Procedure and inpatient cost data were collected at Center level. Non medical costs included costs for travel and accommodation, domestic help and informal care. The loss of productivity of patients have been estimated using the human capital approach. The incremental effects of having comorbidities on social costs were estimated by multivariate Generalized Linear Models (log link, Gamma family) adjusting for gender, age, BMI, type of intervention and complications. Costs are expressed in Euro 2013. Results: Among 301 patients enrolled, 108 (36%) had hypertension, 53 (18%) T2DM and 47 (16%) ADD. The raw social cost of intervention were € 8,749 (± € 2,359), € 9,511 (± € 2,292) and € 8,999 (± € 2,275) for patients with hypertension, T2DM and ADD. A significant incremental effect of having T2DM was found on social cost of intervention (€ 751, 95%CI: 242-1,259, p= 0.004). 1 year after intervention reductions of 48%, 81% and 15% were observed for hypertension, T2DM and ADD. The raw social annual costs estimated were € 2,461 (± € 1,490) for hypertension, € 2,424 (± € 951) for T2DM and € 3,582 (± € 2,017) for ADD. Direct non medical costs and indirect costs represent the main component of social cost in patients with hypertension and ADD. Conclusions: Bariatric surgery led to reductions of obesity-related comorbidities. One year after, the economic burden is mainly sustained by patients, their families and the productivity system. Objectives: Although acromegaly is acknowledged as requiring resource-intensive treatment, its ultimate economic burden is unclear. As an extension of work presented at ISPOR 2013 International Conference (New Orleans, US), the objective of this research is to measure the annual economic burden of acromegaly in Italy using a case-review methodology with a prevalence-based sample of patients diagnosed with acromegaly. Methods: A case-review method was used with a sample of 22 endocrinologists reviewing 86 patient cases (4 cases per physician) diagnosed with acromegaly. The patient case histories included: resource utilization including office visits and hospitalization, diagnostic procedures and labs, medications prescribed, medical procedures preformed, and an estimate of lost productivity. A micro-costing analysis was conducted to obtain costs in the prior 12 months for each patient case reviewed, using published literature, medical fee schedules, and pharmaceutical cost databases to assign costs to treatments and medical procedures identified in the survey data. Annual costs were examined across a broad range of patients of different ages, gender and time from diagnosis. Two biomarkers were used to categorize acromegaly patients as Controlled vs. Uncontrolled: Insulin Growth Factor-1 (IGF-1) and Growth Hormone (GH). Several patient characteristics were used as control factors when comparing annual economic impact: age, sex, and time from diagnosis. Statistical tests and confidence intervals were calculated to determine the significance of patient characteristic effects on economic burden. Results: Three patient subgroups were used to classify uncontrolled acromegaly patients: IGF-1, GH and both IGF-1 and GH. The annual per-patient economic burden of disease costs ranges from € 18,600 to € 38,000 across these groups. These cost ranges are benchmarked to other studies to provide context and validity. Conclusions: The total economic burden acromegaly in Italy is significant. Understanding the factors impacting burden of illness will inform future improvements in treatment practice. PSY41 Cost of Illness Analysis of Duchenne Muscular Dystrophy In Italy PSY39 Costs of Absenteeism In Psoriatic and Enteropathic Arthropathies Based On Real-Life Data From Poland’s Social Insurance Institution Database In 2012 PSY42 The Burden of Myelofibrosis In Greece Kawalec P , Malinowski K Jagiellonian University Medical College, Krakow, Poland . . Objectives: The aim of this study was to assess the indirect costs caused by absenteeism associated with psoriatic and enteropathic arthropathies from the perspective of the Social Insurance Institution (ZUS) in Poland. Methods: The estimates were based on data published by ZUS referring to year 2012 and concerning absence from work due to the illness (sick leave), the amount of short term disability, the sufferers of which claim rehabilitation benefit, and the amount of permanent (or long term) disability, the sufferers of which claim disability pension. Costs were calculated with Human Capital Approach methodology taking into account Gross Domestic Product (GDP) per capita equaled 41 398 PLN; results were presented in Polish zloty (PLN). Results: Total indirect costs of psoriatic and enteropathic arthropathies in the year 2012 calculated using GDP per capita in Poland were 9 312 773 PLN. The highest component of indirect costs of psoriatic and enteropathic arthropathies was sick leave (82%). Long and short term disability costs constitute 9% and 8% of total indirect costs of psoriatic and enteropathic arthropathies, respectively. In 2012 Poland’s Social Insurance Institution database reported 1 900 patients on sick leave, 54 patients with short term disability and 3 patients with long term disability. Indirect costs per patient associated with sick leave were 4 037 PLN calculated using GDP per capita. Indirect costs per patient associated with short term disability were as high as 16 227 PLN and associated with long term disability were as high as 255 288 PLN. Conclusions: Psoriatic and enteropathic arthropathies in Poland generated high indirect costs. The main component was sick leave; rehabilitation benefit and disability pension generated much lower costs of lost productivity. PSY40 Social Costs of Different Procedures In Bariatric Surgery In Patients With Obesity-Related Comorbidities Bellelli S 1, Armeni P 2, Tarricone R 3, Turchetti G 1 1Scuola Superiore Sant’Anna, Pisa, Italy, 2Bocconi University, Milano, Italy, 3Bocconi University, Milan, Italy . . . . Objectives: To estimate the social cost of bariatric surgery techniques in obese patients with hypertension, diabetes mellitus (T2DM) and anxiety-depression disorders (ADD). Methods: A longitudinal multicenter study was conducted by enrolling obese adult patients in charge to 6 Hospital in Italy at time of intervention of gastric banding, gastric by-pass, and sleeve gastrectomy and following up to 1 year. Fabriani V , Marcellusi A , Mennini F S , Giannantoni P University of Rome “Tor Vergata”, Rome, Italy . . . . . Objectives: The objective of this study is to estimate the average annual direct and indirect costs associated with Duchenne muscular dystrophy (DMD) in Italy considering both National Health System (NHS) societal perspective. Methods: A probabilistic prevalence-based cost of illness model was used to estimate the economic impact of a rare disease as DMD. All the costs were determined through a survey that families registered with the Muscular Dystrophy Association “Parent Project onlus” completed on-line. NHS and family prospective has been analyzed dividing the patients into three age groups (< 8, 8 – 16 and > 16). Human capital approach was used to determine loss of productivity due to absenteeism, while the bottom up approach was used to calculate direct costs. Furthermore, a probabilistic sensitivity analysis with 5,000 Monte Carlo simulations was performed, in order to test the robustness of results and define the 95%CI. Results: Indirect costs are those that weigh more on the total expenditure of the NHS with € 474.634.836 (95%CI: € 300.028.168 - € 698.965.090) per year, while the direct health care costs are € 7.475.596 (95%CI: € 5.124.369,29 - € 10.263.785) and nonmedical costs are € 12.944.879 (95% CI: € 7.925.699 - € 19.175.331). Patients with more than 16 years spend more than those between 0 and 7 years old, and even more than those between 8 and 15. For what concern the private expenditure, the model estimated € 2.910.506 (95%CI: € 345.231,83 - € 718.786) for the direct costs, and € 185.333.744 (95%CI: € 114.177.282 - € 273.446.219) for the nonmedical costs. Conclusions: Although DMD is a rare disease, its economic impact on NHS is quite remarkable. Furthermore, the most of the impact relies on families and society. Kousoulakou H 1, Symeonidis A 2, Kyriakou D 3, Sotiropoulos D 4, Gigantes S 5, Delimbasi S 5, Tsirigotis P 6, Hatzikou M 7, Geitona M 1 1University of Peloponnese, Corinth, Greece, 2University of Patras Medical School, Patras, Greece, 3University Hospital of Larissa, Larissa, Greece, 4G. Papanicolaou Hospital, Thessaloniki, Greece, 5Evangelismos Hospital, Athens, Greece, 6Attikon Hospital, Chaidari, Greece, 7Novartis Hellas, Metamorfosis, Greece . . . . . . . . . Objectives: To estimate the burden of myelofibrosis (MF) in Greece, focusing on epidemiological data, quality of life (QoL), direct and indirect costs. Methods: A 17-page questionnaire was developed, validated, and completed with the Delphi technique. It included questions on epidemiological, resource use, QoL and socioeconomic data. An expert panel with 9 KOL haematologists was convened consisting of experts from the largest Haematology Units of Greece, covering geographically six out of seven Regional Health Authorities. Unit costs in 2014 prices were taken from officially published sources. The societal perspective was adopted. Results: Prevalence and incidence rates of MF in Greece are approx. 2.5: 100,000 and 0.7: 100,000 people respectively, corresponding to approx. 270 patients (71.7% with primary and 28.3% with secondary) and 76 new cases every year; 92% of the patients present with splenomegaly at diagnosis, 1/3 of which reduce their daily activities. Current treatment options in Greece are ruxolitinib and best supportive care (BSC). 72.6% of the primary and 65% of the secondary MF patients treated with ruxolitinib show improvement of splenomegaly vs. 23% and 7%, respectively for patients treated with BSC. Ruxolitinib patients show QoL improvement and less splenectomies (< 1%) compared with BSC patients (~3%). Work loss days associated with ruxolitinib are estimated at 23 days per year (51 days for BSC), and 20% of them return to work after treatment (5% for BSC). The annual direct cost of managing all MF patients in Greece is estimated at € 1.65 million, including pharmaceutical, hospital, follow-up costs, blood transfusions, and management of infections. Productivity losses are estimated at € 217,975 per year, resulting in a total annual burden of approx. € 1.87 million. Conclusions: MF is associated with significant burden to patients, their families, and to the society. Treatment with ruxolitinib appears to improve patients’ QoL and reduce indirect costs, mainly through reduction of splenomegaly and splenectomies. PSY43 The Indirect Costs of Multiple Sclerosis Associated With Absenteeism In Poland Mocko P 1, Kawalec P 2 University Medical College, Krakow, Poland, 2Jagiellonian University Medical College, Krakow, Poland . 1Jagiellonaian . A529 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: The aim of this study was to assess the indirect costs associated with multiple sclerosis (MS) from the perspective of Social Insurance Institution (ZUS) in Poland. Methods: The estimates were based on data from the Social Insurance Institution (ZUS) referring to year 2012 and focused on absenteeism due to the illness: costs of sick leave as well as the short-term disability due to rehabilitation benefit and the burden of permanent (or long term) disability due to disability pension in Poland. Cost analysis was performed based on the Human Capital Approach taking into account Gross Domestic Product (GDP) per capita equaled 41 398 PLN and Gross Value Added (GVA) per worker equaled 99 697 PLN. Costs were presented in polish zloty (PLN). Results: Total indirect costs of multiple sclerosis associated with absenteeism in the year 2012 in Poland were 39 870 385 PLN calculated using GDP per capita and 96 018 323 PLN as a GVA per worker. The predominant component of absenteeism of MS was sick leave, which accounted for 69%. Long and short term disability costs constituted 21% and 9% of total indirect costs of multiple sclerosis associated with absenteeism, respectively. One sick leave of person with multiple sclerosis generated the cost of lost productivity equal 1 866 PLN and 4 490 PLN calculated using GDP per capita and GVA per worker, respectively. Indirect cost of short-term disability for one entitlement to the benefit of rehabilitation were 17 077 PLN and 41 125 PLN, respectively. Cost of one long term benefit were much higher than short term benefit and equaled 41 398 PLN and 99 697 PLN, respectively. Conclusions: Multiple sclerosis in Poland generated very high indirect costs. The main component was sick leave; disability pension and rehabilitation benefit generated lower costs of lost productivity. PSY44 Costs of Absenteeism In Ankylosing Spondylitis Based On Real-Life Data From Poland’s Social Insurance Institution Database In 2012 Malinowski K 1, Kawalec P 2 1Jagiellonian University Medical Collage, Krakow, Poland, 2Jagiellonian University Medical College, Krakow, Poland . . Objectives: The aim of this study was to assess the indirect costs caused by absenteeism associated with ankylosing spondylitis (AS) from the perspective of the Social Insurance Institution (ZUS) in Poland. Methods: The estimates were based on data provided by ZUS referring to year 2012 and concerning absence from work due to the illness (sick leave), the amount of short term disability, the sufferers of which claim rehabilitation benefit, and the amount of permanent (or long term) disability, the sufferers of which claim disability pension. Costs were calculated with Human Capital Approach methodology taking into account Gross Domestic Product (GDP) per capita equaled 41 398PLN and Gross Value Added (GVA) per worker equaled 99 697PLN and were presented in 2012 prices in Polish zloty (PLN). Results: Total indirect costs of AS in the year 2012 calculated using GDP per capita and GVA per worker in Poland were 21 299 578PLN and 51 294 959PLN, respectively. The highest component of indirect costs of AS was sick leave (56%). Long and short term disability costs constitute 6% and 39% of total indirect costs of AS, respectively. In 2012 Poland’s Social Insurance Institution database reported 3 500 patients on sick leave, 72 patients with short term disability and 7 patients with long term disability. Indirect costs per patient associated with sick leave were 3 400PLN and 8 188PLN calculated using GDP per capita and GVA per worker, respectively. Indirect costs per patient associated with short term disability were 16 602PLN and 39 983PLN respectively and associated with long term disability were as high as 1 171 958PLN and 2 822 381PLN, respectively. Conclusions: AS in Poland generated high indirect costs. The main component was sick leave; rehabilitation benefit and disability pension generated lower costs of lost productivity. PSY45 The Economic Burden of Systemic Lupus Erythematosus: A Structured Literature Review Jugrin A V 1, Sun Y 2, Cox F 3 1IMS Health RWE Solutions and HEOR, Vilvoorde, Belgium, 2UCB Pharma, Brussels, Belgium, 3UCB Pharma, Smyrna, GA, USA . . . . Objectives: To assess the direct and indirect costs of systemic lupus erythematosus (SLE), analysis of which is complex due to the heterogeneity of the disease, accumulative organ damage and associated comorbidities. Methods: Direct and indirect costs attributable to SLE were extracted from studies published January 2004–March 2013, identified using Medline, EMBASE, EconLit and NHS EED. Eligible studies involved adults with active SLE. Results: The direct cost associated with SLE management was € 3,691/yr in Europe.1 In the US, annual health care costs were $10,984 higher for SLE patients vs non-SLE controls.2 Organ damage and disease activity were key drivers of annual cost variations;1–4 in Europe, cardiovascular/respiratory (€ 596/yr) and renal involvement (€ 511/yr) appeared the most costly organs.1 Estimated direct annual cost of treating any flare in the EU was ~€ 399, whilst for severe flares was ~€ 1,002.1 In the US, renal involvement was the most costly organ: cost ratios were 3.1 (95% CI: 2.3–4.2) in newly diagnosed and 2.7 (95% CI: 2.4–3.0) in existing SLE patients.4 Annual costs for mild and severe flares in the US were $129/flare and $11,716/flare, respectively.2 The indirect economic burden is also substantial, as SLE typically occurs during the peak of working years and in young women with child-bearing potential (around the age of 20 in Black females and 30 in White females).5,6 Predictors of future work loss in SLE included older age, shorter job tenure, thrombotic and musculoskeletal manifestations, greater number of manifestations and high disease.7,8 Conclusions: The direct and indirect economic burden is prominent within SLE patients. Organ damage is a substantial “hidden cost” and should be considered when assessing the economic impact References: 1. Doria A. Ann Rheum Dis 2014; 73 (1): 154-160, 2. Kan H. Biomed Res Int 2013; 808391, 3. hamashta M. Lupus 2014; 23 (3): 273-283, 4. Furst D. Lupus 2013; 22 (1): 99-105, 5. Baker K. Lupus 2009; 18 (14) 1281-1288, 6. Somers E. Arthritis Rheum 2007; 57 (4): 612-618, 7. Yelin E. Arthritis Care Res 2013; 64 (2): 169-175, 8. Oglesby A. Arthritis Rheum 2012; 64 (10): S934 PSY46 The Cost of Active Systemic Lupus Erythematosus in Greece Results From the Lycos Study Athanasakis K 1, Karampli E 1, Psomali D 2, Perna A 3, Kyriopoulos J 1 School of Public Health, Athens, Greece, 2GlaxoSmithKline, Halandri, Greece, 3GlaxoSmithKline, Brentford, UK . . . . . 1National Objectives: Systemic Lupus Erythematosus (SLE) is an autoimmune disease, characterized by periods of remissions and flares, with significant clinical and economic burden. The primary study objective was to estimate the 1-year direct medical cost for adult patients with active, autoantibody-positive SLE in Greece. Methods: This is a national, multi-centre, retrospective study. Data were abstracted from patient records in 6 hospital centers specialized in SLE management. Starting with the patient with the most recent visit, patients with consecutive visits (backwards in time) were considered for inclusion, provided they met specific criteria. In order to estimate costs per disease severity, a stratification criterion was applied. Patient data were collected for a 1-year period starting from the inclusion date (January -September 2011) and moving forward. Data included patient characteristics and health care resource utilization. In addition, all SLE patients fulfilling the inclusion criteria and followed-up in the participating centers during a 3-month retrospective period were recorded. For cost calculation, official 2013 list prices were used. Results: 215 patients (30% severe according to the stratification criterion) were included in the study. Mean direct medical costs were estimated at € 1,225 for patients with non-severe and at € 3,741 for patients with severe active SLE. Laboratory and imaging tests, medicines, physicians’ visits, and hospitalization costs represented 10.5%, 51.7%, 1.2%, 36.5% of mean cost respectively. Costs were statistically significantly higher for severe SLE patients. The total number of patients visiting the participating clinical sites during a 3-month period was 318 (19% with severe SLE). The weighted mean annual direct medical cost of SLE in Greece was estimated at € 1,703. Conclusions: Direct medical cost of SLE in Greece is significant, especially for patients with severe disease. An estimation of indirect costs would provide a comprehensive picture of the societal burden of the disease. PSY47 Romiplostim Cost Per Response In Itp Treatment In the Brazilian Health Care System Pepe C 1, Teich V 1, Coutinho M B 2, Almeida S 3 - Grupo Resulta, São Paulo, Brazil, 2Amgen Brazil, Sao Paulo, Brazil, 3Amgen, São Paulo, Brazil . . . . . 1NewBD/Medinsight Objectives: Immune thrombocytopenia (ITP) is characterized by isolated thrombocytopenia with no underlying cause. It manifests clinically as mucocutaneous bleeding caused by decreased platelets. Recently, two thrombopoietin receptor-agonists have emerged as an important therapeutic options: romiplostim and eltrombopag. Since these medications have different mode of administration, safety and efficacy profiles, the present study was carried out in an attempt to investigate which drug would be more cost-effective in the Brazilian setting. The objective was to perform an economic analysis evaluating the cost per response of romiplostim versus eltrombopag in adult patients with chronic ITP and refractory to other treatments as corticosteroids and immunoglobulins in Brazil health care private system. Methods: Two economic analyses were performed in order to study the use of romiplostim and eltrombopag for the treatment of adults with chronic refractory ITP, based on annual treatment costs and cost per response. The average body weight adopted was 74,6 kg. Efficacy data were obtained from the product inserts, as well as from scientific publications. In cost per response, it was considered a 6-month analysis, which corresponds to the overall platelet response in clinical trials. The outcome of these trials was the global response to treatment. The cost of each drug presentation were based on ex-factory price (VAT 18%) and obtained from the official price list (CMED; April, 2014). Results: The cost of treatment with romiplostim showed an annual saving of R$7,724 over eltrombopag within the payer perspective, which may be further improved if we assume that drug-food interactions of eltrombopag in some patients lead to reduced medication adherence and loss of response. The cost per response with romiplostim was 26% lower than with eltrombopag. Conclusions: Romiplostim was more cost-effective than eltrombopag for the treatment of chronic refractory ITP in adult patients and may represent savings to the Brazilian health system. PSY48 Direct and Indirect Costs Associated With Increasing Body Mass Index (Bmi) In the Eu5 Richard L 1, Gupta S 2, Pomerantz D 2, Forsythe A 2 1Eisai Europe Ltd, Hatfield, UK, 2Kantar Health, Princeton, NJ, USA . . . . Objectives: This study evaluated the impact of BMI category on health utilities, health care resource-utilisation, productivity, activity impairment, and associated costs. Methods: Results were from the 2013 EU5 National Health and Wellness Survey (N= 62,000), a nationally representative, online survey of respondents aged ≥ 18 years. This analysis focused on normal weight (BMI≥ 18.5 & < 25 kg/m2), overweight (BMI≥ 25 & < 30 kg/m2), obese class (OC) I (BMI≥ 30 & < 35 kg/m2), OC II (BMI≥ 35 & < 40 kg/m2), and OC III (BMI≥ 40 kg/m2) respondents. Patients provided information on SF-6D (health utility) scores, productivity loss (Work Productivity and Activity Impairment questionnaire) and resource-utilisation (type/number of visits) in the past six months. Direct and indirect costs were estimated from the literature. Generalised linear models predicted resource use and productivity as a function of BMI category, adjusting for covariates (e. g., age, gender, comorbidities). Results: Among 58,364 respondents, 46.9% were normal weight, 34.5% were overweight, 12.5% were OC I, 4.0% were OC II, and 2.1% were OC III. Metabolic comorbidities increased as BMI increased. Adjusting for covariates, health utility (normal weight: 0.720; overweight: 0.718; OC I: 0.703; OC II: 0.683; OC III: 0.662) scores declined with an increase in BMI (all p<0.05 vs. normal). Among employed patients (57.7%), overall work impairment increased as BMI increased (normal weight: 17.9%; overweight: 18.4%; OC I: 19.0%; OC II: 21.4%; OC III: 26.7%, p< 0.05 all OCs vs. normal). Normal A530 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 (vs. all OCs) had lower activity impairment and fewer provider visits, lower indirect costs (normal weight: € 7,974; overweight: € 7,825; OC I: € 8,465; OC II: € 9,394; OC III: € 10,437), and lower total direct costs (normal weight: € 516; overweight: € 553; OC I: € 583; OC II: € 605; OC III: € 717), all p< 0.05. Conclusions: Increased BMI is associated with higher direct and indirect costs and worse health utilities. The findings highlight the impact of obesity on health outcomes. PSY49 Cross-Country Comparison of Medical Resource Utilisation In Patients With Autosomal Dominant Polycystic Kidney Disease In Europe Blais J 1, Krasa H B 2, Szende A 3, Colman S 4, Schaefer C 5, Dale P 6, Robinson P 6, O’Reilly K 6 Pharmaceutical Development & Commercialization, Inc., Princeton, NJ, USA, 2Otsuka Pharmaceutical Development & Commercialization, Inc., Rockville, MD, USA, 3Covance, Leeds, UK, 4Covance Pty Ltd, North Ryde, Australia, 5Covance, Gaithersburg, MD, USA, 6Otsuka Pharmaceutical Europe Ltd, Wexham, UK . . . . . . . . . 1Otsuka Objectives: Autosomal dominant polycystic kidney disease (ADPKD) is the most common genetic kidney disease. Currently there is little published information on medical resource utilisation (MRU) in European ADPKD patients. This study aimed to estimate the health care resource use of ADPKD patients across six European countries. Methods: A retrospective review of medical charts was conducted via an online physician survey in France, Germany, Italy, Spain, Sweden and the United Kingdom (UK). A total of 353 physicians were recruited to review MRU over the previous 24 month period from the records of 1,055 patients, with each clinician selecting the last three patients representing three different chronic kidney disease stages. Data collected included patient characteristics and MRU such as tests, visits, hospitalisations and medications. Results: The proportion of patients who had a kidney ultrasound during the 24 months study period varied from 20.0% in Sweden to 68.0% in Germany. In all countries, except Germany (66.0%), at least 90% of patients had one or more contact with a specialist. Annualised specialist visits varied from 2.7 (SD 7.9) in Germany to 6.0 (SD 10.7) in Sweden. Germany had the largest proportion of patients with hospitalisations related to ADPKD (25.7%) whereas Sweden had the lowest (13.3%). Hospital length of stay was longest in Spain (mean, 8.5 days (SD 6.2)) and shortest in the UK (mean, 4.7 days (SD 3.8)). In terms of medication use, 32.7% of German patients were prescribed loop diuretics whereas this proportion was 9.0% in the UK. The UK had the highest rate of opioid use for renal pain overall (8.7%) and among patients on dialysis (27.3%). Conclusions: This is the first study to provide comparative evidence on MRU in European patients with ADPKD. Results highlight some country-specific differences in treatment patterns, including rates of hospitalisation, frequency of specialist visits and medication use. PSY50 An Evaluation of Medical Resource Utilisation In Patients With Autosomal Dominant Polycystic Kidney Disease In Europe Blais J 1, Krasa H B 2, Szende A 3, Colman S 4, Schaefer C 5, Dale P 6, Robinson P 6, O’Reilly K 6 Pharmaceutical Development & Commercialization, Inc., Princeton, NJ, USA, 2Otsuka Pharmaceutical Development & Commercialization, Inc., Rockville, MD, USA, 3Covance, Leeds, UK, 4Covance Pty Ltd, North Ryde, Australia, 5Covance, Gaithersburg, MD, USA, 6Otsuka Pharmaceutical Europe Ltd, Wexham, UK . . . . . . . . . 1Otsuka Objectives: Autosomal dominant polycystic kidney disease (ADPKD) is the most common type of polycystic kidney disease (PKD) and the fourth leading cause of endstage renal disease. Currently, there is no published information on medical resource utilisation (MRU) in patients with ADPKD in Europe. This study aimed to better understand MRU associated with ADPKD by disease stage in six European countries (Germany, France, the United Kingdom, Italy, Spain and Sweden). Methods: This study was a retrospective review of medical charts collected via an online physician survey. Participating physicians abstracted data for the last three eligible ADPKD patients treated in their practice, each within a different Chronic Kidney Disease (CKD) stage. Data collected over the past 2 years included socio-demographics, clinical characteristics and MRU, such as diagnostic tests, specialist visits, dialysis, emergency department visits, inpatient admissions and medications. Results: A total of 1,055 ADPKD patients were enrolled. The mean (SD) age of the sample was 50.4 (14.4) years, 53.6% were male, with 39.4% employed full time. Only 40% of patients had a known PKD genotype, of which 83% had PKD-1 and 17% had PKD2. Almost all patients (99.1%) experienced at least 1 ADPKD-related complication, most commonly hypertension (75.4%). On average (SD), patients had 4.4 (9.3) annual specialist visits. Approximately 19.7% of patients experienced at least 1 hospitalisation over the 2 year study period with a mean (SD) length of stay of 7.0 (8.5) days. Patients in more advanced disease stages reported higher mean (SD) annual specialist visits and more disease-related hospitalisations, 1.5 (1.1) and 9.2% in CKD Stage 1 compared to 10.0 (17.2) and 32.4% in Dialysis, respectively. Conclusions: This is the first study to provide evidence on MRU among patients with ADPKD in Europe. Results demonstrate that ADPKD patients require substantial MRU, which accumulate with disease progression. PSY51 Cost-Consequences Analysis of the Long-Term Prophylaxis In A Type 1 Von Willebrand Disease Patient With Recurrent Bleedings In Italy Cipolla A 1, Cultrera D 1, Teruzzi C 2, Mantuano M 3 1Dept. of Hematology, Haemophilia Regional Reference Centre, University of Catania, Catania, Italy, 2Temas - A Quintiles Company, Cassina de’ Pecchi, Italy, 3Temas. A Quintiles Company, Milan, Italy . . . . Objectives: Von Willebrand Disease (VWD) is the most common inherited bleeding disorder, caused by a deficiency or abnormality of the Von Willebrand factor (VWF). Long-term prophylaxis (LTP) with Factor VIII-containing VWF product (FVIII/VWF) is used for patients with severe VWD. However, FVIII/VWF is not enough effective in some cases. Furthermore repeated infusions of FVIII/VWF may increase the risk of thromboembolic events. The objective of the analysis was to assess the economic impact of VWF concentrate almost devoid of FVIII as an alternative to FVIII/VWF for a patient on LTP with recurrent bleedings. Methods: A cost-consequences analysis was adopted to assess the economic impact from the NHS and Society perspectives. The cost analysis was based on one patient case (type 1 VWD, recurrent gum bleedings), treated with FVIII/VWF and then with VWF concentrate almost devoid of FVIII. The costs included direct costs (drug acquisition, hospital admissions, outpatient visits, red blood cells units) and indirect costs (working days lost). Data were gathered from a questionnaire (Hemophilia Reference Center of Catania). The health care costs were calculated by DRG analysis that assessed DRG refund value for day hospital and standard hospitalization. The indirect costs were calculated on the basis of the annual income average. Results: The analysis showed a reduction of the number of bleedings when treating with VWF concentrate almost devoid of FVIII (30 vs 0), minimizing the cost per bleeding episodes (€87,957 vs €0), hospitalizations (€3,252 vs 0), monitoring visits (€331 vs €165) and the number of working days lost (26 vs 2). The annual health care costs and indirect costs avoided were € 49,684 and €3,734 respectively. Conclusions: The replacement therapy with VWF concentrate almost devoid of FVIII decreased the consumption of hospital resources and reduced the number of working days lost per bleedings together with the discomfort related to bleedings. PSY52 Cost Per Response Analysis for Thrombopoietin Receptor Agonists (Tpo-Ras), In the Treatment of Adult Chronic Immune Thrombocytopenia (Itp) In Mexico Alva M E 1, Rivera R 1, Arocho R 2, Campos I 3, Campioni M 4 City, Mexico, 2Amgen, Inc., Barcelona, Spain, 3Amgen, Zug, Switzerland, 4Amgen (Europe) GmbH, Zug, Switzerland . . . . . . 1Amgen, Mexico Objectives: TPO-RAs, represent an evidence-based treatment option in the public setting for adult patients with chronic ITP in Mexico. TPO-RAs have not been directly compared in head-to-head randomized controlled trials (RCTs); however an indirect comparison was undertaken using Bayesian metaregression: the overall platelet response was significantly higher in patients receiving Romiplostim than in those receiving Eltrombopag, estimating an Odds ratio (OR) of eltrombopag vs romiplostim as 0.15 (95% CI: 0.02, 0.84). The objective of this study is to compare the cost per response of TPO-RAs in chronic adult ITP in Mexico. Methods: A cost per response analysis was developed. Dose was derived from RCTs and Summary of Product Characteristics. Median dose for romiplostim was 2.5 mcg/kg/week and average dose for eltrombopag was 55 mg/day (21.5% of patients received 25mg; 41% 75mg; and remaining 37.5% 50mg). Costs for romiplostim were based on vials used, considering a representative Mexican patient (65kg); for eltrombopag, it was based on milligrams needed. Cost assessment included cost of medication, cost of administration and in case of eltrombopag, cost of liver monitoring, expressed in Mexican pesos. Crude Overall Response Rate (ORR) for romiplostim was 83%. Placebo adjusted ORR for eltrombopag was 42% calculated by applying the OR estimated from the Bayesian indirect comparison performed by the NICE Evidence Review Group. Results: Romiplostim generates a cost per overall platelet response of $219,690.80, while eltrombopag yields $374,137.72. Conclusions: Within the TPORAs, romiplostim generates a lower cost per response than eltrombopag, in adult patients with chronic ITP in Mexico. PSY53 Cost-Effectiveness Analysis of Belimumab In the Treatment of Adult Systemic Lupus Erythematosus (Sle) Patients With Positive Biomarkers In Spain Vallejo-Aparicio L A 1, Díaz-Cerezo S 1, Parrondo J 1, García-Aparicio A M 2 España, Tres Cantos, Spain, 2Hospital VIrgen de la Salud, Toledo, Spain . . . . . . 1GSK Objectives: Belimumab is a novel biological treatment specifically developed for the treatment of active, autoantibody positive SLE patients. The purpose of this study is to estimate the cost-effectiveness of belimumab for SLE patients from the Spanish societal perspective. Methods: A UK cost-effectiveness microsimulation model was adapted to the Spanish setting. The analysis compared standard of care (SoC) vs. belimumab plus SoC with 2 years maximum treatment duration and a life-time horizon. Disease activity reduction seen with belimumab in the BLISS-52 and BLISS76 trials, was extrapolated using data from the US Johns Hopkins Lupus Cohort to predict impact on long-term organ damage and mortality.. Utility values and unit direct costs (treatment, administration, patient´s follow-up and organ damage costs) were obtained from UK and Spain published data, respectively. Indirect costs were calculated using the Human Capital Approach method. A discount rate of 3% was applied to costs and outcomes. Results were expressed as € 2014. Deterministic and probabilistic sensitivity analyses (PSA) were conducted to determine the robustness of the results. An additional analysis was conducted to compare societal and Spanish National Health System (NHS) perspectives. Results: The cost per life year gained (ICER) and cost per QALY (ICUR) for Belimumab were € 16,647 and € 23,158 respectively. In 68% of scenarios plotted in the PSA, belimumab was a cost-effective alternative considering a € 30,000/QALY threshold. From NHS perspective (indirect costs excluded), ICER and ICUR were € 25.619 and €35.640. Conclusions: Base-case results show that belimumab is cost-effective from the Spanish societal perspective. From the NHS perspective, the model provides results that fall within an acceptable threshold considering the prevalence and the severity of the disease. These results highlight the importance of adopting a societal perspective, especially in pathologies such as SLE which affect young people of working age. PSY54 Cost-Effectiveness Analysis of Maintenance Treatment With Rituximab In Patients With Follicular Lymphoma Responding To First Line Induction Therapy In Portugal Pereira C 1, Rubio Terres C 2, Rubio Rodríguez D 2 Farmacêutica Química, Lda., Amadora, Portugal, 2Health Value, Madrid, Spain . . . 1Roche Objectives: Evaluate the efficiency of rituximab in maintenance treatment of patients with follicular lymphoma (FL) who respond to first line induction with rituximab versus observation in Portugal. Methods: Cost-effectiveness (Life A531 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Years Gained - LYG) and cost-utility analysis (Quality- Adjusted Life years – QALYs) were performed for a time horizon of 10 years according to a Markov economic model with four health states - “progression free survival (PFS) in first and second lines”, “progression” and “death” - and monthly cycles. Health states transition probabilities were obtained from two randomized controlled clinical trials: PRIMA (Salles G. et al 2010) and EORTC 20981 (van Oers M. et al 2010). Health state utilities were obtained from literature (Pettengell R. et al 2008). Resource consumption was estimated by a Portuguese expert’s panel. Costs were calculated considering the Portuguese Health System perspective through official data (unit costs: € in 2014). Costs and consequences were discounted at 5% per annum. Deterministic and probabilistic (Monte Carlo simulation) sensitivity analyses were performed for several assumptions namely time horizon, PFS supportive care and progression costs; adverse events costs; health states utilities values and costs and benefits annual discount. Results: For a 10 years’ time horizon, the cost per LYG and QALYs gained was € 10,630 and € 10,674 respectively. Sensitivity analyses confirmed the base case results for time horizons of 20 and 30 years, ranging between € 7,430 and € 7,155 per QALY gained, respectively. Probabilistic sensitivity analysis confirmed the robustness of the model with a cost per QALY gained of € 10,657. The incremental cost-effectiveness acceptability curve shows that rituximab maintenance therapy would be cost effective from a willingness to pay of € 12,000 per QALY gained. Conclusions: According to the present model rituximab maintenance treatment of FL patients who respond to first line induction therapy compared with observation is a cost-effective strategy in Portugal. PSY55 The Cost-Effectiveness of Expanding the Nhs Newborn Bloodspot Screening Programme To Include Homocystinuria (Hcu), Maple Syrup Urine Disease (Msud), Glutaric Aciduria Type 1 (Ga1), Isovaleric Acidaemia (Iva), and Long-Chain Hydroxyacyl-Coa Dehydrogenase Deficiency (Lchadd) Bessey A 1, Chilcott J 2, Pandor A 1, Paisley S 1 University of Sheffield, Sheffield, UK, 2University of Sheffield, Sheffield, UK . . . . 1The Objectives: The NHS newborn bloodspot screening programme currently screens all babies in England for five rare conditions. The objective of this study was to assess the cost-effectiveness of expanding the screening programme to include five new rare conditions all inborn errors of the metabolism; HCU, MSUD, GA1, IVA, and LCHADD. Methods: A decision tree model was built to estimate the cost-effectiveness of the expanded newborn screening programme. Estimates of the prevalence of the five conditions and the test characteristics of screening were taken from the literature. Survival and morbidity estimates for the screened and unscreened populations were estimated from published case series. Quality adjusted life years (QALYS) were estimated from the extended EQ-5D+ (C) which includes a cognitive dimension in order to capture the impact of neurological impairment and developmental delay which are known sequelae of the five conditions. Costs related to the marginal cost of the expanded screening programme, management costs of the conditions, and costs associated with the sequelae of the conditions were estimated from the pilot study of the expanded screening, case reports from the pilot, expert elicitation, published guidelines and estimates from the literature. Costs and QALYs were multiplied by survival and morbidity estimates to give lifetime estimates for the screened and unscreened populations. A probabilistic sensitivity analysis (PSA) was conducted. Results: The results from the deterministic analysis and PSA suggests that screening for all five conditions is cost-saving with screening associated with lower total costs and higher total QALYs compared to no screening. The incremental net benefit for all five conditions, at a threshold of £25,000 per QALY, was between £0.46 for IVA and £5.94 for GA1. Conclusions: Screening for MSUD, HCU, IVA, GA1 and LCHADD are each estimated to be potentially cost saving and result in increased quality of life compared to no screening. PSY56 Cost-Effectiveness of Capsaicin 8% Patch (Qutenzatm) Compared With Pregabalin for the Treatment of Patients With Peripheral Neuropathic Pain (Pnp) In Scotland Patel S 1, Trueman D 2, Bentley A 2, Poole C 1, Chambers C 1 Pharma Europe Ltd, Chertsey, UK, 2Abacus International, Bicester, UK . . . . . 1Astellas Objectives: PNP is a high-burden disease exacerbated by poor tolerability of conventional oral therapies. Capsaicin 8% patch is a well-tolerated cutaneous treatment for PNP in non-diabetic adults either alone or in combination with other therapies. We evaluated the cost-effectiveness of capsaicin 8% patch versus pregabalin in patients with PNP from the perspective of NHS Scotland. Methods: A decision tree model was developed considering patients with PNP who had neither achieved pain relief nor tolerated conventional first-/second-line treatments. After 8 weeks’ treatment with capsaicin 8% patch or pregabalin, patients remained on therapy or discontinued due to intolerable adverse events. Patients continuing on therapy were classified as either responders (≥ 30% decrease in pain from baseline) or non-responders. Last-line therapy was given to non-responders and those who discontinued treatment. The base-case time horizon was 2 years. Effectiveness, discontinuations and quality of life utilities were estimated from a recent head-tohead study (ELEVATE; NCT01713426). Other inputs were obtained from published sources or clinical expert opinion. All costs were based on GBP 2013/14. The results were presented as incremental cost-effectiveness ratios (ICERs), i. e. cost per qualityadjusted life-year (QALY) gained. Model assumptions were tested with scenario analyses. Parameter uncertainty was tested using one-way and probabilistic sensitivity analyses. Results: Compared with pregabalin, capsaicin 8% patch was dominant versus pregabalin (total cost difference, –£11 and total QALY gains, +0.049). Using a 1-year time horizon, the ICER increased to £1,242/QALY. The model was most sensitive to variations in time to capsaicin 8% patch retreatment (worse case ICER, £7,951/QALY). Capsaicin 8% patch was dominant in six/seven scenario analyses. At a willingness-to-pay threshold of £20,000/QALY gained, the probability of costeffectiveness for capsaicin 8% patch versus pregabalin was 97%. Conclusions: Capsaicin 8% patch is cost-effective compared to pregabalin for patients who have failed one or more previous systemic treatments for PNP. PSY57 Cost Effectiveness Analysis Evaluating Factor Viii As Primary Prophylaxis Treatment for Patients With Severe Haemophilia A In the Netherlands Kip M 1, van den Bosch M 2, Fischer K 3, Tamminga R 4, Lepage-Nefkens I 1 Netherlands, 2Bayer, Mijdrecht, The Netherlands, 3Van Creveldkliniek & Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, The Netherlands, 4Beatrix Childrens Hospital, University Medical Center Groningen, Groningen, The Netherlands . . . . . 1PANAXEA, Enschede, The Objectives: Multiple regimens are used in the treatment of severe haemophilia A in the Netherlands. Most patients receive clotting factors intravenously 2-3 times weekly to prevent bleedings: intermediate dose prophylaxis. Given the high utilization of prophylaxis treatment, budget restraints might hinder the availability of prophylaxis for patients in the nearby future. Other treatment regimens are ondemand (OD) treatment, administering clotting factors in case of bleedings, and prophylaxis treatment with a switch to OD at 18 years. This analysis estimates the cost-effectiveness of Dutch prophylaxis treatment for severe haemophilia A patients compared to other treatment regimens. Methods: A Markov model is developed with the health stages ‘Alive’, ‘Severe joint damage’ and ‘Death’. Bleeding rates of individual patients are simulated over lifetime, including a probability of inhibitor development. A higher joint bleed rate is accompanied by increased joint damage, increasing the chance of joint surgery. Disease progression, within the Alive health state, is modeled with the Pettersson Score (PS). The PS indicates the radiographic arthropathy. Increased joint damage is associated with physical limitations and decreased QoL. Because the chosen treatment regimen affects both the joint bleed rate and inhibitor development, it also affects the HRQoL. The analysis was performed from a societal perspective. Results: Prophylaxis treatment was associated with the greatest QoL. The cost-effectiveness acceptability curve shows a probability of 90% for prophylaxis treatment to be cost-effective at a threshold of € 0, - compared to OD treatment. Compared to prophylaxis with a switch to OD at 18 years, prophylaxis treatment has a 50% probability of being cost-effective at a € 80.000, - threshold. The model outcome is sensitive for variations in bleeding rate, prophylaxis dosage, inhibitor development and utilities. Conclusions: Based on our model, treatment of severe haemophilia A patients with lifetime prophylaxis is cost-effective compared to OD treatment. PSY58 Cost-Effectiveness of the Lidocaine 5% Medicated Plaster VERSUS Pregabalin and Amitriptyline for the Treatment of Post-Herpetic Neuralgia In the Netherlands Obradovic M 1, Vanden Baviere H 2, Liedgens H 3 1GfK Market Access, Nuremberg, Germany, 2Grünenthal, St Stevens Woluwe, Belgium, 3Grünenthal GmbH, Aachen, Germany . . . Objectives: The objective of the analysis was to evaluate costs and outcomes of treating post-herpetic neuralgia (PHN), a chronic disease with severe burden for patients, in the Netherlands with lidocaine 5% medicated plaster compared to pregabalin and amitriptyline. Methods: A Markov model was used to extrapolate outcomes beyond the time horizon of the available trial data and to allow for the fact that patients may discontinue treatment at any point during treatment. Costs and effects, expressed in terms of the quality-adjusted life-year (QALY) gained, were calculated for each treatment strategy over a period of 6 months. The study included direct costs related to PHN. Indirect costs were not included as most patients with PHN are older and retired. Transition probabilities were based on the comparative and long-term clinical trials. Utilites were identified through a literature review. Resource utilization was obtained from a two-step Delphi study with pain specialists, cost data were obtained from the official price tariffs/lists. Extensive sensitivity and scenario analyses were performed to explore robustness of the results. Results: In 6-month time horizon, treatment with the lidocaine plaster yielded 0.4283 QALYs. For pregabalin and amitriptyline the total effect was 0.3390 QALYs. The mean costs per patient treated with lidocaine plaster (1.71 plasters/day) were 1,082 € . For pregabalin (488 mg/day) and amitriptyline (25 mg/day) the mean costs were 912 € and 346 € , respectively. Therefore, the lidocaine plaster compared to pregabalin and amitriptyline had an incremental cost-effectiveness ratio of 1,907 € /QALY and 8,246 € /QALY, respectively. Probability of the lidocaine plaster being cost-effective versus pregabalin and amitriptyline exceeded 90% when considering a threshold of 30,000 € per QALY gained. Extensive scenario and one-way sensitivity analyses confirmed robustness of the results. Conclusions: The lidocaine 5% plaster is a highly cost-effective treatment for PHN in the Netherlands. PSY59 Cost-Effectiveness Analysis of Amfepramone (Diethylpropion) for the Obesity Treatment In Mexico Soto Molina H 1, Díaz Martínez J P 1, Escobar Juárez Y 1, Fernández del Valle C 2 Estudios Farmacoeconómicos, Mexico City, Mexico, 2Instituto Politécnico Nacional, Mexico city, Mexico . . . . . 1HS Objectives: The main objective was to perform a pharmacoeconomic analysis to find out the cost effectiveness of diethylpropion (DEP) with diet and exercise. (DEP+DaE), compared against Diet and Exercise (DaE) in the treatment for obesity from the institutional point of view in Mexico. Methods: Effectiveness data from a mexican clinical trial (Morin, 2007) was used to populate a decision tree model to estimate the cost-effectiveness of DEP+DaE and its comparator DaE. The target population were men and women over 18 years with BMI > 30 kg/m2. Principal outcome was the reduction of the Body Mass Index (BMI); benefit was expressed as the percentage of patients who reduced more than 10% of their initial weight. Only direct medical costs were used, such as medications and adverse events; these were obtained from the portal shop by IMSS and also from their unitary costs. To prove the A532 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 strength of the analysis, deterministic and probabilistic sensitivity tests were performed; all the quantities are expresed in Mexican pesos (MXP) at 2013. Results: Both groups lose weight and reduce their BMI. However, these changes were earlier and more pronounced in the (DEP+DaE) group. DeP+DaE presented a significant higher percentage of patients reducing 10% of their intial weight (37% vs 17%, p-value< .05). Incremental cost was $1,455.09 MXP and incremental effectiveness was 20% in reducing the initial weight. The incremental cost-effectiveness ratio of (DEP+DaE) relative to DaE was calculated to be $7,374.07 MXP per additional percentage of reduction suggesting that DEP+DaE is cost-effective compared with DaE as a treatment for obesity in Mexico. Conclusions: The combination of DEP+DaE provides a cost effective improvement to the treatment of patients with a risk profile for obesity in Mexico. PSY60 Cost-Effectiveness Analysis of Oxycodone Lp An Opioid Analgesic for Patients With Moderate To Severe Pain Secondary To Cancer In Mexico Soto Molina H 1, Sanchez K 1, Escobar Juárez Y 1, Constanzo A 1, Fernandez Z 2, Melendez C 2 Estudios Farmacoeconómicos, Mexico City, Mexico, 2Psicofarma S.A. de C.V., Mexico City, Mexico . . . . . . 1HS Objectives: To perform a complete economic evaluation type using oxycodone (Endocodil XR ®) versus morphine, buprenorphine (transdermal patches) and fentanyl (transdermal patches) on moderate to severe pain secondary to cancer type in Mexico, from the point of view of public health. Methods: A cost-effectiveness analysis was made, a decision tree model was developed to simulate the costs and benefits of health outcomes of each opioid analgesic, the probabilities were obtained through a systematic review of the literature. The model has 13 cycles of 28 days under a time horizon of 364 days, the main measure of effectiveness was determined as days without severe adverse events and pain controlled, only measured direct medical costs and the ratio was obtained incremental cost-effectiveness, further a deterministic sensitivity analysis, probabilistic sensitivity analysis and budget impact was performed. Results: The results demonstrate that oxycodone (Endocodil XR ®) is a dominant option respect to Buprenorphine (PT), fentanyl (PT), the results showed 309 days without severe adverse events and controlled pain compared to 219 and 255 respectively Oxycodone had a cost of $ 57,702.34 under $ 58,254.43 and 129,906.49 costing of their comparators respectively, compared to morphine oxycodone is more effective but more expensive, it had an ICER of $ 202.05, sensitivity analysis and budget impact showed that oxycodone remains an option cost-saving. Conclusions: The results demonstrate that oxycodone (Endocodil XR ®) is a cost-effective option for the treatment of patients with pain of moderate to severe secondary to cancer in Mexico, it is an option that effectively combines price and fails to meet the necessary standards for palliative care and pain management. PSY61 Comparative Pharmacoeconomic Analysis of the Application of Posaconazole, Fluconazole and Itraconazole With the Purpose of Primary Prevention of Invasive Fungal Infection In Patients With Neutropenia During Chemotherapy for Acute Myelogenous Leukemia Or Myelodysplastic Syndrome Krysanov I 1, Krysanova V 2 1Postgraduate Medical Institute, Moscow National University of Food Production, Moscow, Russia, 2I.M. Sechenov First Moscow State Medical University, Moscow, Russia . . Objectives: to perform ñomparative pharmacoeconomic analysis of alternative schemes for prevention of invasive mycosis with fluconazole, itraconazole and posaconazole in patients with neutropenia, after chemotherapy for acute myelogenous leukemia or myelodysplastic syndrome. Methods: were reviewed research on the clinical effectiveness and safety of use of Posaconazole. Assess of the quality of research and level of evidence obtained in these results was performed. The model is constructed on the basis of the results of a multicenter randomized trial Cornely O. A. et al., 2007. The model calculated the differences in direct medical costs for the use of drugs, as well as the cost of medical treatment cases invasive mycosis the ineffectiveness of primary prevention. The duration of preventive treatment, the probability of various outcomes correspond to the data of specified clinical study. Results: analysis of the evidence has shown that on patients with neutropenia antifungal prophylaxis with posaconazole is more effective than fluconazole/itraconazole, and significantly reduces the risk of developing invasive mycosis, and associated mortality. Total costs for the use of posaconazole was 13169 USD, that by 24.6 % higher than the use of fluconazole (total costs 9932 USD) and 24.3 % higher than itraconazole (total costs 9965 USD). Also use of posaconazole has increased the total number of LYG by 6%. The ICER was $ 21339 USD and 21117 USD compared to fluconazole and itraconazole, and lower than the estimated threshold of willingness to pay in the Russian Federation equal 38390 USD. One-way sensitivity analysis showed that in case of changes in the cost of posaconazole from 75% to 125 %, the ICER for a one LYG do not exceed the threshold of willingness to pay. Conclusions: use of Posaconazole for prevention of invasive mycosis in patients with neutropenia is economically justified. PSY62 Cost-Effectiveness Analysis of Eltrombopag As Support Treatment In Chronic Hcv Infected Patients With Thrombocytopenia To Enable Interferon-Based Regimens Parrondo J 1, Rincón Rodríguez D 2, Romero Gómez M 3, Sola Lamoglia R 4, Roldán C 1 1GSK Spain, Tres Cantos, Spain, 2Hospital General Universitario Gregorio Marañón, Madrid, Spain, 3Hospital Universitario de Valme. Universidad de Sevilla, Sevilla, Spain, 4Hospital del Mar IMIM - Universitat Autonoma de Barcelona, Barcelona, Spain . . . . . Objectives: Thrombocytopenia limits the use of interferon (INF) based regimens in chronic hepatitis C virus (HCV) patients. Eltrombopag, a thrombopoietin-receptor agonist, effectively elevates platelet count allowing optimal INF-treatment. The objective of the present study was to assess the cost-effectiveness as support treat- ment in chronic HCV infected patients with thrombocytopenia to enable INF-based treatment regimens from the Spanish Health System perspective. Methods: A two-phase individual-level model was developed to evaluate the cost-effectiveness of eltrombopag treatment in thrombocytopenic HCV-patients over a lifetime horizon. Individual-level models are more flexible and provide more accurate estimations than Markov approaches. One million patients were simulated using data from trials ENABLE 1 and 2 and local studies. In the first phase, a discrete event simulation was used to recreate patient events during INF-treatment. When eltrombopag was considered, patients underwent an initial pre-INF treatment with eltrombopag. Those that failed to reach INF-label platelet count did not receive INF-based regimens. In the second phase, a microsimulation was used to emulate each patient from treatment discontinuation to death. Health states included fibrosis (F0, F1/2, F3, F4), liver decompensation, hepatocelullar carcinoma, liver transplant and death. Transition probabilities for each 1-month cycle, utilities and direct health care costs (€ 2014) were obtained from literature and national databases. A 3% annual discount was applied to costs and health outcomes. Sensitivity analysis with 0% and 5% discount rates were performed. Results: Eltrombopag was associated with an average increment of 0.58 quality-adjusted life years (QALY) and an additional cost of € 17,084.47/patient. The average incremental cost effectiveness ratio (ICER) was 29,808.26 € /QALY. Considering a € 30,000 threshold, eltrombopag was cost-effective in 59.12% of cases. This proportion remained similar with 5% (56.7%) and 0% (63.25%) discount rates. Conclusions: With the premises considered in this study, eltrombopag in HCV patients could be considered cost-effective from the Spanish Health System perspective. PSY63 Cost-Effectiveness of Romiplostim for the Treatment of Chronic Immune Thrombocytopenia In Portugal Augusto M 1, Gouveia M 2, Borges M 3, Campioni M 4 Care Research and Consulting, Lda, Lisboa, Portugal, 2Católica Lisbon School of Business and Economics, Lisbon, Portugal, 3University of Lisbon, Lisbon, Portugal, 4Amgen (Europe) GmbH, Zug, Switzerland . . . . 1Evigrade, Health Objectives: To assess the cost-effectiveness of romiplostim in the treatment of Adult Immune Thrombocytopenia (ITP) in Portugal, versus eltrombopag and current medical standard of care (SoC). Methods: A lifetime treatment-sequence cost-utility Markov model with an embedded decision tree was developed from the Portuguese Healthcare Payer perspective to compare romiplostim with eltrombopag and SoC in splenectomized patients, non-splenectomized patients and a combined population. Treatment sequences and health care utilization were validated by a Portuguese expert in ITP. Outcomes included incremental cost-utility ratio (ICUR). Costs (€, 2014) included drug acquisition costs and costs associated with monitoring patients and managing complications. Results: In the combined population, romiplostim versus eltrombopag had incremental costs (IC) of 13.848€ and a quality-adjusted life-years (QALY) gain of 0,566, yielding an ICUR of 24.451€ . Compared with SoC, romiplostim had IC of 18.622€ and a QALY gain of 0,938, yielding an ICUR of 19.848€ . In splenectomized patients, the most prevalent sub-population in Portugal (65% of adult chronic ITP patients), romiplostim had an ICUR of 6.304€ versus eltrombopag and an ICUR of 3.179€ versus SoC. One-way sensitivity analysis showed that the model was most sensitive to variations in the drug doses/ percentage of utilization and costs (romiplostim, eltrombopag and intravenous immunoglobulin) and to the utility of patients responding to ITP treatments. In the combined population the probabilistic sensitivity analysis showed that romiplostim is likely to be cost-effective in 66% and 84% of samples versus eltrombopag and versus SoC at a willingness-to-pay threshold of 30.000€/QALY, respectively. Conclusions: Use of romiplostim in the ITP treatment pathway, compared with eltrombopag or SoC, is likely to be cost-effective in Portugal. PSY64 Belimumab for the Treatment of Systemic Lupus Erythematosus (Sle) In Greece: A Cost-Effectiveness and Cost-Utility Analysis Athanasakis K 1, Karampli E 1, Ollandezos M 1, Igoumenidis M 1, Karabela P 2, Psomali D 2, Kyriopoulos J 1 1National School of Public Health, Athens, Greece, 2GlaxoSmithKline, Halandri, Greece . . . . . . . Objectives: Systemic Lupus Erythematosus (SLE) is a chronic autoimmune inflammatory disease, associated with significant health and socioeconomic burden. Current treatment of SLE involves glucocorticoids, antimalarials, non-steroid antiinflammatory drugs, and immunosuppressive agents. Belimumab, a human IgG monoclonal antibody specific for soluble human B lymphocyte stimulator protein, is a novel pharmaceutical treatment approved as an add-on therapy in adult SLE patients with highly active disease (autoantibody-positive and low complement levels) despite treatment. The study objective was to estimate the incremental cost-effectiveness ratio (ICER) and incremental cost-utility ratio (ICUR) of adding belimumab to the Standard-of-Care (SoC) treatment of SLE patients with high disease activity in the Greek health care setting. Methods: The analysis is based on the local adaptation of a micro-simulation model. The model follows individual patients over a lifelong period. Data on short-term outcomes were sourced from two randomized controlled trials (BLISS 72/7614). Long-term outcomes were estimated via natural history models developed on the basis of data from the John Hopkins cohort of SLE patients. Direct costs consisted of short-term disease activity related costs, organ damage costs, SoC treatment and belimumab administration costs. Short-term costs and costs of treatment were calculated on the basis of resource utilisation elicited from a panel of experts and using list prices (2013 prices). Organ damage costs were identified through a literature review. Results were discounted at 3.5% for both costs and effects. The study was performed from the perspective of the health care payer. Results: Treatment with belimumab+SoC resulted in 0.81 added life years and 0.377 QALYs (Quality-Adjusted Life Years). This resulted in 18,350€ / LYG (Life-Year-Gained) and 27,254€ /QALY. Conclusions: Cost-effectiveness and cost-utility ratios of belimumab compared to SoC treatment are below internationally applied thresholds. Belimumab can be considered as an add-on therapy to SoC for the treatment of SLE patients with highly active disease in Greece. A533 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PSY65 Cost-Utility Analysis of Bosutinib for Previously Treated Chronic Myeloid Leukemia (Cml) In Portugal Paquete A T 1, Inês M 2, Duarte I 3, Almeida A 4, Esteves S 5, Miguel L S 1 - ISEG/UL, Lisboa, Portugal, 2Instituto de Medicina Molecular, Lisboa, Portugal, 3Pfizer Oncology, Porto Salvo, PA, Portugal, 4Instituto Português de Oncologia de Lisboa, Lisboa, Portugal, 5IPOLFG, Lisboa, Portugal . . . . . . . . 1CISEP Objectives: To assess the incremental cost-utility ratio (ICUR) of bosutinib for CML as a third-line (3L) treatment for chronic phase (CP) Philadelphia chromosomepositive (Ph+) patients in Portugal compared to hydroxycarbamide, in the societal perspective. Methods: A survival analysis model was adapted to the Portuguese setting. Bosutinib overall survival was based on Study 200, an open-label phase I/ II single-arm study of Ph+ CML patients. Overall survival with hydroxycarbamide and time spent in the accelerated and blastic phases, that are assumed to be independent of treatment, were estimated by an expert panel of five Portuguese haematologists. Resource consumption was elicited by this expert panel, being unit costs based on Portuguese official sources. Utility values were adapted from the IRIS study. A 5% discount rate was applied to both costs and consequences on a 50-year time horizon. Results: The use of bosutinib allows an increase of 4.1 life years (LY) and 3.5 quality adjusted life years (QALY), being associated to an additional cost of 88,319€ . Consequently, cost per LY is 21,465€ and cost per QALY is 25,252€ . Sensitivity analysis shows that results are mainly driven by survival time gained with bosutinib. Conclusions: Bosutinib for CML 3L treatment for CP patients in Portugal represents a substantial added benefit relative to hydroxycarbamide although with added costs per LY and per QALY. These ratios are generally accepted in Portugal, both below 30,000€ willingness to pay threshold. PSY66 Cost Effectiveness of Romiplostim for the Treatment of Immune Thrombocytopenia (Itp) Patients In the Czech Republic Brezina T 1, Klimes J 2, Dolezal T 2, Maskova H 1, Campioni M 3, Kutikova L 3 1Amgen s.r.o., Praha 1, Czech Republic, 2VALUE OUTCOMES, Prague, Czech Republic, 3Amgen (Europe) GmbH, Zug, Switzerland . . . . . . Objectives: To assess the cost-effectiveness of romiplostim in the treatment of adult patients with ITP in the Czech Republic, in comparison with the medical standard of care (SoC) and eltrombopag. Methods: A lifetime treatment sequence cost-utility Markov model was developed from the health care payer perspective. The model was based on the treatment sequences that reflect current practice for ITP management and was driven by platelet response (platelet count ≥ 50x109/L), which determined effectiveness and progression along the treatment pathway, need for rescue therapy and risk of bleeding. Costs were derived from reimbursement lists available in January 2013. Four scenarios were conducted where romiplostim was compared with SoC with rituximab, SoC without rituximab, SoC without both rituximab and mycophenolate mofetil (MMF), and eltrombopag. Patients were evaluated by splenectomy status and for the combined population (CP). Results: Compared to SoC with rituximab, romiplostim was dominant in splenectomised patients and CP, with cost savings of 2,203,982CZK and 1,078,899CZK and gains of 1.58 and 1.81 quality-adjusted life-years (QALYs), respectively, and was cost-effective in non-splenectomised patients with an ICER of 44,107 CZK/QALY. Compared to eltrombopag, romiplostim was dominant in splenectomised patients and CP, with cost savings of 1,626,409CZK and 741,613CZK and gains of 1.12 and 1.21 QALYs, respectively, and was cost-effective in nonsplenectomised patients with an ICER of 74,266 CZK/QALY. Similar results were shown in the other two scenarios: romiplostim was dominant in splenectomised patients and CP compared to SoC without rituximab and SoC without both rituximab and MMF, and was cost-effective in non-splenectomised patients with ICERs of 51,011CZK/QALY (SoC without rituximab) and 41,192CZK/QALY (SoC without rituximab and MMF). Conclusions: Romiplostim was less costly with higher QALY gain compared to SoC and eltrombopag in splenectomised patients and CP. Romiplostim was cost-effective with ICERs between 41,000 and 74,000CZK/QALY in all non-splenectomised populations in all four scenarios. PSY67 Cost-Effectiveness Analysis of Celecoxib In the Treatment of Patients With Chronic Pain In Japan Kawaguchi I 1, Kamae I 2, Soen S 3, Sakamoto C 4 Japan Inc, Tokyo, Japan, 2The University of Tokyo, Graduate School of Public Policy, Tokyo, Japan, 3Nara Hospital, Kinki University School of Medicine, Nara, Japan, 4Nippon Medical School, Graduate School of Medicine, Tokyo, Japan . . . . 1Pfizer Objectives: To evaluate the cost-effectiveness of celecoxib compared with loxoprofen for the treatment of patients with chronic pain caused by osteoarthritis, rheumatoid arthritis and low back pain in Japan. Methods: Using the Markov model, long-term simulations were conducted with a 3-month Markov cycle for each regimen of celecoxib vs. loxoprofen in the perspective of Japanese health care payers. The direct medical costs and quality-adjusted life years (QALYs) were estimated as outcome measures throughout the lifetime of patients. The base case population was aged 57 years which was the weighted mean age from the cited literature. All the costs were expressed as of March 2014, and annual discount rates of 2% were applied for both costs and health benefits. Results: The incremental effectiveness of celecoxib compared with loxoprofen was 0.024 QALY while the incremental cost was JPY 73,785 (USD 716, USD 1 = JPY 103), indicating the incremental cost-effectiveness ratio (ICER) was JPY 3,131,480 (USD 30,403) per QALY gained. Sensitivity analyses identified two factors substantially sensitive to ICER: 1) decrease in efficacy of celecoxib in the recurrence of symptomatic ulcer, and 2) increase in utility of the primary disease that causes chronic pain while the administration of the drug halts. The literature review, however, supported that both of those sensitive cases are not so much likely to happen. Conclusions: Celecoxib was considered cost-effective compared with loxoprofen in patients with chronic pain in Japan. PSY68 Population-Based Cost-Efficiency Simulation of Partial Versus Complete Thromboprophylaxis In Hospitalized Patients In Saudi Arabia: Application of A British Model Alomi Y 1, Alharbi E 2, Alshayban D 3, Khoshaim M 1, Alhowasi M 1, Zamakhshary M 1, Alhumaidi A 1, Alesseimi M 1, Abraham I 3 1Ministry of Health, Riyadh, Saudi Arabia, 2King Saud Medical City, Riyadh, Saudi Arabia, 3University of Arizona, Tucson, AZ, USA . . . . . . . . . Objectives: Venous thromboembolism (VTE) is a frequent but preventable complication in hospitalized patients. The objective of this population-based simulation was to estimate, for Saudi Arabia and using a British (NICE) model, the cost-efficiency of partial versus complete thromboprophylaxis of hospitalized adult patients at medium-to-very-high (M2VH) -VTE risk. Methods: Simulation including all 2012 Saudi adult medical (n= 2,129,866) and surgical (n= 964,033) hospital admissions, expressed in Saudi riyals (SAR), using three scenarios based on published realworld thromboprophylaxis studies: (1) international study and (2) Saudi subsample thereof (Cohen et al, Lancet 2008), and (3) Middle-Eastern study (Mokhtari et al, J Thromb Haemost 2011). Variables included rates for: high bleeding risk (mechanical versus pharmacoprophylaxis); major bleeding episode (MBE) secondary to pharmacoprophylaxis; deep vein thrombosis (DVT; pulmonary embolism (PE); and 12-month DVT/PE readmission following such discharge diagnosis. Costs considered were: thromboprophylaxis; management of MBE, DVT, PE; and 12-month DVT/PE readmission. Results: Scenario (1): total costs for partial and complete prophylaxis were SAR631,462,830 and SAR508,957,517 for a complete-over-partial cost-avoidance by 24.1% of SAR122,505,313/year (SAR92.14 per patient and SAR167.83 per patient at VTE risk). Scenario (2): total costs for partial and complete prophylaxis were SAR712,441,824 and SAR600,483,584 for a complete-over-partial cost-avoidance by 18.6% of SAR111,958,239/year (SAR77.35 per patient and SAR127.58 per patient at VTE risk). Scenario (3): total costs for partial and complete prophylaxis were SAR749,939,936 and SAR647,300,590 for a complete-over-partial cost-avoidance by 15.9% of SAR102,639,346/year (SAR62.77 per patient and SAR97.84 per patient at VTE risk). Conclusions: In this population-based cost-efficiency simulation for Saudi Arabia, thromboprophylaxis of all admitted M2VH-VTE risk versus partial methods is associated with significant cost savings per year. Considering the associated reductions in DVT, PE, and readmissions, even under increased pharmacoprophylaxis-related MBE risk and costs, complete thromboprophylaxis of (M2VH) -VTE risk patients achieves the Institute for Healthcare Improvement Triple Aim of better care, better patient outcomes, and reduced costs. PSY69 Cost Minimization Analysis of Activated Prothrombin Complex Concentrate (Apcc) Compared To Recombinant Factor Viia (Rfviia) for Hemophilia Patients With Inhibitors Mlcoch T 1, Klimes J 2, Dolezal T 2 1VALUE OUTCOMES, s.r.o., Prague, Czech Republic, 2VALUE OUTCOMES, Prague, Czech Republic . . . Objectives: Approximately 15-35% of patients with hemophilia A develop inhibitory antibodies to factor VIII. There are currently two bypassing agents to treat inhibitor patients in the case of bleeding episodes (BE) and preemptively before and during major surgeries. Both APCC and rFVIIa demonstrated similar efficacy and safety results. Our aim was to compare the costs of bypass strategy based on APCC or rFVIIa while we included only drug acquisition costs, and from them we derived costs of bleeding episodes and surgery costs in Hungary, Slovakia, Slovenia and Serbia. Methods: For the purpose of cost comparison, we developed a model using cost-minimization approach which included only the costs of APCC or rFVIIa. We assessed the cost differences for two basic scenarios - patients with bleeding episodes and patients undergoing major surgeries. The doses used in model were: i) BE APCC 62.5 U/kg, ii) surgery APCC 85 U/kg, iii) BE and surgery rFVIIa 90 μ g/kg. The doses are based on the SmPCs (middle of given dose intervals) and published literature (in the case of APCC used in surgery which is administered in higher doses in clinical practice compared to declaration in SmPC). Results: In the scenario of bleeding episodes, the use of APCC instead of rFVIIa brings the potential savings € 4,596-8,704; € 5,135-9,347; € 5,738-10,807; € 4,369-8,527 respectively in Hungary, Slovakia, Slovenia and Serbia (depending on the dosing scheme). While using APCC during major surgeries, the savings are equal to € 157,159 (if compared to rFVIIa therapy) and € 15,616 (if compared to the combination therapy of APCC and rFVIIa) in Hungary; € 180,653 and € 16,558 in Slovakia; € 196,965 and € 19,362 in Slovenia; and € 145,979 and € 15,422 in Serbia. Conclusions: The potential savings by using APCC instead of rFVIIa are significant for health care systems, APCC thus indirectly enables more treated patients for the same health care costs consumption. PSY70 Cost-Minimization Analysis of Methadona Opioid Analgesic for Mexican Patients With Acute and Chronic Secondary Cancer As Rotation Option In Severe Pain Soto Molina H 1, Sanchez K 1, Constanzo A 1, Escobar Juárez Y 1, Fernandez Z 2, Melendez C 2 Estudios Farmacoeconómicos, Mexico City, Mexico, 2Psicofarma S.A. de C.V., Mexico City, Mexico . . . . . . 1HS Objectives: Perform a full economic evaluation of cost minimization of the use of Methadone versus oxycodone (extended release), buprenorphine (transdermal patches), fentanyl (transdermal patches) and hydromorphone in opioid rotation for patients with acute severe pain secondary to cancer and chronic type as rotation option in Mexico, from the point of view of public health. Methods: A systematic literature review was conducted to identify articles and extract data of safety and efficacy and compare the available alternatives: methadone, oxycodone, buprenorphine (PT), fentanyl (PT) and hydromorphone. Based on the results of this systematic review was identified that 5 alternatives presented safety and efficacy profiles without significant differences. Economic evaluation corresponded to a cost minimization, the analysis was performed with three subgroups with rotation of opioids, each subgroup received a certain amount of morphine (80mg, 160mg, A534 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 480 mg respectively) before rotation the number of milligrams of each opioid was calculated based on equianalgesic tables using only direct medical costs, further a deterministic sensitivity analysis was performed. Results: The results demonstrate that methadone (Amidone ®) generates average savings per patient of $ 84.37 to $ 5,817.00 compared to oxycodone, buprenorphine (PT), fentanyl (PT) and hydromorphone in the three subgroups analyzed, the sensitivity analysis shows that methadone remains a cost-saving option. Conclusions: Using Methadone (Amidone ®) is a cost-saving option for patients with acute and chronic severe pain secondary to cancer in Mexico, from the institutional point of view. PSY71 Real-World Cost-Utility Evaluation of Multiple Myeloma Treatments In Stem Cell Transplanted Patients Gilabert Sotoca M , Schoenenberger Arnaiz J A , Pons Llobet N , Martinez Sogues M , Mangues Bafalluy I , Martinez Castro B Hospital Universitari Arnau de Vilanova, Lleida, Spain . . . . . . . Objectives: Multiple myeloma (MM) is an incurable disease with an incidence of 4-7 new cases by 100.000 people. In the setting of the study, 28% of the yearly 25 news cases of MM are candidates to autologous stem cell transplant (ASCT). Because of the lack of real-world economic studies, authors sought to describe the cost per quality-adjusted-life-year (QALY) of the MM treatment in a group of transplanted patients. Methods: An observational retrospective study was performed and included detailed clinical data from a transplanted cohort of patients with MM. All patients received bortezomib based treatments. Costs were evaluated from the payer’s perspective and included total drug costs and hospital admission costs. QALYs values were obtained from the CEA Registry of Tufts University. Four health states were considered: complete response (CR), partial response, stability and progressive disease. Time between state transitions was used to calculate QALYs for each patient. Results: The study included 17 patients with a mean age of 61,2 years; 12 of them (Group 1) were followed during 2,77±0,24 years and 5 (Group 2) during 1,6±0,19 years. Global complete response rate one year after ASCT was 8/17 (47%) and dropped to 5/12 (41,6%) at two years. For the whole cohort the median of QALY’s cost was 56.198€ (IQ range 36.391-70.339). For patients with CR one year after ASCT, the median of QALY’s cost was 46.358€ (IQ range 36.299-67.537). For all other patients the median of QALY’s cost was 56.676€ (IQ range 39.114-73.613). For Group 1 patients, the median of QALY’s cost was 51.278€ (IQ range 36.345-73.613). Conclusions: These results reveal that the actual costs of MM treatment using protocols that include bortezomib are above the 30.000-50.000€ threshold generally admitted in cost-effectiveness studies. PSY72 Orphan Drug Pricing In France: Influence of Main Factors GRAND H 1, Samson A L 2, Aulois-Griot M 1 Bordeaux, Bordeaux, France, 2Université Paris Dauphine, Paris, France . . . . 1Université Objectives: Orphan drugs (OD) require considerable expenditures, which causes difficulties in their market access. For several years, the price of these new therapies has often been criticized and considered as too high. However few studies about OD pricing mechanisms are available. The aim of the paper is to highlight the main factors that influence OD pricing in France. Methods: We collected public prices of 37 products designated as OD and approved by the European Medicines Agency until January 2014. Given that an OD can have several indications, our database contains 49 observations. For each observation, we calculated the ex-factory price without tax (defined per daily dose) defined by summary products characteristics. We also collected different characteristics of these products: improvement in actual clinical benefit (IACB), actual clinical benefit (ACB), number of OD indications, number of comparators drugs, number of competing orphan drugs, therapeutic use, target population, inclusion on the list of medicines reimbursed by National Health Insurance. Ordinary least squares are used to analyse the determinants of OD prices. Results: The distribution of OD prices is very heterogeneous, with a minimum of 2.34 euros and a maximum of 2882.08 euros. The average (SD) price is 380.675 euros (687.414). Our main results are that OD prices are significantly lower for OD that do not improve actual clinical benefit (-136%, p= 0.024), for OD with only one orphan indication (-80%, p= 0.047), and for OD with a high target population (-251%, p= 0.000). Conclusions: There is a need to understand OD pricing mechanisms. Our study shows that innovation and research efforts are encouraged by pricing policy. PSY73 Evaluation of Use of Belimumab In Clinical Practice Settings (Observe Study) In Spain: Health Resource Utilization and Labour Absenteeism Cortés J 1, Andreu J L 2, Calvo J 3, García-Aparicio A M 4, Coronell C G 5, Díaz-Cerezo S 5 1Hospital Vall d´Hebrón, Barcelona, Spain, 2Hospital Puerta de Hierro, Majadahonda (Madrid), Spain, 3Hospital de Sierrallana, Torrelavega, Spain, 4Hospital VIrgen de la Salud, Toledo, Spain, 5GSK España, Tres Cantos, Spain . . patients presented an overall clinical improvement of ≥ 20%, ≥ 50% and ≥ 80%. These improvements were associated with a reduction of steroid use (75% of patients on steroids at belimumab-initiation decreased mean dose from 14.8 to 6.8mg/day; p< 0.001) and HRU between the pre/post index periods: emergency-room visits 1.65 to 0.41; p= 0.001; unscheduled visits to treating-physician 1.02 to 0.03; p< 0.001, visits to other specialists (1.64 to 1.06; p= 0.017) and antibody tests (7.78 to 7.53; p= 0.47). An increase in HRU was observed for hematological and renal tests (3.14 to 3.52; p= 0.045) and (5.95 to 6.59; p= 0.024), respectively. Working patients (39%) also showed a reduction in the LA days between the pre/post index periods (25.6 to 5.7 days; p= 0.025). Conclusions: Belimumab treatment yielded improved clinical outcomes and a reduction in HRU directly related with SLE management, as corticoid use. Mean number of LA days also showed a substantial reduction, especially important in SLE, mostly affecting young patients. . . . . . . . Objectives: To analyze the health resource utilization (HRU) and labour absenteeism (LA) in Systemic Lupus Erythematosus (SLE) patients treated with belimumab in the Spanish clinical setting. Methods: OBSERVE is a multicenter retrospective medical chart-review study. Twenty-five rheumatologists from Spanish hospitals with > 10 SLE patients annually and >5-years of practice experience identified adult SLE patients who had received 6-months of belimumab treatment. In the 6-months pre- and 6-months post-index periods physicians assessed: demographics, comorbidities, SLE disease characteristics, treatment clinical outcomes, HRU and LA data. Index-date is the date of the first infusion. OBSERVE primary endpoint was overall clinical response per physician judgment. Statistical analyses included appropriate tests for paired-samples (parametric/ non-parametric). Two-way P-values 0.05 were considered statistically significant. Results: A total of 64 patients were eligible for analysis: mean age 42.7±12 years; female 89%; hypocomplementemia 70% and high anti-dsDNA 69%. After receiving 6-months therapy, 72%, 52% & 27% of SYSTEMIC DISORDERS/CONDITIONS – Patient-Reported Outcomes & Patient Preference Studies PSY74 Impact of Patient Programs On Adherence In Inflammation and Immunology: A Global Systematic Review and Meta-Analysis of Published Evidence Burudpakdee C 1, Khan Z M 2, Gala S 1, Nanavaty M 1, Kaura S 2 1MKTXS, Raritan, NJ, USA, 2Celgene Corporation, Summit, NJ, USA . . . . . . Objectives: Patient adherence is important for successful treatment in chronic conditions, including inflammation and immunology (I&I) diseases, to improve patient outcomes. Programs and interventions that aim at improving medication adherence play an essential role in optimizing care. This review and metaanalysis assessed the effectiveness of different types of adherence programs in I&I. Methods: A global systematic literature search was conducted and studies were identified from PubMed, conference proceedings, and grey literature. Selection criteria included studies of patient programs in I&I diseases published in English language between January 2008 and September 2013 that reported % of adherent patients. A meta-analysis was performed using a random effects model. Weights, odds ratios, 95% CI, and forest plots were developed for behavioral, informational, and combination interventions. Results: Of 29 studies screened, a total of 13 studies were eligible for inclusion in the meta-analysis. Seven studies were in patients with osteoporosis, 4 in ulcerative colitis, 1 in childhood-onset systemic lupus erythematosus, and 1 in rheumatoid arthritis / psoriasis. Overall, patient programs increased adherence (OR = 2.48, 95% CI = 1.68 - 3.64, P < 0.00001) as compared to standard of care or no intervention. Combination interventions that used both informational and behavioral strategies were superior in increasing adherence (OR = 3.68, 95% CI = 2.20 - 6.16, P < 0.00001) compared to interventions using solely a behavioral strategy (OR = 1.85, 95% CI = 1.00 - 3.45, P = 0.05) or only an informational strategy (OR = 2.16, 95% CI = 1.36 - 3.44, P = 0.001). Conclusions: Patient programs and interventions can significantly improve adherence in I&I diseases as compared to standard of care or no intervention. Programs employing a multimodal approach are more effective in improving adherence than either informational or behavioral strategies alone. This in turn may improve patient outcomes. PSY75 Adherence To Anticoagulant Therapy In Children Hospitalized for Pulmonary Embolism and Deep Vein Thrombosis Singh R R 1, Gupte K P 1, Wilson J P 1, Moffett B S 2 University of Texas at Austin, Austin, TX, USA, 2Baylor College of Medicine/Texas Children’s Hospital, Houston, TX, USA . . . . . . . . 1The Objectives: The American College of Chest Physicians Guideline recommends anticoagulant therapy for at least three months in children with venous thromboembolism. The objectives of the study were to evaluate the medication utilization patterns, and the predictors of adherence to anticoagulant therapy in the pediatric population. Methods: Texas Medicaid medical and prescription claims from June 01, 2007 to September 31, 2012 were extracted for children (< 18 years) hospitalized for Pulmonary Embolism (PE) or Deep Vein Thrombosis (DVT). The index date was defined as the date of the first prescription of an anticoagulant warfarin (oral) and/or enoxaparin (injectable) given within 14 days after discharge from hospitalization. Patients hospitalized for atrial fibrillation, air/fat embolism, bleeding/ coagulation disorder within 90 days of discharge were excluded. Proportion of days covered (PDC≥ 80% vs. < 80%) was used to assess adherence to anticoagulants while controlling for demographics, cause of hospitalization, history of NSAID use, anticoagulant use, malignancy, drug type, and Charlson comorbidity index (CCI). A multivariate logistic regression analysis was used. Results: The patients (n= 57) had a mean (±SD) age of 14.1 (±4.9) years, were primarily female (54.4%), African American (61.4%), enoxaparin users (54.4%), and had a mean (±SD) CCI of 19.7 (±39.4). The mean (±SD) adherence rates for warfarin and enoxaparin were 85.6% (±22.3%) and 78.2% (±22.7%), respectively. 66.7% were adherent (PDC≥ 80%) to anticoagulant therapy. The median (Mean±SD) persistence with anticoagulant therapy was 84.6 (71.9±33.3) days. Logistic regression showed that increasing age was significantly associated with adherence to anticoagulant therapy (Odds Ratio= 1.3, p= 0.0158), after controlling for covariates. Conclusions: Nearly one third of the pediatric patients on anticoagulant therapy after discharge from hospitalization with PE or DVT were non-adherent. Further research is needed to underline the factors responsible for non-adherence in pediatric patients. PSY76 New Observer-Reported Outcomes To Measure Treatment Satisfaction, Compliance, Palatability, and Gi Symptoms for Patients Needing Iron-Chelation Therapy Lasch K 1, Horodniceanu E G 2, Carter J A 2, Dhatt H 2, Bal V 3, Côté I 3, Constantinovici N 4, Herranz R M 4, Malet I 4 . . . . . . . . . . . . A535 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1Pharmerit International, Cambridge, MA, USA, 2Pharmerit International, Bethesda, MD, USA, Pharmaceuticals Corporation, East Hanover, NJ, USA, 4Novartis Pharma AG, Basel, Switzerland outcomes and greater indirect costs. The distribution of BMI is similar to the 5 E.U. Finding suggest the need for additional weight-loss options. Objectives: Adherence to iron chelation therapy (ICT) is essential for patients with transfusion-dependent anemia (TDA) (e. g., sickle cell disease [SCD] and thalassemia) to increase survival and minimize disease-related sequellae. Adherence has been noted as especially problematic for children and adolescents. In order to measure compliance and treatment satisfaction in young children, the Satisfaction with Iron Chelation Therapy (SICT) instrument (initially developed for use by adults and adolescents) was modified to be an observer-reported outcome (ObsRO) measure to be administered electronically to caregivers. In addition, three other ObsROs were developed measuring compliance, palatability, and gastrointestinal (GI) symptoms related to treatment benefit of a new ICT formulation. Methods: Subjects included 10 caregivers of children with TDA. Informed consent was obtained. Two sets of face-to-face cognitive interviews were conducted iteratively with modification to items and further debriefing of modifications. Interviews began with an open-ended question to elicit caregivers’ and their reports of their child’s experiences with ICT. Interviews were audio recorded and transcribed. Data were analyzed using ATLAS. ti software. Results: Three interviewers conducted 10 caregiver interviews in 6 US cities. Caregivers were 90% female aged 35-65 (mean= 48). Children of caregivers included those with SCD (80%) and thalassemia (20%) and were 60% female, aged 2-17 (mean= 9). Responses to the open-ended question confirmed several concepts in the modified SICT and the need for new items. The resultant questionnaires included the modified SICT (17 items), Compliance (2 items), Palatability (4 items), and GI Symptom Diary (6 items). Changes were made to each to ensure comprehension, relevance, lack of redundancy, and appropriate response options. Conclusions: Evidence supports the content validity of the modified SICT, Compliance, Palatability, and GI Symptom Diary questionnaires. Use in clinical research awaits tests of validity, reliability, and responsiveness. PSY79 The Multicentric Castleman’s Disease (Mcd) -Symptom Scale (Mcd-Ss): Development and Validation of A Patient-Reported Outcome (Pro) Measure for An Ultra-Orphan Disease 3Novartis PSY77 Impact of Pulmonary Exacerbations On Eq-5d Measures In Patients With Cystic Fibrosis Solem C T 1, Vera-Llonch M 2, Liu S 1, Botteman M 1, Lin F J 1, Castiglione B 2 International, Bethesda, MD, USA, 2Vertex Pharmaceuticals, Boston, MA, USA . . . . . . . . 1Pharmerit Objectives: Pulmonary exacerbations (PEs) in patients with cystic fibrosis (CF) have been reported to impact health-related quality of life. This analysis examined the impact of PEs on EQ-5D measures in patients with CF using trial-based data. Methods: In a 48-week randomized, placebo-controlled study of ivacaftor in patients ≥ 12 years with CF and a G551D-CFTRmutation, the EQ-5D questionnaire, including visual analog scale (EQ VAS), was administered at baseline, 15 days, 8 weeks, and every 8 weeks thereafter. UK preference weights were employed to derive the EQ-5D index. PE was defined as a change in antibiotic therapy for ≥ 4 of 12 signs or predefined symptoms. EQ-5D measures collected in the 8-week pre- and post- exacerbation periods were examined after pooling observations across study treatments and visits. Results: A total of 146 PEs were experienced by 72 (44.7% of total 161) patients, including 52 (35.6%) PEs that required hospitalization. Mean (±SD) duration was 30.0±22.2 days for PEs requiring hospitalization (n= 48) and 20.6±11.6 days for those not requiring hospitalization (n= 89) (9 PEs had missing end dates). For PEs requiring hospitalization, mean (±SD) EQ-5D index/VAS scores within 1-8 weeks before PE start were 0.91 (±0.13)/75.5 (±14.2), respectively. The lowest average EQ-5D index/VAS scores were reached within 1 week of PE start; mean scores within 1 week were 0.76±0.33 (p= 0.044 vs. 1-8 weeks prior to PE) /63.7±24.1 (p= 0.039). Corresponding values for PEs not requiring hospitalization were 0.89±0.16/73.4±16.1 within 1-8 weeks before PE start, and 0.90±0.13 (p= 0.777) /68.4±17.4 (p= 0.124) within 1 week of PE start. Conclusions: In a clinical study of patients with CF (≥ 12 years of age and a G551D-CFTR mutation), the PEs were associated with lower EQ-5D and VAS scores, primarily among those requiring hospitalization. Reducing PEs requiring hospitalization is likely to improve health-related quality of life among these patients. PSY78 Is Obesity A Problem In Brazil? Richard L 1, Gupta S 2, Pomerantz D 2, Forsythe A 3 1Eisai Europe Ltd, Hatfield, UK, 2Kantar Health, Princeton, NJ, USA, 3Eisai Inc., Woodcliff Lake, NJ, USA . . . . Objectives: This study investigated the effect of BMI on quality of life, productivity, activity impairment, health care resource-utilisation, and associated costs. Methods: Results were from the 2011-2012 Brazil National Health and Wellness Survey (N= 24,000), a nationally representative, online survey of respondents aged≥ 18 years. Analysis focused on normal weight (BMI≥ 18.5 & < 25 kg/m2), overweight (BMI≥ 25 & < 30 kg/m2), obese class (OC) I (BMI≥ 30 & < 35 kg/m2), OC II (BMI≥ 35 & < 40 kg/m2), and OC III (BMI≥ 40 kg/m2) respondents. Outcomes included quality of life (SF-36v2), health utilities (SF-6D), productivity loss (Work Productivity and Activity Impairment questionnaire) and resource utilisation (type/number of visits) in the past six months. Direct and indirect costs were estimated from public sources. Generalised linear models predicted outcomes as a function of BMI category, adjusting for covariates (e. g., age, gender, comorbidities). Results: Among 22,871 respondents, 46.4% were normal weight, 34.9% were overweight, 12.9% were OC I, 3.7% were OC II, and 2.0% were OC III. Adjusting for covariates, mental, physical component summary (normal weight: 51.8; overweight: 51.4; OC I: 50.3; OC II: 48.5; OC III: 46.0), and health utilities (normal weight: 0.73; overweight: 0.72; OC I: 0.71; OC II: 0.70; OC III: 0.67) decreased as BMI increased (OCs vs. normal, p< 0.05). Amongst the employed (66.1%), impairment while working (normal weight: 15.5%; overweight: 15.6%; OC I: 16.9%; OC II: 18.2%; OC III: 22.4%, OCs vs. normal, p< 0.05) increased as BMI increased. Normal weight (vs. OCs, p≤ 0.05) had lower activity impairment, fewer provider visits, and lower indirect costs (normal weight: R$333; overweight: R$338; OC I: R$385; OC II: R$395; OC III: R$458). Conclusions: Over 50% of Brazilians are overweight or obese. Increase in BMI is associated with poorer Casper C 1, Van Agthoven M 2, Rothman M 3, Fleming S 4, Ho K F 4, Qi M 4, Vermeulen J 5, Cavet J 6 1Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 2Janssen-Cilag BV, Tilburg, The The Netherlands, 3Janssen Global Services, LLC, Spring House, PA, USA, 4Janssen R&D, US, Spring House, PA, USA, 5Janssen, Leiden, The Netherlands, 6Christie Hospital, Manchester, UK . . . . . . . . . Objectives: MCD, an ultra-orphan lymphoproliferative disorder with heterogeneous symptomology, including fatigue, malaise, sweats, and nodal/extranodal masses, is driven by dysregulated interleukin-6 signaling. Siltuximab, an antiinterleukin-6 antibody, is approved in the US and EU for treatment of human immunodeficiency virus (HIV) - and human herpes virus-8 (HHV-8) -negative adult MCD patients. This study aimed to fill an unmet need for a validated PRO tool to quantify MCD symptoms and responses to therapy. Methods: Using FDA guidance, an MCD symptom scale (MCD-SS) was developed by collecting and evaluating symptoms in 12 patients from 2 US centers, with further validation using additional patients from 1 US center. Item performance, scale structure and scoring, reliability, validity, responsiveness, and clinically meaningful change were evaluated in a randomized, placebo-controlled, double-blind study of siltuximab (N=79) in HIV-, HHV-8-negative MCD patients (Wong Blood 2013:122 (21)). Results: Patient-reported MCD symptoms were: cough, dizziness, drowsiness, dry mouth, fever, headaches, hot flashes, itching, skin lesions, loss of appetite, nausea, numbness/tingling, pain, rash, shortness of breath, sweating, swollen lymph nodes, swelling/edema, tiredness/fatigue, vomiting, weakness, weight gain, and weight loss. Following exploratory factor analysis, 3 domains were created: fatigue, rash/itching, and sweats (total score also calculated). The final MCD-SS comprised 16 questions with 6 response options: “did not experience,” “very mild,” “mild,” “moderate,” “severe,” and “very severe” (score, 0-5). Subjects reported a mean (SD) and median (range) of 9.2 (3.76) and 9.5 (1-16) symptoms, respectively. Test-retest reliability was acceptable (r> 0.70) for fatigue, sweats, and the total score but lower for rash/itching (r= 0.65). Internal consistency exceeded 0.70 for all. Lower ECOG scores were associated with lower MCD-SS scores. MCD-SS total scores were moderately correlated (r= 0.48) with baseline clinicianreported signs and symptoms. Conclusions: This is the first known attempt to develop a PRO instrument for MCD. These data support further use of the MCD-SS in the target population. PSY80 “I Don’t Know How It Happened Or When Everything Changed. It’s Like I Blinked and All of A Sudden, I Didn’t Recognise My Own Body”: Using Qualitative Insights To Develop A Conceptual Model To Understand the Lived Experience of Patients With Systemic Sclerosis Willgoss T G 1, Humphrey L 1, Blankenburg M 2 1Abacus International, Manchester, UK, 2Bayer HealthCare Pharmaceuticals, Berlin, Germany . . . . Objectives: Systemic sclerosis (SSc) is a rare, multisystem chronic disease characterised by fibrotic changes affecting all or some bodily organs. SSc is associated with high morbidity and a significant effect on patients’ health-related quality of life. Qualitative insights can provide rich context to the patients’ experience of a disease. Conceptual models based on qualitative data are a valuable way to “explain, with graphically or narrative form, the main things to be studied - the key factors, concepts and variables – and the presumed relationships between them”. Despite the existence of SSc-specific measures, no conceptual model of SSc exists. This research aimed to develop a qualitatively-derived, patient-centred, conceptual model of SSc. Methods: To identify patient-reported SSc symptoms and impacts, we reviewed qualitative literature (published since 2000 to limit the search) in which experiences of living/coping with SSc are described. We also reviewed social media blogs/forums to identify additional concepts and provide supporting quotes. Concepts were identified by independent researchers who collaboratively developed the model. Results: Twelve qualitative studies and 150 social media posts were reviewed. The review identified 56 symptom concepts, which were categorised into 13 domains (peripheral, cognitive, pain, neurological, cardiorespiratory, ophthalmological, gastrointestinal, fatigue, nasopharyngeal, weight, oesophageal, dizziness, nausea) and 48 impact concepts, which were categorised into 8 domains (daily living, diet, social, clothing, work, physical and psychological functioning). Of note, nausea and dizziness were identified only through the social media review. Conclusions: A conceptual model for SSc was developed based on qualitative insights. The model depicts the diverse range of symptoms and impacts experienced by patients. By incorporating a social media review, relevant symptoms, which would not have otherwise been identified, were found and included in the model. This research is an important first step in identifying the most relevant and conceptually comprehensive clinical outcomes assessments for clinical research/practice. PSY81 Prevalence of Neuropathic Pain and Its Disease Burden In Korea Patients With Lumbar Spine Surgery Cho Y E 1, Moon S H 2, Whang C J 3, Kim H J 4 1Gangnam Severance Spine Hospital, Yonsei University College of Medicine, Seoul, South Korea, 2Department of Orthopedics, College of Medicine, Yonsei University, Seoul, South Korea, 3Asan Medical Center, College of Medicine, University of Ulsan, Seoul, South Korea, 4Pfizer Pharmaceuticals Korea Limited, Seoul, South Korea . . . . . . . . Objectives: The objectives of this study were to investigate the prevalence of Neuropathic Pain (NP) and newly occurred (de novo) NP postoperatively. This study also aimed to identify the disease burden (pain severity and quality of life (QoL)) of NP. Methods: This study was a nationwide, multi-centered, prospective, and A536 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 observational study. It was conducted from September 2011 to December 2012 and included a total of 1,109 patients who were scheduled for lumbar spinal surgery from 44 spinal centers (both orthopaedic surgery and neurosurgeons). Patients were diagnosed of having NP if the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS) pain scale criteria were ≥ 12 points. The patients were investigated to assess their pain severity using pain numeric rating scale (NRS) and quality of life using EuroQol (EQ)-5D at baseline, after 1 week and 3 months of the surgery. Results: Among 1,109 patents, at baseline, NP was identified in 404 (36%) patients. After 1 week and 3 months of the surgery, NP was found in 8.6% and 4.0% patients respectively. Among the 705 patients without NP preoperatively, the prevalence of de novo NP occurred in the 1 week and 3 months of post-surgery was 3.1% and 2.3% respectively. At baseline, NP patients showed lower QoL compared with non-NP patients (0.49 vs 0.53 p< .001). However, NP patients improved more their QoL compared to non-NP patients after 3 months (0.86 vs 0.84 p=.029). Among the de novo NP patients at 3 months after surgery (n= 16), the pain severity was not improved after 1 week and 3 months of the surgery. Conclusions: In Korea, NP patients were suffered from severe pain and lower QoL than non-NP patients. De novo NP caused severe pain which may not easily be handled. Those study findings highlight that timely diagnosis and management of NP are required in patients with lumbar spine surgery. PSY82 The Pain Assessment for Lower Back Symptoms (Pal-S): Refinement of A New Pro Instrument Through A Mixed Methods Approach McCarrier K P 1, Bushnell D M 1, Ramasamy A 2, Liedgens H 3, Blum S I 4, Cano S 5, Martin M L 1, Patrick D L 6 1Health Research Associates, Inc., Seattle, WA, USA, 2Forest Research Institute, Jersey City, NJ, USA, 3Grünenthal GmbH, Aachen, Germany, 4GlaxoSmithKline, Collegeville, PA, USA, 5ScaleReport, Stotfold, UK, 6University of Washington, Seattle, WA, USA . . . . . . . . . . . . . Objectives: The Pain Assessment for Lower Back Symptoms (PAL-S) is a Patient Reported Outcome (PRO) instrument being developed to assess the key symptoms of chronic low back pain (cLBP). Qualitative development included both concept elicitation and cognitive interviews. As part of the ongoing development of the instrument, we further evaluated and refined the PAL-S using a mixed methods approach. Methods: Adults self-reporting a clinical diagnosis of cLBP were recruited from an existing US-based commercial survey panel to participate in a pilot quantitative study. Qualifying participants completed a web-based survey consisting of the 14-item PAL-S and items assessing clinical, treatment, and demographic characteristics. Study data was analyzed to assess item- and scale-level performance of the PAL-S using Rasch Measurement Theory analyses. Following analysis and modification, two waves of cognitive interviews were conducted to evaluate respondent understanding of the revised PAL-S. Results: The dataset included 598 respondents (mean age: 55.5±12.6; 67.9% female; 88.0% white; and 54.0% married) who had cLBP for mean of 15.2±11.5 years. The Rasch analyses item threshold maps showed only two items having ordered thresholds, suggesting that respondents experienced increased difficulty distinguishing between options at the lower levels of the 0-10 scale. Simulations collapsing the responses to a bestfit 4-point response scale resulted in improved ordering of thresholds, suggesting a more optimal response option structure. Based on these findings, the numeric response scale of the PAL-S items was replaced with a 4-point verbal rating scale incorporating response choices such as not at all, slight, moderate, and severe. Findings from eight cognitive interviews confirmed patient comprehension and relevance of the revised instrument. Conclusions: The mixed-methods approach proved valuable to the ongoing development of the PAL-S, as Rasch analyses identified a need for refinement of the response scale. The measurement properties of the revised PAL-S will be evaluated in additional web-based and clinic-based quantitative studies. PSY83 Impacts of Lower Back Pain: Refinement of the Pain Assessment for Lower Back-Impacts Questionnaire (Pal-I) Using a Mixed Methods Approach Bushnell D M 1, McCarrier K P 1, Ramasamy A 2, Liedgens H 3, Blum S I 4, Cano S 5, Martin M L 1, Patrick D L 6 1Health Research Associates, Inc., Seattle, WA, USA, 2Forest Research Institute, Jersey City, NJ, USA, 3Grünenthal GmbH, Aachen, Germany, 4GlaxoSmithKline, Collegeville, PA, USA, 5ScaleReport, Stotfold, UK, 6University of Washington, Seattle, WA, USA . . . . . . . . . . . . . Objectives: The Pain Assessment for Lower Back-Impacts (PAL-I) is a patientreported outcome (PRO) instrument being developed to assess key impacts, e. g. walking, sitting, standing, etc. associated with chronic low back pain (cLBP). Following a mixed methods approach, the PAL-I development included qualitative work (both concept elicitation and cognitive interviews) and now a quantitative “pilot study” evaluation for further content validity. Methods: Adults self-reporting a clinical diagnosis of cLBP were recruited from a US-based commercial survey panel. Qualifying participants completed a web-based survey consisting of the 13-item PAL-I and items assessing clinical, treatment, and demographic characteristics. Study data was analyzed to assess item- and scale-level performance of the PAL-I using Rasch Measurement Theory analyses. Following analysis and modification, cognitive interviews were conducted to evaluate patient understanding of the revised PAL-I. Results: The 598 subjects in the pilot study reported having cLBP (mean of 6.1 on 11-point numerical rating scale, 0= no pain). Subjects experienced cLBP for 0.3 to 66 years (mean 15.2, SD 11.5), were 55.5 years old (SD 12.6), 67.9% female, 88.0% white and 54.0% married. The Rasch item threshold map showed only 2 items having an ordered threshold identifying problems with the response categories. Category probability curves indicated subjects had “difficulty” endorsing items specifically in relation to the extreme options. Based on these findings, four items were removed and the response options were modified for the remaining items (from 6-point scale 4-point: Not at all limited, Limited a little, Limited a lot, Did not do). Comprehension of the revised instrument was evaluated and confirmed during eight individual cognitive interviews. Conclusions: The mixed-methods approach provides valuable support in the development of a fit-for-purpose instrument assessing impacts of cLBP. Upon testing this revised PAL-I in a second pilot quantitative study, the final measure will undergo formal validation including sensitivity to change. PSY84 Pro Claims In Orphan Medicines Approved By the European Medicines Agency (Ema) for the Treatment of Lymphoproliferative Disorders Acquadro C 1, Perret C 2, Arnould B 3 Research Trust, Lyon, France, 2MAPI Research Trust, Lyon, France, 3Mapi, Lyon, France . . . 1Mapi Objectives: 1) To identify orphan medicines indicated for lymphoproliferative disorders approved by the European Medicines Agency (EMA); (2) To identify medicines for which a PRO evaluation was performed; (3) To list those with a PRO labeling claim, and (4) To identify reasons for not granting a PRO claim. Methods: The search was performed on the EMA website (06/21/2014). The products were browsed by type (i. e., orphan medicines). Products refused and withdrawn were excluded. The PROLabels database was searched for each product retrieved to identify any PRO claim in the label. Summary of Product characteristics (SMPc) and CHMP Assessment Reports (AR) were retrieved for each product and analyzed to find out about PRO evaluation reported in the AR and not reported in the label. Results: Thirteen orphan medicines indicated in lymphoproliferative disorders were identified, representing three main indications: lymphomas (Hodgkin, systemic anaplastic large cell, T-cell lymphoblastic, mantle-cell), leukemias (chronic lymphocytic, hairy cell, acute lymphoblastic) and multiple myeloma. Only one product had a PRO claim: ofatumumab (resolution of constitutional symptoms). The label of another product (brentuximab vedotin) indicated “resolution of B symptoms.” However, there was no mention in the AR on how the symptoms were collected (patient or clinician). For one product (pomalidomide), a HRQL evaluation was mentioned in the AR, but not reported in the label. However, there was no information about this evaluation in the AR and the reader is left to wonder about the HRQL results and the reasons for not including them in the label. Conclusions: The percentage of PRO claims in orphan medicines (7.7%) is inferior to the percentage of PRO claims in all EMA products (26%). This is remarkably low considering the profound effect of lymphoproliferative disorders on patients’ life. Efforts should be made to improve the reporting of PRO data in the CHMP Assessment Reports. PSY85 Psychometric Validation of the Newly Developed Phenylketonuria– Quality of Life (Pku-Qol) Questionnaires Assessing the Impact of Phenylketonuria and Its Treatment On Patients’ Quality of Life Bosch A M 1, Burlina A 2, Cunningham A 3, Bettiol E O 4, Moreau-Stucker F 5, Benmedjahed K 6, Regnault A 6 1University Hospital of Amsterdam, Amsterdam, The Netherlands, 2University Hospital of Padova, Padova, Italy, 3Tulane University School of Medicine, New Orleans, LA, USA, 4University of Geneva Hospitals and Faculty of Medicine, Geneva, Switzerland, 5EMD Serono Inc, Billerica, MA, USA, 6Mapi, Lyon, France . . . . . . . . . Objectives: Phenylketonuria (PKU) is a rare genetic disorder impacting phenylalanine (Phe) metabolism. Treatment involves a lifelong Phe restricted diet that is strict and socially demanding. Even when treated early and well, mild cognitive abnormalities have been seen. PKU can affect quality of life in individuals and their families. The phenylketonuria–quality of life (PKU-QOL) questionnaires are the first PKU-specific QOL questionnaires ever developed. The study aimed to perform the psychometric validation of these questionnaires. Methods: An observational study was conducted in France, Germany, Italy, The Netherlands, Spain, Turkey and the UK to finalize and validate PKU-QoL questionnaires in individuals with treated PKU aged 9–11, 12–17 and ≥ 18 years, and in parents of individuals < 18 years. Questionnaires were assessed for reliability (internal consistency, test–retest), concurrent validity (using three generic questionnaires adapted to the respondent: PedsQOL, SF-36 for adults and CHQ-PF28) and clinical validity (using PKU severity and overall assessment of patient health status). Results: In total, 559 participants (306 individuals, ages 9–45 years; 253 parents, ages 24–66 years) were included in the analysis. Return rate and quality of completion of the questionnaires were good, indicating good acceptability. Scores were defined to assess all relevant aspects of experiences: PKU symptoms, impact of PKU, dietary protein restriction and supplementation. Reliability and validity were satisfactory overall for the adolescent, adult and parent PKU-QoL questionnaires, and slightly weaker but acceptable for the child version. Conclusions: The four PKU-QOL questionnaires are valid and reliable instruments for assessing the specific quality of life aspects that are affected in individuals with PKU of different age groups (children, adolescents and adults) and their parents, and are available in seven languages. They are very promising tools for focused evaluation of PKU impact on individuals and parents in different countries, and for monitoring the efficacy of therapeutic strategies. PSY86 Evaluating Relationship Between White Blood Cells and Platelets During Recovery Phase In Dengue Hemorrhagic Fever Cases In Punjab, Pakistan: A Retrospective Study Rasool F 1, Ahmad M 2, Masood I 2, Khan H M S 3 of Punjab, Lahore, Pakistan, 2The Islamia University of Bahawalpur, Bahawalpur, Pakistan, 3Islamia University Bahawalpur, The Islamia University of Bahawalpur, Pakistan . . . . . . 1University Objectives: Dengue infection is a major cause of disease in tropical areas with an estimated 50 million infections occurring each year and more than 2.5 billion people being at risk of infections. The main objective of this study was to investigate relation between white blood cells and platelets during recovery phase in dengue hemorrhagic fever. Methods: A retrospective multi-center study was conducted on 1000 seropositive cases of dengue fever. Results: More prevalence has been observed in male 880 (88%) as compared to female 120 (12%). A rapid fall in white blood cells count (WBC) was observed in initial CBC reports at start of disease then A537 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 in platelet count. During recovery phase WBC count increased first followed by platelets count production after 3-4 days. Among 1000 confirmed dengue fever patients 812 were considered dengue hemorrhagic fever cases on the basis of clinical finding. In most of these cases (n= 783; 96.47%), directly proportional relation was observed between WBC and platelets count. Conclusions: It is wrongly perceived in community that dengue virus infection is still progressing when platelets count is below normal even the white blood cells counts is getting better during recovery phase. White blood cells production during recovery phase is a good indicator about recovery of disease rather than focused on platelets counts production. PSY87 Physicians’ and Patients’ Preferences Over the Attributes of Biological Agents Used In the Treatment of Rheumatic Diseases In Spain: A Conjoint Analysis Martin E 1, Rodriguez M 2, Ibero I 3, Raya E 4, Nolla J M 5, Nocea G 6, Aragon B 6, Lizán L 7, Paz S 7 La Paz, Madrid, Spain, 2Complejo Hospitalario de Ourense, Ourense, Spain, 3Hospital General de Alicante, Alicante, Spain, 4Hospital Universitario San Cecilio, Granada, Spain, 5Hospital Universitario de Bellvitge, Barcelona, Spain, 6MSD, Madrid, Spain, 7Outcomes’10, Castellon, Spain . . . . . . . . . . 1Hospital Objectives: To define the importance values assigned to the attributes of biological agents (BA) by Spanish rheumatologists and patients with rheumatic diseases: rheumatoid arthritis (RA), ankylosing spondylitis (AS) and psoriatic arthritis (PA). Methods: Observational, cross-sectional design based on conjoint analysis. RA, AS and PA patients diagnosed at least 2 years prior and currently or previously (≤ 1 year ago) receiving BA for a minimum of 1 year were consecutively recruited. Rheumatologists with at least 3 year experience on BAs participated. A literature review and 4 focus groups were undertaken to identify attributes and levels. Scenarios were selected using orthogonal design. Participants ranked 8 scenarios from 1 (most preferred) to 8 (least preferred). Relative importance (RI) of attributes was calculated. Multivariate regression analysis was performed for each attribute. Results: 488 patients [male= 50.9%; mean (SD) age= 50.6 (12.06) years; RA= 33.8%, AS= 32.4%, PA= 33.8%; mean time from diagnosis= 12.6 (8.2) years] and 136 rheumatologists [male= 50.4%; mean age= 46.4 (9.1) years; mean time of practice= 16.7 (8.8) years] took part. Most important attributes selected by patients and physicians, respectively, were: ‘Pain relief and improvement of the functional capacity’ (RI= 49.1% and 48.9%); ‘Risk of adverse events’ (RI= 31.8% and 31.5%), ‘Administration method’ (RI= 10.2% and 11.4%) and ‘Time to perceive the need for a new dose’ (RI= 9.0% and 8.2%). The ideal BA, for patients and physicians, should allow pain relief and an improvement of the functional capacity, with a low risk of adverse events, a long time prior to perceiving the need for a new dose and self-administration at home, when possible. Conclusions: Although efficacy and safety are key for patients with rheumatic diseases and rheumatologists to make a choice over a BA, the need for a low frequency of administration and the administration method also play an important role as preference attributes for BAs in Spain. PSY88 Preferences of Spanish Patients Over the Attributes of Biological Agents for the Treatment of Rheumatic Diseases Depending On the Administration Route Rodriguez M 1, Ibero I 2, Martin E 3, Nolla J M 4, Raya E 5, Aragon B 6, Nocea G 6, Lizán L 7, Aceituno S 7 1Complejo Hospitalario de Ourense, Ourense, Spain, 2Hospital General de Alicante, Alicante, Spain, 3Hospital La Paz, Madrid, Spain, 4Hospital Universitario de Bellvitge, Barcelona, Spain, 5Hospital Universitario San Cecilio, Granada, Spain, 6MSD, Madrid, Spain, 7Outcomes’10, Castellon, Spain . . . . . . . . . . Objectives: Biological agents (BA) to treat rheumatic diseases are commonly administered by either the subcutaneous or the intravenous route. The objective of this study was to assess rheumatoid arthritis (RA), ankylosing spondylitis (AS) and psoriatic arthritis (PA) Spanish patients’ preferences over BA considering the administration route. Methods: Observational, cross-sectional design. Participants were RA, AS and PA patients (diagnosed ≥ 2 years prior to study entry; currently or previously (≤ 1 year ago) receiving BA for a minimum of 1 year. Conjoint analysis was performed to define preferences over 8 scenarios combining 4 attributes [‘Administration method’ (ADMINISTRATION), ‘Pain relief and improvement of the functional capacity’ (RELIEF), ‘Risk of adverse events’ (AE) and ‘Time until perceiving the need for a new dose’ related to the frequency of administration (TIME)]. Relative importance (RI) was calculated for patients on subcutaneous or intravenous administration, respectively. Results: A total of 488 patients [male= 50.9%; mean (SD) age= 50.6 (12.06) years; RA= 33.8%, AS= 32.4%, PA= 33.8%; mean time from diagnosis= 12.6 (8.2) years; receiving currently BA= 98.2%] were included. The patients currently receiving subcutaneous (n= 305) or intravenous (n= 174) administration with BA gave highest importance to ‘RELIEF’ (45.4% and 46.7%) and ‘AE’ (28.2% and 31.8%), followed by ‘ADMINISTRATION’ (19.7% and 10.2%) and ‘TIME’ (6.7% and 11.3%, respectively). Both groups of patients preferred to stay on the same route of administration, either subcutaneous or intravenous, they had been on. Moreover all patients considered most crucial a longer time until perceiving the need for a new dose (8 over 4 over 2 over 1 week). Conclusions: Spanish patients with rheumatic diseases placed high importance on pain relief and risk of AEs as preference attributes for BAs. The frequency of administration (time until perceiving the need for a new dose) also plays a crucial role as all patients indicated their preference for lower vs. higher frequencies of BA administration. PSY89 Preference for Rituximab Subcutaneous (Sc) and Intravenous (Iv) Among Patients With Cd20+ Non-Hodgkin’s Lymphoma (Nhl) Completing the Rasq Measure In Randomized Phase Iii Studies Prefmab and Mabcute Rule S 1, Briones J 2, Smith R 3, Theodore Oklota C 4, Ngoh C A 3, Osborne S 3, Capochiani E 5, Rummel M 6 1Derriford Hospital, Plymouth, UK, 2Santa Creu i Sant Pau Hospital, Barcelona, Spain, 3F. Hoffmann-La Roche Ltd, Basel, Switzerland, 4Genentech, South San Francisco, CA, USA, . . . . . . . . . 5Centro Aziendale di Ematologia, Livorno, Italy, 6University Hospital Giessen and Marburg, Giessen, Germany Objectives: Rituximab SC reduces administration times (~5 minutes) compared with the IV route (~4 hours). We examined the extent of patient preference for rituximab SC versus IV in the PrefMab and MabCute studies using the Rituximab Administration Satisfaction Questionnaire (RASQ). Methods: In PrefMab (NCT01724021) patients with untreated CD20+ DLBCL/FL received 1 cycle of IV rituximab (375mg/m2) followed by either SC rituximab (1400mg, x3) then IV rituximab (x4), or IV rituximab (x3) then SC rituximab (x4), with chemotherapy. In MabCute (NCT01461928) patients with relapsed/refractory CD20+ indolent NHL received induction rituximab IV (375mg/m2; 1 cycle) then rituximab SC (1400mg; cycles 2–8) plus 6–8 chemotherapy cycles. RASQ evaluated preference by assessing patients’ perceptions of the impact of administration route and treatment satisfaction. Conceptual validation of RASQ has been conducted and psychometric data will be reported. Results: Median RASQ scores for PrefMab were: convenience IV (n= 211) 58.3 (interquartile range: 41.7–75.0) and SC (n= 207) 83.3 (75.0–91.7); satisfaction: IV (n= 211) 75.0 (62.5–87.5), SC (n= 208) 87.5 (75.0–100.0); impact on daily life: IV (n= 145) 50.0 (41.7–83.3), SC (n= 163) 83.3 (83.3–100.0); physical impact IV (n= 211) 83.3 (66.7–100.0), SC (n= 208) 83.3 (75.0–100.0); psychological impact IV (n= 211) 80.0 (65.0–90.0), SC (n= 208) 88.8 (75.0–95.0). SC administration was preferred by 80.3% and 85.9% of patients with IV or SC as most recent dose, respectively. Results were similar irrespective of treatment sequence. Median RASQ scores for MabCute (n= 92) were: convenience: IV 58.3 (33.3–66.7), SC 83.3 (66.7–83.3); satisfaction: IV 62.5 (50.0–87.5), SC 87.5 (75.0–100.0); impact on daily life: IV 58.3 (41.7–66.7), SC 83.3 (66.7–83.3); physical impact IV 75.0 (66.7–91.7), SC 83.3 (66.7–91.7); psychological impact IV 70.0 (60.0–85.0), SC 85.0 (75.0–95.0). Conclusions: Consistent patient satisfaction and preference for SC versus IV rituximab was demonstrated in the PrefMab and MabCute studies. RASQ is a reliable and valid measure of patient treatment preference. PSY90 A Systematic Literature Review of the Humanistic Burden of Multiple Myeloma Rizzo M 1, Xu Y 2, Panjabi S 3, Iheanacho I 1 . . . . 1Evidera, London, England, 2Evidera, Lexington, MA, USA, 3Onyx Pharmaceuticals, Inc., an Amgen subsidiary, South San Francisco, CA, USA Objectives: We conducted a systematic literature review to identify published evidence from observational studies on the humanistic burden of multiple myeloma (MM). Methods: We searched MEDLINE, Embase, EconLit, and the Cochrane Library for English-language articles and analysed these qualitatively. Results: The review identified 20 publications based on 18 observational studies; these were mainly cross-sectional in design (n= 14). Most studies (n= 15) examined populations with MM in general, typically without stating the proportion with relapsed (R) or relapsed and refractory (RR) MM. Fewer studies examined patients with newly-diagnosed MM (n= 2) or RMM/RRMM (n= 1). Health-related quality of life (HRQoL) was typically assessed using validated cancer-specific instruments (e. g., the EORTC-QLQ-C30: n= 9; and the MM-specific EORTC-QLQ-MY20: n= 4). On average, patients with MM had poorer HRQoL compared to the general population (n= 6), and compared to patients with certain other hematologic cancers (n= 2). MM patients whose disease duration ranged from diagnosis to 11 years had greater physical function impairment (p< 0.001) than other hematologic cancer patients. Other complaints included fatigue, bone pain, tingling, and non-specific pain. Patients on active treatment had worse side effects than those in a first treatment-free interval (p< 0.001); the latter had better scores on HRQoL dimensions including role and social function, future perspectives, and body image (all p< 0.05). Between baseline and one year, patients experienced worsening on the EORTC-QLQ-C30 global health scale (p< 0.001) and in fatigue, nausea/vomiting, and pain scores (all p< 0.05). Symptomatic patients had lower physical functioning scores (p< 0.05) than asymptomatic patients; those with severe symptoms had lower EORTC-QLQ-C30 global health scores (p< 0.05; mild/moderate symptoms: p= NS). Fatigue, bone pain, and anaemia were associated with lower. Conclusions: HRQoL in patients with MM deteriorates with disease duration, symptom severity, disease progression, or development of complications. This evidence suggests substantial unmet needs in MM patients. PSY91 Patient-Reported Outcomes In Moderate To Severe Hemophilia Patients: Finding From A Cross-Sectionalstudy In Korea Lee K S 1, Cha J H 2 1Kyungpook National University School of Medicine, Daegu, South Korea, 2Pfizer Pharmaceuticals Korea Ltd., Seoul, South Korea . . . . Objectives: There are approximately 2,000 hemophilia patients in Korea, but patient-reported outcome (PRO) studies involving a large number of hemophilia patients have been rarely studied. The aim of this study was to assess PROs in moderate to severe hemophilia patients in Korea. Methods: It was a crosssectional, multi-centered and observational study. Moderate to severe, male hemophilia patients aged 8 to 65 were recruited at 2 of Korea Hemophilia Foundations and 3 other pediatrics from November 2012 to September 2013. All patients completed self-reported questionnaires to measure patients’ characteristics and PROs including health-related quality of life (HRQoL) and productivity loss. HRQoL was examined using EQ-5D, ranged 0-1, which higher values indicate better HRQoL and Heamo-QoL, ranged 0-100, where higher values imply lower HRQoL. Productivity loss was estimated with absenteeism and presentism in terms of lost productivity time (LPT). Results: For a total of 605 patients (mean age, 29.3 years; 88.6% with severe hemophilia) enrolled in this study, the mean scores of heamo-QoL and EQ-5D were 32.28 and 0.68 respectively. The mean scores of EQ-5D in this study are comparable to 0.68 in rheumatoid arthritis patients from Korea Observational Study Network for Arthritis. Significantly lower EQ-5D was found in patients with the following clinical factors compared to those with reverse conditions; joint bleedings (0.68 vs. 0.73, p= .001), A538 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 hemophilic joint health (0.67vs. 0.73, p< .001) or disability (0.65 vs. 0.70, p< .001) as similar as in the results of Haemo-QoL. With patients who were either on a job or students (n= 467, 77.2%), LPT was estimated at 127.81 hours per month on average. Of 467, patients with inhibitor or disability showed higher LPT compared to those without inhibitor (130.61 vs. 126.61 hours per month, p= 0.486) or disability. (132.27 vs. 124.95 hours per month p=.087). Conclusions: The study findings suggest that patients’ clinical characteristics should take into account for the management of hemophilia given patient-reported outcomes differed by clinical manifestations. PSY92 Burden of Lupus Nephritis (Ln) Among Patients Managed In Routine Clinical Practices In Europe (Eu) Narayanan S 1, Hutchings R 2, Lu Y 2 1Ipsos Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, London, UK . . . Objectives: To assess the burden of LN in comparison to SLE patients without Nephritis (Non-LN) in routine clinical practices in EU. Methods: A multi-center medical chart-review of adult (16-89yrs) SLE patients was conducted among rheumatologists and internal medicine physicians in UK/France/Germany/Italy/Spain (5EU). Physicians were recruited from a geographically representative sample in each country. Approx. 5 consecutive eligible persistent active or relapse remitting SLE patients currently managed as part of usual care were identified. Physicians abstracted de-identified patient data on disease characteristics, lab values and treatment patterns. LN and Non-LN cohorts were compared using descriptive statistics. Results: 168 LN patients and 569 non-LN patients with SLE were analyzed. Patient characteristics included (LN/Non-LN): age (yrs): 40.2/42.7; % Female: 82.7/79.4; % Caucasian: 82.7%/88.8%; % full-time employment: 32.1/36.0; % part-time employment: 19.0%/22.0%. Among LN/Non-LN cohorts, frequency of clinic visits was: 9.5wks/10.6wks, % currently receiving care in in-patient setting was: 14.9/8.8, % hospitalized >=1 in the past-year was: 49.4/29.7; mean # of organ manifestations was: 3.9/3.0. Top-5 organ manifestations were (% LN/Non-LN): musculoskeletal: 85.1/90.8, mucocutaneous: 85.1/84.8, haematologic: 51.8/52.2, renal: 100.0/5.9, pulmonary: 16.7/16.2; % experiencing a flare was (LN/Non-LN): 23.2/16.6. Renal biopsy was performed in 87.5% (LN) and 4.0% (Non-LN) of patients. Steroids were used by 82.1% (LN) and 69.4% (Non-LN). % patients with low C3 and C4 was LN: 66.3/48.1/Non-LN: 60.2/50.0; mean ESR scores were 40.2 (LN) and 36.8 (Non-LN). Among patients with available data, SELENA-SLEDAI scores were 11.5 (LN) and 8.5 (Non-LN). Humanistic burden (reported via physician ratings, on a scale of 0 (most impact) to 7 (least impact)) was (LN/Non-LN, mean scores): ability to perform every-day tasks: 5.0/5.3, ability to interact fully with family and friends: 5.4/5.6, and ability to work/keep employment: 4.5/5.0. Conclusions: LN cohorts had higher clinical and humanistic burden in 5EU in comparison to their non-LN SLE counterparts. Factors influencing the observed burden, including the therapeutic strategies used in these geographies warrant further investigation to manage SLE, and LN in particular, optimally. PSY93 Quality of Life (Qol) With Psoriasis: Ethnography Study Evaluating the Impact of Psoriasis On Moderate To Severe Patients In Europe (Eu), From A Patient’s Perspective Narayanan S 1, Franceschetti A 2 Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, London, UK . . 1Ipsos Objectives: To qualitatively assess the impact of Psoriasis on patient QoL. Methods: An ethnographic study with moderate/severe Psoriasis patients was conducted in 4EU (UK/France/Spain/Italy) and the US to explore patients’ views on treatment options and Psoriasis impact on QoL. Anthropologists and ethnographers spent several-hours with consented patients and filmed their behaviours inside and outside their homes in everyday situations. 175 hours of recordings/notes were analysed to identify QoL-related themes: self-image, psychological effects of psoriasis (e. g., anxiety/depression), life-style changes due to Psoriasis, relationship with family, friends and colleagues. Results: Study included 35 adult patients (4EU: 20, US: 15; mean age: 39 yrs; female: 65%). Patients described their appearance with a sense of disgust and self-loathing. Feelings of frustration were often expressed due to a perceived lack of control of their lives. Prior to biologic initiation, daily rituals absorbed good part of their day, including waking up earlier to apply creams, going to work earlier to check their appearance and cover patches of dead skin. Due to lack of cultural discourse and patient’s difficulty in articulating Psoriasis’ impact, partners and family did not know how to react nor did they realize the full extent of the problem. Difficulty in getting appropriate psychological support needed left them with a feeling of resignation. As a result, majority dealt with their issues in isolation. Most patients experienced social discriminations due to Psoriasis which led some to lie about their disease, keeping it a secret. Biologic experienced patients noticed a significant improvement physically, but psychological scarring remained. Despite their regained confidence owing to the effectiveness of biologics, they still did not discuss their condition with family and friends. Conclusions: Patients with Psoriasis experienced significantly lower quality of life and high psychological scarring. Ethnographic study design vividly depicted the unarticulated and emotional impact of Psoriasis on patients’ everyday lives. PSY94 Health Related Quality of Life of Patients and Their Caregivers In Rare Diseases Results of the Burqol-Rd Project In Hungary Péntek M 1, Baji P 1, Pogány G 2, Brodszky V 1, Boncz I 3, Gulácsi L 1 University of Budapest, Budapest, Hungary, 2Hungarian Federation of People with Rare and Congenital Diseases, Budapest, Hungary, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . 1Corvinus Objectives: The Social Economic Burden and Health-Related Quality of Life in Patients with Rare Diseases in Europe (BURQOL-RD, http://www.burqol-rd.com/) project aimed to investigate disease burden and self-percieved health outcomes of patients and their caregivers in rare diseases in eight EU countries (Bulgaria, France, Germany, Hungary, Italy, Sweden, Spain, UK). Methods: An online questionnaire survey was developed for patients (adults/children) and caregivers. Patients were recruited by patient organisations in cystic fibrosis (CF), Prader–Willi syndrome (PWS), haemophilia (HEMO), Duchenne muscular dystrophy (DMD), epidermolysis bullosa (EB), fragile X syndrome (FXS), scleroderma (SCL), mucopolysaccharidosis (MPS), juvenile idiopathic arthritis (JIA) and histiocytosis (HIS). Demography amd main characteristics were recorded. Patients’ and caregivers’ health state was assessed by the EQ-5D-5L, disability and caregivers’ burden by the Barthel Index and Zarit Burden Interview questionnaires. Results: In Hungary, 296 Hungarian patients (children: 161, 54%) participated in the study (CF 110, PWS 5, HEMO 58, DMD 57, EB 6, FXS 12, SCL 38 and MPS 10 patients), no data were obtained in JIA and HIS. Mean age among adults/children was 37.0 (SD 16.1) / 9.6 (SD= 4.5) years, and disease duration was 18.5 (SD= 14.4) / 7.1 (SD= 4.5) years, respectively. The lowest average EQ-5D-5L score was found in MPS (0.134 / 0.070) and DMD (0.310 / 0.198) just alike with the Barthel Index (MPS: 35.6; DMD 58.0). Caregivers’ (N=95) mean age was 39.2 (SD= 8.0) years and their EQ-5D-5L score (0.797, SD= 0.251) was not significantly different from the general populations’ average. Satisfaction with health care (1-10 Likert scale) was the lowest in MPS, DMD and EB (mean 4.1, 5.2 and 4.3). Conclusions: Rare diseases induce substantial deterioration of patients’ quality of life and impose burden on caregivers. Taking an integrated approach our results can serve for international comparisons and facilitate further investigations in other orphan diseases. PSY95 Is the Disease-Specific Lupusqol Sensitive To Changes of Disease Activity In Systemic Lupus Erythematosus Patients After Treatment of A Flare? McElhone K 1, Burnell J 2, Sutton C 2, Abbott J 3, Lanyon P 4, Rahman A 5, Yee C S 6, Akil M 7, Ahmad Y 8, Bruce I N 9, Gordon C 10, Teh L S 1 Blackburn Hospital, Blackburn, UK, 2UCLAN, Preston, UK, 3University of Central Lancashire, Preston, UK, 4Queen’s Medical Centre, Nottingham, UK, 5University College London, London, UK, 6Doncaster Royal Infirmary, Doncaster, UK, 7Royal Hallamshire Hospital, Sheffield, UK, 8Betsi Cadwaldr University Health Board, Llandudno, UK, 9University of Manchester, Manchester, UK, 10University of Birmingham, Birmingham, UK . . . . . . . . . . . . . . . 1Royal Objectives: The aim of the UK multi-centre LupusQoL Sensitivity Study is to assess whether the LupusQoL, a systemic lupus erythematosus (SLE) specific, health-related quality of life measure is sensitive to change when disease activity improves or deteriorates. Methods: Patients with SLE experiencing a flare (baseline) & requiring an increase in treatment were recruited. Assessments were undertaken at baseline & monthly for 9 months & included BILAG-2004 disease activity index & the LupusQoL with 8 domains and scores ranging from 0 (worst) to 100 (best HRQoL). LupusQoL domain scores when disease activity improved or deteriorated between consecutive time-points are reported as mean changes, with 95% CI constructed using robust standard errors to account for repeated patient assessments. Results: Mean (SD) age was 40.9 (12.8) & duration since diagnosis was 9.3 (8.1) years for the 101 patients recruited; 94% females, 62.6% white Caucasians. At baseline all mean LupusQoL domain scores were < 52. Scores for LupusQoL physical health (+4.0, 95%CI 1.9 to 6.1), pain (+7.7, 95%CI 4.8 to 10.5) & fatigue (+4.1, 95%CI 1.7 to 6.5) increased when BILAG improved. Scores for physical health (-4.9, 95% -9.4 to -0.4) and pain (-6.9, 95%CI -12.9 to -0.8) decreased with a major BILAG deterioration but changes with a minor deterioration were small and non-significant. The effects of improvements & deterioration in BILAG on the other LupusQoL domain scores were smaller. Conclusions: Improvement and deterioration of LupusQoL domain scores for physical health, pain & fatigue domain scores was seen in patients with significant changes in disease activity over 1 month. Sensitivity to change of other LupusQoL domains in relation to changes in disease activity may need to be evaluated over a longer interval as the more emotive type of response to the disease & its consequences may be latent and therefore not evident at monthly intervals. PSY96 The Quality of Life of Patients Treated With Robotic Versus Traditional Surgery Results From An Italian Observational Multicenter Study Turchetti G , Pierotti F , Palla I , Manetti S , Cuschieri A Scuola Superiore Sant’Anna, Pisa, Italy . . . . . Objectives: To assess the Quality of Life (QoL), focusing on level of pain, of patients submitted to surgical interventions with robotic technique in comparison with traditional approach in the Italian setting. Methods: The prospective multicentre study analysed the QoL of 699 patients submitted to surgical interventions, enrolled in 8 Italian Hospitals for the period February 2011-May 2014. The specialties were general, thoracic, gynecological surgery performed with open, manual laparoscopic or robotic technique. Patients completed two questionnaires: one related to pain intensity during the hospitalization and one fulfilled at home related to pain intensity and its impact on daily activities, mood, relationship, sleep. Pain was measured using the scale of facial expressions corresponding to the Visual Analog Scale. For each intervention and specialty, linear regression for repeated measure, corrected by length of stay and use of analgesics, were performed to explain level of pain during hospitalization and at home, respectively. Results: Level of pain during hospitalization is significantly (p= 0.05) lower in general and gynecological robotic surgery versus the open technique, but not versus laparoscopic, while robotic thoracic surgery presents significant differences with laparoscopic but not with open surgery (p= 0.059). Level of pain at home is significantly different for robotic interventions versus open both in general and gynecological surgery; moreover, with respect to laparoscopic interventions, the differences are significant both in gynecological and in thoracic but not in general surgery. Conclusions: The study gives us insightful knowledge about QoL, focusing on pain, of patients submitted to surgical interventions with robotic technique vs traditional surgery. Further analysis are in progress to combine results on QoL, not only referred to level of pain, with clinical severity and other indicators of clinical efficacy as length of stay, operating A539 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 time, medical/surgical complications to refine a robust measure of effectiveness useful to perform cost effectiveness analysis. PSY97 Characteristics of Patients With Systemic Lupus Erythematosus (Sle) Currently On Remission, With Active Disease But Not Experiencing Flare, and Those Experiencing Flares In Clinical Practices In Europe Narayanan S 1, Hutchings R 2, Lu Y 2 1Ipsos Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, London, UK . . . Objectives: To assess the characteristics of SLE patients currently on Remission (Group-1), currently with active disease but not experiencing a flare (Group-2) and those who are experiencing flares (Group-3) in Europe (EU). Methods: A multicenter medical chart review of adult (16-89 yrs) SLE patients was conducted in 1Q2014 among rheumatologists/internal medicine physicians in UK/France/ Germany/Italy/Spain (5EU). Physicians were recruited from a geographically representative sample in each country. Approx. 5 consecutive eligible persistent active or relapse remitting SLE patients currently managed as part of usual care were identified. Physicians abstracted de-identified patient data on disease characteristics, lab values and treatment patterns. Patient disease status and humanistic burden was assessed by physician per clinical judgment & patient interaction. Results: 747 SLE patient charts were abstracted in 5EU (UK: 156/France: 149/Germany: 148/Italy: 146/ Spain: 148); Group-1: 25.7% (range: 19.9% (Italy) –29.7% (Germany)), Group-2: 56.2% (range: 42.3% (UK) –66.4% (Italy)), Group-3: 18.1% (range: 13.5 (Germany) –28.8 (UK)). Patient characteristics included (Group-1/Group-2/Group-3): age (yrs): 41.4/43/40.6; % female: 81.8/80.2/77.8; % full-time employment: 38.0/34.5/32.6; % part-time employment: 25.5/20.5/17.8; % on sick leave: 3.6/7.6/17.0; % currently receiving treatment in in-patient setting: 3.1/7.6/28.1; % hospitalized > = 1 in past-year: 23.4/32.1/55.6. Top-5 organ manifestations were (% Group-1/Group-2/Group-3): musculoskeletal: 90.6/89.5/88.1, mucocutaneous: 86.5/84.0/85.2, haematologic: 45.8/53.8/55.6, renal: 23.4/26.7/33.3, pulmonary: 15.1/13.6/26.7. In Group-1/Group-2/Group-3, % patients with low C3 and C4 were 33.3/53.9/73.4 and 31.6/53.2/77.5 and % anti-ds-DNA positive were 54.7/71.0/78.5. Humanistic burden (reported via physician ratings, on a scale of 0 (most impact) to 7 (least impact)) was (Group-1/Group-2/Group-3, mean scores): patient ability to perform every-day tasks: 6.11/5.17/4.37, patient ability to interact fully with family and friends: 6.20/5.45/4.90, and patient ability to work/ keep employment: 5.76/4.80/3.87. Conclusions: Over half of the SLE patients had an active disease while one-in-five were experiencing a flare in this study cohort, with significant variations observed within 5EU. Clinical and humanistic burden varied based on patient disease status, with highest burden observed among those experiencing flares. of 16 criteria. Methods: 85 experts were interviewed to estimate the importance of each criterion in the decision-making on financing MT for rare diseases. We used 10-point scale, where 10 points mean major importance to the priority indicator, and 1point means minor importance. Mean estimates were calculated using descriptive statistics, then means were normalized. Results: Respondents were 41 years on average (ranging from 23 to 64 years), and included 20 public servants, 16 health administrators, 32 practitioners, and 14 researchers. 44 respondents had a scientific degree. The most important criteria were characteristics of treatment “Effect of treatment on quality of life” and “Effect of treatment on life expectancy” with 1 points each. The least important criteria were both characteristics of the disease - “Cognitive disorders as manifestations of the disease and “Additional burden on the daily lives of care-givers” with 0.28 and 0.1 respectively. Conclusions: Characteristics of treatment turned out to be more important for respondents than characteristics of disease, therefore characteristics of treatment should be given consideration when evaluating rare diseases to determine priority financing. PSY100 Patterns and Trends In Opioid Use In Iran From 2007 To 2011 Taheri F Minitry of Health, tehran, Iran . Objectives: Opioid analgesics are proven to be safe and effective in malignant or nonmalignant pains. The consumption trend of six opioids (Morphine, Codeine, Oxycodone, Fentanyl, Pethidine, and Methadone) in Islamic Republic of Iran assumed as an indicator for prescription pattern and is evaluated during five year period (2007-2011) as the aim of this study. Methods: The data of opioid analgesic consumption were collected from FDO (Food and Drug Organization) of Iran. The collected data were converted to DDD (Define Daily Dose) for each of six selected opioids in order to be compared accurately. Results: Overall consumption of opioids was grown during the period of investigation. Putting six selected opioids into two groups – synthetic and non-synthetic -, the growth rate of synthetic ones are obviously higher than those of non-synthetics. Opioid analgesics consumption in 2011 was shown to be 4 times more than the opioid consumption in 2007. The CAGR (Compound Average Growth Rate) between 2007-2011 were reported 111.27%, 33.11%, 16.48%, 3.91%, 3.76%, and -41.63% for Oxycodone, Methadone, Fentanyl, Morphine, Pethidine and Codeine respectively. The growth rate of mentioned above opioids for the last year of investigation was reported 41.12%, 16.54%, 29.99%, -0.38%, 7.66%, 23.67% respectively. Conclusions: Like other low consuming countries, consumption of opioid analgesics in Iran is limited by means of different kinds of barriers. There are also parameters which set a specific orientation in opioid consumption. Analyzing this trend, results in defining the barriers and other parameters clearly. PSY101 Orphan Drug Approvals In Europe: Historical Review and Trends SYSTEMIC DISORDERS/CONDITIONS – Health Care Use & Policy Studies Rodrigues J 1, Korchagina D 1, Rémuzat C 1, Brunet J 2, Tavella F 3 Publique des Hôpitaux de Marseille, Marseille, France, 3Creativ-Ceutical, London, UK . . . . . 1Creativ-Ceutical, Paris, France, 2Assistance PSY98 Orphan Drug Policy: Approaches To Market Access In Multiple Countries Shih D Y , Jarrett J MAPI, London, UK . . . Objectives: Despite increasing policy guidance to encourage the development of and access to orphan drugs, the policy landscape indicates a lack of transparency and consistency across countries. The objective of this study is to compare the orphan drug policies of Japan, South Korea, and Taiwan to the policies in the EU. Methods: A targeted literature review was conducted to identify papers pertaining to orphan drugs in Japan, South Korea and Taiwan, with a focus on quantitative analysis for policy-making related to pricing and reimbursement from the payer perspective. No limits were placed on language. Country specific policy websites were hand-searched. Results: A total of 3465 abstracts were identified for screening. Of those, 104 were eligible for full-text screening. The definition of prevalence with regard to rare diseases in Japan, South Korea and Taiwan was stricter than in the EU. All of them had introduced regulations, guidelines and incentives to the development of orphan-designated drugs. Strict HTA requirements were waived for rare diseases, although cost-effectiveness data are referenced in South Korea. In Japan and Taiwan, prices were reduced for existing drugs with new orphandesignated indications. Within the existing benefits of national health insurance schemes or under the coverage of rare disease regulations, full reimbursement was given, which is different from the partial reimbursements observed in certain European countries. In some cases, the application of drug treatment can be approved prior to or without market authorisation. The availability of orphan drugs to patients was not inferior to that of the EU. Conclusions: While access in Japan, South Korea and Taiwan is seemingly no different to the EU, there are no consistent or transparent policies in place in these countries. Policy makers should prepare innovative schemes that offer warranties to both payers and patients and establish a systematic evaluation procedure for manufacturers. PSY99 Mcda Approach To Ranking Rare Diseases In Russia: Preliminary Results Fedyaeva V K 1, Omelyanovsky V V 1, Rebrova O 2, Khan N 1, Petrovskaya E V 3 1The Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia, 2Pirogov Russian National Research Medical University, Moscow, Russia, 3Samara State Medical University, Samara, Russia . . . . . . . . Objectives: The ranking and prioritization of rare diseases are crucial in order to define for which of them state support measures are justifiable. For this purpose the set of 16 criteria to assess rare diseases (8 related to the characteristics of disease, and 8 to the characteristics of treatment) was proposed by experts, so multi-criteria decision (MCDA) analysis approach could be useful. The aim of the study was to assess the reliability of the criteria set and to determine the relative importance Objectives: In Europe, orphan designation has been granted by European Medicines Agency since 2000. Molecules with orphan designation can benefit from a number of incentives to guarantee return on investment for manufacturers. Since introduction of orphan legislation, the number of Orphan Drugs (OD) has significantly increased. In 2012, total OD sales reached 13% of the whole pharmaceutical market. This study aims to analyse current situation and trends in OD approvals. Methods: All ODs approved gaining marketing authorization was identified through EMA website. Extensions to a new disease were considered as an independent approval, while extensions of indication for the same condition were not considered. Information on drug status, indications, therapeutic area, and authorisation date was extracted. Results: 77 orphan molecules were identified of which two have been withdrawn, and one orphan designation has expired. Yearly approvals continuously grow (from 2 ODs approved in 2002 to 12 and 8 drugs in 2012 and 2013 respectively). By mid-2014, already 6 were approved. There is a high imbalance within therapeutic area with 36% for oncology, 27% for endocrine, nutritional and metabolic disorders whereas the remaining 37% cover 11 therapeutic classes. No product has been approved for 7 therapeutic areas despite existence of rare disorders. Moreover, oncology is associated with the highest growth. Conclusions: Orphan drug market is rapidly growing but targeting selective therapeutic areas with high potential return on investment. The incentive policies happen to well function but discriminate among diseases. A review of orphan drug policies might be warranted to ensure patient equity in development of new OD. PSY102 Comparative Analysis of Hta Decisions, Price and Reimbursement Level of Orphan Drugs In France and Italy Korchagina D 1, Tavella F 2, Rémuzat C 1, Kornfeld A 3, Toumi M 4 1Creativ-Ceutical, Paris, France, 2Creativ-Ceutical, London, UK, 3Creativ Ceutical, Paris, France, 4University of Marseille, Marseille, France . . . . . Objectives: While there exist a number of incentives to stimulate research and development of orphan drugs (OD), the Health Technology Assessment (HTA) agencies do not offer specific path for OD in most countries where it is left to payer value judgement. This leads to a high inequity in patient access to OD. The study aims at comparing the HTA decisions, price and reimbursement level of OD in France and Italy. Methods: All OD assessed since 2000 were analysed. Prices, reimbursement rates and decision details were extracted for each drug using Farmadati Italia database for Italy and Transparency committee reports, AMELI’s national health insurance and Thériaque databases for France. Results: Among 74 OD approved in Europe 66 are available in Italy compared to 53 in France. All ODs available in France are officially available in Italy. The average delay between the market authorization and the price and reimbursement decision was about 16 and 17 months in France and Italy, respectively. In France all available drugs are 100% reimbursed through hospital, 36 molecules are available in retail pharmacy with reimbursement from A540 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 15% to 100%. In Italy only 9 molecules are reimbursed in retail pharmacy and other 10 are not reimbursed at all. In 63% cases ex-factory prices in Italy are higher than in France. The average price difference in price is 12 % with 50% of products sharing almost same price (less than 5% difference). No information is available in France on managed entry agreement, while it is publicly available in Italy. This prevents fair price comparison. Conclusions: OD are more available in Italy, but reimbursement is poorer than in France. Prices are slightly higher in Italy but France displays multiple confidential rebates making it impossible to compare net prices. In Italy the actual accessibility depends a lot on regional level unlike France. Orphan Drugs, only the Eculizumabe for the treatment of PNH has no approved registration by the National Health Surveillance Agency (ANVISA). Conclusions: Considering its high cost, high judicial demand and limited availability of scientific evidence, orphan drugs represent a challenge for researchers and decision makers. Clinical benefit, disease severity, availability of therapeutic alternatives, ethical, political and social aspects should be considered. It is necessary to start a multidisciplinary reflection on the development of HTA models and policies regarding rare diseases and innovative treatments in the SUS, as well as fostering the primary researches in this field. PSY103 Health Technology Assessment, Price and Reimbursement Review for Orphan Drugs In Italy PSY106 To What Extent Do Disease and Treatment Characteristics Influence Hta-Based Recommendations for A Sample of Orphan Drugs In Three Countries, and Could These Indicate Whether Orphan Drugs Have A “Special Status”? Tavella F 1, Korchagina D 2, Rodrigues J 2, Rémuzat C 2 . . . . 1Creativ-Ceutical, London, UK, 2Creativ-Ceutical, Paris, France Objectives: In Italy drug Health Technology Assessment (HTA) is conducted by the Scientific Technical Commission of Italian Medicines Agency (AIFA) with further negotiation between the manufacturers and the AIFA’s Pricing & Reimbursement Committee on price and reimbursement. After the decision is taken it is published in the official journal (Gazetta officale), the assessed drug is formally available for Italian patients next day. There does not exist a specific procedure for orphan drugs (OD), they are evaluated under the same conditions as drugs for common diseases. Pharmaco-economic studies are recommended for innovative drugs. The objective of the study is to review HTA decisions, prices and reimbursement of OD in Italy. Methods: All OD assessed in Italy since 2000 were identified. Prices, reimbursement rates and decision details were extracted for each drug using Farmadati Italia database. Results: Among 74 OD approved in Europe 66 molecules are officially available in Italy. It took 5-10 months from granting market authorization to final decision on pricing and reimbursement and publication in ‘Gazetta officale’. The mean time was about 17 months shorter than for common drugs. Only 9 of available molecules are fully reimbursed in Italy, reimbursement of 30 drugs is restricted to hospital use, 10 are not reimbursed and 5 are in a special class waiting a decision on reimbursement rate. For other drugs information is not available. Postmarketing surveillance studies were requested for a half of indications, AIFA registries were reported in 25% of cases. Annual treatment ex-factory price of OD varies from € 2 500 to almost € 1 000 000. Conclusions: Almost 90% of approved OD are available in Italy. However, the actual situation often differs a lot from the official data as some drugs are unavailable from retail or hospital pharmacies. In addition, regional authorities contribute to inequity in access especially for “expensive drugs”. PSY104 Health Technology Assessment, Price and Reimbursement Review for Orphan Drugs In France Korchagina D 1, Rémuzat C 1, Rodrigues J 1, Kornfeld A 2, Toumi M 3 1Creativ-Ceutical, Paris, France, 2Creativ Ceutical, Paris, France, 3University of Marseille, Marseille, France . . . . . Objectives: In France Orphan Drugs (OD) undergo the same Health Technology Assessment (HTA) procedure as other drugs. The evaluation is performed by the Transparency Committee (TC). Two scores are assigned and further used for pricing & reimbursement decision: drug’s medical benefit (SMR) and improvement in medical benefit (ASMR). OD can be eligible to an accelerated procedure established for innovative products. The study aim is to analyse HTA decisions, prices and reimbursement of OD in France. Methods: Exhaustive prices, reimbursement and HTA opinions were extracted using Transparency Committee Website, AMELI’s national health insurance and Thériaque databases. Results: Among 74 OD approved in Europe, 21 drugs are not available in France. 6 were recently assessed by TC but are not yet reimbursed, 14 were not assessed and one got a negative opinion. Reimbursement process took between 2 and 5 months after TC opinion. 88% of drugs were considered as bringing a substantial medical benefit. At the same time, more than a half of medicines were graded as providing a major (9%), significant (27%) or moderate (20%) improvement in actual benefit leading to opportunity for premium prices, 30% were rated minor improvement, and about 14% no improvement according to TC opinion. Annual treatment prices of OD varie from € 1,500 to almost € 1,000,000. Conclusions: About a third of approved ODs are not available in France. Most of them were authorized recently and are might become available after the HTA and pricing process. In comparison to non OD, the ASMR scoring is outstanding. Lack of alternative treatment and severity of the condition appear to be the drivers of high SMR-ASMR score. PSY105 Hta Studies On Orphan Drugs By Rebratsxmembers Souza K M 1, Gonçalves L 2 1Brazilian Ministry of Health, Brasília, Brazil, 2Brazilian Ministry of Health, Brasília-DF, Brazil . . . Objectives: In Brazil, the studies produced by members of the Brazilian Network for Health Technology Assessment (REBRATS) have contributed in a significant way in the process of management and incorporation of technologies in the Brazilian Public Health system (SUS). However, there is still a lack of a Pharmaceutical Assistance and HTA Policy for rare diseases and the evidence for orphan drugs are limited and lower. The coverage for these drugs, is frequently done through judicial orders, political and social pressure, with no support of evidence-based medicine. The objectives is to evaluate the production of HTA studies for orphan drugs made by REBRATS members. Methods: Query to the REBRATS database and the internal production of the HTA coordination, prioritizing 6 major diseases: Gaucher disease, Fabry disease, Mucopolysaccharidosis Type I (MPS I); Mucopolysaccharidosis Type II (MPS II); Mucopolysaccharidosis Type VI (MPS VI); Paroxysmal Nocturnal Haemoglobinuria (PNH). Results: Five HTA studies were found, which are: a Rapid Response for Gaucher disease; a study on Health Technology Management for Fabry disease; an Economic Evaluation for MPS I, II and VI; a Systematic Review for Mucopolysaccharidosis Type II; and one Rapid Review for PNH. For the evaluated Nicod E London School of Economics and Political Science, London, UK . Routine HTA methods may not adequately capture all the important considerations of a treatment’s value and the impact of the condition on the patient given that evidence is often incomplete. This study aims to explore the influence of broader considerations of scientific and social value judgments on reimbursement decisions for a sample of orphan drugs.Objectives: To identify and compare the extent to which these broader considerations not captured by the incremental cost-effectiveness ratio (ICER) influenced HTA decision-making process in three countries; and, on this basis, explore whether orphan drugs have a “special status”. Methods: Countries included were England, Scotland and Sweden. Ten drug-indication pairs with EMA orphan designation and all appraised by NICE were selected. Publicly available HTA reports were coded using thematic analysis to systematically identify and compare these broader considerations across countries using an existing analytical framework. Results: 108 different other considerations were identified and grouped into 15 clusters based on the information provided. The most common related to the nature of the disease, and considerations based on rarity or unmet need. 52% were one of the main reasons for the decision, and in some cases, were also a pivotal factor in accepting high and uncertain ICERs. Categorising these as social or scientific value judgments was done to identify areas where further elicitation of societal preferences, and where more consistency and transparency in their use are needed, respectively. Each of these was then compared to determine whether they pertained specifically to orphan drug or rare disease characteristics. Conclusions: Considerable variation was seen in the application of these broader considerations. Identifying these is a way forward to highlight areas where more research, or consistency and transparency are needed. Some of these other considerations may also favour orphan drugs, furthering the debate around whether orphan drugs deserve special status. PSY107 Why Are There Differences In Hta Recommendations Across Countries? A Systematic Comparison of Hta Decision Processes for A Sample of Orphan Drugs In Four Countries Nicod E London School of Economics and Political Science, London, UK . HTA reimbursement recommendations often result in different outcomes across countries despite the same evidence being appraised for a same technology. There is a need to understand the reasons for these differences. Objectives: To systematically compare HTA processes for a sample of orphan drugs across four countries (England, Scotland, Sweden, France): to identify the use and interpretation of the evidence appraised, and highlight differences across countries. Methods: Ten orphan drug-indication pairs were selected and systematically compared using a previously validated framework. An exploratory sequential mixed methods design divided the research into two stages: (1) qualitative in-depth analysis of the decision-making processes; and (2) quantitative identification of agency-specific risk preferences and agreement levels across countries. Results: Differences at each step of the decision-making process were identified. The same pivotal trials were appraised but with varying levels of detail in reporting the clinical outcomes, explaining some of the reasons for differing HTA recommendations. Agency-specific risk preferences were identified through correspondence analysis as drivers of these decisions, further explaining some of these differences. Poor to moderate agreement in the interpretation of the evidence was measured using Cohen’s kappa scores. This reflected situations where the countries interpreted the same evidence differently and situations where differences in the handling of the same uncertainties were seen, including differences in the extent to which stakeholder input influenced a decision. Conclusions: This research systematically compared HTA processes in different countries, facilitating the understanding of these complex processes including how different HTA bodies conduct value assessments. It enabled to raise awareness around the reasons for differences across countries, and highlight areas for potential methodological improvements in HTA. Further application of this framework to other disease areas and countries is a way forward to improving the drivers of coverage decisions while better understanding the settings and limitations of HTA. PSY108 Top 20 Orphan Drugs Availability, Pricing and Reimbursement In Slovakia: 2005-2012 Review Babela R 1, Uraz V 2, Babelova O 2, Slezakova Z 1 University, BRATISLAVA, Slovak Republic, 2St. Elizabeth University, Bratislava, Slovak Republic . . . . 1St. Elizabeth Objectives: Orphan drugs are highly priced and top 20 orphan drugs create almost 2,5% of total drug expenditure in Slovakia. We conducted 8 years review of government and literature sources to provide insight into pricing, reimbursement and availability situation surrounding top 20 orphan drugs in Slovakia from the health care payer perspective. Methods: We provide analysis of official prices, reim- A541 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 bursement status and availability of top 20 orphan drugs in Slovakia from 2005 till 2012. Data were obtained from government sources. Results: We considered orphan drugs list (Cote and Keating, 2012) that exceeded 1 billion $ sales in 2008 (globally) and compared molecules’ availability in Slovakia. Same molecules are among best selling 20 orphan drugs in Slovakia, with highest sale of 95 million EUR (Bevacizumab, 2005-2012) compared to lowest sale of 19 million EUR (Tacrolimus, 2005-2012). It took from 1 (Imatinib) to 19 years (Glatiramer acetate) to be launched in Slovakia after orphan designation. Top 20 orphan drugs had average DOT 472 EUR compared to total pharma market DOT average 0,43 EUR. From selected orphan drugs 60% had full (100%) reimbursement status and 40% were fully covered by hospital budgets. Only 4 of them were launched in Slovakia since 2005 (included), 16 of them were launched from 1990 till 2004. Prices ranged from 330 EUR to 5800 EUR (ex-factory, one package, 2012). Conclusions: There are highly valuable incentives for industry to invest in to development of orphan drugs in EU. Current context of economic constraints in EU however justifies the need to pay close attention to the rationale of maintaining such incentives in the context of potential return on investments of companies offering high priced drugs. Top 20 orphan drugs in Slovakia have prices high above average and also full reimbursement status. We expect more restrictive drug policy measures in this field. PSY109 Effect of Excluding Non-Patient Benefits As An Element On Acmg Newborn Screening (Nbs) Recommendations Mahida S , Algarni M , Rittenhouse B MCPHS University, Boston, MA, USA . . . Objectives: In 2006 the American College of Medical Genetics (ACMG) developed a stakeholder survey to make recommendations for 84 rare conditions to be considered for mandatory newborn screening (NBS). Scores of 19 different surveyed attributes for each condition were totaled. These scores determined an entry point to an algorithm (EPA) that determined final recommendations (Core conditions, Secondary Targets or Not Recommended). Among these attributes was a controversial one - non-patient benefits (NPB). Such attributes have historically not been considered in mandatory screening recommendations. This analysis examines how the exclusion of this attribute would affect recommendations. Methods: The ACMG report provided scores for individual attribute survey responses. We deleted the attribute score for NPB (0 to 100 points) and rescored the totals for each condition. We then assessed whether score changes were sufficient to alter the EPA and whether a different EPA would result in changes to the recommendations. Results: Six conditions had missing data. Of the 78 remaining, there were 15 conditions (19%) whose initial total scores were capable of changing by enough in this exercise (maximum 100 point reduction) to change the EPA. Of those, 10 (67%) did change EPA and, of those, 3 (30%) changed final category (in all cases from Core to ST). Of the initial 29 recommended Core conditions, there would be a 10 percent reduction to 26 and an increase in Secondary Targets from 25 to 28. Conclusions: Including screening benefits to non-patients (family or society) is controversial and has not been standard in the past. We have shown that in the ACMG recommendations, had no changes been made to the algorithm consequent to dropping non-patient benefits from consideration, 3 conditions would have changed from a Core recommendation for screening to only a Secondary Target. PSY110 Access To Orphan Drugs In Greece During Economic Crisis Tsiantou V 1, Mylona K 1, Karampli E 1, Boubouchairopoulou N 1, Kyriopoulos I I 1, Athanasakis K 1, Gabriel E 2, Makridaki D 3, Kyriopoulos J 1 1National School of Public Health, Athens, Greece, 2Institute of Pharmaceutical Research and Technology, Pallini, Greece, 3Panhellenic Association of Hospital Pharmacists, Marousi, Greece . . . . . . . . . . Objectives: Orphan Drugs (ODs) are medicinal products intended for diagnosis, prevention or treatment of rare diseases. Access to ODs is crucial for patients’ health and quality of life. The aim of this study was to identify current problems and future challenges of patients’ access to ODs in Greece. Methods: A qualitative study took place between December 2013 and January 2014. Data were retrieved through semistructured interviews with six representatives of key stakeholders in Greece and policy documents identified through web searches using keywords “orphan drugs” and “rare diseases” in Greek. Web-based documents and transcribed interviews were content analyzed. Results: Delays in pricing and reimbursement of ODs in the Greek pharmaceutical market, budget cuts in hospitals and absence of patient registries constitute according to the analysis the greatest barriers in patients’ access to ODs. There are two main channels through which the patient can have access to an OD and it depends whether it is licensed in Greece or not. In the first case the patient can take the drug through the hospital or the pharmacy of EOPYY if it is not available at the hospital pharmacy and in the second case through a public sector organization (IFET). All cases are characterized by extensive bureaucracy and involvement of up to three organizations in order to receive the approval, a procedure creating delays in patients’ access and risking their health. Also, the absence of a well-described procedure and lack of cooperation between the organizations and committees create further delays. Conclusions: Ensuring patients’ access to ODs in Greece is challenging especially during the economic crisis. Financial constraints and continuous legislative changes in the health system and the pharmaceutical market constitute important barriers to patients’ access. There is a need to describe, organize and communicate the pathway of patients’ access to ODs. PSY111 Self Reported Health Care Resource Use and Indirect Economic Burden of Opioid Induced Constipation (Oic) Alemayehu B 1, Coyne 1AstraZeneca . K S 2, King . . F3 . Pharmaceuticals LP, Wilmington, DE, USA, 2Evidera, Bethesda, MD, USA, 3AstraZeneca, Gaithersberg, MD, USA Objectives: To describe the health care resource utilization associated with the diagnosis, treatment, and general management of opioid-induced constipation (OIC) and events attributed to OIC including the negative impact on job-related activities. Methods: A prospective longitudinal study conducted in the United States (US), Canada (CN), UK (UK), and Germany (GE) of patients with OIC who have been on opioid therapy for at least four weeks was conducted. OIC related medical history and health care resource use was collected from participants self report. The number of hours missed from work and the extent to which the work productivity and regular daily activities were affected was collected using the WPAI-SHP. Results: A total of 489 eligible participants (US: 238; CN: 38;, GE: 115; UK: 98). Back pain (77%) and joint pain (52%) were the most common pain diagnosis with an average duration of chronic pain and opioid medication use of 10 and 6 years respectively. 27% of participants were currently employed. 18% of participants used at least one prescription laxative; 70% reported using at least one over-the-counter (OTC) laxative with the most common being stimulant laxatives (20%), osmotic laxatives (15%) and stool softeners (7%). 63% of participants reported discussing OIC with a health care provider, 3.0% reported a visit to the emergency room and 2.0% reported being admitted to a hospital because of their OIC during a 6-month time period. 9% of employed participants reported missing an average of 4.6 hours per week because of problems associated with constipation and 32% reported impairment while working due to constipation. Conclusions: The cumulative impact of OTC use, physician office visits and the negative impact on work-related activities for OIC may be substantial. PSY112 Health Care Utilisation and Selected Expenditures Associated With Neuroblastoma In England George S 1, Buckle J 2 . . 1Milliman, Tampa, FL, USA, 2Milliman, London, UK Objectives: Neuroblastoma (NB) is a rare cancer of childhood, with nearly 90% of cases diagnosed by age 5 (ACS 2013). Our objective was to report the utilisation and cost of hospital services related to patients who have a diagnosis of NB and High Risk NB (HRNB) reported in an England dataset from a Clinical Commissioning Group (CCG) perspective. Methods: We used an England dataset covering hospital events (April 2010 - September 2013). Patients were included if they were under age 18 and had a hospital event with a primary or secondary diagnosis coded as International Classification of Disease 10thEdition (ICD10) C749. Newly diagnosed patients were identified if they had no hospital events in the first 4 months of the study period. From this newly diagnosed cohort we identified a HRNB cohort (bone marrow transplant) that included patients who had a high risk procedure that did not occur in the last 12 months of the study period. Cost and utilisation is reported from hospital admissions, emergency department visits, and outpatient attendances. Results: We observed 336 patients as newly diagnosed and an additional 33 patients were identified as HRNB. Newly diagnosed population inpatient admits were 12 per patient, compared with 22 per patient for the HRNB population. Total costs associated with the 336 newly diagnosed patients were £24.3m. Total costs associated with the 33 HRNB patients were £4.3m. Costs per HRNB patient (£130,303) were almost double the costs per newly diagnosed patient (£72,321). The average length of stay was 6 days for both sets of patients. Conclusions: To our knowledge this is the first retrospective analysis of NB cost and utilisation using encounter data from England. While it does not capture the entire costs to the England health care system, it indicates the level of resource intensity and cost at the CCG level. PSY113 Database Analysis On Patients Using Immunobiological Drugs In A Brazilian Private Health Care Plan: A Real World Data Analysis Reis H P L C 1, Viana A D J R 1, Magalhães D D P 1, Alcantara A C D C 1, Sartori D P 1, Vieira J B 1, Ferreira J 2, Rodrigues E X 1, Filho N G 1 1Unimed Fortaleza, Fortaleza, Brazil, 2UFC, Fortaleza, Brazil . . . . . . . . . . . . . . . . . . . . . . . . Objectives: Patients’ profile undergoing intravenous immunobiological treatment is very limited. This study aimed to describe this information from the perspective of a Brazilian health plan, located in Fortaleza. Methods: This was a cross sectional study with data obtained from the HMO database as presented by Reis H et al at ISPOR 18th Annual Meeting. Eligible criteria for data analysis were patients being treated for rheumatoid arthritis (RA), ankylosing spondylitis (AS), psoriatic arthritis (PsA) or Crohn’s disease (CD) who have received at least one dose of immunobiological drug between March/2012 and October/2013. Data was stratified by indication (RA, AS, PsA and CD), patient weight and treatment profile (naïve versus non-naïve patients). Results: A total of 118 patients had been analyzed, with an average age of 51 years, and 66.9% (n = 79) of them being women. RA (n = 53,44,9%) and AS (n = 49,41,5%) were the most prevalent diseases being treated, followed by PsA (n = 13,11%) and CD (n = 3,2, 5%). The average weight of patients varied according to the disease being treated: 67 kg for RA and CD, and 70 kg for AS and PsA. It was observed that 65.5% patients were naïve to immunobiological drug, of which 73% initiated treatment with an anti-TNFa, being infliximab the most commonly prescribed one (85.2%). As for patients who had already been previously treated, golimumab and abatacept were the most commonly prescribed drugs (23%), whereas 48.7% out of total were receiving the third immunobiogical drug and 35.9% were receiving the second one. Conclusions: The knowledge of patients profile and treatment information is the basis for any planning strategy in an HMO. Associated with costs, this data is crucial in supporting HMO board decisions on best treatment alternatives and so optimize the provided care. PSY114 Multi-Criteria Decision Analysis for Reimbursing Orphan Drugs: A Dutch Demonstration Study Using the Analytic Hierarchy Process Method Trip A M 1, Tsiachristas A 1, Koenders J M 2, Kanters T A 1 1Erasmus University Rotterdam, Rotterdam, The Netherlands, 2Novartis Pharma B.V., Arnhem, The Netherlands . . . . . . . A542 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Reimbursement decisions are often unsystematic and lack transparency, especially for orphan drugs. The objective of this study was to demonstrate whether multi-criteria decision analysis (MCDA) can support rational and explicit reimbursement decision process for orphan drugs in the Netherlands. Methods: An Analytic Hierarchy Process (AHP) framework was used in which Health Economics students were asked to weigh criteria used in drug reimbursement decisions through a web-based survey. Criteria were identified by a systematic literature review. Three different orphan drugs (alglucosidase alfa in infantile Pompe disease, canakinumab in cryopyrin-associated periodic syndromes and investigational product in rare disease) were also assessed by the students on their performance on these criteria. Criteria weights and performance scores were aggregated to an overall score for each orphan drug. Rank-ordering on overall scores prioritized the reimbursement of the three drugs. The students were also asked to assess the AHP survey on feasibility. Results: Nine criteria were identified and categorized in four domains; disease (burden of illness without treatment, life-threatening nature of the disease), drug (availability of other treatments, effectiveness of the drug, side effects and safety of the drug), financial aspects (annual costs of the drug per patient, budget impact, cost-effectiveness) and quality of evidence. The criterion ‘life-threatening nature of the disease’ was given the highest importance weight and budget impact the least. Alglucosidase alfa for treatment of infantile Pompe disease ranked highest of the three orphan drugs examined, particularly due to its performance in the disease and drug domains. The AHP survey was perceived as difficult by the respondents, which was confirmed by poor values for consistency ratios. Conclusions: Performing MCDA can enable explicit, transparent and auditable reimbursement decision-making for orphan drugs. However, its feasibility and applicability needs further investigation. PSY115 Orphan and Rare Diseases - the Payer Perspective Spoors J , Rietveld A RJW & Partners, Royston, UK . . Objectives: To look at the affordability of orphan medications across Europe and whether payer attitudes to high-price medications are changing in the face of rising health care expenditure and tighter budgets. Methods: We conducted an online semi-quantitative survey of 10 European markets and the USA to understand how payers views and attitudes are changing in response to new treatments coming to market for rare and ultra-rare conditions. The payers selected for the survey hold or have held senior positions within their respective market institutions. The USA was included to provide international context to the European results. Results: 82% of payers surveyed believe that the current approach to orphan drug pricing is unsustainable in the future and all respondents predict a tougher approach from payers going forward. 73% of payers do not believe that patent expiry alone will free up the necessary space for innovative orphan and ultra-orphan products. 82% of the payers surveyed believed that less than half of all orphan and ultra-orphan drugs coming to market are supported by an adequate evidence base for reimbursement. Although payers view rare diseases as a relatively high priority to fund, they are still behind therapy areas such as oncology and cardiovascular disease. Conclusions: As the financial performance of European countries begins to diverge, so do attitudes towards the funding of orphan medicines. The increasing number of rare diseases is forcing payers to view orphan drugs in a new light and they are becoming increasingly sceptical about the prices charged in relation to the clinical benefit offered. There is space for innovation; and patent expiry is freeing up funds, but rare diseases are competing with other therapy areas for limited budget. The bottom-line is that as rare disease spending becomes a higher proportion of pharmaceutical budgets, payers will take action to curb this trend. PSY117 Sources of Information and Pharmacists’ Knowledge Regarding Rare Diseases and Orphan Drugs: Cross-Sectional Study In Serbia Arsic J 1, Krajnovic D 2, Tasic L 2, Marinkovic V 2, Djordjevic J 3 1Pharmacy Vranje, Vranje, Serbia and Montenegro, 2University of Belgrade – Faculty of Pharmacy, Belgrade, Serbia and Montenegro, 3Alvogen Pharma d. o. o, Serbia, Belgrade, Serbia and Montenegro . . . . . Objectives: The lack of information and scientific knowledge of rare diseases (RDs) and orphan drug (ODs) could affect the quality of health care delivered to patients suffering from rare diseases. The aim of this study was evaluation of the level of the general epidemiological knowledge among pharmacists regarding RDs and ODs as well as how that knowledge is influenced by information sources, education level and years of experience. Methods: The research design was based on a descriptive cross-sectional study. A questionnaire previously used in a pilot KAP study in Serbiain2012 was applied. The respondents were 182 pharmacists from public pharmacies in seven of 29 districts in Serbia. Individual level of knowledge was assessed by total number of correct answers from a maximum of 9, and overall knowledge was an average of the individual level of knowledge. Results: In total, 155 pharmacists were included in the full analysis set (response rate was 86.3%). Overall, the mean age was 43.4 years, and 94% were women. The average number of information sources regarding RD was 1.7%, and mostly one source out of five was used (56.1%). Pharmacists who were engaged in post-graduate programmes or completed such programmes tended to use more sources of information (69.2%) than those who were not involved in any such programme (41.9%). The mean value of correct answers about pharmacists’ knowledge regarding RD and OD was 4 ± 1.77. Most pharmacists (n = 30, 19.35%) replied correctly to 6 questions. Conclusions: The results indicate that years of experience and age among pharmacists do not have influence to the overall knowledge about RD. The positive impact of education was evidently, and for the better pharmaceutical care of RD patients the training of pharmacists to proper use of professional sources of information should be usefully. PSY118 Behavior Therapy for Obesity Treatment Considering Approved Drug Therapy – An Update Kossmann B 1, Wasem J 2, Buchberger B 1 . . . 1-Universität Duisburg-Essen, Lehrstuhl für Medizinmanagement, Essen, Germany, 2University of Duisburg-Essen, Essen, Germany Objectives: Many obesity-associated diseases require intensive medical treatment and are cause of a large proportion of health-related expenditures in Germany. Treatment of obesity includes nutritional, exercise and behavior therapy, usually in combination. The goal of behavior therapy for obesity is to bring about a long-term alteration in eating and exercise habits of overweight and obese individuals. Depending of the severity of obesity, drug treatment may be indicated. To evaluate the clinical and economic effectiveness of behaviour therapy for obesity considering approved drugs reducing weight, a Health Technology Assessment was carried out in the year 2008. This HTA was updated with publications up to 12/2013, along with new developments in behavior therapies and drugs. Methods: A systematic review was carried out using relevant electronic literature databases Publications chosen according to predefined criteria were evaluated by approved methodological standards of evidence-based medicine and health economics systematically and qualitatively. Results: Nine randomized controlled trials showed moderate but statistically significant reduction of weight in the intervention groups compared to control groups between 1.1 kg (at month 4) and 6.6 kg (at month 9). Studies with several examination time points resulted in statistically significant differences in the first evaluation time point (month 6) but not in the subsequent time points (month 12, 18, 24). The most frequent approach used for behavior therapy, was per phone or Email, two studies offered behavior therapy face-to-face. New behavior therapy approaches applied were techniques such as “Motivational Interviewing” and “Transtheoretical model”. No study was identified examining behavior therapy in combination with approved drug therapy. Two identified studies evaluating cost-effectiveness of behavior therapy per Email or phone showed cost-effectiveness for this kind of intervention but the results are biased due to a high rate of drop-outs. Conclusions: Behavior therapy considering new approaches is an effective method to reduce weight. PSY119 Comparison of Treatment Patterns and Disease Severity Among Patients With Psoriatic Arthritis (Psa) Receiving Their First Biologic, Treated By Rheumatologists and Dermatologists In Europe (Eu) Narayanan S 1, Hautamaki E 1, Lu Y 2, Franceschetti A 2 Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, London, UK . . . . 1Ipsos Objectives: To compare rheumatologists and dermatologists in terms of treatment patterns and disease severity among PsA patients receiving their first biologic in 5EU (UK/Germany/France/Italy/Spain). Methods: A medical chart-review study of psoriasis and PsA patients was conducted among rheumatologists and dermatologists in hospitals and private practices to collect de-identified data on disease and treatment characteristics. Physicians were screened for duration of practice (3-30yrs) and patient volume (≥ 2 Psoriasis/PsA biologic patients/month) and recruited from a large panel to be geographically representative in each country. Physicians abstracted charts of the next 5 consecutive Psoriasis/PsA patients in their respective sites. Treatment patterns and disease severity among PsA patients on their first line of biologic therapy treated by rheumatologists and dermatologists respectively were compared using descriptive statistics. Results: In Q42012, 337 rheumatologists abstracted 527 PsA patient-charts (mean-age: 47.4yrs, male: 51.4%) and 225 dermatologists abstracted charts of 109 psoriasis patients with PsA (mean-age: 49.0yrs, male: 56.0%; 55.1% were managed in conjunction with a rheumatologist; 67% were referred by GP/another dermatologist). Time to first biologic since diagnosis was 41.0mo/20.8mo for the rheumatologist/dermatologist-treated cohorts; disease severity at biologic initiation per physician judgment was (mild/ moderate/severe): rheumatologist-treated-cohort: 2.7%/60.2%/37.2%, dermatologisttreated-cohort: 1.96%/46.08%/ 51.96%. In rheumatologist-treated-cohort: treatment naïve-12.9%, non-biological DMARDS-experienced: 74.0%; in dermatologist-treated cohort: treatment nainve-6.5%, 1-or-2 systemic-treatment-experienced prior to their first biologic-initiation: 69.5%. 38.7%/61.3% and 27.5%/72.5% had moderate-severe/ remission-mild disease-status among rheumatologist- and dermatologist-treated cohort respectively. Average current PASI score was higher among rheumatologisttreated-cohort (18.7 vs. 10.3). Conclusions: Across the EU5, PsA treatment patterns and disease severity varied based on physician specialty (rheumatologist vs. dermatologist); at biologic initiation dermatologists reported a significantly higher proportion of moderate/severe PsA patients than rheumatologists, but once first biologic treatment was well established dermatologists reported a lower disease burden then rheumatologists. Factors influencing these observed variations, including optimal therapeutic approaches and care coordination between specialties to alleviate patient burden may warrant further scrutiny. PSY120 Variations In Treatment Patterns and Disease Severity Among Patients With Psoriasis Receiving Their First Biologic Therapy In Europe (Eu) Narayanan S 1, Franceschetti A 2 Healthcare, Columbia, MD, USA, 2Ipsos Healthcare, London, UK . . 1Ipsos Objectives: To assess treatment patterns and disease severity of psoriasis patients receiving their first biologic-therapy in EU. Methods: A medical chart-review study of psoriasis patients was conducted in 4Q2013 in EU5 (UK/Germany/France/Italy/ Spain) among dermatologists to collect de-identified data on disease/treatment characteristics. Physicians were screened for duration-of-practice (3-30yrs) and patient-volume (≥ 2psoriasis biologic patients/month) and recruited from a large panel to be geographically representative in each country. Physicians abstracted the charts of next 5 consecutive psoriasis patients in their center/practice. Results from patients on their first biologic treatment were analyzed and comparisons made to EU5-averages. Results: 877 patient-charts were abstracted; 702 (80.0%) were on their first biologic (mean age: 47.3yrs, male: 64.1%). Prior to initiating biologic therapy, immunomodulators/phototherapy were more widely used in UK than in other countries (90.3%/43.1% vs. 76.6%/30.4% overall, respectively); in Germany, fumarates/ A543 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 corticosteroids were more widely used (46.5%/26.8% vs. 12.3%/11.0%) and retinoids were less commonly used (5.1% vs. 18.0%). UK physicians were less likely to prescribe a biologic-treatment before trying >=1 other systemic-treatment (1.8% vs. 8.5%), and to wait longer after diagnosis to initiate biologic-therapy (25.8movs. 20.1mo); in Germany, patients were more likely to have tried 3-5 other systemic-treatments before initiating biologic-treatment (28.0% vs. 14.2%); patients in France were the most likely to initiate a biologic without trying another systemic-treatment first (11.2% vs. 8.5%). Patients in Germany were started on biologics later on average (25.8mo after diagnosis vs. 14.0mo). The average flares in past-year was highest in Germany (1.4 vs. 0.9). Average current PASI-score (26.4 vs. 16.0) and BSA-score (22.8 vs. 19.5) were highest in Germany. In France, 41.2% of patients had not had a PASI score done past year (vs. 23.3%). In Germany, 56.0% of patients had severe/terminal disease severity at biologic-therapy-initiation (vs. 47.8%). Conclusions: Among psoriasis patients receiving their first biologic-therapy, treatment patterns and disease severity varied across the EU5. Factors influencing the observed variations in treatment patterns and outcomes warrant further scrutiny to decrease patient disease burden. PSY121 Apoptosis and Oxidative Stress Induced By Exposure of Microwave Radiation In Rat Thymus: Modulatory Effect of Melatonin Sokolovic D 1, Sokolovic D M 2 of Nis, Faculty of Medicine, Nis, Serbia and Montenegro, 2Institute for Blood Transfusion in Nis, Nis, Serbia and Montenegro . . . 1University Objectives: Exposition to microwave radiation (MW), from mobile phones, satellite communications, radio relays, radars and microwave devices in medicine induce disturbances in thymus. The pineal secretory product, melatonin (Mel), exerts a veriety of effects on the immune system. The aim of the present study was to evaluate the effect of melatonin on apoptosis and oxidative stress parametars in thymus tissue of rats after 40 days long exposure to MWs. Methods: Wister rats were divided in 4 experimental groups: I (control), II (Mel group) - rats treated with Mel every day (2 mg/kg b.w., i.p), III (MW group) - rats exposed to MW (4 h/day), IV (MW+Mel) - rats treated with Mel every day (2 mg/kg b.w., i.p) and exposed to MW radiation (4 h/day). Ten animals from 4 group were successively sacrificed after 40 days of the experiment. MW was produced by a mobile test phone (SAR = 0.0430.135 W/kg). Results: The current study results demonstrate that MW significantly increased thymocyte apoptosis, detected using the Annexin V-FITC/PI detection kit (p< 0.001). DNA fragmentation in thymocytes injury of MW is probably triggered by the increase activation of alkaline-DNase I (caspase 3-activated) and acid-DNase II (p< 0.05). A significant increase in the thymus malondialdehyde (MDA) and carbonyl group concentration (p< 0.001), and decreased activity of catalase (p< 0.001) was registered during exposure. Melatonin was found to be effective on rat thymocyte: (1) decreased apoptotic rate of thymocytes (p< 0.001), (2) effect on terminal apoptotic reaction, because of the decrease DNase I and DNase II activity (p< 0.01), (3) decreased MDA and carbonyl group levels (p< 0.01), (4) increase activity of catalase (p<0.05), comparated with MW group. Conclusions: Having in the mind obtained results we can conclude that melatonine exerts protective effects on rat thymocyte by preventing apoptosis and oxidative stress disturbances in rats’ thymus under exposure of MW. RESEARCH POSTER PRESENTATIONS - SESSION IV RESEARCH ON Methods STUDIES RESEARCH ON METHODS – Clinical Outcomes Methods PRM1 Symtomatic Factors in Patients With Major Depressive Disorder (MDD): Results from an Observational Study Novick D 1, Montgomery W 2, Aguado J 3, Dueñas H 4, Haro J M 5 Lilly Holdings Limited, Windlesham, UK, 2Eli Lilly Australia, Sydney, Australia, 3Parc Sanitari Sant Joan de Deu, CIBERSAM, Sant Boi de Llobregat, Spain, 4Eli Lilly de Mexico, Mexico City, Mexico, 5Parc Sanitari Sant Joan de Déu, CIBERSAM, Universitat de Barcelona, Barcelona, Spain . . . . . . 1Eli Objectives: To explore the existence and clinical implications of symptomatic factors in patients with major depressive episodes. Methods: Data are from a 6-month prospective, non-interventional, observational study that included 1,549 MDD patients without sexual dysfunction in twelve countries. Depression severity was measured using the Clinical Global Impression (CGI) and the 16-item Quick Inventory of Depressive Symptomatology Self-Report (QIDS-SR16). Pain and quality of life were measured using the pain related items of the Somatic Symptom Inventory (SSI) and the EuroQoL-5D, respectively. The QIDS-SR16and the SSI items were jointly included in a factor analysis. Exploratory factor analysis (EFA) was conducted in a randomly selected half of the sample and confirmatory factor analysis (CFA) in the remaining half. Results: The EFA showed that a four factor model explained the data apropriately (RMSEA 0.041, 90%CI 0.034- 0.048; CFI 0.979). The four factors were mood (feeling sad, concentration/decision making, self criticism, suicidal thoughts, interest in people or activities, energy/fatigability, psychomotor retardation and agitation); sleep (initial, middle insomnia, early awakening and sleeping too much); appetite and weight, and pain (muscle soreness, cramps in abdomen, pain in lower back, pain in heart or chest, pain in joints, neck pain, headache). The CFA showed good fit indexes for this four-factor model (RMSA 0.054, 90% CI 0.049-0.059; CFI 0.954). There was a highly statistical significant correlation (Spearman) between each of the four factors and CGI severity score and quality of life at each of the visits, with higher scores in the factors (higher severity) associated with higher CGI and lower quality of life (p< 0.001, all comparisons). Conclusions: Considering the results presented, the data reasonably support that pain symptoms be included in the evaluation of patients with major depression. More severe pain symptoms are associated to higher severity of depression and lower quality of life. PRM2 Predictors of Functional Disability in Patients With Chronic Lumbosacral Radicular Pain Kanukula R 1, Bansal D 2, Ghai B 3 Institute of Pharmaceuitical Education and Research, Punjab, India, 2National Institute of Pharmaceuitical Education and Research, Mohali, India, 3Postgraduate Institute of Medical Education and Research, Chandigarh, India . . . 1National Objectives: Chronic lumbosacral radicular pain has significant morbidity and burden to the society. The objective of this study was to assess the functional disability and factors affecting it in patients with chronic lumbosacral radicular pain. Methods: We performed an observational cross sectional study in a public tertiary care hospital in north India. Adult patients (18 and 75 years), with >12 weeks of low back pain, without any co-morbidities were included in this study. Data regarding socio-demographics, duration of low back pain, prescribing pattern and depression collected at baseline. Pain assessed using visual analogue scale (VAS), functional disability using modified oswesrty disability questionnaire (MODQ). Patients also asked for health care utilization at the end of study. Predictors of high disability were analysed using multivariate regression analysis. Results: A total of 246 patients (51% males and 49% females) with mean age of 44.9 (12.25) years were included for final analysis. Mean VAS and MODQ scores at baseline are 72.3+12.5 and 48.3+11.2 respectively. Based on disability scores, 62% of patients found to be crippled whereas 62% and 24% of patients fall in severe and moderate disability category respectively. VAS and MODQ scores were positively correlated (r=0.84, p<0.05). Multi factorial analysis reveals that severe pain (higher VAS scores), high duration of pain, older age, over-weight, patients from urban region and depression were significantly associated with high disability. Conclusions: Our study results suggest that chronic lumbosacral radicular pain patients suffer with sever disability. Severity of pain was significantly correlated with levels of disability. PRM3 Hierarchical Network Meta-Analysis Incorporating Ordering Constraints on Increasing Doses of Interventions - Application to Overactive Bladder Syndrome Owen R K , Tincello D G , Bujkiewicz S , Abrams K University of Leicester, Leicester, UK . . . . . . Background: For the conservative treatment of Overactive Bladder (OAB) symptoms, the National Institute for Health and Care Excellence (NICE) in the UK currently recommends a course of supervised pelvic floor muscle training, behavioural therapy, anticholinergic medication, sacral nerve stimulation, and more recently, botulinum toxin type A (BoNTA) and Mirabegron. Given the large number of interventions and relatively few primary trials, network meta-analyses (NMAs) produce considerable uncertainty in the estimated treatment effects and consequently, there is little evidence of the most clinically effective intervention. Objectives: To evaluate the use of hierarchical NMAs incorporating ordering constraints on increasing doses in order to identify the most effective intervention for the treatment of OAB symptoms. Methods: Using Bayesian Markov Chain Monte Carlo methods, we apply a 3-level hierarchical NMA that accounts for both the correlation between treatments within the same class, as well as the residual between-study heterogeneity. We further extend this model to incorporate ordering constraints on increasing doses of the same intervention. We apply the methods to a dataset obtained from a systematic literature review of randomised controlled trials evaluating interventions for OAB syndrome. The primary outcomes of interest were mean change from baseline for voiding, urgency, and incontinence episodes. Results: The dataset includes 78 trials comparing 39 interventions that can be further categorised into 10 classes of interventions, including placebo. For voiding, and urgency episodes, BoNTA 200u was the most effective intervention with estimated mean reduction of -2.24 (95% CrI: -2.95, -1.48), and -2.6 (95% CrI: -3.46, -1.7) episodes relative to placebo, respectively. BoNTA 300u was the most effective intervention for reducing incontinence episodes with an estimated mean reduction of -1.81 (95% CrI: -2.39, -1.33) episodes relative to placebo. Conclusions: Use of hierarchical NMAs, incorporating ordering constraints, increases the precision in the effect estimates but maintains the interpretability of individual interventions. BoNTA was found to be the most effective intervention for reducing symptoms of OAB. PRM4 Treatment Effect Heterogeneity in Clinical Trials: An Evaluation of 13 Large Clinical Trials Using Individual Patient Data Kent D M 1, Nelson J 1, Altman D G 2, Hayward R A 3 Medical Center/Tufts University School of Medicine, Boston, MA, USA, 2University of Oxford, Oxford OX2 6UD, UK, 3University of Michigan, Ann Arbor, MI, USA . . . . . . . 1Tufts Objectives: Using randomized clinical trials (RCTs) for clinical decision-making necessitates making decisions for individuals based on average treatment effects. While many assume important patient variation in treatment effects, identifying patients most likely to benefit is problematic. Stratifying patients by their risk of the primary outcome was proposed as a method to identify high versus low benefit patients. Methods: From publically available sources, we identified 13 large RCTs with greater than ~1000 enrollees and overall statistically significant results. We derived Cox or logistic regression models using established risk factors blinded to treatment assignment and stratified the patient population into quartiles of risk for the outcome. Treatment effect within each risk quartile was estimated on relative and absolute scales. Heterogeneity of treatment effect (HTE) was evaluated statistically by testing for an interaction between treatment and the linear predictor of risk, and by comparing hazard (or odds) ratios and absolute risk reduction in the extreme risk quartiles. Results: Among 19 unique treatment comparisons analyzed, there was no apparent relationship between baseline risk and the hazard (or odds) ratios across trials; only 1 of 19 analyses had a significant interaction between treatment and baseline risk on the proportional scale. The difference in the log hazard ratio between the extreme risk quartiles ranged from -0.89 to 0.60 (median= 0.03; inter-quartile range (IQR) = -0.4- A544 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 0.2). However, absolute risk reduction was generally higher in high risk strata, ranging from -1.4 to 18.3% (median= 4.6%; IQR= 0.8-6.1%) in quartile one and from 0.8 to 35.0% (median= 11.5%; IQR= 3.3-19.8%) in quartile four. The difference in the absolute risk reduction between the extreme risk quartile ranged from -3.2 to 28.3% (median= 7.7%; IQR= 0.3-11.3). Conclusions: Clinically significant HTE is common even in phase 3 “efficacy” trials on the absolute scale. A multivariate risk stratified approach to subgroup analysis is feasible and often clinically informative when assessing treatment efficacy. PRM5 Non-Treatment Specific Parameter Value Estimates: Relationship Between BMI and Utility Halfpenny N J A 1, Quigley J M 1, Donatti C 2, Hawkins N S 1 Health Economics, Oxford, UK, 2Janssen-Cilag UK, High Wycombe, UK . . . . . . . . 1ICON Objectives: Cost-effectiveness models are an important component of health economic evaluation. In addition to estimates of treatment effects (typically estimated for RCTs), cost-effectiveness estimates may be sensitive to estimates of non-treatment specific parameters that describe the relationship between model variables. These may be estimated from epidemiological studies that themselves include covariable adjustment in order to account for potential confounding. We use the example of the estimated relationship between BMI and utility as to illustrate methods for the review meta-analysis of parameter estimates arising from multiple studies and issues around the selection of appropriate estimates. Methods: A targeted search was carried out in MEDLINE and EMBASE for studies with utility data on BMI. The outcome was utility change per unit increase in BMI. Study characteristics recorded included the utility instrument used, study location, diabetes status and number of covariates. Fixed and random effects models as well as graphical methods were used to investigate the influence of study characteristics. Results: Several utility scales were used throughout with some using multiple utility scales within the study to assess quality of life. EQ5-D and SF-6D were the most commonly used utility scales. Using a random effects model we observed a change in utility per unit increase in BMI of -0.0054 [-0.0077; -0.0031]. However there was significant heterogeneity between studies. The number of covariates ranged greatly between the studies and appeared predictive of the magnitude of effect of BMI on utility. Conclusions: We illustrate methods for meta-analysing multiple parameter estimates and discuss the selection of appropriate parameter estimates for the inclusion in cost-effectiveness models. In particular we illustrate the relationship between the selection of appropriate parameter estimates in terms of which covariables were included in the originating studies and the cost-effectiveness model structure in terms of which independent causal effects are modelled. PRM6 Comparison Of Iqwig and G-BA Benefit Ratings in Oncology Schuchardt M 1, Khoury C 2, Friedmann B 3, Haigh J 4 1Quintiles Consulting, Hoofddorp, The Netherlands, 2Quintiles Consulting, Neu-Isenburg, Germany, 3Quintiles Commercial, Mannheim, Germany, 4Quintiles Consulting, Reading, UK . . . . Objectives: This research was conducted to understand key reasons why the G-BA came to a different benefit rating than IQWiG during oncology HTAs. Methods: Searching the G-BA and IQWiG homepages, oncology HTAs between 1st of January 2011 and 31st of December 2013 have been identified. Assessments for which the G-BA attested a benefit rating divergent from the one proposed by IQWiG have been analyzed to reveal the key reasons for difference. Results: In the observed time frame, 18 HTAs of oncology products have been conducted by IQWiG and G-BA in Germany. IQWiG and G-BA were aligned in their benefit ratings in 14 reports (78%). During four assessments (22%) the G-BA came to a different conclusion than IQWiG. These are the assessments of crizotinib, eribulin, pertuzumab and the resubmission for vandentanib. One reason for different ratings observed in three assessments is that the G-BA looked at time-adjusted analyses to correct for different lengths of treatment in the trial arms. The extend of benefit that is extracted from the same data also potentially differs. This is exemplified by the case of vandentanib, where the G-BA was more convinced by the data on “time to worsening of pain” than IQWiG and attested a “minor benefit” (as compared to “no added benefit”). In the case of eribulin, G-BA and IQWiG agreed on benefit based on survival data. Due to the potential side effects however, G-BA demoted to “minor”, while IQWiG initially demoted to “no added benefit”. Conclusions: The translation of clinical evidence into a benefit rating and weighing of positive versus negative effects is highly complex. We see that G-BA and IQWiG can come to different conclusions. In all cases it is important that the manufacturer shows evidence against the specified comparator. The benefit rating plays an important role in the reimbursement amount negotiations. PRM7 Survival Status in (Pharmaco) Epidemiological Studies can be Successfully Investigated Using Administrative Residential Registries Potthoff P , Eichmann F , Klamert A Kantar Health Germany, Munich, Germany . . . Objectives: Study subject survival is a key outcome and endpoint in pharmacoepidemiological studies. When participants in long-term studies drop out, administrative residential registries can be a useful source to investigate subjects’ vital status. The present contribution reports procedures and results of vital status inquiries in German administrative residential databases. A pharmacoepidemiological and an environmental exposure study are used as case studies. Methods: For both case studies, residential addresses of study participants were submitted to the responsible official residential register with the purpose of collecting information about vital status or – in case of address changes – the new address was collected. For persons with multiple address changes up to 7 inquiry loops were necessary until vital status could be ascertained. Results: In the environmental case study assessing effects of an urban chemical accident, 20.170 addresses of German citizens (5.574 exposed and 14.596 not-exposed) were submitted to residential registries. The vital status of 96,6% of the study subjects could be confirmed by information of the registries. 80,7% were still alive 15 years after the accident (81% exposed, 80% not-exposed). 24,6% had died from cancer, 40,9% from cardiovascular diseases and 31,8% from other causes of death. No effects of the exposure on the vital status could be determined (Cox-Regression). In the pharmacoepidemiological case study evaluating long term safety of hormone replacement therapy, the survival status of 2.485 participants was investigated in country-wide residential databases. The vital status of 90,8% these participants could be confirmed using the registry information. 2.250 were still alive after 4-5 years of inclusion into the study, only 6 deaths were identified and for only 9% the status could not be determined. Conclusions: Official residential registries can be a valuable source for investigating the survival status and the causes of death of study subjects in pharmaco- and general epidemiological studies. PRM8 Recruiting Myelofibrosis Patients for Clinical and Health Outcome Studies Using Managed Physician Panels in 5 EU Countries Eichmann F 1, Potthoff P 1, Brown C 2, Cholmakow-Bodechtel C 1 Health Germany, Munich, Germany, 2All Global, London, UK . . . . 1Kantar Objectives: Myelofibrosis (MF) is a chronic blood cancer with an estimated prevalence in the US population of 4-5 patients per 100.000 inhabitants. Among the sources to recruit MF patients for post-approval studies in a real life environment, oncological or cardiological sites are of primary importance. The objective of the present contribution is to assess the number of MF patients in these specialties and the willingness of the physicians to enroll MF patients for clinical and health outcomes studies. Methods: In 2011, a feasibility survey assessing the incidence of MF patients in the sites of 94 oncologists, 72 hematologists and 65 cardiologists in 5 EU countries (UK, GER, FR, IT, ES) was run, using the All Global managed physician panel. The willingness of these specialist groups to participate in clinical and health outcomes studies and to recruit patients for these studies was estimated among 335 oncologists/hematologists and 208 cardiologists. Results: 88% of the oncologists and 99% of the hematologists treat Myelofibrosis patients (88% oncologists; 99% hematologists). Cardiologists are involved in MF treatment to a smaller degree (75%). In 25% of all practices, less than 10 patients are treated per year, in 36%, 11 to 50 patients, and in 38%, more than 50. Approximately 90% of the oncologists/ hematologists and 88% of the cardiologists are experienced in clinical and health outcomes studies; 80% of the oncologists/hematologists and 60% of the cardiologists are willing to participate in future studies and 92% resp. 94% in this group are willing to enroll patients for clinical and health outcome studies, including management of informed consent and ethics procedures as required for the study type. Conclusions: Myelofibrosis patients can be successfully recruited for clinical and health outcomes studies using managed panels. Oncologists/hematologists and cardiologists are experienced in these study types and willing to participate in future studies. PRM9 Can Gastric Cancer Patients be Successfully Recruited for Clinical Phase III/IV and Health Outcome Studies Using Managed Physician Panels? Eichmann F 1, Potthoff P 2, Brown C 3 Health, Munich, Germany, 2Kantar Health Germany, Munich, Germany, 3All Global, London, UK . . . 1Kantar Objectives: Gastric cancer (GC) is a cancer arising from parts of the stomach. To recruit GC patients for post-approval studies in a real world environment, oncological, hematological, and internist sites might be of primary importance. The present contribution assesses the number of GC patients in these sites and the willingness of physicians to enroll GC patients for health outcome or clinical phase III/IV studies. Methods: In 2011 a feasibility study about the number of GC patients was run in the sites of 63 oncologists, 23 hemato-oncologists and 26 internists in UK (16), GER (19), FR (11), IT (27), ES (27), and US (12). Physicians reported about the number of “new patients with unresectable, locally advanced or metastatic gastric or gastroesophageal junction (GEJ) adenocarcinoma”. The willingness of these specialist groups to participate in clinical phase III/IV or health outcome studies was also estimated in a different survey. Results: In total these 112 specialists have seen approximately 3,200 new gastric cancer patients in the last 6 months. In average (median values) oncologists have seen 30 new gastric cancer patients in 6 months, hematologists 20 patients and internists 6 patients. Most of the patients (2,600) have been treated with chemotherapy. Approximately 90% of the oncologists/hematologists and 88% of the cardiologists are experienced in clinical phase III/IV or health outcome studies; 80% of the oncologists/hematologists and 60% of the cardiologists are willing to participate in future studies and 92% resp. 94% in this group are willing to enroll patients for clinical phase III/IV or health outcome studies and ask them for written informed consent. Conclusions: Patients suffering from Gastric Cancer can be successfully recruited for clinical phase III/IV or health outcome studies using managed panels. Oncologists/hematologists and cardiologists are usually experienced in these studies. PRM10 Assessing the Relationship Between Treatment Effect and Baseline Risk in Network Meta-Analsysis of Moderate to Severe CHRONIC Plaque Psoriasis Trials Smiechowski B 1, Cope S 2 1Mapi, Boston, MA, USA, 2Mapi, Toronto, ON, Canada . . Objectives: To investigate differences across randomized controlled trials (RCTs) evaluating biological therapies for the treatment of moderate to severe chronic plaque psoriasis in terms of baseline risk in psoriasis area severity index score (PASI) and the association with treatment effects by means of network meta-analysis (NMA). Methods: 25 RCTs reporting the proportion of patients experiencing a A545 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 75% improvement from baseline in PASI (PASI 75) at the end of the trial were identified from two recently published NMAs focused on the efficacy of biologics in the treatment of moderate to severe psoriasis based on a systematic literature review. Differences in baseline risk of PASI 75 were explored graphically. The association between the treatment effect of 19 different biologics and baseline risk of PASI 75 with placebo was assessed by conducting a Bayesian NMA with baseline risk covariate adjustment using a model assuming a constant treatment by covariate interaction, and a normal distribution for baseline risk. The model also allowed for the baseline risk adjustment of RCTs that did not include a placebo arm. Results: Across the RCTs, PASI 75 at the end of the trial period for patients in the placebo arm ranged from 2% to 19%. The coefficient for baseline risk was a median of -0.33 (95% credible intervals: -0.79, 0.21) which suggested baseline risk was not significantly associated with the proportion achieving PASI 75 across the RCTs. Conclusions: Based on a NMA of RCTs regarding the efficacy of biologics in terms of PASI 75, it is unclear if baseline risk of PASI 75 acts as a treatment effect modifier. Further analyses are required to assess whether baseline risk may explain differences in the classes of biological therapies used to treatment moderate to severe psoriasis. PRM11 Riding the E-Publication Wave Quigley J M , Thompson J C , Halfpenny N , Scott D A , Hawkins N S ICON Health Economics, Oxford, UK . . . . . . . . . Objectives: Submissions to Health Technology Assessment (HTA) bodies must include the most up-to-date information on comparator technologies. This enables HTA bodies to make fully informed decisions. In order to ensure up-to-date information is obtained, all relevant data sources should be searched. At present HTA method guidelines generally specify that a core set of databases comprising EMBASE, MEDLINE, MEDLINE-in-process and the Cochrane Library are searched. However, in an attempt to manage the increasing quantity of research, journals are publishing more articles ahead of print as online e-publications. E-publications ahead of print are identifiable in PubMed but are not indexed in EMBASE, MEDLINE or Cochrane. Methods: We selected a NICE guideline published in June 2014 for ablation techniques in atrial fibrillation which searched for randomised clinical trials (RCT) in the common databases. We updated the searches in EMBASE, MEDLINE, MEDLINE-in-process and the Cochrane Library from the date of the last search, to identify any studies published since the guideline was published. We concurrently searched PubMed using a filter to identify studies only available as e-publications ahead of print. Results: We identified one RCT meeting the inclusion criteria that was retrieved by the search in PubMed but was not identified by searching in EMBASE, MEDLINE or Cochrane (all searches being run on the same date). Conclusions: If studies are missed from systematic reviews there can be important implications on clinical efficacy comparisons. To produce the most upto-date systematic review, with all available full papers it is necessary to search for e-publications ahead of print in PubMed as well as for all other publications in EMBASE, MEDLINE and Cochrane. PRM12 Individualised Growth Response Optimisation (IGRO): A MultiLanguage Software Medical Device to Predict Growth Response in Children Treated With Growth Hormone (GH) Loftus J 1, Lindberg A 2, Aydin F 2, Rawlings F 3, Ranke M 4, Camacho-Hubner C 5 1Pfizer Ltd, Tadworth, UK, 2Pfizer, Sollentuna, Sweden, 3Pfizer Inc, Collegeville, PA, USA, 4University Children’s Hospital, Tuebingen, Germany, 5Pfizer Inc, New York, NY, USA . . . . . . Objectives: To produce a modern, accessible, quick and intuitive growth prediction tool in which growth prediction models (GPMs) were clinically validated. Methods: GPMs support physicians to optimise and monitor GH treatment and response. They are validated, providing accurate predictions of growth response following GH treatment in children with idiopathic growth hormone deficiency (IGHD), Turner syndrome (TS) and short children born small for gestational age (SGA). By using GPMs, realistic expectations of treatment outcomes are conveyed to patients and parents (important as children respond differently to treatment). A modern platform, a website, will host iGRO; the software should ensure web browser compatibility in European endocrine clinics. An internal paediatric endocrinologist robustly tested the GPMs. User acceptance testing (UAT) for acceptability and usability was conducted with external paediatric endocrinologists. These outputs contribute to the Class I medical device Technical File; a requirement for iGRO. Results: The clinical validation performed under simulated conditions using the final medical device software passed, consistent with regulatory expectations. GPMs for IGHD, SGA, TS were tested for prediction accuracy. Validation acceptance criteria and test results were included in the test script records. UAT was designed and executed following a rigorous software development lifecycle methodology to ensure completeness of the tool and the specified business and user requirements. UAT was certified through role based testing scenarios (requirements and processes for daily use). The tool, accessed via a secure password protected website, was easily navigated and provided a clear interface requiring little training. Data entry was quick and growth outcomes (i.e. graphical comparisons between patient’s actual growth response and one-year predictions) were easily viewed. Conclusions: iGRO, an accessible webbased software medical device, provides accurate growth prediction which enable physicians to monitor patients’ response to GH treatment. It may potentially be used to detect discrepancies early. PRM13 Strengthening Evidence Base for Traditional Medicine in Asean, Quality of Reporting of Randomised Controlled Trials of Herbal Interventions in Asean Plus Six Countries: A Systematic Review Pratoomsoot C 1, Sruamsiri R 2, Dilokthornsakul P 2, Chaiyakunapruk N 3 1Discipline of Applied Thai Traditional Medicine, Faculty of Public Health, Naresuan University, Phitsanulok, Thailand, 2Department of Pharmacy Practice, Faculty of Pharmaceutical Sciences, Naresuan University, Phitsanulok, Thailand, 3Monash University Malaysia, Selangor, Malaysia . . . . Objectives: Traditional Medicine and the use of herbal interventions are regarded as an integral part of health care among countries of Association of Southeast Asian Nations (ASEAN). To date numerous randomised controlled trials (RCTs) of herbal interventions have been conducted within ASEAN. It is recognised that good quality reporting of RCTs is crucial to the assessment of clinical significance. ASEAN has placed strong emphasis on the facilitation of research in Traditional Medicine, especially in the strengthening of evidence base for herbal medicines. The objective of this study was to systematically review the quality of reporting of RCTs of herbal interventions conducted in ASEAN Plus Six Countries. Methods: Searches were performed using PubMed, EMBASE and MEDLINE, The Cochrane Library, and Allied and Complementary Medicine (AMED), from inception through October 2013. The following limits were applied: Human; RCTs. Herbal species search terms were based on those listed in the National List of Essential Medicines [NLEM (Thailand, 2011)]. Studies conducted in ASEAN Plus Six Countries, published in English were included. Quality of reporting was assessed according to the 22-item Elaborated CONSORT statement for reporting RCTs of herbal interventions. Results: Seventyone articles were identified, of which thirty RCTs (42.25%) were conducted in ASEAN Countries, whereas 41 RCTs (57.75%) were from the Plus Six Group. Adherence to the recommended CONSORT checklist items for reporting of RCTs of herbal interventions among ASEAN Plus Six Countries ranged from 0% to 97.18%. Less than half of the RCTs reported methods used to generate random sequence allocation (item 8,47.89%), and implement random allocation sequence (item 9 allocation concealment, 29.58%). Less than a quarter of RCTs (18.31%) reported information on standardisation of herbal products. Conclusions: The present study highlighted the need to improve reporting quality of RCTs of herbal interventions across ASEAN Plus Six Communities. PRM14 Endpoints in Pain: the Suitability for Health Economic Evaluation of Endpoint Designs in Chronic Pain Studies Rycroft C 1, Hirst M 2, Dunlop W 2, Pirk O 3, Mullins C D 4, Akehurst R 5 International Limited, Cambridge, UK, 3Olaf Pirk Consult, Nürnberg, Germany, 4University of Maryland School of Pharmacy, Baltimore, MD, USA, 5University of Sheffield, Sheffield, UK . . . . . . . 1BresMed, Sheffield, UK, 2Mundipharma Objectives: A wide range of instruments are used to measure outcomes in clinical studies of chronic pain. However, the suitability of study outcomes for economic evaluations in chronic pain is limited by variability in the instruments used and the failure of those that are in common use to adequately capture the dimensions of a patient’s experience of pain, including function, quality of life and tolerability. The current study aims to identify the pain instruments and study endpoints most commonly used in the clinical trial setting. Methods: A structured, comprehensive literature review of ongoing registered trials (trial registries) and published clinical studies (PubMed) in the areas of chronic pain (cancer, non-cancer and neuropathic), pain in elderly patients, and breakthrough pain was conducted. Inclusion criteria were: interventional study, pain population of interest and pain measured within primary endpoint. Data were extracted from identified publications focusing on primary and secondary endpoints reported, and primary pain instrument used. Results: Of 1,256 citations retrieved, 132 were included as they reported large clinical studies in pain populations. The majority of primary endpoints were pure pain measures (112), most commonly: numerical rating scale (NRS) (44), visual analogue scale (VAS) (34), and brief pain inventory (BPI) (18). Other outcome domains were typically limited to secondary endpoints and were not consistently applied across the studies. These include measures of function (47), quality of life (36), safety/tolerability (33), emotional wellbeing (26) and sleep (20). Conclusions: The majority of studies still use simple pain measures as their primary outcomes. Useful economic outcomes such as quality of life and function are relegated to secondary endpoints and are inconsistently used. Types of pain instruments currently used in registered, ongoing trials will also be reported. Further research is needed to consider which existing endpoint designs constitute best practise and if new designs are required. PRM15 Comparative Real World Effectiveness of Novel Agents Versus Conventional Therapies in Multiple Myeloma Patients in Sweden Thilakarathne P 1, Diels J 2, van Sanden S 3, Liwing J 4, Van Agthoven M 5, Chirita O 6, Nahi H 7 1Janssen Pharmaceutica N. V., Beerse, Belgium, 2Janssen Research & Development, Beerse, Belgium, 3EMEA HEMAR Analytics, Janssen EMEA, Beerse, Belgium, 4Janssen-Cilag AB, Sollentuna, Sweden, 5Janssen-Cilag BV, Tilburg, The Netherlands, 6Janssen, High Wycombe, UK, 7Karolinska University Hospital, Solna, Sweden . . . . . . . Objectives: Comparing outcomes between different treatments based on real world evidence can be challenging, first because of confounding due to lack of randomization at treatment initiation, and secondly due to selection bias induced by selective treatment switching. To account for both sources of bias, we explored the Inverse probability of censoring weights analysis (IPCW) to account for this bias on a retrospective dataset of multiple myeloma patients in Sweden. Methods: IPCWapproach was used to additionally adjust for bias induced by selective treatment switching in 2 steps. First, time-varying weights were estimated using multivariate logistic regression, including age, gender, stage, M-protein type, creatinine-clearance as baseline covariates and M-protein as time-varying covariate. Secondly, these time-dependent weights were incorporated in a proportional hazards model, including treatment and baseline characteristics, with patients censored at initiation of subsequent therapy. ITT-estimates were generated based on multivariate proportional hazards model, including the same baseline covariates. Results: In the total MM population (n= 1638), the non-transplant population was 1125, of which 31% (n= 349) received novel therapies (bortezomib, lenalidomide and thalidomide based) while 69% had conventional therapies in frontline. Mean age was 71/75 respectively. The mean baseline serum m-protein level was 30.9 g/L for both groups. The ITT HR in frontline was 0.56 [0.45, 0.70] and IPCW-adjusted HR is 0.31 [0.22, 0.42]. ITT vs IPCWbased HR for 2nd and 3rd line of therapy were 0.71 [0.58, 0.87] vs 0.667 [0.487, 0.914] A546 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 and 0.87 [0.68, 1.11] vs 0.655 [0.455, 0.944], respectively. Conclusions: This analysis shows the improved survival of patients who received novel therapies as compared to conventional therapies, across the different therapy lines. Additionally, results illustrate the impact of selection bias induced by selective treatment switching, and the need to apply novel approaches as IPCW to make additional adjustments, for which traditional statistical techniques cannot be used for. PRM16 Comparing the Use of Patient-Level Data to An Average Patient Profile Within a Type 2 Diabetes Simulation Model McEwan P 1, Bennett H 2, Ward T 2, Bergenheim K 3 Economics and Outcomes Research Ltd, Cardiff, UK, 2Health Economics and Outcomes Research Ltd, Monmouth, UK, 3AstraZeneca, Mölndal, Sweden . . . . 1Health Objectives: Despite significant patient heterogeneity and complex treatment pathways, averages are commonly relied upon when defining patient populations and treatment effects within type 2 diabetes modeling. As a result, clinicians may struggle to relate results to the clinical setting. This study compares outcomes when using patient-level and average cohort inputs within a published simulation model, based on the UKPDS68 outcomes equations. Methods: UK patient data (2,251 patients initiating dual therapy) were obtained from The Health Improvement Network (THIN). Simulations, performed over a medium-term horizon of 20 years, utilised either patient-level data, collating outputs over all replications, or average cohort data. The outputs (total costs, benefits and complication rates) were then compared. Results: Average baseline characteristics were: age: 63.36 (±11.14) years; HbA1c: 8.39% (±1.23); total cholesterol: 4.18 (±0.92) mmol/L; systolic blood pressure: 135.07 (±14.76) mmHg; weight: 89.85 (±19.01) kg. The mean treatment effect was a reduction in HbA1c of 1.01 (±1.23) %. Over 20 years, fewer macrovascular and microvascular events (-82/1,000 patients) and higher all-cause mortality (+17/1,000 patients) were predicted when using patient-level data compared to the average profile. Differences in the frequency and timing of deaths were driven primarily by variation in age and led to fewer estimated life-years (-0.66), quality-adjusted life-years (QALYs; -0.59) and costs (-£551) per patient. Patients estimated to have lower costs and higher QALYs than those associated with the average profile were younger, with higher HbA1c and cholesterol but lower blood pressure at baseline. Conclusions: Modelling results differ depending on the use of patient-level or average cohort model inputs. Patient-level data may provide insight into the type of patients in whom therapy is likely to be most beneficial. Furthermore, it enables the accurate simulation of correlation between patient characteristics and treatment effect, which are rarely accounted for as part of a standard probabilistic sensitivity analysis. PRM17 Quantifying Nonlinear Effects in Stochastic Markov Simulation Using UKPDS 68 and Ukpds 82 Equations in Type 2 Diabetes Modeling Analysis With the IMS Core Diabetes Model (CDM) McEwan P 1, Grant D 2, Lamotte M 3, Foos V 4 1Health Economics and Outcomes Research Ltd, Cardiff, UK, 2IMS Health, London, UK, 3IMS Health Consulting, Brussels, Belgium, 4IMS Health, Basel, Switzerland . . . . Objectives: Previous studies have demonstrated incorporating parameter sampling (PS) is crucial to capture nonlinear effects (NE) in cost effectiveness modeling. NE are, among other causes, driven by the degree through which the symmetric sampling of a risk factor is translated into non-symmetrically distributed probabilities generated by the applied risk equations (RE). This study sought to assess degree by which the incorporation of NE through PS alters event rate predictions from the UKPDS 82 (UK82) and UKPDS 68 (UK68) RE in a set of selected validation studies conducted with the CDM. Methods: A total of 50 validation simulations were performed to data from ACCORD, ADVANCE, VADT, ASPEN, DCCT and UKPDS. Simulations mirroring cohort baseline characteristics of each of the trials were conducted with and without PS using UK68 and UK82 REs. Predicted versus observed macrovascular (MAC) and microvascular (MIC) complications and all cause mortality (ACM) were assessed using the coefficient of determination (R2) goodness of fit measure. Results: When the CDM was run without PS, validation studies produced an R2 statistic of 0.898 using UK68 and 0.853 using UK82 RE. This compared to R2 statistics of 0.876 and 0.791 in analysis with PS for UK68 and UK 82 REs, respectively. Overall, PS caused end point predictions for MAC, MIC and ACM to increase. Internal validations against UKPDS 80 demonstrated that PS increased event rate predictions for myocardial infarction (MI), stroke, MIC and ACM by 4.4%, 21.5%, 19% and 16.4% when UK68 RE were applied and 26.3%, 64.7%, 14.9% and 34.8% with UK82 RE, respectively. Conclusions: The findings from this study have shown that external validity declined with PS in simulations using UK68 RE and UK82 RE. The degree by which PS increased end point predictions was considerable stronger in UK82 RE predictions for MAC and ACM but lower for MIC. PRM18 Inverse Probability of Censoring Weighted Analysis to Adjust the Treatment Effect on Overall Survival for Subsequent Therapy: A Case Study in a Clinical Trial in Multiple Myeloma Thilakarathne P 1, Palumbo A 2, Diels J 1, Delforge M 3, van Sanden S 4, Mateos M V 5, Chirita O 6, Dimopoulos M A 7, van de Velde H 8, San Miguel J F 9 1Janssen Pharmaceutica N. V., Beerse, Belgium, 2University of Torino, Torino, Italy, 3University Hospital Leuven, Leuven, Belgium, 4EMEA HEMAR Analytics, Janssen EMEA, Beerse, Belgium, 5Hospital Universitario de Salamanca, Salamanca, Spain, 6Janssen, High Wycombe, UK, 7Alexandra Hospital, Athens, Greece, 8Johnson & Johnson ORD, Beerse, Belgium, 9Hospital Universitario Salamanca, Salamanca, Spain . . . . . . . . . . . . . Objectives: ITT-analyses of oncology trials tend to underestimate the treatment effect on overall survival, due to the impact of subsequent therapy. Inverse probability of censoring weighted analysis (IPCW) was explored to estimate an adjusted treatment effect on OS in VISTA, a phase III randomized clinical trial comparing melphalan and prednisone with or without bortezomib (VMP vs MP) in previously untreated multiple myeloma patients ineligible for stem cell transplantation. Methods: The IPCW consisted of 2 steps. First, time-varying weights were estimated using multivariate logistic regression, including age, gender, stage, M-protein type, creatinine-clearance as baseline covariates and M-protein as time-varying covariate. In a second step, these time-dependent weights were incorporated in a proportional hazards model, including the same baseline characteristics, with patients censored at initiation of subsequent therapy. Results: 338/344 patients received up to nine 6-week cycles of VMP or MP respectively, with median follow-up of 44.2 months. 68% of MP-patients received subsequently therapy, compared to 58% in the VMP-arm. Age< 75, creatinine-clearance 30-60ml/min, stage III, and increasing M-protein measures over time were additional drivers for treatment-switching. The IPCW-approach generated an adjusted hazard ratio of 0.584 [0.406, 0.839], compared to the ITT-estimate of 0.704 [0.576, 0.860]. Conclusions: In oncology, particularly in early line treatment, it is common that patients receive subsequent treatment lines. This typically happens more frequently and earlier in the comparator arm, which may bias the estimate for the treatment effect on OS. The IPCW-approach was explored to adjust for this bias, which resulted in an increased estimate of the treatment-effect on OS of VMP vs MP, compared to the original ITT-analysis. With overall survival being a key input in economic evaluation, estimating the accurate effect on OS is key. Employing this type of approaches may result in more accurate cost effectiveness results and thus more consistent/appropriate Health Technology Assessment recommendations. PRM19 Sharing of Information Across Studies to Inform Choice of Functional Form When Conducting Parametric Survival Analysis Parker C 1, Hawkins N S 2 1ICON Clinical Research, Oxford, UK, 2London School of Hygiene and Tropical Medicine, London, UK . . . Objectives: To explore the sharing of information across multiple studies in order to inform the choice of functional form when conducting parametric survival analysis. Methods: A set of four clinical trials in advanced soft tissue sarcoma were identified from a published systematic review. Individual patient data for overall survival were estimated from digitised Kaplan-Meier curves using a published algorithm. A range of parametric survival models (exponential, Weibull, Gompertz, log-normal, log-logistic, Gamma and Generalised Gamma) were fitted. Two approaches were explored for identifying the preferred parametric model: (i) selecting models independently for each study (ii) selecting a common model across all the studies. Models were selected using the Bayesian Information Criterion (BIC). For approach (ii) a single BIC statistic was calculated by summing the components of the BIC (n, k and ln (L)) across studies. Estimates of mean survival were derived for each model and a bootstrap analysis was conducted to estimate both the uncertainty in model selection and the variance in mean survival estimates. Results: Independent selection led to different functional forms being selected for each study with considerable uncertainty regarding the choice of model (the bootstrap estimation for the probability that the optimum model had been selected varied between 16 to 84% across studies). The choice of model influenced mean survival predictions. Selecting a common model across studies was found to reduce the uncertainty in model selection and variance of the estimated mean survival (by up to 65%) compared to selecting models independently. Conclusions: Use of multiple studies to inform choice of functional form can improve the efficiency of survival estimates and hence reduce uncertainty of cost-effectiveness estimates. Given the considerable uncertainty in selecting survival models within individual studies, it may be reasonable to treat information on functional form as exchangeable between studies and to ‘borrow’ strength across studies. PRM20 Predictive Modeling to Assess Predictors of Treatment Success and Failure Among Combination Statin Therapy Patients Nyandege A 1, Burudpakdee C 1, Philip S 2 1MKTXS, Raritan, NJ, USA, 2Amarin Pharma Inc., Bedminster, NJ, USA . . . Objectives: Combination statin therapy may help to further lower low-density lipoprotein cholesterol (LDL-C) better than monotherapy alone. The objective of this study was to apply predictive modeling methodology to determine the predictors of success and failure in achieving LDL-C goals after combination statin-fibrate therapy in patients diagnosed with hypertriglyceridemia (HTG). Methods: A large claims database was used to identify patients initiating a fibrate between January 2011 and December 2011 (index date). Diagnosis of HTG and the use of statins were confirmed within 6 months before the index date. A total of 622 patients were selected for the current analysis. Patients were categorized into very high risk, high risk, moderate risk, and low risk groups. Logistic regression and two-group discriminant analysis models based on 17 potential predictors for treatment success or failure were constructed. Results: At index, the median triglyceride (TG) level among all patients was 95.5 mg/dL, LDL-C level was 92 mg/dL, and high-density lipoprotein (HDL) was 40 mg/dL. The mean age was 54 years. Two predictors were associated with combination statin-fibrate treatment success or failure and accounted for 5.3% of variance between groups. Low HDL (defined as < 40 mg/dL) (OR= 0.35; 95% CI, 0.20-0.59) and peripheral arterial disease (OR=0.10; 95% CI, 0.02-0.38) were significantly associated with treatment failure. Low HDL variable was the key discriminator. Conclusions: Analytic insights enabled by predictive models may help researchers gain information on discriminating factors about certain target treatment groups and drug classes. A set of key predictors may suggest opportunities to understand and predict treatment success and failure of targeted groups and/or drug classes. These predictors may be useful in developing treatment strategies that will optimize outcomes. PRM21 Predictive Modelling for Optimal Target Population and RealWorld Study Design: An Example In Mother-To-Child Transmission of HIV Amzal B LASER Analytica, London, UK . A547 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: This research will present the Bayesian decision analytic framework for late phase study simulation and calculation of chance of study success applied to optimal target population and study design. A real-world example in motherto-child transmission of HIV conducted in Thailand will be used to illustrate the concepts throughout. Methods: Predictive models describing virtual cohorts over time under various care strategies can be informed via Bayesian inference using all relevant data available on associations between population characteristics, relative drug efficacy, drug uses and design parameters. In the proposed example, historical transmission data from 3,876 Thai women were modeled via mixed-effect logistic regression adjusted for viral load, gestational age, CD4 count at delivery and infant treatment duration. Viral load was described as an exponential function of prophylaxis duration. Monte Carlo simulations were used to predict intrapartum transmission rates with and without single dose nevirapine (sdNVP) added to the standard of care (antenatal prophylaxis), and predict chance of success of a naturalistic Phase IIIb study in Thailand under various assumptions on target population and adaptive study design. Results: In women with short prophylaxis durations (<8-weeks) estimate of intrapartum transmission rate was 2.6% (95%-Credibility Interval=0.5%-9.2%) with the standard of care and 0.8% (95%-CI=0.1%-2.9%) with sdNVP added to standard care, corresponding to a risk ratio of RR=3.9 (95%-PI=2.1-9.7). Study simulations showed that a single-arm study with stopping rules at N= 58, 118, 275, and 410 has 78% (resp. 68%) probability of evidencing RR>1.3 (resp. RR>2). Conclusions: When the association between outcomes, design parameters and the main effectiveness drivers can be informed by historical data, Bayesian predictive models can be powerful decision support tools for optimal target population and late-phase or pragmatic study design. Retrospective database studies of PAH using US payer claims data have limitations due to lack of specific ICD-9 codes for PAH and ability to identify patient severity. Previous studies used an algorithm which includes patients with non-specific PH codes, along with a claim for an advanced PAH drug therapy. This study attempts to validate the algorithm and identify patient disease severity through linkage to data abstracted from medical charts. Objectives: To evaluate validity of a retrospective review of payer database along with chart abstraction for confirmation of PAH diagnosis and identification of World Health Organization Functional Class (FC). Methods: Medicare patients who received an (1) endothelin-receptor antagonist, phosphodiesterase type 5 inhibitor, or prostacyclin AND (2) had a diagnosis of pulmonary hypertension, other chronic pulmonary hypertension or chronic pulmonary heart disease OR (3) medical claim indicating right heart catheterization (RHC) were identified from pharmacy and medical claims data. A random sub-sample of 110 patients was chosen and the providers contacted to provide medical charts. Charts were reviewed to abstract data indicating PAH diagnosis, FC, and/or symptoms, diagnostic tests, and treatments to enable classification. Results: Of 110 charts requested, 41 were received and abstracted. Twenty-one charts (51%) came from a specialist. All 41 charts documented a confirmed diagnosis of PAH. Of those, 18 (44%) explicitly identified PAH class. Physical symptoms were reported, with dyspnea (66%) being most frequent, while walk test results, documentation of RHC and pulmonary diagnostic tests were reported in less than 20% of cases. Conclusions: The identification algorithm successfully identified diagnosed, confirmed cases of PAH. Refinements to provider selection algorithm could result in an increase in provider response rate, charts with documented FC and overall chart data quality. PRM22 The statistical analysis of delayed effects in survival outcomes for immunotherapies. Estimation of time-delay and application of weighted log rank PRM25 Quality Assessment of Controlled Trials Evaluating Chinese Herbal Medicine in Patients With Rheumatoid Arthritis: A Systematic Review Luaces P , Sánchez L , Viada C , Frias A , Alvarez M , Rodríguez P C Center of Molecular Immunology, Havana, Cuba Pan X 1, Lopez-Olivo M A 2, Nayak P 2, Suarez-Almazor M E 2 Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, China, 2The University of Texas, MD Anderson Cancer Center, Houston, TX, USA . . . . . . . . Objectives: The aim of the study was to assess the delayed-time effect on survival of the immunotherapy with a cancer vaccine by comparing the conventional logrank test vs a weighted logrank in presence of non-proportional hazards. Methods: Data from a multicenter, open-label and randomized phase III clinical trial with an EGF-based cancer vaccine in advanced NSCLC. The diagnosis of the delayedtime effect was done and the time-delay was estimated. The non-proportional hazards were also confirmed and tested delayed effect on survival of the treatment. Weighted logrank tests was applied and the results were compared with those obtained using the conventional logrank test. The R software was used in the analysis. Results: The time-delay was estimated in 28 months and a significant effect after this time was verified. The proportional hazard assumption was not satisfied. The median survival for the vaccinated arm was 10.37 months vs. 8.93 months for non-vaccinated arm. The difference was statistically significant by weighted logrank (p=0.04) and the conventional logrank test does not detect this difference. Conclusions: Weighted logrank is substantially more efficient than the conventional logrank statistic in those situations in which non-proportional hazards are foreseen. This analysis is recommended for immunotherapies where the appearance of a late biological effect is displayed several months after randomization. PRM23 Impact of Single Risk Factor Changes on Long Term Outcomes and Cost in a Type 2 Diabetes Modeling Study Contrasting Projections With UKPDS 68 Versus UKPDS 82 Risk Equations Foos V 1, McEwan P 2, Grant D 3 1IMS Health, Basel, Switzerland, 2Health Economics and Outcomes Research Ltd, Cardiff, UK, 3IMS Health, London, UK . . . Objectives: The degree to which predefined risk factor (RF) changes alter long-term clinical and cost outcomes in the IMS-CORE-Diabetes-Model (CDM) was reported in earlier publications. Since this time the CDM has undergone a series of updates including the inclusion of recently published UKPDS-82 risk equations (UK-82-RE). The objective of this study was to project the lifetime benefits and total lifetime costs (TLC) associated with a range of selected RF changes opposing results from CDM projections utilizing UKPDS-68 risk equations (UK-68-RE) vs. UK-82-RE. Methods: The CDM was applied to project the lifetime benefits (life years (LYs), quality adjusted life years (QALYs)) and TLC (£GBP) associated with baseline RF changes for HbA1c, body-massindex (BMI), systolic blood pressure (SBP), high-density-lipoprotein (HDL) and low-density-lipoprotein (LDL). A intermediate risk type-2 diabetes cohort (age 52 years, HbA1c 8%, SBP 140 mm-Hg, BMI 30 Kg/m2, HDL 50 mg/dl and LDL 150 mg/dl) was projected over lifetime to explore the sensitivity of undiscounted LYs, QALYs and TLC for selected RF ranges (A1c+/-2%, SBP+/-20 mmHg, BMI+/-2 Kg/m2, HDL+/-10 mg/dl, LDL+/-20 mg/ dl). Linear regression models were fitted to assess the degree of end point sensitivity per unit RF change. Results: When UK-68-RE were applied, projected changes in benefits were 0.264, 0.094, 0.050, -0.611 and 0.162 LYs and 0.288, 0.111, 0.123, -0.358, and 0.113 QALYs associated with unit RF reductions of 1% point (A1c), 10 mm-HG (SBP), 1 KG/m2 (BMI), 10 mg/dl (HDL) and 10 mg/dl (LDL), respectively. This compared to changes of 0.161, 0.079, 0.053, -0.262, 0.311 (LYs) and 0.215, 0.099,0. 116, -0.178, 0.208 (QALYs) utilizing UK-82-RE. TLC decreased by £1’105, £298, £115, £680 and £38 utilizing UK-68-RE and £1’073, £309, -£71, -£221, -£7 with UK-82-RE. Conclusions: The degree to which RF changes are translated into benefits and costs may change considerably dependent on the choice of selected risk equations. PRM24 A Chart Abstraction Based Method to Classify Real World Patients With Pulmonary Arterial Hypertension Based on Who Functional Classification Stemkowski S 1, Pruett J 2, Dufour R 1, Lane D C 1, Raspa S 2, Drake W 2 Health Insights, Humana, Louisville, KY, USA, 2Actelion Pharmaceuticals US Inc., San Francisco, CA, USA . 1Comprehensive . . . . . . . . . . 1Shuguang . Objectives: We conducted a systematic review to appraise the methodological quality of controlled clinical trials evaluating the efficacy and safety of Chinese herbal medicine (CHM) patients with rheumatoid arthritis (RA). Methods: We searched electronic databases (Medline, EMBASE, The Cochrane Library, and Web of Science) from inception until May 2014. Study selection was performed by 2 independent reviewers. The methodological quality of the trials was assessed using the Cochrane risk of bias tool for randomized trials and Newcastle Ottawa Scale for controlled non-randomized studies. Results: 54 studies were included (51 randomized trials; 3 non-randomized studies) evaluating 7,792 patients. Only one study was conducted in the US, the remaining in China. There were 3,446 patients receiving CHM. In the control groups 2,283 patients received a disease modifying anti-rheumatic drug (DMARD), 182 non-steroidal anti-inflammatory drugs (NSAIDs), and 164 inert placebo. For the randomized studies, when evaluating selection bias 54% of the studies were judged to have an adequate random sequence generation, but 77% had inadequate allocation concealment. 79% had a high risk of performance bias (not blinding participants and/or personnel) and detection bias was unclear in 56% of the studies; 62% of the studies reported how missing data was handled, therefore attrition bias was judged to be low. In 87% no disclosure of interest or source of founding was reported. For non-randomized studies, all the studies were representative of RA patients, had an adequate ascertainment of intervention with comparable groups, but only one demonstrated that the outcome of interest was not present at start of study or provided the rate of lost to follow-up. Conclusions: Studies evaluating CHM often fail to meet expected methodological criteria, and high quality evidence is lacking. Future studies of CHM should be methodologically robust and adhere to reporting guidelines such as the CONSORT statement for TCM. RESEARCH ON METHODS – Cost Methods PRM26 Validation of The Hospital Episode Statistics Outpatient Dataset in England Thorn J C 1, Turner E 1, Hounsome L 2, Walsh E 1, Down L 1, Donovan J 1, Verne J 2, Neal D 3, Hamdy F 4, Martin R M 1, Noble S 1 1University of Bristol, Bristol, UK, 2Public Health England, Bristol, UK, 3University of Cambridge, Cambridge, UK, 4University of Oxford, Oxford, UK . . . . . . . . . . . . . Objectives: Health economists are being encouraged to use routine datasets for resource-use measurement and costing purposes in economic evaluations alongside clinical trials to reduce research burden. The Hospital Episode Statistics (HES) dataset, which records all NHS hospital-based activity in England, is one such dataset. However, its validity for research purposes has not been established. This study aims to assess the validity of the HES outpatient dataset. Methods: Men who died of, or with, prostate cancer were selected from a prostate-cancer screening trial (CAP, Cluster randomised triAl of testing for Prostate cancer). Details of visits that took place after 1/4/2003 to hospital outpatient departments for conditions related to prostate cancer were extracted from medical records (MR). Data from the HES outpatient dataset were obtained for the same men. Appointments for visits extracted from MR were sought in the HES dataset. The matching procedure was repeated for periods before and after 1/4/2008 (when the dataset was accredited as a national statistic). Results: 4922 outpatient appointments were extracted from MR for 370 men between 2003 and 2012. 4086 appointments recorded in MR were identified in the HES dataset (83.0%; 95%CI 81.9–84.1). Allowing a +/-2 day tolerance for the appointment date resulted in a slight improvement to 4171 (84.7%; 95%CI 83.7–85.7) matches (p= 0.5). For appointments occurring when the dataset was considered experimental (prior to 1/4/2008), 2194/2754 (79.7%; 95%CI 78.1–81.2) matches were observed, while 1892/2168 (87.3%; 95%CI 85.8–88.6) appointments occurring after 1/4/2008 were identified (p= 0.03). Conclusions: The HES outpatient data- A548 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 set appears reasonably valid for research, particularly following accreditation. The dataset may be a suitable alternative to collecting MR data manually within a trial, although caution should be exercised with earlier data. Further work is ongoing to establish the nature of the missing data and the implications for cost differences. PRM27 Can Using a Resource Use Log in an Economic Evaluation Alongside a Randomised Controlled Trial Reduce the Amount of Recall Bias? Noble S , Tudge I , Wylde V , Lenguerrand E , Marques E M University of Bristol, Bristol, UK . . . . . . Objectives: To determine whether giving patients a resource use log (RUL) at hospital discharge reduces recall bias in a follow-up resource use questionnaire (RUQ). Methods: Within the APEX randomised controlled trials (RCTs), 86 patients undergoing joint replacement were randomised to receive or not receive an RUL at hospital discharge (The RUL trial). A postal RUQ was then administered to all participants at 3-months after surgery. Resource use data (visits to GPs, GP home visits and telephone calls; GP practice nurse visits and telephone calls; and prescribed medication) in relation to the patient’s joint replacement were extracted from GP records from hospital discharge until completion of the 3-month RUQ by a blinded researcher. Data from both sources were coded into use of resource and number of contacts. For each resource use category, descriptive statistics were calculated by data source and RUL trial arm. Kappa statistics and Concordance Correlation Coefficients (CCC) were calculated as appropriate. Results: GPs were contacted for 67/86 patients originally randomised to receive or not receive an RUL (one had surgery delayed, 3 died, 5 withdrew, 6 had GP practices outside of area, 4 did not complete the 3-month RUQ). Information was then extracted for 66/67 patients. There was evidence of improved recall in favour of the RUL arm in relation to visiting a GP (Kappa= 0.5312 vs. -0.0161). There was some slight evidence in favour of the non-RUL arm with regards to having a GP home visit (Kappa= 0.335 vs. -0.0937). The RUL arm showed more agreement than the non-RUL arm between data sources in terms of number of: visits to GPs (CCC= 0.581 vs. -0.013); GP telephone calls (CCC= 0.564 vs. 0.173) and prescriptions (CCC= 0.418 vs. -0.13). Conclusions: Although based on small numbers, our study found some evidence that provision of an RUL reduces recall bias in relation to visits to GPs. PRM28 Systematic Review and Critique of Health Economic Models on Relapsing-Remitting Multiple Sclerosis in the UK Kusel J 1, Maruszczak M 1, Montgomery S 1, Allen F 2, Adlard N 2 1Costello Medical Consulting Ltd., Cambridge, UK, 2Novartis Pharmaceuticals UK Limited, Surrey, UK . . . . PRM29 Should Changes in Drug Price Over Time be Considered in CostEffectiveness Analyses? Millier A 1, Briquet B 1, Aballea S 1, Toumi M 2 of Marseille, Marseille, France . . PRM30 Estimating Costs in A Cost-Effectiveness Analysis: Adherence to HTA Guidance Mauskopf J A 1, Mitchell S E 2, Samuel M 2 Health Solutions, Research Triangle Park, NC, USA, 2RTI Health Solutions, Manchester, UK . . . . . 1RTI Objectives: Since the results of a cost-effectiveness analysis (CEA) are generally sensitive to the input cost parameter values selected for difference disease-related outcomes a systematic approach should be used to derive these estimates as suggested in HTA guidance. To determine the extent to which a systematic approach was used to select disease-related cost estimates for inclusion in CEAs for new hepatitis C treatments. Methods: A systematic literature review of primary costing studies and of the cost data used in published CEAs was performed for different stages of liver disease for those with chronic hepatitis C infection. The process described in the cost-effectiveness analyses by which they selected the input base-case cost values as well as the ranges used in the sensitivity analyses was reviewed to determine whether or not a systematic approach was used to identify primary cost studies and whether or not the a rationale was supplied for the values selected. Results: The hepatitis C systematic review focused on US costs and cost-effectiveness analyses. In most of the hepatitis C cost-effectiveness analyses, the cost estimates used were either taken directly or derived from recent primary cost studies. However, a systematic review was not generally used to identify the recent primary cost studies. In addition, the method used to adapt the data from the selected studies for use in the CEA was either not explained and/or appeared to be incorrect in some of the CEAs. In most of the CEAs, sensitivity analyses assumed arbitrary ranges for the cost estimates (for example, plus or minus 50%) rather than using ranges from alternative cost studies. Conclusions: Very little detail is provided in published CEAs about the methods used to identify primary disease-related cost studies and a rationale for selection of the costs is generally not provided. . Objectives: Several new disease modifying therapies have recently received marketing authorisations for the treatment of relapsing-remitting multiple sclerosis (RRMS). Given the recent appraisal by NICE of these therapies, the objective of this study was to systematically review and critically evaluate the techniques used in modelling relapsing-remitting multiple sclerosis in the UK. Methods: Embase, Medline, Cochrane Library and the NICE website were searched systematically on 03.03.14 to identify articles relating to cost-utility models in RRMS with a UK perspective. Data sources, techniques and assumptions of the included models were extracted, compared and critically evaluated. Results: Of 385 search results, 25 full texts were evaluated and 17 articles (relating to 12 different models) were included. Early models varied considerably in method and structure but convergence was apparent over time towards a Markov model with states based on disability score, a 1-year cycle length and a lifetime time horizon. More recent models also allowed for disability improvement within the natural history of the condition. Considerable variety remains, however, with an increasing number of comparators over time, the need for treatment sequencing and different assumptions around efficacy waning and treatment withdrawal. Additionally, modelling techniques were sometimes implemented inappropriately. Confidential data sources were frequently used, especially within the models submitted to NICE. Conclusions: Despite a convergence over recent years to a similar Markov structure, there are still significant discrepancies between the models simulating the course of RRMS in the UK. Differing methods, assumptions and data sources make the comparison of models, and their results, problematic. The Markov structure commonly used also leads to problems such as an incapability to deal with heterogeneous populations and multiplying complexity with treatment sequences; these would best be solved by using alternative model types such as discrete event simulations. . dynamic incremental cost-effectiveness ratio (ICER), at the time of market entry and by year thereafter. While experts were not aware of any existing guidelines, the predominant view was that although using the brand price for the studied drug would be a conservative approach, it is reasonable to account for price reductions after patent expiration for all drugs considered. Conclusions: Drug price variations may introduce a source of uncertainty in CEA, as both timing of entry and level of generic drug pricing are unknown. There is currently no consensus on how this should be considered. Failure to incorporate generic drug entry in CEA is likely to yield overestimates of ICER for treatments used over long-term. . PRM31 Identifying the Broader Value of Vaccines in Low and Middle Income Countries Van der Putten I M 1, Hiligsmann M 1, Paulus A T G 1, Hutubessy R 2, Evers S M 1 University, Maastricht, The Netherlands, 2World Health Organization, Geneva, Switzerland . . . . . . . . . 1Maastricht Objectives: Current economic evaluations of vaccine immunization strategies mainly concentrate on immediate health gains (measured in metrics such as QALYs or DALYs) and household cost savings. Vaccine immunization strategies, however, often take place within a broader societal context. In order to financially sustain these strategies, economic evaluations should not only encompass immediate health gains and household costs but also the ‘broader value of vaccines’. This study aims to identify the relevance of information with regard to the broader value of vaccines for decision makers in low and middle income countries. Methods: Several methods were used to identify the broader value of vaccines including a literature review, a survey, interviews and consultations with experts. The long-term effects of those who were vaccinated and the effects experienced by society as a whole, including non-vaccinated community members, were included in a framework. Results: In total, twenty broader values in five different domains were identified. The first domain included long-term productivity gains. These gains refer to the individual long-term productivity due to better physical and mental health as well as to the economic consequences of decisions made by households due to improved child survival. The second domain consists of ecological values which are related to the decline of prevalence and incidence of vaccine related diseases. The third domain encompasses different types of equity considerations. The fourth domain includes the impact of vaccine strategies on other health interventions. Finally, the fifth domain includes macroeconomic effects, such as the impact of vaccine immunization strategies on GDP tax revenues and overall government savings. Conclusions: Several broader economic values outside the health care sector were identified. These results provide the input for the incorporation of these values in economic evaluations. Further research is needed to identify the most important broader values for national decision makers. 1Creativ-Ceutical, Paris, France, 2University Objectives: Cost-effectiveness analyses (CEA) are used to support funding decisions for new drugs by estimating their clinical and economic value. While prices of drugs may fall over time due to market competition, entrance of generic drugs, or negotiated price cuts, this is rarely accounted for in CEA. The objective is to review pharmacoeconomic guidelines and current practice around drug price evaluation in CEA. Methods: Pharmacoeconomic guidelines were reviewed to identify countries in which a drug price modification over the time horizon is allowed or recommended in CEA. Then methodological articles and published CEA using price modifications were identified. Finally, several health economics experts were interviewed. Results: Only 3 pharmacoeconomic guidelines report recommendations around price adjustment in CEA (Norway, New-Zealand): modellers should take into account changes in drug price over time. In France, it is possible to include a generic price in sensitivity analysis. In other countries, this was not mentioned in the guidelines. Methodological articles mentioned the possibility to use an estimated 4% decrease over time in UK. In most of published CEA incorporating price modifications, this was performed as secondary analyses. Other CEA reported PRM32 Proposal for a Comprehensive Definition of Budget Impact Analysis Bierbaum M Friedrich-Alexander-Universität Erlangen-Nürnberg, Nuremberg, Germany . Objectives: To our knowledge in most articles BIA is only defined as what it does. Some authors have tried to define it by comparing it to cost effectiveness analysis. But still there is no common stand-alone definition of the term Budget Impact Analysis available. Our aim is to provide such a definition. Methods: In the course of a PhD thesis we conducted a systematic literature review in order to identify methodological articles regarding budget impact analyses. We searched pudmed and seven other databases to identify relevant articles. From the eligible articles the different understandings and definitions of BIA were extracted and synthesized into a comprehensive definition. Results: Our search delivered 223 articles from which 28 met our inclusion criteria. 15 different approaches to describe BIAs were identified. Over the years (2001 to today) there was a constant improvement and increase of complexity in the descriptions. Nevertheless most of the late definitions are based on the work of A549 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Mauskopf et al. in 2005 and 2007. Conclusions: Based on the results we suggest the following definition for BIA: “A budget impact analysis is a form of health economic evaluation. Its purpose is to predict the financial consequences of the introduction or removal of an intervention from the current health care setting. Therefore BIA is a framework to synthesize the best available evidence. Rather than calculating a precise impact number it provides a valid model to the decision maker, enabling him to understand the relative effects of his decisions. The analysis is undertaken by modeling two scenarios where the reference scenario is the status quo and the second scenario a simulation of the decision to be made. The model parameters are chosen according to the framework requirement of the decision maker.” PRM33 Episodes of Care and their Costs Based on ICPC-2 Classification: Three Month Follow-Up Study in Finland Heinonen J 1, Soini E 2, Ryynänen O P 3, Koskela T 1 1Department of General Practice, University of Tampere, Tampere, Finland, 2ESiOR Oy, Kuopio, Finland, 3Department of Public Health and Clinical Nutrition, University of Eastern Finland, and General Practice Unit, Kuopio University Hospital, Kuopio, Finland . . . . . Objectives: To explore patient characteristics, resource use and costs related to different episodes of care (EOC) in Finnish health care. EOC is a health problem needing testing, diagnosis, care or follow-up from its first presentation by the patient to health care until the completion of the last health care contact for it. Literature around costs of episodes (COE) is scarce. Methods: Primary and secondary care data was collected during the three months prospective, nonrandomized follow-up study (Effective Health Centre) using questionnaires and electronic health record. Setting included three primary health care practices in Pirkanmaa, Finland. 622 (41% of potential) patients were recruited during one week period. Patients that had doctor/nurse appointment on the recruiting day and agreed to participate were included. Patients visiting specialized health guidance clinic for pregnant women, children and mothers were excluded. The main outcome measures were patient characteristics, resource use and costs classified based on the International Classification of Primary Care (ICPC-2) episode title codes. Resource use was valued with health care providers’ 2012 unit costs. Social Insurance Institution costs (e.g. outpatient drugs) were excluded. Results: On average, patient had 1.22 EOCs during the three months. Patient characteristics and resource use differed between the EOC-classes. Class L ‘Musculoskeletal’ had the highest number of episodes (17%). The most common (8%) single EOC was ‘upper respiratory infection’. The mean COE was € 390 (SE € 61) and the median COE was € 165 (IQR € 118-289) during the three month follow-up. The most expensive class was K ‘Circulatory’, with a mean COE of € 910. The most expensive single COE (€ 32,546) was in the group K. The most expensive one percent of COEs summed up covered 36% of total COEs. Conclusions: Patient characteristics, resource use and costs differed between the ICPC-2 classes, which could be taken into account in evaluations, planning and pricing. PRM34 Do the Us Panel Recommendations Hold for Europe? Investigating the Relation Between Quality of Life Versus Work-Status, Absenteeism and Presenteeism Knies S 1, Boonen A 2, Severens J L 3 Health Care Institute, Diemen, The Netherlands, 2Maastricht University Medical Center, Maastricht, The Netherlands, 3Erasmus University Rotterdam, Rotterdam, The Netherlands . . . . 1National Objectives: In the last twenty years there has been an intense debate on how to value lost productivity in economic evaluations. According to the Washington panel, lost productivity influences health-related quality of life (HRQoL) and should thus be considered a health effect instead of a cost to avoid double counting. Until now empirical evidence on the inclusion of income loss when valuing health states is not decisive. We examine the relationship between three aspects of lost productivity (work-status, absenteeism and presenteeism) and patient or social valuation of health-related quality of life. Methods: Cross-sectional survey data from a total of 830 respondents with a rheumatic disorder from four Western-European countries. Health-related quality of life was expressed in either the European societal utility using EQ-5D-3L or the patient valuation using EQ-VAS. Linear regression analyses were performed to examine the impact of work-status (four categories), absenteeism (absent from paid work during the past three months), and presenteeism (QQ method) on EQ-5D utilities and VAS scores taking demographic characteristics and disease severity (duration, pain and restriction) into account. Results: The relationship between work-status, absenteeism or presenteeism and HRQoL is stronger for patient valuation than societal valuation. Compared to work-status and presenteeism is the relationship between absenteeism and HRQoL even less explicit. However, results for all measures of work are only marginal significant and negligible compared to the influence of restriction due to disease we studied. Conclusions: In four European countries, analyses among patients with a rheumatic disorder do not fully support the claim of the Washington panel that lost productivity has a significant relationship with HRQoL, and this is even more apparent for absenteeism than for work-status and presenteeism. Therefore absenteeism should continue to be included in the costs and not in the QALY. Findings need to be confirmed in other disease areas. PRM35 Cost of Previously Treated Chronic Lymphocytic Leukemia (CLL) and Indolent Non-Hodgkin’s Lymphoma (INHL) in the United Kingdom (UK) Cognet M 1, Druais S 1, Gervais F 1, Gauthier A 1, Abrams K R 2 of Leicester, Leicester, UK . . . . . . 1Amaris, London, UK, 2University Objectives: Accurate cost data are required to inform cost-effectiveness assessments of novel treatments in the UK for NICE appraisals. Up to now, levels of resource use to manage CLL and iNHL have mainly been based on clinical expert opinion. However, recently, two key primary care databases in the UK (THIN and CPRD) were linked with Hospital Episode Statistics (HES) at the patient level, thereby providing additional information on secondary care in England. This study aimed to generate more accurate resource use for the management of CLL and iNHL by using the THIN-HES database. Methods: First, a MEDLINE and UK Health Technology Appraisals (HTAs) reviews were undertaken to identify studies documenting the cost of previously-treated iNHL and CLL in the UK. Then, to collect patients health care resource use, THIN database linked to the HES dataset was analysed. Results: Three HTAs were identified as relevant, and cost estimates relied on assumptions from clinical experts. Assumptions varied as TA193 related to relapsed CLL assumed that health care visits were three times more frequent post-progression (3 consultations/month: £86) than pre-progression (1 consultation/month: £28.67) while another, TA202, assumed a rather constant number of visits across the two health states (1 clinic visit per month: £121.11). Therefore, analyses of the THIN database linked to HES were undertaken, including more than 1 000 patients. OPCS4 codes and READ codes in the HES and THIN databases respectively, were used to identify treatments prescribed and procedures undertaken. Costs were estimated by applying unit costs from national references. Conclusions: To our knowledge, this analysis is the first retrospective observational study to assess the cost of managing previously-treated CLL and iNHL in the UK. This study will serve as an important resource in the health economic evaluation of emerging therapies. This method suggests a greater standardization of disease management costs across HTAs. PRM36 A Systematic Review of Methods to Assess the Economic Impact of Air Pollution Conti S , Fornari C , Madotto F , Cesana G University of Milano - Bicocca, Monza, Italy . . . . Objectives: Despite the fact that short and long-term effects of the exposure to air pollution on health have been extensively analyzed, estimates of the health care economic impact of such effects are still limited. We therefore carried out a systematic review of the literature, with the aim of identifying the current major research focuses in the field and the topics that will need to be addressed in the future. Methods: We searched the electronic databases MEDLINE and EMBASE, in which we applied respectively the following algorithms: 1) “((“cost of illness” [MeSH Terms] OR “health care costs” [MeSH Terms] OR “health expenditures” [MeSH Terms]) AND “environmental pollution” [MeSH Terms]) OR (Pollution [Title/Abstract] AND (Expenditure [Title/Abstract] OR Expenditures [Title/Abstract] OR cost [Title/ Abstract] OR costs [Title/Abstract]) AND (health [Title/Abstract] OR health care [Title/ Abstract]))“; 2) “‘health care cost’/exp AND ‘pollution’/exp”. Searches were limited to article written in English and Italian, without any date restriction. Results: The initial selections identified 775 records in MEDLINE and 466 in EMBASE, 149 of which were classified as relevant. They focused on a wide range of pollutants, including volatile organic compounds, nitrogen dioxide, pesticides, ozone, particulate matter and tobacco smoke. Most of the studies assessed the health impact of environmental pollutants using direct and indirect cost estimates acquired from literature, mainly relying on cost of illness methods; 27 papers used an individual direct health care costs approach, but they usually didn’t involve indirect costs in the final computations. Finally, only a few studies distinguished between short-term and long-term effect of air pollution. Conclusions: The results of our review identified two main topics that deserve further research: future health impact assessments should integrate indirect costs estimates with information from the direct modeling of real-life health care costs; the short- and long-term economic impacts should be clearly separated. PRM37 Cost-Effectiveness Analysis of Ipilimumab in Previously Untreated Patients With Unresectable Malignant Melanoma in Scotland Lee D 1, Porter J 1, Hatswell A J 1, Hertel N 2, Walker A 3 Myers Squibb, Uxbridge, UK, 3University of Glasgow, glasgow, UK . . . . . . 1BresMed, Sheffield, UK, 2Bristol Objectives: This analysis assessed the cost effectiveness of ipilimumab 3mg/ kg as first-line treatment for metastatic melanoma. As ipilimumab has an existing second-line recommendation, the decision problem is ipilimumab first-line followed by best supportive care (BSC), compared with Scottish clinical practice - dacarbazine or vemurafenib first-line followed by ipilimumab. Methods: In line with SMC requirements, an area under the curve model was built comparing firstline ipilimumab, dacarbazine and vemurafenib. The model utilised progression, survival and utility data from CA184-024 for ipilimumab/dacarbazine and dacarbazine, survival data from MDX010-20 for ipilimumab second-line, and survival and progression data from BRIM-3 for vemurafenib. MDX010-20 and observational data, using the approved regimen, were tested within scenario analyses assessing the performance of ipilimumab 3mg/kg at first-line. 2013 costs were taken from Scottish or UK official sources. Results: Economic analysis, including patient access schemes for ipilimumab and vemurafenib, shows that ipilimumab firstline followed by BSC is cost-effective versus dacarbazine first-line followed by ipilimumab (incremental costs: £10,502, incremental quality-adjusted life-years [QALYs]: 0.33, incremental cost-effectiveness ratio [ICER]: £31,481). Compared with ipilimumab first-line followed by BSC, the sequence vemurafenib first-line followed by ipilimumab is associated with incremental QALYs (0.26) but also incremental costs (£33,306), resulting in a not cost-effective cost/QALY trade-off (ICER = £126,482), i.e. ipilimumab first-line should be the preferred option. A scenario analysis that compared ipilimumab first-line with vemurafenib first-line alone resulted in ipilimumab being the dominant treatment option. Comprehensive sensitivity analyses identified survival parameters as having the largest impact on model results. Ipilimumab remained cost-effective at a threshold of £50,000 per QALY gained against both comparators. Conclusions: First-line ipilimumab treatment for melanoma is cost-effective, and as a first-line option it would expand clinician choice, enabling selection of the most appropriate therapy for patients depending on their disease characteristics and BRAF mutation status. A550 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PRM38 Clusters of Health-States Valuations Garcia-Molina M 1, Chicaiza-Becerra L A 1, Rincon C J 2, Romano G 2 1Universidad Nacional de Colombia, Bogotá, Colombia, 2Universidad Nacional de Colombia, Bogota, Colombia . . . . . . Objectives: To identify groups of countries with similar health preferences. Methods: Cluster analyses were performed for the 242 states of the general population EQ-5D valuations for 13 published studies based on Time Trade Off; and for 10 studies based on the Visual Analogue Scale. The perfect health state was not included. Cluster and their optimal number were identified by means of the Ward algorithm with the Euclidean measure and the hierarchical clustering technique. The identified clusters in each case are compared in order to find out whether they coincide. Results: 3 clusters were identified for TTO: 1) Germany, Argentina, Poland, The Netherlands, Denmark; 2) Japan, South Korea, USA, Hispanic USA, Zimbabwe; 3) Spain, Chile, UK. 4 clusters were identified for VAS: 1) Belgium, New Zealand, Germany; 2) Europe, UK, Spain; 3) Denmark, Slovenia; 4) Finland, Argentina. Countries are not in the same clusters for the two methodologies. Only the UK and Spain belong in the same groups in both cases. Conclusions: Healthstate valuations tend to be clustered in a few groups of countries but the groups differ according to the methodology. This suggests that Visual Analogue Scale results may not be a good approximation to Time Trade Off. PRM39 IMTA Productivity Cost Questionnaire (IPCQ) . . . . PRM42 A Methodology for Estimating the Population of Advanced or Metastatic EGFR M+ Non-Small Cell Lung Cancer Patients in the UK and Ireland Mildred M Boehringer Ingelheim Ltd, Bracknell, UK . Bouwmans C 1, Krol M 2, Brouwer W 2, Severens J L 1, Koopmanschap M A 2, Hakkaart L 1 1Erasmus University Rotterdam, Rotterdam, The Netherlands, 2Erasmus University, Rotterdam, The Netherlands . parameters that can be combined to produce suitable measures of cost (c) and clinical benefits (e) associated with an intervention. Within the Bayesian framework (which is the natural environment for BCEA), this amounts to estimating a posterior distribution for the pair (e, c). Health economic evaluations then proceed by computing some relevant summaries of the resulting decision process: is the innovative intervention t1 more “cost-effective” than the standard intervention t0?. Methods: BCEA provides a set of functions that can be used to produce a standardised analysis, by synthesising the decision process given the current evidence and uncertainty, as well as producing several indicators that can be used to perform Probabilistic Sensitivity Analysis (PSA) to parameter and model structure uncertainty. These include the Cost-Effectiveness Acceptability Curve and the analysis of the Expected Value of Information, which can be used to prioritize research. Results: BCEA uses as inputs vectors of simulations from the distributions of the average costs and benefits. This naturally fits the Bayesian framework, but a frequentist analysis can also be carried out by using tools such as the bootstrap. There is scope for linking R and programs such as Excel to facilitate a comprehensive health economic analysis, including extensive PSA. Conclusions: In this talk, I will present the main feature of BCEA and its applicability to the wider context of health economic evaluation and cost-effectiveness analysis. . . . Objectives: Productivity costs often reflect a large part of the total cost in economic evaluations adopting a societal perspective. Currently, no consensus exists on how productivity losses are best measured. We aimed to develop a standardized instrument for measuring productivity losses to enhance the comparability and generalizability of the outcomes of economic evaluations. Methods: A focus group of well-experienced researchers in the field of measuring and valuing productivity losses for use in economic evaluations assessed the instruments’ main quality criteria including: building on pre-knowledge and evidence on items’ reliability and validity, inclusion of all relevant domains of productivity losses, allowing for quantifying productivity losses suitable for self-report. A feasibility study was performed to check on consistency and intelligibility of the questionnaire and applicability for different valuation methods. Results: The focus group identified three separate aspects of productivity losses leading to three modules in the iPCQ. Questions for measuring absenteeism and presenteeism are evidence- based originating from the Short-Form Health & Labour Questionnaire and PRODISQ. As evidence regarding measurement of losses of unpaid work is lacking, the questions of this module were developed during brainstorm sessions, based on similar questions on paid work. To enhance the instruments’ feasibility and responsiveness the draft version was translated into language level 1 by an agency specialized in language and clear writing. The feasibility study included 195 respondents aged > 18 years. Five percent identified problems while filling in the iPCQ, including the questionnaire’s instructions and routing (n=8) and wording (n=2). Conclusions: The iPCQ is based on previously available instruments and satisfies the current scientific state of play in productivity cost measurement and valuation. The instrument is understandable for the vast majority of the general public including low-educated people. To enhance the applicability of the iPCQ for national and international studies a translation in English is performed. Objectives: Budget impact models (BIMs) which demonstrate the economic impact of introducing or increasing the use of specific treatments are routinely used to assist the NHS with financial planning. A core component of any BIM is the estimation of the eligible patient population. The objective of this study was to identify an appropriate methodology for estimating the size of the stage IIIb/IV EGFR M+ nonsmall cell lung cancer (NSCLC) patient population eligible for first-line treatment with a tyrosine kinase inhibitor such as afatinib (GIOTRIF®). Methods: A review of the approach taken by NICE in the costing statements of all treatment options for patients with advanced (stage IIIb) or metastatic (stage IV) EGFR M+ NSCLC was conducted. The costing statements of tyrosine kinase inhibitors afatinib, erlotinib and gefitinib were reviewed, as was the costing statement for the chemotherapy agent pemetrexed. Results: Based on the reviewed approaches, the calculation can be broken down into six discrete steps from the estimation of the general population to the target population: (1) Incidence of lung cancer; (2) Proportion of NSCLC; (3) Proportion with stage IIIb/IV NSCLC; (4) Proportion who receive first-line chemotherapy; (5) Proportion with EGFR mutation status; and (6) Proportion who are EGFR M+. A detailed breakdown of the methods used to calculate the patient population eligible for treatment with afatinib was not available in the respective NICE costing statement; however the eligible population estimated by NICE validates that this approach is reasonable. Conclusions: The methodology employed by NICE to estimate the proportion of stage IIIb/IV EGFR M+ NSCLC patients was broadly consistent across all costing statements considered. Is it reasonable to assume that this approach, used to estimate the population of stage IIIb/IV EGFR M+ NSCLC patients in England and Wales is also applicable in Scotland and Ireland. PRM43 Are Care-Seekers Good Candidates for Subgroups Cost-Effectiveness Analyses? Rapp T 1, Sirven N 2 of Paris Descartes, Paris, France, 2Université Paris Descartes, Paris, France . PRM40 Feasibility of the Headroom Analysis in Early Economic Evaluation of Innovative Diagnostic Technologies With no Immediate Treatment Implications Van Nimwegen K J Radboud university medical center, Nijmegen, The Netherlands . . Objectives: There is a growing need for early evaluation of innovative technologies to prevent ineffective and expensive technologies to be widely diffused in health care. The headroom method was introduced for early determination of the potential value of new technologies. In this study we explore the feasibility and usefulness of the headroom method in the early assessment of diagnostic technologies with no immediate treatment implications. Methods: We applied the headroom method to the implementation of whole exome sequencing (WES) into the current diagnostic trajectory of complex pediatric neurology. We determined the room for improvement regarding health-related quality of life (HRQoL), diagnostic yield and the duration of the current diagnostic trajectory. Results: The headroom in a certain diagnostic trajectory can be calculated after the so-called effectiveness gap is established and monetised. The preferred measure for the effectiveness gap is HRQoL expressed in quality-adjusted life years (QALYs). Since the direct product of diagnostics is information, and not improved health, no impact on HRQoL is expected. Other measures, such as diagnostic yield, can also be used to calculate the effectiveness gap. Unlike QALYs, these appeared difficult to monetise, however. Despite this difficulty, effectiveness gap calculation using these effect measures is very informative on the room for improvement in current clinical practice. In combination with foreseeable downstream costs and savings due to a new technology it gives an idea of the potential societal value of this technology. Conclusions: Despite some methodological challenges, the headroom method proved to be potentially useful in early health economic evaluation of diagnostic technologies with no immediate treatment implications. PRM41 BCEA: A R Package to Perform Bayesian Cost-Effectiveness Analysis Baio G University College London, London, UK . Objectives: BCEA is a R library specifically designed to post-process the result of a health economic model. Typically, this consists in the estimation of a set of relevant . 1University Objectives: There is a growing need to consider heterogeneity in cost-effectiveness analyses (CEA). To capture heterogeneity, subgroups analyses have been performed using various socio-demographic and clinical variables. However, the results of these subgroups CEA can be considered inequitable. Consequently, there is need to find new subgroups that can be used for decision-making. Methods: We explore whether subgroups defined by care-seeking behaviors are good candidates for CEA subgroup analysis. Care-seekers are defined as patients who received both an early diagnosis and an early treatment. We use data from the PLASA study, a French randomized controlled trial designed to reduce the rate of functional decline in Alzheimer’s disease: 1,131 patients were randomized in an intervention group and in a control group and were followed during a 2-year period. We use a sample selection model to explore whether the unobserved heterogeneity associated with the early diagnosis decision is correlated with the unobserved heterogeneity associated with the early treatment decision. We use a fixed-effect model to explore whether the rate of functional decline was lower within the care-seekers subgroup. Results: Our theoretically grounded selection model shows that the care-seeking behavior is associated with unobserved preferences, motivating the need to run subgroup analyses within a subgroup of care-seekers. Our fixed-effect model results show that on average, the clinical intervention was not effective. However, the intervention was effective within the subgroup of care-seekers. Care-seekers who received the intervention did not face a significant decline in their functional status over the 2-year study period. On the contrary, care-seekers in the control group lost on average 9 points of ADCSADL per year (p< 0.01). Conclusions: Stratifying CEAs by care-seekers subgroups seems relevant. Our analyses can be easily implemented by adding three questions in the clinical protocol. PRM44 A Review of The Utility Values Used in Published Cost-Effectiveness Analyses of Angiotensin-Converting Enzyme Inhibitor or Angiotensin Receptor Blocker Therapy in Patients With Diabetic Nephropathy Paczkowski R 1, Kennedy-Martin T 2, Rayner S 2 Lilly and Company, Inc., Indianapolis, IN, USA, 2Kennedy Martin Health Outcomes, East Sussex, UK . 1Eli . . A551 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Diabetic nephropathy (DN) is a progressive kidney disease that occurs in around 40% of patients with diabetes, and the recommended treatment is an angiotensin-converting enzyme inhibitor (ACEi) or angiotensin receptor blocker (ARB). Individuals’ health-related quality of life (HRQoL) can be summarised by utility values, which reflect preferences for different health states, and are used in cost-utility analyses (CUAs). This literature review identified the utilities used in CUAs of ACEi/ARB treatment in patients with DN. Methods: A combined protocol was developed to identify CUAs of ACEi/ARB treatment in DN as well as studies that estimated utilities in DN (for a separate review). We searched electronic databases (MEDLINE, EMBASE, NHS EED, HEED, CEA Registry, EconLit, RePEc, HTA) from 1 January 2000 to 31 July 2013. Search results were assessed for relevance by two reviewers. For eligible CUAs, data extracted included health states with assigned utilities. Results: Of 5236 references identified from the combined search, 28 CUAs remained after review of titles/abstracts, and eight after full text review. The CUAs assessed treatment (ACEi n= 3; ARB n= 4; both n= 1) in patients with type 1 diabetes (n= 2), type 2 diabetes (n= 4) or unspecified diabetes (n= 2). Seven CUAs used a Markov model and one a decision-analysis tree. Health states in the models included microalbuminuria, proteinuria, nephropathy, end-stage renal disease (ESRD), nonESRD, dialysis and transplant. ESRD was the most frequent, with utilities ranging 0.53-0.69 (n= 6). Utility data used in the models were sourced from a number of different studies. Conclusions: For CUAs of ACEi/ARB therapy in patients with DN, few health states were included, and there was variation in values for the same health state due to a range of different reference sources being used. It is important for future economic models of DN therapies that robust utility values are available for included health states. PRM45 Modelling Dependence Between Disability Status and Health Service Costs of People With Rheumatoid Arthritis in Hungary Rakonczai P 1, Nagy B 1, Rojkovich B 2, Gáti T 2 Consulting Ltd., Budapest, Hungary, 2Buda Hospital of Hospitaller Brothers of St. John, Budapest, Hungary . . . . 1Healthware Objectives: The main objective of this study is to estimate the impact of the level of functional status and disability on health service costs related to rheumatoid arthritis (RA) disease in Hungary. It is straightforward to think that higher disability implies higher costs, where the nature of the relationship is unclear. In order to explore the dependence structure a novel approach is proposed. Instead of fitting trend lines for the cost by regression methods the entire bivariate distribution was modelled. Methods: Health Assessment Questionnaire’s disability index data were collected for 497 RA patients (with 2594 observations) treated in the Arthritis Center (AC) Buda Hospital of Hospitaller Brothers of St. John from 1st January 2005 to 31st July 2013. The same patients were also found in the database of National Health Insurance Fund Administration (NHIFA) and further parameters as e.g. relevant treatments and health service costs (in- and outpatient) were collected. After merging AC and NHIFA database the 2 dimensional patterns of the HAQindex measurements versus costs (sum of relevant costs in the following quarter year) become available for bivariate modelling. The ingredients were the empirical distribution of HAQ-index and quarter year cost, and some parametric copula families (elliptical or Archimedean) capturing the (possibly non-linear) dependence structure. The performance of the different model assumptions were compared by goodness-of-fit procedures. Results: The fitted bivariate distribution based on the best-performing dependence model are shown in the original “HAQ-index vs. cost” scale. The differences of average costs for low/medium/high HAQ-index values are summarized and the conditional distribution functions of costs are presented, respectively. Conclusions: It has been proved that there is a significant positive dependence between the disability status of RA and health service costs. The dependence cannot be considered as linear but this non-linearity can be tackled easily by using copula methods. PRM46 Pharmacy Cost Calculator for Hepatitis C Virus Patients in Turkey Baser E 1, Baser O 2, Altinbas A 3, Kariburyo M F 4 1STATinMED Research and Gazi University, Ankara, Turkey, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA, 3Diskapi Yildirim Beyazit Education and Research Hospital, Gastroenterology Clinic, Ankara, Turkey, 4STATinMED Research, Ann Arbor, MI, USA . . . . . Objectives: To design a user-friendly cost calculator to estimate and project health care costs of patients diagnosed with hepatitis C virus (HCV) infection in Turkey. Methods: We used Visual Basic in a Microsoft Excel to program complicated models within an easy-to-use framework for providers and payers to calculate pharmacy costs for HCV patients in Turkey. The calculator, first, uses the starting year to the end year as an input. The user then enters prevalent and incident patient numbers. Medication types, medication names, prescription numbers per year and prescription costs per medication are also entered. The calculator allows combination therapies. Depending on response rates, rates of incident and prevalent patients and medications type, prescription rates from 2014 and forward (until 2020) can then be calculated and projected. The calculator can be used for any country as long as the inputs are available through literature from real-world or simulation studies. Results: We used a hypothetical example to project pharmacy costs in Turkey in 2014 and 2015. We assumed 10,000 patients in 2014 (20% were assumed to be treatment naïve) and expected 2,000 new patients in 2015. We assumed ribavirin and telaprevir combination therapy for 2014 treatment naïve patients, and peginterferon alfa-2a and ribavirin for prevalent patients. The response and treatment rates were assumed at 80% for patients who were eligible for treatment. The recurrence rate was assumed to be approximately 10% after treatment. Pharmacy costs for HCV were calculated at € 7 million in 2014 and approximately € 2 million in 2015, using these rates. Conclusions: The economic burden of HCV in Turkey is significant. A simple-to-use calculator that uses real-world data and econometric models to estimate health care costs of patients with HCV can help payers and providers to make improved evidence-based health care decisions. PRM47 A De-Novo Economic Model to Assess Clinical and Economic Consequences of Bronchiectasis Bhattacharyya S B 1, Calado F 2, Priedane E 3, Shirore R M 4, Haworth C S 5, Flume P A 6, Sonathi V 1, Thomas S K 7 1Novartis Healthcare Pvt. Ltd., Hyderabad, India, 2Novartis, East Hanover, NJ, USA, 3Evidera, London, UK, 4ConvergeHEALTH, Bangalore, India, 5Cambridge Centre for Lung Infection, Cambridge, UK, 6Medical University of South Carolina, Charleston, SC, USA, 7Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA . . . . . . . . . . . . . Objectives: Bronchiectasis (BE) is characterised by permanent dilation of bronchi with destruction of elastic and muscular components of their walls. Prophylactic antibiotic treatment with acute management of exacerbation episodes is an important component of treatment. Currently there are no approved therapies for BE and inhaled antibiotics and nebulized solutions are used off-label. The objective of this study was to predict clinical and economic consequences associated with BE in adult patients over their lifetime. Methods: We developed a Markov cohort model with a cycle length of 4 weeks from UK health care perspective. The model included 4 Markov states: one stable disease state defined as confirmed non - cystic fibrosis bronchiectasis and without exacerbations in the previous 4 weeks and three exacerbation states defined on the basis of severity of exacerbation and associated health care resource utilization. Disease state specific costs included costs of drugs, consultations/visits, diagnostic test/procedures, physiotherapy, emergency room visits and hospitalizations. Patient baseline characteristics, exacerbation rates and utilities were estimated from literature. Results: The model suggests that ability to reduce hospitalizations and limit impact on quality of life are the major value drivers. The model is also sensitive to baseline risk of exacerbations. A treatment for BE that results in 35% reduction in exacerbation rates and related hospitalizations may result in a reduction of 15.53 exacerbations and 3.23 hospitalizations per patient over lifetime. Reduction in number of hospitalization will result in fewer long-term complications, leading to reduction of 1 death per 600 patients over a period of 10 years. The treatment could result in reduction of medical cost by £10,777 and 0.11 QALY gain per patient over lifetime. Conclusions: Any health care intervention that could reduce the number of exacerbations and related hospitalizations in the BE patients can have significant beneficial impact on clinical outcomes, costs and quality of life. PRM48 Different Strategies for Latent TB Assessment in Patients Undergoing Anti-TNF Treatment: an Economic Model Pierotti F , Trieste L , Turchetti G Scuola Superiore Sant’Anna, Pisa, Italy . . . Objectives: Since treatment with biologics may reactivate latent TB, testing and prophylaxes before initiating therapy are mandatory. However, there is not a unique solution for the number and type of tests to be used for detecting the presence of TB, and for the definition of when prophylaxis should be prescribed. Which alternative should be preferred depends on the effectiveness of treatment and also on the cost associated. This methodological study aims to define a general economic model to assess different protocols for latent TB detection in patients undergoing anti-TNF treatment. Methods: A decision tree approach has been designed for comparing alternative protocols in terms of (a) expected costs of different strategies for latent TB assessment in patients undergoing anti-TNF treatment; b) the economic convenience of the prophylaxis therapy in presence of a negative chest X-ray. Uncertainty is expressed by the probability of being positive and negative to the tests. In the NHS perspective the model considers: costs of tests; costs of TB onsets for rheumatologic patients who undergo biologics; costs of extending anti TB prophylaxis also to false negative patients; costs related to 1-month biologic therapy delay because of prophylaxis; costs for adverse events therapy and/or test related. Indirect costs are also considered in the broader societal perspective. Results: Costs assessed are useful to choose the less costly alternatives. The comparison also considers the reduction of false negatives to the tests which do not follow prophylaxis while being affected by latent TB. Conclusions: The model, that can represent a useful tool both for clinical and health policy decision making, is a general one; it can be applied to any country by inserting the country specific epidemiological, clinical, and economical data, and to any anti-TNF drug, by using the specific biologic drugrelated risk factor. PRM50 An Efficient Design for Cost-Effectiveness Studies of Personalized Medicine Strategies Leunis A 1, Redekop W K 2, Lowenberg B 3, Uyl-De Groot C A 4 1Institute for Medical Techonology Assessment (iMTA), Rotterdam, The Netherlands, 2Erasmus University, Rotterdam, The Netherlands, 3Erasmus Medical Center, Rotterdam, The Netherlands, 4Institute for Medical Technology Assessment (iMTA), Erasmus University, Rotterdam, The Netherlands . . . . . . Objectives: The aim of this study is to assess how cost-effectiveness analyses of personalized medicine approaches (PMA) can be efficiently designed by using a case study in the field of acute myeloid leukemia. Methods: The cost-effectiveness analysis of the PMA was performed in two steps. As a PMA often only causes a treatment change in a specified subgroup of the patients, the cost-effectiveness of the identified treatment change was estimated as a first step (restricted analyses). Subsequently, a full cost-effectiveness analysis was performed which included all patients and the costs of identifying the specified subgroup. The relationship between the full and restricted analyses was evaluated by varying different input parameters. Results: It was found that the full cost-effectiveness of a PMA approach could be described as a function of the (cost-) effectiveness of the treatment change in the specified subgroup, the costs of identifying the subgroup and the size of the subgroup. If no additional costs are associated with the identification of the subgroup, the incremental cost-effectiveness ratio (ICER) of the full analysis is identical to the ICER of the restricted analysis. Otherwise, the ICER A552 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 increases with higher test costs, smaller subgroup and smaller incremental effects due to the treatment change. Conclusions: The data collection of costs, survival and quality of life of PMAs should be restricted to the subgroup for whom treatment changes. Therefore, cost-effectiveness analyses of PMAs should start with identifying that subgroup. Once the size of the subgroup, the cost-effectiveness of the treatment change and the test costs are known, the full cost-effectiveness can be easily calculated. The described function can also be used to assess the potential cost-effectiveness of future PMAs. If a PMA causes a treatment change in a small subgroup, the PMA can only be cost-effective with low test costs or large incremental effects. PRM51 Methods for Health Economic Evaluations of Vaccines – Results from an International Expert-Workshop Ultsch B 1, Damm O 2, Beutels P 3, Bilcke J 3, Brüggenjürgen B 4, Gerber-Grote A U 5, Greiner W 2, Hanquet G 6, Harder T 7, Hutubessy R 8, Jit M 9, Knol M 10, Kuhlmann A 11, von Kries R 12, Levy-Bruhl D 13, Perleth M 14, Postma M J 15, Salo H 16, Siebert U 17, Wasem J 18, Weidemann F 19, Wichmann O 7 1Robert Koch Institute / Charité University Medical Center, Berlin, Germany, 2School of Public Health, Bielefeld University, Bielefeld, Germany, 3University of Antwerp, Antwerp, Belgium, 4Steinbeis University Berlin (SHB), Berlin, Germany, 5Institute for Quality and Efficiency in Health Care (IQWiG), Cologne, Germany, 6Belgian Health Care Knowledge Centre (KCE), Brussels, Belgium, 7Robert Koch Institute, Berlin, Germany, 8World Health Organization, Geneva, Switzerland, 9London School of Hygiene and Tropical Medicine / Public Health England (PHE), London, UK, 10RIVM - Centre for Infectious Disease Control, Bilthoven, The Netherlands, 11Leibniz Universität Hannover, Hannover, Germany, 12Ludwig-Maximilians-University Munich (LMU), Munich, Germany, 13Institut de Veille Sanitaire, Saint-Maurice Cedex, France, 14Gemeinsamer Bundesausschuss (G-BA), Berlin, Germany, 15University of Groningen, Groningen, The Netherlands, 16National Institute for Health and Welfare, Helsinki, Finland, 17Medical Informatics and Technology, and Director of the Division for Health Technology Assessment and Bioinformatics, ONCOTYROL, Hall i. T, Austria, 18University of Duisburg-Essen, Essen, Germany, 19Robert Koch Institute / Charité Berlin, Berlin, Germany . . . . . . . . . . . . . . . . . . . . . . . . Objectives: Health economic evaluations (HEEs) of vaccines are commonly considered during immunization introduction decision-making processes in most industrialized countries. Despite the availability of guidelines advocating more standardization for such HEEs, there are still several infection/immunization-specific particularities that are debated in the scientific community. An international expert-workshop was convened to identify good practices for (i) how to conduct HEEs of vaccines and (ii) how to consider results of HEE in vaccine introduction decision-making. Methods: A systematic literature search was conducted to identify prevailing opinions and remaining issues of HEE in vaccination. Twentytwo experts in the field of health economics and immunization decision-making were invited to a workshop and were asked to answer a survey-questionnaire based on the systematic literature search beforehand to inform the preparation of group work sessions (GWS). In GWS, issues focusing on ‘mathematical modeling’, ‘health economics’, and ‘decision-making’ were discussed and summarized. Results: The GWS (based on systematic literature search) included topics such as cost-components, quality of life (QoL), discounting, and perspectives leading to suggestions such as including caregiver QoL impact and applying decreasing time-related discount rates. Since vaccination often causes indirect effects, the use of dynamic models is required and exceptions should be justified. In order to facilitate transparent decision-making, the results of HEE should present parameter and methodological uncertainty as well as cumulative and time-specific figures. The majority of countries in Europe use results from HEEs in an informal judgment-process without willingness to pay (WTP) threshold. The expert-group emphasized that transparency should be maximized in decisionmaking process. Conclusions: The deliberations led to suggestions on several HEE issues. However, vaccines not always need to be considered differently in HEE since other interventions might share similar characteristics. Transparency in the conduct and presentation of HEE, and subsequent decision-making is essential, especially in the absence of explicit WTP thresholds. across every stage, all of which are fully transparent, provides a unique combination of broadness, resilience and inclusiveness making it an ideal decision making tool. PRM53 Comparison of Generic, Condition-Specific and Mapped Health State Utility Values for Pediatric Asthma Gialetti S 1, Trieste L 2, Pierotti F 2, Turchetti G 2, Silvestri M 1, Della Casa Alberighi O 1 1Institute Giannina Gaslini, Genova, Italy, 2Scuola Superiore Sant’Anna, Pisa, Italy . . . . . . Objectives: Many aspects of QALY measurement in children are not yet fully developed. This study is aimed at contextualizing the mapping methodology in a field not yet covered: paediatric asthma. The objective is to derive utility values from non-preference-based questionnaires, the estimated utility values will be useful for QALY assessment by means of: i) evaluation of a linear mapping between the generic preference-based EQ-5D-3L (self- and proxy-versions) and the condition-specific non-preference-based Paediatric Asthma Quality of Life Questionnaire (PAQLQ) and Paediatric Asthma Caregiver’s Quality of Life Questionnaire (PACQLQ), respectively; ii) assessment of the capability of mapping to discriminate for disease severity. Methods: Either PAQLQ or PACQLQ and EQ-5D-3L will be administrated to 170 asthma children (7-17 years of age) during a 24-month multicenter randomized placebo-controlled trial of allergen specific sublingual immunotherapy coded EudraCT No. 2012-005678-76/FARM94793N at baseline, month 12, and month 24 visits. Level of severity will be assessed using the Asthma Control Test (ACT) and the Childhood-ACT (C-ACT) questionnaires. The possibility of a linear mapping will be evaluated through a Tobit model, where either PAQLQ/PACQLQ will be tested as predictors of EQ-5D-3L answers. Capability of mapping to be sensible in changes of disease severity will be measured through a Pearson’s correlation between changes in estimated EQ-5D-3L scores and changes in ACT and C-ACT answers. Results: Linear mapping, if feasible, applies the statistical relationship between either PAQLQ or PACQLQ and EQ-5D-3L, being PAQLQ/PACQLQ answers predictors of EQ-5D-3L score. EQ-5D-3L and PAQLQ/ PACQLQ, as well as EQ-5D-3L built on PAQLQ/PACQLQ answers by applying the linear mapping is expected to discriminate for both patients’ level and changes in disease severity. Conclusions: This study seeks to inform policy makers in their choice of the source of utility values and their discrimination across severity groups and responsiveness in asthma children. PRM54 Estimating Means from Medians: A Case Study With Treatments for Metastatic Colorectal Cancer (MCRC) Ozer-Stillman I 1, Whalen J D 2, Mendivil J 3, Villegas-Sánchez J 3, Chang J 4 1Evidera, Lexington, MA, USA, 2Evidera, Inc., London, UK, 3Bayer Hispania, Barcelona, Spain, 4Bayer HealthCare, Whippany, NJ, USA . . . . . . PRM52 Applying Multiple Criteria Decision Analysis in the Context of Health Technology Assessement: an Empirical Case Study Objectives: Guidelines for economic evaluation from ISPOR and NICE recommend the use of mean values rather than medians, but do not offer guidance for situations when mean values are not available. This study evaluated the impact of different methods of estimating means from medians, within the context of estimating treatment duration for drugs used to treat mCRC. Methods: Clinical trials and prescribing information for drugs used to treat mCRC were reviewed for information on the mean, median, and range of treatment cycles. Various approaches were used to estimate mean values, including direct use of the median, a published equation considering the median and range, confidence intervals and interquartile range, and the use of distributions commonly used in survival analysis. Where possible, the estimated means were compared with reported means from clinical trials. Results: Very few studies reported both median and mean treatment duration; direct use of the median under-predicted the mean by 23-39% and the published equation over-predicted the mean by 19-28%. Simple assumptions about the distribution of treatment durations performed best, predicting the reported mean within ±12%. The use of progression-free survival as a proxy for treatment duration over-predicted treatment duration by 5-38%, although estimates were improved by accounting for early discontinuation. Conclusions: By only considering the 50th percentile, the median may not provide an accurate representation of the outcomes in a population. It is important that researchers and budget-holders are aware of the limitations in the use of medians, and that they consider multiple estimation methods to estimate mean values for economic analyses. Angelis A , Kanavos P London School of Economics and Political Science, London, UK, Past research has indicated that Multiple Criteria Decision Analysis (MCDA) approaches could be used as an alternative methodology for assessing the value of medical technologies from a Health Technology Assessment (HTA) perspective. PRM55 Adopting an Evidence Synthesis Approach for Assessing CostEffectiveness of Screening Strategies for Prostate Cancer in Ireland . . Objectives: To apply in practise an MCDA framework for the value assessment of a set of therapeutic options in regards to an oncology indication through an HTA simulation exercise. Methods: Using MCDA principles, past research outcomes and the clinical and economic literature, a disease specific value tree was constructed incorporating the values concerned as criteria. The alternative options were scored against the criteria through the development of value functions, weights were assigned to the criteria using a swing method, scores and weights were aggregated using a linear additive model, and sensitivity analysis of the results was conducted. All the stages were informed by extensive stakeholder engagement through their participation at a decision conference workshop. Results: Value parameters considered included burden of illness, therapeutic, safety, innovation and socioeconomic criteria. Overall value scores were produced reflecting the performance of the options against the criteria while considering their relative importance. Hypothetical payer’s resource allocation decisions on the coverage of the options were made on “value for money” grounds through the use of “cost-perunit of value” that was derived by incorporating purchasing costs. Conclusions: MCDA possesses the prerequisites of a value based assessment methodological framework. The multiplicity of criteria that can be incorporated to assess value, the weights that can be applied to the criteria, and the stakeholders’ involvement Burns R M 1, Leal J 1, Wolstenhome J 1, O’Neill C 2, Sullivan F J 3, Drummond F J 4, Sharp L 4 1University of Oxford, Oxford, UK, 2National University of Ireland Galway, Galway City, Ireland, 3Prostate Cancer Institutue, Galway, Ireland, 4National Cancer Registry Ireland, Cork, Ireland . . . . . . . . . . Objectives: Prostate cancer (PCa) incidence has been steadily increasing over the last twenty years, resulting in Ireland having the highest incidence rate in Europe in 2008. The main driver of this is wide-spread use of prostate specific antigen (PSA) testing in primary care as an ad-hoc detection mechanism. The objective of this research was to undertake a cost-effectiveness analysis of organised screening using PSA testing in Ireland. Methods: Using a Bayesian Multi-Parameter Evidence Synthesis (MPES) framework, non-cost parameters were synthesised, informed by incidence, mortality and clinical data from the National Cancer Registry Ireland for men diagnosed with PCa in 2009. The MPES framework estimates unobserved parameters using available evidence and so facilitates analysis of interventions where observational data is limited. Average costs were estimated using projectspecific survey costs, Irish hospital costs (HSE Casemix) and costs published in the literature. Effectiveness of PSA testing was sourced from the literature. Utility scores were collected from 2,500 PCa survivors. The cost-effectiveness analysis employed a cohort, semi-Markov model following men from age 30 to death. A range of PSA screening strategies were compared with no PSA testing and an extensive series of A553 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 scenario analysis was employed. PSA cut-off levels were varied between > 3ng/ml and > 4ng/ml reflecting European guidance and practice variation in Ireland. Costs and benefits were discounted at 5% per annum. Results: Extensive probabilistic sensitivity analyses highlighted wide variation in incremental cost-effectiveness ratios (ICERs). PSA testing may be cost effective using a once-off test at age 50 or age 55 depending on the ceiling ratio incorporated. Results for the ≥3ng/ml PSA cut-off consistently dominated those for the ≥ 4ng/ml PSA cut-off. Conclusions: This analysis illustrates the value MPES methods for economic modelling of interventions. The results contribute to the ongoing accumulation of evidence on the costs and benefits of PSA testing internationally. RESEARCH ON METHODS – Databases & Management Methods PRM56 Preliminary Steps in the Development of an Algorithm for Identifying Relapsed CLL Patients in Secondary Data Foley K A 1, Princic N 1, Bizier R 1, Hansen L G 2, Huse D M 1 Health Analytics, Cambridge, MA, USA, 2Truven Health Analytics, Northwood, NH, USA . . . . . . . . clarifying and updating ClinRO information. Methods: PROQOLID was searched on April 9, 2014 to retrieve current information about ClinROs using an advanced search engine. Results: The ClinRO information was found under the category “mode of administration” in the subcategory “clinician-rated.” Out of the 801 questionnaires in the database, fifty-two (6.5%) were identified as ClinROs. Out of these 52 questionnaires, nine were generic. Eight different therapeutic areas were identified (i.e., digestive system diseases, musculoskeletal diseases, neoplasms, nervous system diseases, respiratory tract diseases, psychiatric disorders, pathological conditions signs and symptoms, and skin and connective tissue diseases), representing 17 different indications, and 33.33% of the therapeutic areas included in PROQOLID (n=24). The most represented therapeutic area was psychiatry (n= 23) followed by nervous diseases (n= 7). Only two questionnaires were specific to children: the Pediatric Evaluation of Disabilty Inventory and the WeeFIM®. To better individualize ClinRO information in PROQOLID, it is proposed to create a new meta-category, i.e., type of COA (PRO, ClinRO, ObsRO and PerfRO). It is also recommended to expand PROQOLID to all COAs. Conclusions: This review has shown that PROQOLID already includes ClinRO information. Recommendations are given on how to modify the organization and content of the database to present information on all COAs. 1Truven Objectives: Despite advances in chronic lymphocytic leukemia (CLL) treatment, roughly 25% of first and 50% of second line patients experience relapse. Relapse, however, is not well coded in claims data and is not well documented in EMR data due to under reporting of patient status, variability in terminology used to report patient status, and change in disease progression over time. The goal of this analysis was to develop an algorithm to identify relapsed patients when patient status is not clearly documented. Methods: CLL patients in the MarketScan® Oncology EMR Database with recorded patient status were identified. Relapse was explored using two methods: 1) recorded patient status of relapse; 2) changes in laboratory data. For the first phase of algorithm development, both indications of relapse were compared to the date of treatment initiation. Laboratory data included lymphocytes, platelets, and hemoglobin. Results: Of 18,334 patients with CLL, 7,865 (43%) had any patient status reported. 528 had any mention of either relapse or remission and 73 (1%) patients had a record for relapse on the same date as a CLL diagnosis and no evidence of any other cancer types. For these 73 patients, the date of new treatment initiation had no relationship with the date of the first recorded relapse. Among these same patients, declines in hemoglobin and platelets, and increases in lymphocytes preceded treatment initiation by several days. Conclusions: Patient status does not appear to be updated regularly and documented status may not indicate decision to treat. This preliminary work suggests that lab data provide a viable source for algorithm development as they are regularly reported in the EMR and for CLL are likely linked to decision to treat. Next steps include determining the specific rule for identifying the change in lab values that triggers treatment initiation or resumption. PRM57 Occurrence, Survival And Annual Cost of Colorectal-, Breast-, Prostate- and Lung Cancer in Hungary Inotai A 1, Abonyi-Tóth Z 2, Rokszin G 2, Voko Z 1 Research Institute, Budapest, Hungary,, 2RxTarget, Szolnok, Hungary . . . . 1Syreon Objectives: Evaluating effectiveness of oncological treatments and their costs becomes more and more important with respect to the high burden of malignant diseases. The aim of this research was to estimate the occurrence, survival and health care cost of colorectal-, breast-, prostate- and lung cancer patients based on the National Health Insurance Fund (NHIF) database. Methods: Survival and cost analyses were performed on the NHIF database. Inclusion criteria: at least two consecutive ICD codes between 2000 and 2012, with a minimum of 30 days difference; or those with one ICD code, followed by death within 60 days. The following ICDs were considered: C18-C20 (colorectal), C33-C34 (lung), C50 (breast), C61 (prostate). 428 860 social security numbers met our inclusion criteria. The following indicators were estimated: number of new cases, mortality, time from diagnosis to treatment, survival and annual costs related and not related to the disease. Results: In 2011, the numbers of new cases were the following: colorectal cancer: 7299 breast cancer: 5842, prostate cancer: 3162, and lung cancer: 5499. The probability of 5-year overall survival from first diagnosis were 41.3%, 75.2%, 62.1% and 17.1%, respectively. Median time from first diagnosis to treatment initiation was less than 1 month in colorectal-, breast- and prostate cancer and less than 2 months in lung cancer. Annual cost of patient was 3166 EUR (colorectal cancer), 2585 EUR (breast cancer), 2833 EUR (prostate cancer) and 4158 EUR (lung cancer), respectively (2011 average exchange rate: 279.21 HUF/EUR). These figures indicate that annual cost of care of these malignant patients are less than half of the annual cost of kidney transplanted and haemophilia patients estimated with similar methodology. Conclusions: Data suggest that payer’s database is suitable for estimating epidemiologic and economic indicators of malignant disorders. Payer’s database analysis can support evidence-based policy-making. PRM58 Update of the Patient-Reported Outcome and Quality of Life Instruments Database (PROQOLID): Integration of The New COA Taxonomy - The Clinro Example Perrier L L , Conway K , Acquadro C Mapi Research Trust, Lyon, France . . . . Objectives: In 2002, PROQOLID was launched to provide an overview of existing PRO instruments. In October 2011, the term Clinical Outcome Assessments (COAs) was introduced to better reflect the importance of the source of information in measurements: patients (PROs), clinicians (ClinROs), and observers (ObsROs). In May 2013, a new category was added: Performance outcome assessments (PerfOs). With this evolving taxonomy, including information about all COAs might become a crucial step in developing PROQOLID. The objective of this study was: (1) To review how ClinROs are currently reported in PROQOLID; and (2) To propose (if needed) ways of PRM59 ECOA Licensing: Lessons Learned from the Copyright of COA Translations and Specificities of ECOAS Anfray C 1, Conway K 2, Acquadro C 2 1Mapi, Lyon, France, 2Mapi Research Trust, Lyon, France . . . Objectives: Electronic Clinical Outcome Assessments (eCOAs) are increasingly being used in clinical trials and their use is encouraged by regulatory authorities. Licensing is a key issue for their appropriate utilization. The objective of this abstract is to make recommendations about eCOA licensing using lessons learned from the COA translation licensing. Methods: Publications about licensing of COA translations were searched and a review of the eCOAs specificities was performed using information available from e-vendors. Results: Very few publications exist about the licensing of COA translations. The ISOQOL TCA SIG has developed a draft reflection paper which considers that translations are derivative work of original questionnaires. As such, they recommend that the copyright of a COA and its translations should be owned by a unique entity, generally the original developer to harmonize and facilitate conditions of access and use. They state that distribution should be centralized to facilitate access to questionnaires, maintain reliable information about them, and control their use. Review of the e-vendors information shows that eCOAs are often customized, with proprietary devices and softwares, and cannot be shared across users. As a consequence, there is a multiplication of e-versions for a same content. Equivalence between paper and e-versions and between e-versions is then a major concern. The review also shows that migration from paper to the electronic platform/device implies changes to the content and format of the paper version. Therefore the eCOAs can be considered as derivative works of an original COA and lessons learned from copyright of translations may apply. Examples will be provided. Conclusions: Centralized copyright ownership by the owner of the original COA and centralized licensing process for eCOAs should be discussed with all stakeholders to help controlling use and users and to protect the integrity of the instrument across e-versions by providing clear rules of e-implementation. PRM60 Mapping European Database Usage: An Analysis of Published Data Types Langham S 1, Pooley N 2, Weir S 1 1PHMR Associates, London, UK, 2PHMR Associates, Newcastle upon Tyne, UK . . . Objectives: To determine how European databases are used to support pharmacoepidemiological research. Randomised, controlled trials remain the gold standard for evaluation of drug efficacy and safety. However, the only way of identifying treatment pathways and improving understanding of real world costs and outcomes at different stages of care is via longitudinal observational studies. Observational data from electronic health records (EHRs) are essential to this pharmaco-epidemiological research. Different European databases have different strengths in terms of data types and availability. Identifying these strengths will help to select the right database for a particular study. In this context, one approach to increasing our understanding is to analyse types of published data to determine how databases have historically been used to support research. Methods: We identified peer-reviewed publications over the last 10 years from one popular longitudinal general practice patient database with some secondary care links. The publications were assigned to disease areas and study types (e.g. prevalence, resource utilisation, treatment patterns, outcomes etc). Results: Based on this mapping exercise, we identified the types of studies and the disease areas that this European database commonly supports. We also highlight gaps in disease area coverage and types of real world evidence studies and discuss potential reasons for this underuse. Conclusions: European observational data from EHRs provide increasingly important information for stakeholders of new treatment, however there are still a number of gaps in terms of disease areas and study types that these databases can support. PRM61 Using an Innovative Approach to Build a Prospective Diabetes Cohort Registry of Patients With Type 2 Diabetes in Germany: DIAREG Garcia Alvarez L 1, Rathmann W 2, Bode-Greuel K 3, Engelhard J 3, Bush S 4, SchröderBernhardi D 3, Hiller J 1 1IMS Health, London, UK, 2German Diabetes Center, Duesseldorf, Germany, 3IMS Health, Frankfurt am Main, Germany, 4AstraZeneca, Hamburg Area, Germany . . . . . . . Objectives: The lack of accessible, comprehensive sources of medical and quality of life data in Germany has partially hindered the ability to research diabetes clinical practice. The aim of this study was to build a prospective, national, multi-centre Type 2 diabetes mellitus (T2DM) registry using an innovative data collection methodology to better understand the disease specific epidemiology, treatment patterns and A554 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 patient reported outcomes (PRO). Methods: From the physician universe of the IMS®Disease Analyzer (1,500 diabetologists and 57,500 general practitioners), a subset of 101 physicians have agreed to participate. Pre-programmed inclusion criteria (i.e. ≥ 18 years old with T2DM diagnosis) triggers consecutive patient selection for further data collection. A custom software-based electronic case report form (eCRF) is completed once per quarter by the physician to capture further diabetes information. Patients complete a validated set of PROs at physician office every 6 months. Anonymised data from the electronic medical record (EMR), eCRF and PRO are linked through a unique process to ensure patient confidentiality. Results: A total of 1,029 T2DM patients are contributing to the registry (at June 2014). Preliminary analysis of the DIAREG cohort has shown comparable representativeness to IMS®Disease Analyzer for several patient characteristics: age, gender, distribution of T2DM treatment and physician type. The DIAREG improves research use of the EMR data by filling in missing information for weight, blood pressure and HbA1C (e.g. completeness of HbA1C records improved from 42.3% to 83.3% of patients). The DIAREG enhances breadth of information by collecting data that is not normally captured in EMR sources, such as physician rationale related to treatment choice or switch, target HbA1C, hypoglycaemic events, occurrence of micro or macrovascular events, and patient reported information (e.g. quality of life, treatment satisfaction and depression). Conclusions: DIAREG enables unique capability to research T2DM in Germany through multiple linked longitudinal data sources. PRM62 Insight In Health Care Databases in Asian Pacific Region Velthuis E J M PPD, Bennekom, The Netherlands . . . Objectives: Health care utilization databases or electronic medical records can be very useful to study the use and outcomes of pharmacological and therapeutical measures. There is little information on the availability and accessibility of these types of databases for health and disease management in the Asian Pacific Region. This study is performed to gain more insight in the availability and accessibility of health care utilization databases in AsiaPac. Methods: Searches were done using Google Scholar, PubMed, ISPOR’s International Digest of Databases and references in publications. Different types of databases were included in the overview, including health insurance databases, claims databases and electronic medical records from primary care of hospitals. Information extracted was: type of database, short description, population covered, start of data collection, variables included, accessibility, URL and English language yes/no. Countries included were Australia, China, Japan, South Korea, Malaysia, Singapore and Thailand. Results: Most of the databases originate in Japan and Australia. Taiwan and South Korea have a large health insurance databases covering ~98% of the population. Limited number of databases is available in the other countries investigated. Accessibility could only be derived from the websites that provide an English translation. From these websites it appears that accessibility to these databases is often limited for health policy makers and researchers due to privacy protection issues. Conclusions: Several valuable health care utilization databases exist in the Asian Pacific region. These databases could be very valuable in drug utilization and health outcomes research if easily available to all researchers. PRM63 Evaluation of Dissemination of Brazilian Network for Health Technology Assessment (REBRATS) L 1, Souza KM2 Gonçalves 1Brazilian Ministry of Health, Brasília-DF, Brazil, 2Brazilian Ministry of Health, Brasília, Brazil . . . has been believed that EMR system improved efficacy of the hospital. But a pure hospital management/financial meaning of EMR is still not clear. It is said that the total costs for maintaining EMR are 2-5% of medical revenue and they are continuously needed. In Japan, in order to increase the efficiency of a clinical trial/research, government is planning to set up one big virtual hospital using EMR and IT technology. In this study, we analyze the change in financial condition and the price of EMR system. Methods: From the hospital website and financial statement, we checked when and whether EMR was introduced. Then we analyzed the change in financial condition of the hospital that introduced EMR with the hospital that did not introduce EMR using the financial statements of 143 hospitals in National Hospital Organization (NHO). The information of the EMR price was obtained from the government website. Results: In 2011, there were 44 hospitals that had introduced EMR. Comparing the financial condition between the hospital with EMR and the hospital without EMR, there was 1% or more of difference in medical revenue. The average price of EMR for 1 year per 1 bed was US$328.8. Conclusions: The result demonstrated that the price of all EMR system did not improve hospital finance and efficiency. Because of the high cost EMR system, making one big virtual hospital using IT technology is still difficult. Therefore, we need more cost effective and easy EMR system. PRM65 A Dutch Administrative Database in Support of Economic Evaluations: A Feasibility Study Tan S S , Bouwmans C , Hakkaart L Erasmus University Rotterdam, Rotterdam, The Netherlands . . . . Objectives: To support tariff setting of Diagnosis Related Groups in the Netherlands, the hospital services of each patient are systematically collected in a national database. This database may also serve as an important data source for economic evaluations. Our study assessed the feasibility of the database to support economic evaluations. Methods: Treatment costs for acute myocardial infarction (AMI) and breast cancer were determined for 2012 from the hospital perspective, with the national database as the primary data source. Results: Using the national database as a primary data source for economic evaluations has several limitations. Firstly, the database does not contain unit costs of hospital services. Thus, we had to rely on alternative data sources such as reference prices. Secondly, the establishment of relevant patient-subgroups is not always possible because only few clinical parameters are recorded. Although we were able to identify relevant patient subgroups for AMI (ST segment vs non-ST segment elevated patients and PCI vs conventional treatment), this was not possible for breast cancer where different stages of disease could not be distinguished. Lastly, some resource quantities, such as medications and medical devices, are not routinely collected. Conclusions: Using the national database as the input of resource quantities in economic evaluations may result in valid cost estimates, especially when relevant subgroups can be identified. The cost assessment of AMI has proven that the database enables the collection of resource quantities for individual patients from a single data source and thus increases the comparability of health economic outcomes. PRM66 Impact of Influenza B in France Lamure M 1, Cohen J M 2, Pribil C 3, Garassus P 4, Auray J 5, Fleming D 6, Pujol P3 1University Claude Bernard Lyon 1, Paris, France, 2Regional Group for the Surveillance of Influenza - GROG, Open Rome, Paris, France, 3GSK, Marly Le Roi, France, 4BAQIMEHP, Paris, France, 5Cyklad Group, Rilleux la Pape, France, 6University Of Surrey, GUILDFORD, UK . . . . . . . Objectives: The Brazilian Network for Health Technology Assessment (REBRATS) spreads the HTA culture in health services and academic institutions, and also supports policy makers and managers in the decision-making process. It works through six working groups that prioritizes HTA settings and themes; develops methodological guidelines; trains professionals; and manage and disseminates information. This study aims to demonstrate REBRATS´s advances based on the productivity of the network members. Methods: Analysis of the network performance through the number of publications available on the REBRATS database; number of accesses to the REBRATS webpage obtained through the access tool and data extraction by Google Analytics, monitoring the participation on the REBRATS social networking and computing new network members. Results: Since REBRATS´ implementation in 2008, it was made available on its website 62 publications including Guidelines, HTA Bulletins and others, and 364 studies in its database. In 2014, 64 institutions are already member of the network, representing an increase of over 45% compared to 2008. Additionally, the site began to be tracked in mid-2012, when the average number of accesses was 36,659 and approximately 2037 per monthly in more than 85 countries. Since March 2012, when REBRATS social network on Facebook was created, there have been more than 300 published news with approximately 43,891 views and an average of 143 visualizations per issue reported by the network. Conclusions: The strategies for advertising the network and disseminate the HTA products produced by its members have contributed significantly to the advance of HTA in Brazil, beyond the reach of people interested in this research field. Efforts should be directed towards advertising the production internationally. Objectives: To evaluate the burden of influenza B in terms of hospitalizations and deaths in France. Methods: The analysis of the impact of influenza B used 3 French databases 1) hospitalization for respiratory diseases, 2) deaths resulting from cardiorespiratory disease, and 3) information on circulating strains. The analysis covered the period 2003 through to 2011. The study is based on an indirect approach of modeling where hospitalization, mortality and virology time series data will be extracted as monthly or weekly aggregated information. The numbers of hospitalizations and deaths attributable to influenza B were determined from positive virological analysis through a regression model (considering all influenza circulating strains classified by type and subtype) firstly over the total French population, and then by age category (0-4 years, 5-12, 15-64, 65 and above). A second method to estimate the excess of hospitalizations and deaths during influenza epidemic periods vs. non-epidemic periods was also used. Results: Over the period 2003-2011, nearly 28,000 hospitalizations and 12,000 deaths on average were associated with influenza each year. The percentages of hospitalizations and deaths attributable to influenza B were 33% (9,200) and 34% (4,000) respectively, amongst the overall population. These were observed predominantly in the elderly (> 65 years): 66% of hospitalizations and 94% of deaths. The two methods used show similar results, with a difference globally of less than 5%. Conclusions: On average each year around 9,200 hospitalizations and 4,000 cardiorespiratory deaths are associated with epidemics of influenza B in France. PRM64 Analysis of the Expenses for the Introduction of Electric Medical Record System in the National Hospital Organization Zulkifly H 1, Dingle K 2, Clavarino A 3 1Universiti Teknologi Mara, Selangor, Malaysia, 2Queensland University of Technology, Kelvin Grove, Queensland, Australia, 3Univesity of Queensland, Woolloongabba, QLD, Australia Nakagawa Y 1, Tomita N 2, Irisa K 1, Ito M 1, Nakagawa Y 1 Hospital Organization Shikoku Medical Center for Children and Adults, Zentuji city, Japan, 2National Institute of Public Health, Saitama, Japan Objectives: To identify the long term effects of common symptoms of pregnancy and whether these increase the risk of cardiovascular disease or symptoms associated with it in women at 21 years after pregnancy. Methods: Data used were from the Mater University Study of Pregnancy (MUSP), a community- based prospective birth cohort study begun in Brisbane, Australia, in 1983. Chi square test and logistic regression analyses were conducted. Results: Data were available for 3692 women. In cross tabulations, morning sickness, heartburn and backache show . . . . . 1National Objectives: Ministry of Health, Labour and Welfare (MHLW) of Japan announced the rule about electronic preservation of a medical record in 1999. Then, introduction of Electric Medical Record (EMR) into a hospital was started. Government set a target of diffusion rate of EMR to 60% in 2006, but the result was only 20-30%. It PRM67 Common Pregnancy Symptoms Increase the Risk Of Cardiovascular Disease . . . A555 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 positive association (p< 0.05) with different cardiovascular outcomes. However, in the multivariate models, only those experiencing heartburn (adjusted OR 1.3, 95% CI 1.0-1.7) during pregnancy were at greater risk of having hypertension 21 years post partum. Women experiencing morning sickness (adjusted OR 1.2, 95% CI 0.82.0) and backache (adjusted OR 1.1, 95% CI 0.6-1.7) were not considered to be at risk for future heart disease. Conclusions: As a whole, our study suggests that most common symptoms of pregnancy are not associated with an increased risk of cardiovascular disease or with hypertension in the long term. PRM68 Increased Accuracy of Distribution Based Missing Value Imputation: An Alternative to Mean Inputation in Real World Environment Survey Research Wasser T , Eisenberg D HealthCore, Inc., Wilmington, DE, USA . . Objectives: Missing values within variables can impede accurate data analysis on many levels including both univariate and multivariate analysis. This research presents distribution-based imputation (DBI), where the distribution of non-missing values is simulated to create a set of values that are then randomly inserted into to the missing values in the actual data and compares this against mean based imputation (MBI). Methods: DBI was compared to MBI in 12 different simulation conditions based on three sample sizes (50, 100, 150 and 200) and three different missing value percentages for each of the sample sizes (10%, 20% and 30%). Each simulation created 1,000 test datasets within each condition for a total of 12,000 simulated datasets. The statistical package, R was used for the sumilation. Results: MBI was biased by simulating smaller Standard Deviations, and less accurate in mean estimation than DBI in all 12 simulation combinations. DBI was more accurate in matching the number of rejected hypotheses as compared to the gold standard. Comparing the calculated p-values for bias where an unbiased estimator would demonstrate a 50/50 split being greater than and less than the gold standard, DBI was closer to the gold standard with at 48.7/51.3 split, as compared to the 25.8/74.2 split of MBI. Conclusions: DBI was found to be more accurate and unbiased as compared to MBI methods. As a result, when studies are small and do not contain a large number of variables, or in situations where more elaborate imputation methods cannot be done, DBI is an accurate and unbiased method. PRM69 INDIRECT COMPARISON OF THE EFFECTS OF ANTI-TNF BIOLOGICAL AGENTS IN PATIENTS WITH ANKYLOSING SPONDYLITIS BY MEANS OF A MIXED TREATMENT COMPARISON PERFORMED ON EFFICACY DATA FROM PUBLISHED RANDOMISED, CONTROLLED TRIALS Migliore A , Bizzi E , Massafra U S. Pietro Fatebenefratelli Hospital, Rome, Italy . . . Objectives: To compare ASAS (Assessment in Ankylosing Spondylitis Response Criteria) 20 response patterns between anti-TNF biological agents in patients with ankylosing spondylitis by means of a mixed treatment comparison of different randomised, controlled trials (RCTs) on the efficacy of biological therapies. Methods: A systematic review of literature was performed to identify a number of similarly designed double-blind, randomized, placebo-controlled trials investigating the efficacy of the TNF-alfa inhibitors etanercept, infliximab, golimumab, certolizumab pegol and adalimumab in the treatment of ankylosing spondylitis patients, conducted over an 18-years period. The endpoint of interest was ASAS20 response criteria at 12 weeks. Results were analyzed simultaneously using Bayesian mixed treatment comparison techniques. Results were expressed as odds ratio (OR) of ASAS20 response and associated 95% credible intervals (CrIs). The probability of being the best treatment was also reported. Results: 6 RCTs were selected for data extraction and further analysis. By mean of MTC, all anti-TNF agents demonstrated to be more efficacious in inducing a ASAS20 response than placebo. Infliximab shows a 67,6% of probability of being the best treatment of all. Adalimumab, golimumab and etanercept show probabilities of 17,7%, 10,6% and 4%, respectively, while certolizumab pegol showed a probability of being the best treatment of 0,1%. No differences were observed when comparing directly an anti-TNFalfa agent against another. Conclusions: Even if the mixed treatment comparisons between infliximab, golimumab, certolizumab pegol, adalimumab and etanercept did not show a statistically significant difference, this analysis suggests that infliximab, compared to placebo, is expected to provide the highest rate of ASAS20 response in SA patients naive to biologic treatments. PRM70 A Tutorial on Dimensionality Reduction in Large Claims Data Sets Juneau P Truven Health Analytics, Boyds, MD, USA . Objectives: The objective of this presentation will be to introduce the audience to various data dimension reduction techniques that may be applied in the setting of a large commercial claims data set to facilitate the task of identifying important factors or key features for use in subsequent analysis. Methods: The author will provide a brief survey of the data dimension reduction literature from areas as diverse as image analysis, neural networks, gene expression microarrays, and high through-put chemistry to demonstrate that despite that many of these techniques have been used in other settings or areas of research, their application to the analysis of health care claims data is relevant and potentially quite useful. Results: One, all-purpose, optimal data dimension technique does not exist for application in the analysis of health care claims data. The analyst needs to weigh the features of the large data set under consideration, the objectives of the downstream or subsequent analysis, and the availability of tools for ease of use and interpretation of results. Conclusions: The number of data dimension reduction techniques available to claims data set researchers is large and diverse; however, keys features of these various approaches can help the analyst make an informed decision that is effective with some simple setting and objectives diagnosis. PRM71 Reimbursement Decisions in Oncology Drugs: An International Analysis Aissaoui A Paris dauphine University, Paris, France Objectives: The aim is to compare the drug reimbursement decisions of innovative anti-cancer drugs in five countries (UK, France, Italy, Spain & Australia). Our approach was to identify both convergence and divergence in these reimbursement decisions and to assess the agreement level between the policy makers. Methods: For our analysis we have used 39 oncology drugs authorized by the EMA between January 2004 and December 2012 covering a total of 65 indications, and we have compared the reimbursement decisions in theses 5 countries. We reviewed the technology appraisal performed by their respective national HTA agencies and their reimbursement decisions. We have also analyzed the level of agreement for reimbursement decisions between pairs of each country with kappa scores. Results: Out of these 39 drugs, only 16 drugs were reimbursed in Australia, 15 in England, 38 in France, followed by Italy and Spain, which respectively reimbursed 29 and 28 drugs. When we have analyzed the common reimbursement decision taken, we have observed that between France and Italy 72% of common positive reimbursement decision were taken, and between France and England only 33 % of positive reimbursement decisions are common. In contrast we have found 49 % of common negative reimbursement decision between England and Australia. Then we have measured the consistency between decision makers with KAPPA scores and it came out that France, Italy and Spain are often in agreement in their reimbursement decisions, and that conversely France and England decisions are significantly in disagreement. Conclusions: This study demonstrates that the discordance between countries reimbursement decisions, in most cases may reflect the differences in the decision making process (Eg. France Vs England). But this analysis cannot be conclusive. This is why we have carried out further researches using larger datasets allowing us to highlight some elements yielding to these reimbursement decision differences between countries. PRM72 Easy Come, Hardly Go: Epidemiological Methods to Evaluate the Effect of Ispor Board of Directors Membership on Publication Activity Merész G 1, Gyurcsán G C 2, Salfer B 3 1Syreon Research Institute, Budapest, Hungary, 2Self-employed, Budapest, Hungary, 3Healthware Consulting Ltd., Budapest, Hungary . . . . Objectives: ISPOR is approaching its 10th anniversary, which offers a suitable occasion to assess the impact of the organization. As the number of members increased, the influence of the Board of Directors on scientific discussions also emerged. The aim of this research is to present what effect being elected an ISPOR Director has on a researcher’s publication activity by using epidemiological methods and data mining techniques. Methods: Data on number of publications by year, co-authors, titles, abstracts of former ISPOR Directors between 1995 and 2012 were obtained from public sources (ISPOR website and PubMed) and analysed by an algorithm developed by the authors in R. A case-only study design was applied by matching the duration spent as a member of the Board of Directors with the same period prior to and after finishing the directorate term. Incidence rate ratios (IRR) were estimated by fitting separate Poisson regression models to correct for the baseline increase in publication activity. The average number of co-authors and probability of the director being the first author was also analysed. Results: The IRR of the period preceding to directorate versus the directorate period was 1.59 (CI 95%: 1.34-1.91), yielding statistically significant association. The IRR of directorate period versus the period succeeding the directorate period was 1.01 (CI 95%: 0.88-1.13). The average number of co-authors was the highest after the directorate term (5.94); the probability of the director appearing as the first author was the highest prior to directorate term (22.91%). Conclusions: As a result of our study, it has been statistically proven that being an ISPOR Director does not only provide leadership in a scientific organization, but can enhance the members’ career as a researcher. ISPOR Directors are more likely to co-author publications even after finishing their directorate term. RESEARCH ON METHODS – Modeling Methods PRM73 Creating Patient Profile in Individual Simulations: A Comparison of Approaches Stern S , Pan F Evidera, Bethesda, MD, USA . Objectives: Individual simulation is increasingly used in economic models, partly because of its capability of predicting event risks based on individual patient characteristics. However, due to lack of individual patient level data, models often use means and standard deviations to create patient profiles. The objective of this study is to evaluate different simulation approaches of creating patient profile at baseline and their impact on model outcomes. Methods: Patient level data (N= 8,857) from National Health and Nutrition Examination Survey (NHANES) was used to evaluate three approaches of creating baseline patient profiles for simulation models. 10 samples of 1000 patients each were created through 1) random sampling from patient level data; 2) using means and standard deviations of the profile variables without correlating the characteristics; 3) using means and variance-covariance matrix among the continuous variable characteristics with cholesky decomposition approach. 10-year cardiovascular diseases (CVDs) rates are estimated using the created patient profiles from these 3 different approaches. Results: The predicted CVD rates based on random sampling are 18.2% for males and 9.7% for females using the random sampling approach, 14.5% for males and 7.9% for females using the mean and standard deviation approach and 16.0% for males and 9.2% for females using the cholesky decomposition approach. The CVD rates using the NHANES entire A556 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 population are 18.5% for males and 9.8% for females. Conclusions: Random sampling from patient level data provided the best approximation of actual NHANES population predicted CVD rates. The cholesky decomposition approach was slightly limited since only continuous variables could be utilized which could explain the deviation from the population predicted CVD rates. Independent sampling underestimated the mean risk by ~20%, an interesting finding as many individual simulation models created patients with this approach. Researchers should be cautious in their use of summary statistics when populating individual simulation models. half-cycle correction method provided more accurate results than calculations without any kind of half-cycle correction with the exception of one set of input parameters. Conclusions: Based on our model the most accurate method for half-cycle correction is Simpson’s method as in most cases it was the closest to real data. It is important to note that with a few exceptions even the standard method’s results were more accurate than in cases where no half-cycle correction was applied. PRM74 Validation of the SPHR Diabetes Prevention Model PRM77 Application of A Model Of Decision Based on Fuzzy Logic to Pharmacoeconomics: Treatment of CROHN’S Disease With Antitnf in Out of Label Use Thomas C , Watson P , Squires H , Chilcott J , Brennan A University of Sheffield, Sheffield, UK 1HOSPITAL . . . . . Objectives: We have developed a model to evaluate type-2 diabetes prevention interventions. We aimed to validate this model against external data to test the accuracy of model predictions. MethodsAn individual patient simulation was developed to predict longitudinal trajectories of HbA1c, 2-hr glucose, FPG, BMI, systolic blood pressure, total cholesterol and HDL cholesterol based on statistical analyses of the Whitehall II longitudinal cohort. Criteria for diabetes diagnosis were flexibly specified. Cardiovascular events were estimated from the QRISK2 algorithm. Microvascular complications of diabetes were estimated from the UKPDS outcomes model. Several validations were performed to compare model outcomes with reported data from external sources. We assessed the predicted diabetes incidence using data from the EPIC Norfolk cohort. Data from the Health Survey for England (HSE) 2003 cohort was simulated for eight years to compare predicted disease incidence and metabolic distributions with HSE 2011 data. We compared microvascular, cardiovascular and mortality outcomes in a diabetic population with those observed in the UKPDS. We assessed the performance of the model in predicting the results of the ADDITION trial for diabetes screening. Results: We found that the model overestimated three-year incidence of diabetes, particularly in high risk (HbA1c>6.0) individuals, but underestimated diabetes incidence in medium risk individuals (HbA1c 5.5-5.9) compared with the EPIC-Norfolk data. Predictions from HSE 2003 were fairly accurate. Predictions for microvascular events were similar to the UKPDS, but cardiovascular disease and mortality were slightly under-predicted. The model replicated the non-significant difference seen between control and intervention arms of the ADDITION trial, but overestimated total mortality and cardiovascular disease. Conclusions: The SPHR Diabetes model appears to be fairly accurate at predicting external data, but has a tendency to overestimate mortality rates in a newly diagnosed diabetic cohort, and underestimate cardiovascular disease and mortality compared with the UKPDS. PRM75 Use Of Model Averaging Techniques in Cost-Effectiveness Analysis in Oncology Le H H , Ozer-Stillman I Evidera, Boston, MA, USA . . . Objectives: Often in cost-effectiveness analysis (CEA) of oncologic drugs, survival data from a randomized controlled trial are extrapolated to a lifetime horizon using parametric regression techniques. To capture parameter uncertainty in the analysis, regression parameters along with other model parameters are varied in probabilistic sensitivity analysis. However, structural uncertainty in the choice of regression model is rarely investigated. This study discusses the use of model averaging and provides an example to address structural uncertainty in CEA. Methods: Using a cohort partition model, the numbers of patients in “progression-free”, “progressed”, and “dead” health states were calculated directly from progression-free survival (PFS) and overall survival (OS) curves. Weibull, exponential, lognormal, log-logistic, generalized gamma, and Gompertz parametric models were used to extrapolated these curves to a lifetime horizon. Total costs, life year (LY), and quality adjusted life year (QALY) for each regression model were estimated. Weighted results across all models were calculated, based on weights that were derived from Akaike’s or Bayesian Information Criterion (AIC or BIC) parameters. Results: Evaluating solely on BIC values, the lognormal distribution was identified as the best model for both survival curves. This resulted in the lowest observed ICERs. When model selection was based on considerations involving the log-cumulative hazard plots, clinical plausibility, and AIC/ BIC for each distribution, the Weibull distribution was selected for both curves, resulting in a 29% and 27% increase in the ICER for QALY and LY, respectively. Similar increases were observed when model averaging was applied using BIC-derived weights. In this case, model averaging produced results that were similar to those where model selection was based on multiple criteria. Conclusions: Choice of parametric models often has the biggest impact on the outcomes in CEAs in oncology. Model averaging takes into account the structural uncertainty surrounding the choice of parametric models. PRM76 Comparing Three Different Methods of Half-Cycle Correction Nemeth B , Szeker V National Institute of Quality- and Organizational Development in Healthcare and Medicines, Budapest, Hungary . . Objectives: To compare three different half-cycle correction methods and their effect on the final results of Markov models. Methods: To assess the relative performance of the alternatives to the standard half-cycle correction we constructed a 5-state Markov model where the courses of the number of patients in health states follow different shapes to represent the most likely cases in modelling practise. We applied three different correction methods (standard half-cycle correction, Simpson’s method and using the mid-cycle values) and we also looked at the results without any correction and with different numbers of Markov cycles. We conducted a sensitivity analysis by changing the input parameters of our model. In total we examined 80 cases. Results: In our Markov model Simpson’s method provided the most accurate results where the difference from real data was less than 0.1% in 67 of the 80 cases. The second most accurate method was using the mid-cycle values. The standard Alonso Herreros J M 1, González-Cuello A 2 LOS ARCOS MAR MENOR, SAN JAVIER (MURCIA), Spain, 2Murcia University, MURCIA, Spain . . . Objectives: We present a model decision based on fuzzy logic, and apply to off label use of antiTNF in Crohn’s disease (CD) (Infliximab (IFB) 10 mg/kg/8 weeks, adalimumab (ADA) 80mg/2 weeks,. Certolizumab (CZB) 200mg/2weeks). The term “fuzzy logic” (FL) was introduced in 1965 by LAZadeh. Compared to traditional logic, FL variables may have a truth value in degree. FL has been applied to many fields, from economic analysis, to artificial intelligence. However it has not been applied so far to pharmacoeconomics. Methods: According to a decision analysis model based on FL four fuzzy variables that affect the choice of treatment are defined: treatment success (expressed as a probability), cost of success, cost of failure (expressed as inverses), and other conditions about the cost (negotiation, handling of drugs...). Based on the value of these fuzzy variables, three linguistic variables (High, Medium, Low) are defined to expressing convenience of choice. The combination of the three possible values for each of the variables gives us 81 possible decision rules, so that the (HHHH) would be the most favorable option and (LLLL) the more unfavorable. So a new fuzzy variable called “ranking” is established for classifying these options with 7 possible values (Very-unfavorable, unfavorable, slightly-unfavorable, neutral, slightly-favorable, favorable, very-favorable). The value of the fuzzy variables for antiTNF at 52 weeks of treatment, were established based recent meta-analysis and reviews. Results: The matrices obtained and corresponding decision rules were: for IFB (0.65, 6.3 10-5, -1.17 10-4, 0.075) / (MMML); For ADA (0.41, 9.21 10-5, 6.4 10-5, 0.075) / (MMML); for CZB (0.52, 1.30 10-4 1.5 10-4 0.075) / (MHHB). Thus the CZB would be the “slightly-favorable” option, versus IFB and ADA (unfavorables). Conclusions: It possible to apply methods of “FL” to pharmacoeconomic studies According to the model, Certolizumab would be a most favorable choice in off-label use for CD. PRM78 Multi-Criteria Decision Analysis (MCDA): Testing a Proposed Mcda Model for Orphan Drugs Schey C 1, Connolly M 2 of PharmacoEpidemiology & PharmacoEconomics, Groningen, The Netherlands,, 2University of Groningen, Groningen, The Netherlands . . 1Unit Objectives: Since the introduction of the orphan drugs in Europe, it has been suggested that the general method of appraising drugs for reimbursement is not necessarily suitable for orphan drugs. The National Institute for Health and Clinical Excellence indicated that several criteria other than cost and efficacy could be considered in reimbursement decisions for orphan drugs. The aim of this study was to apply a MCDA framework that was proposed by Hughes-Wilson et al (2012) to a range of orphan drugs in different diseases to test the correlation between drug price and aggregated MCDA scores for each product. Methods: A MCDA framework was developed using the nine criteria suggested by Hughes-Wilson et al. A supplementary literature review was conducted to identify other attributes described in the application of MCDA in rare diseases. A numerical scoring system on a scale of one to three was developed for each criterion. Correlations between the average annual cost of the drugs and aggregate MCDA scores were tested and plotted graphically. Different weightings for each of the attributes were also tested. A further analysis was conducted to test the impact of including the drug cost as an attribute in the aggregate index scores. Results: The literature review identified further commonly cited criteria: ‘convenience of administration’, ‘age of the target population’, ‘quality of life’, and ‘drug innovation’ that were added to the aggregate index scores. In the drugs studied, the R 2was 0.808 and 0.704 when costs were included and not included, respectively. The standard error of the slope varied from 7711.9 to 11413.3 when costs were included and not included, respectively. Conclusions: This quantitative study provided insight into using MCDA and its relationship to annual costs. Further work should explore the potential for therapy-specific MCDAs and how to inform value-based pricing assessment. PRM79 Advishe: a New Tool to Report Validation of Health-Economic Decision Models Vemer P 1, Corro Ramos I 2, Van Voorn G 3, Al M J 2, Feenstra T L 4 Groningen, Groningen, The Netherlands, 2Erasmus University, Rotterdam, The Netherlands, 3Wageningen University & Research, Wageningen, The Netherlands, 4University of Groningen, University Medical Center Groningen, Groningen, The Netherlands . . . . . . . 1UMC Background: Modelers and reimbursement decision makers could both profit from a more systematic reporting of the efforts to validate health-economic (HE) models. Objectives: Development of a tool to systematically report validation efforts of HE decision models and their outcomes. Methods: A gross list of model validation techniques was collected using a literature review, including sources outside the HE field. A panel then selected the most important items. Based on the Delphi method, the panel members could score items in three e-mail rounds. Participants were HE modelling experts, covering various nationalities and work environments. They could comment on relevance, feasibility and formulation of the items and received feedback on comments from others. This resulted in a draft tool of selected items, which was tested and improved in two further rounds. In A557 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 addition, the Dutch National Health Care Institute commented on usefulness for decision makers, while a separate group of 50 HE experts could comment during a workshop at ISPOR Montreal 2014. Results: 35 Validation techniques were identified and grouped into four categories: conceptual model validation, computerized model validation, data validation and operational validation. Around 30 HE experts commented in each of the first three Delphi rounds, resulting in a 15 item draft tool. The Dutch health care advisory institute suggested to add one more item. Participants from the ISPOR workshop delivered 19 filled-in questionnaires. A fourth round resulted in 17 responses. This led to a refined version containing 16 items, which is currently sent out for a final, fifth round. Conclusions: When filled out by the modellers, AdVISHE (Assessment of the ValIdation Status of HealthEconomic decision models) supports model users in assessing the validation status of a model It will be useful as part of reimbursement dossiers, by providing systematic and transparent insight into the validation efforts performed and their results. PRM80 Modelling Survival in the Presence of Different Mechanisms of Action: Ipilimumab and Vemurafenib in Advanced Melanoma Lee D 1, Porter J 1, Hertel N 2, Hatswell A J 1 1BresMed, Sheffield, UK, 2Bristol Myers Squibb, Uxbridge, UK . . . . . Objectives: Traditional indirect treatment comparison methods assume the underlying survival profiles of treatments are similar (i.e. proportional hazards). This assumption is unlikely to hold for the comparison of ipilimumab and vemurafenib: Whereas vemurafenib exhibits improved short-term survival compared with ipilimumab, pooled study data for ipilimumab consistently show that patients achieve durable long-term survival. We present a method to compare across trials with differing survival profiles by accounting for follow-on treatments and different patient baseline characteristics. Methods: Comparative survival estimates for ipilimumab and vemurafenib were produced using patient-level data from trial CA184-024 for ipilimumab and published survival curve fits from BRIM-3 (along with registry data) for vemurafenib. The BRIM-3 vemurafenib overall survival curve was adjusted to account for (a) the effect of second-line ipilimumab (via a tunnel-state methodology) and (b) differences in patient baseline characteristics between BRIM-3 and CA184-024, by means of a model (Korn model), constructed to predict the outcomes for dacarbazine-treated patients. The resulting survival estimates were compared with naïve unadjusted survival curve fits, and estimates produced using a hazard ratio (from an indirect comparison) to the ipilimumab data. Results: Estimated survival for ipilimumab was 3.3 years (mean). Predicted survival for vemurafenib, using a naïve comparison, was 3.0 years (mean). Adjusting for second-line ipilimumab and different baseline characteristics resulted in an estimate of 2.8 years for vemurafenib. When a hazard ratio was applied to the ipilimumab data, which underlies the here strong assumption that the vemurafenib overall survival profile is similar to that of ipilimumab, predicted survival for vemurafenib increased to 4.2 years.Conclusions: Depending on the methodology used, the mean predicted survival for vemurafenib varied from 2.8 to 4.2 years. Alternative methods that incorporate the long-term survival profile of ipilimumab (naïve comparison or more sophisticated adjustment methodology) demonstrate a higher number of life years with ipilimumab versus vemurafenib. PRM81 Health Economic Models in Alzheimer’s Disease: A Critical Assessment Walzer S 1, Droeschel D 1, Kaier K 2 Market Access & Pricing Strategy GmbH, Weil am Rhein, Germany, 2University of Freiburg, Freiburg, Germany . . . 1MArS Objectives: Alzheimer’s Disease destroys brain cells, causing problems with memory, thinking, and behavior severe enough to affect work, family and social relationships, and, eventually, the most basic activities of daily living. Different treatment options have been introduced and evaluated from a health economic perspective. However, given the specific characteristics of the disease an evaluation of existing models is needed. Methods: The following databases were searched systematically: PubMed, Health Technology Assessment Database, NHS Economic Evaluation Database, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, DAHTA-database, PSYNDEX and PsycINFO. For the abstracts that met the pre-defined inclusion criteria, full text articles were obtained and evaluated for inclusion in the assessment. Results: After eliminating duplicates the search indicated yielded 1’219 articles of which another 940 were excluded based on the title selection. Finally 59 articles have been reviewed in full text after abstract review. Out of those articles 39 were deemed to be relevant based on the research question. The majority of models (48%) have been Markov models, other methods being used were various statistical analysis applications, micro-simulation, and discrete-event simulations. Limitations of existing models include the following: Focus on cognitive function as disease progression only; lack of inclusion of correlation between disease progression and other factors (e.g. residential status); lack of complete structure of diagnosis and treatment of disease (e.g. including non-drug treatments). Based on the Drummond checklist for health economic models the quality of models proved generally to be high but the majority of those lack presenting a comprehensive pathway of the natural history of the disease. Conclusions: Current models do not allow decision makers optimally characterizing the disease, to better assess the costs and benefits of a wide range of potential interventions. Potential new models need to take the disease characteristics and specifics more appropriate into account. PRM82 Approaches Used to Model the Relationship Between ProgressionFree Survival (PFS) / Time-To-Progression (TTP) And Overall Survival (OS) Within Health Economic Models of Cancer Therapies Rafia R , Ward S E University of Sheffield, Sheffield, UK . . . Objectives: Within health economic models of metastatic cancer therapies assumptions on the relationship between progression-free survival (PFS) / time-to-progression (TTP) and overall survival (OS) are typically required; notably when OS data are immature or unavailable. A review was undertaken to identify the methods that have been used within health economic models regarding this relationship and to identify the rationale given for the approach taken, specifically in those situations where OS data were not available or immature. Methods: All NICE technology appraisals in the advanced and/or metastatic cancer setting completed by December 2013 were reviewed. The review included all relevant appraisal documents publicly available on the NICE website containing information on the methods used and/or rationale for the approach taken to model the relationship between OS and PFS/TPP within the health economic model. This included the sponsor submission and updated analyses, the independent Assessment Report, and other reports/analyses in relation to the appraisal process. Results: In those instances where OS data were immature or not available, PFS/TTP was typically assumed to be a valid surrogate of OS. Justification for this assumption was inconsistently reported. In some health economic models a quantification of the assumed relationship was informed by published evidence and/or expert judgement. In some cases attempts were made to explore the potential impact of this relationship in sensitivity analysis. Conclusions: The methods and/ or rationale given for the approach used to model the relationship between OS and PFS/TTP in health economic models has been inconsistently reported and justified. Whilst some health economic models attempted to quantify this relationship, further transparency is required. A consensus needs to emerge on the most appropriate approaches to be used within health economic models to quantify this relationship, specifically when OS data are not available or immature and to identify the circumstances when particular approaches may be most relevant. PRM83 Comparison of Methods to Estimate Health State Utilities in Metastatic Breast Cancer (MBC) Hudgens S 1, Briggs A 2, Tremblay G 3, Forsythe A 3, Lloyd A 4 Outcomes Solutions, Tucson, AZ, USA, 2University of Glasgow, Glasgow, UK, 3Eisai Inc, Woodcliff Lake, NJ, USA, 4ICON Plc, Oxford, UK . . . . . 1Clinical Objectives: Patient-level utility values for different stages of MBC and toxicities commonly associated with chemotherapy regimens are useful for health economic assessments. Three methods to estimate utilities exist when direct utility data are not available: utility ‘mapping’ from existing disease-specific scales, vignette studies that describe the health states; or derivation of preference-based measures from an existing condition-specific scale. This study compares utility estimates in MBC utilizing the above methods. Methods: Based on data from a phase 3 clinical trial in MBC (N= 1102) utility mapping was conducted using a published regression algorithm to convert the EORTC QLQ-C30 questionnaire to the EQ-5D utility. Mean utility values were estimated for relevant health states: stable disease (SD), tumor response (TR), disease progression (DP) and common toxicities. Results were compared to previously published values obtained for a vignette study conducted in one hundred members of the general public. Results: Observed MBC utilities were similar in mapping vs. vignette studies for SD: 0.697 vs. 0.715, and TR: 0.782 vs. 0.790. General public respondents in the vignette study assigned much lower utility to symptomatic DP (0.443) vs. imaging-based DP in mapping study (0.679); and disutility for toxicities: vomiting: 0.103 vs. 0.050; fatigue 0.115 vs. 0.029; febrile neutropenia 0.150 vs. 0.012 (vignette vs. mapping respectively). Hand-foot syndrome, stomatitis and hair loss were not associated with disutility in the mapping study (potentially due to small sample size) while disutility of 0.116; 0.151; and 0.114 were reported by the vignette study. Conclusions: Utilization of different methods to estimate utilities in MBC may lead to a wide range of estimated values with potentially significant implications for health economic evaluation. Caution must be exercised when comparing utility values derived using different methods. It is preferable to collect such data from patients directly and use vignettes as a last resort. PRM84 Cost-Effectiveness Models for Chronic Obstructive Pulmonary Disease (COPD): Cross-Model Comparison of Hypothetical Treatment Scenarios Hoogendoorn M 1, Feenstra T 2, Asukai Y 3, Borg S 4, Hansen R N 5, Jansson S A 6, Samyshkin Y 3, Wacker M 7, Briggs A 8, Lloyd A 3, Sullivan S D 9, Rutten-van Mölken MP1 1Erasmus University Rotterdam, Rotterdam, The Netherlands, 2RIVM /UMCG, Bilthoven, The Netherlands, 3IMS Health, Economics and Outcomes Research, London, UK, 4The Swedish Institute for Health Economics, Lund, Sweden, 5School of Pharmacy, University of Washington, Seattle, WA, USA, 6The OLIN Studies, Luleå, Sweden, 7Helmholtz Zentrum Munchen, Neuherberg, Germany, 8University of Glasgow, Glasgow, UK, 9University of Washington, Pharmaceutical Outcomes Research and Policy Program, Seattle, WA, USA . . . . . . . . . . . . . . Objectives: To compare different COPD cost-effectiveness models with respect to structure and input parameters and to cross validate the models by running the same hypothetical treatment scenarios. Methods: COPD modeling groups simulated four hypothetical interventions with their model and compared the results with a reference scenario of no intervention. The four interventions modeled assumed: 1) 20% reduction in decline in lung function, 2) 25% reduction in exacerbation frequency, 3) 10% reduction in all-cause mortality and 4) all these effects combined. The interventions were simulated for a five-year and lifetime horizon with standardization, if possible, for sex, age, COPD severity, smoking status, exacerbation frequencies, mortality due to other causes, utilities, costs and discount rates. Furthermore, uncertainty around the outcomes of intervention four was compared. Results: Seven out of nine contacted COPD modeling groups agreed to participate. Differences in 5-year QALY gains ranged from 0.00020 to 0.039 for intervention one, 0.0089 to 0.075 for intervention two and 0.017 to 0.048 for intervention three. The difference in costs ranged from € 561 to € 912 for intervention one, € 739 to € 1350 for intervention two and € 1140 to € 1618 for intervention three. The 5-year cost-effectiveness ratios (ICERs) for the most comprehensive intervention, intervention four, was € 17,000/QALY for two models, € 25,000-€ 28,000/QALY for three models A558 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 and € 47,000/QALY for the remaining two models. Differences in the outcomes could mainly be explained by differences in input values for disease progression, exacerbation-related mortality and all-cause mortality with high input values resulting in low ICERs and vice versa. Lifetime results were mainly influenced by the input values for mortality. The probability of intervention four to be cost-effective at a willingness-to-pay of € 50,000/QALY was 90-100% for five models and about 70% and 50% for the other two models. Conclusions: Mortality was the most important factor determining the differences in cost-effectiveness outcomes between models. PRM85 A DE-NOVO Model to Predict Outcomes of a New Hypothetical Intervention to Reduce CV Risk in Post Mi Patients Jain M 1, Sonathi V 1, Rathi H 1, Bakuli A 1, Thomas S K 2, Mollon P 3 1Novartis Healthcare Pvt. Ltd., Hyderabad, India, 2Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA, 3Novaris Pharma AG, Basel, Switzerland . . . . . . . Objectives: The risk of cardiovascular (CV) events in post myocardial infarction (MI) patients poses a significant burden on the UK health care system despite the current standard of care (SoC). The objective of this analysis was to develop a model to quantify the relationship between efficacy and outcomes of a new hypothetical drug given in addition to current SoC to reduce the risk of CV events in post-MI patients when compared to SoC. Methods: A 6-state lifetime Markov model with a 1-year cycle length was developed from a UK health care perspective. Recurrent MI, stroke and CV death were modeled. The hypothetical drug was assigned efficacy values for its ability to reduce the incidence of CV events. The outcomes were measured in terms of QALYs and LYs. A linear regression model was fitted to estimate the expected outcome/patient based on relative risk reduction (RRR) in the incidence of CV events. All outcomes were discounted at 3.5% annually. Results: The model structure addressed some of the limitations of previous economic models, namely increased risk due to stroke in MI patients and increased risk of subsequent events in the first year. For identical cohorts, the outcomes from the model compared well with other published studies. For a cohort of patients aged 40-years, the model predicted on an average, LYs of 17.6 and QALYs of 13.64. A hypothetical drug achieving a 5% RRR in CV events resulted in an incremental LYs of 0.28 and QALYs of 0.23. The increase in incremental QALYs and LYs per percentage point reduction in relative risk as compared to SoC was estimated to be 0.043 and 0.054 respectively. Conclusions: A de-novo economic model quantifies the relationship between the efficacy and outcomes of a hypothetical drug when compared to the SoC to reduce the risk of CV events in post-MI patients. PRM86 Estimating the Lifetime Health Outcomes of Type 2 Diabetes Mellitus (T2dm) Patients Inadequately Controlled on Metformin Plus Sulphonylurea Receiving Either Canagliflozin or Sitagliptin Using the UKPDS Outcomes Model V1.3 Nuhoho S 1, Worbes-Cerezo M2 1Janssen-Cilag A/S, Birkerød, Denmark, 2Janssen-Cilag UK, High Wycombe, UK . Objectives: The goals of type 2 diabetes management are to control glycaemia and other micro- and macrovascular risk factors such as weight, blood pressure and lipids in order to prevent death and other complications due to the disease. The natural history of the disease makes it challenging to estimate the effects of treatments on these long term complications and mortality from standard clinical trials. Modelling is therefore a key bridging tool for predicting long term health outcomes from intermediate endpoints. The objective of this analysis was to estimate the relative effects of canagliflozin 300mg and sitagliptin 100mg on mortality, micro- and macrovascular complications in T2DM patients in triple line as add on to metformin plus sulphonylurea using the UKPDS Outcomes Model v1.3. Methods: A probabilistic patient generator was developed which generated 10,000 patients with applied treatment effects based on data from head to head randomised clinical trials. Upon loss of glycaemic control (HbA1c ≥7%), patients were assumed to switch to insulin. 1% point reduction in HbA1c was applied on rescue. For model stability, patients were looped 1,000 times (creating 10 million patients in each arm) with 100 bootstrap simulations. Outcomes were discounted 3.5% annually. Results: At the end of the 40 years simulation, patients initiating canagliflozin 300mg had 49 more survivors and 16,918 fewer diabetes-related deaths. Micro- and macrovascular complications were estimated in fewer patients on canagliflozin 300mg than on sitagliptin 100mg (between 5,948 fewer renal failures and 41,157 fewer myocardial infarctions). There were discernible relative risk reductions in all complications and diabetes-related death ranging from 1.40% (heart failure) to 2.96% (amputation). Conclusions: Results of the analysis using the UKPDS Outcomes Model v1.3 suggest that canagliflozin 300mg compared with sitagliptin 100mg as add on to metformin plus sulphonylurea reduces long-term diabetes-related mortality and complications. PRM87 All-Cause Mortality Validation of the Core Diabetes Model Against Predictions of the Charlson Comorbidity Index Foos V 1, McEwan P 2, Lamotte M 3, Grant D 4 1IMS Health, Basel, Switzerland, 2Health Economics and Outcomes Research Ltd, Cardiff, UK, 3IMS Health Consulting, Brussels, Belgium, 4IMS Health, London, UK . . . . Objectives: All cause mortality (ACM) validations with the IMS CORE Diabetes Model (CDM) have demonstrated below average fit when compared to overall validation scores from 96 validation end points (EP) with a R2-statistic of 0.651 (vs. 0.93 All-EP). Lack of fit was associated with a model overestimation of ACM when compared to contemporary outcome studies (ACCORD, ADVANCE, VADT). The objective of this investigation was to put these findings into perspective by comparing the model to mortality risk predictions from the Charlson-Comorbidity-Index (CCI). Methods: The CCI was applied to predict the 10 year mortality risk for diabetes patients with age of 50,60,70 and 80 years and four different co-morbidity levels: no complications (NC), myocardial infarction (MI), MI and stroke (MI+S), MI+S and heart failure (MI+S+HF) and MI+S+HF and renal failure (MI+S+HF+RF). CCI mortality scores were compared to corresponding 10 year ACM predictions from the CDM. Base case (BC) analyses applied UKPDS-68 risk equations (UK68-RE) for CV risk and mortality. Two sets of sensitivity analyses were conducted using UK68-RE for CV risk but mortality tracked individually per complication event (non combined mortality approach) (SA1) and UKPDS-82 risk equations (UK82-RE) applied for CV risk and mortality (SA2). Results: Across all age and co-morbidity states, CDM simulations demonstrated the closest match to CCI-scores in SA1 with an R2-statistic of 0.877. This compared to R2-statistics of 0.757, and 0.851 for BC and SA2, respectively. BC and SA2 analyses noteworthy underestimated ACM risk in analyses with increased co-morbitity level by 68% (BC) and 49% (SA2) vs. 17% (SA1) in (MI+S+HF) and 44% (BC) and 36% (SA2) vs. 3% (SA1) in (MI+S+HF+RF). Conclusions: The CDM demonstrated a closer match to CCI mortality scores (vs. outcome studies) with a trend to underestimate ACM. This trend increased with baseline age and (only BC and SA2) co-morbidity level. PRM88 Deterministic Versus Stochastic Prediction of Risk for Cardiovascular Events Villa G 1, Lothgren M 1, Gandra S R 2, Lindgren P 3, van Hout B 4 (Europe) GmbH, Zug, Switzerland, 2Amgen, Inc., Thousand Oaks, CA, USA, 3IVBAR, Karolinska Institutet, Stockholm, Sweden, 4University of Sheffield, Sheffield, UK . . . . . . 1Amgen Objectives: Multivariate functions can be used to predict individual risk for cardiovascular (CVD) events and also to estimate baseline risk in economic models. We present a comparison of deterministic versus stochastic risk predictions using Framingham’s [D’Agostino 2008] and REACH’s [Wilson 2012] functions. Stochastic risk prediction accounts for patient-level heterogeneity, but involves a number of issues including increased complexity, data requirements, need for assumptions and computational burden. To our knowledge, this topic has not been studied in the CVD setting. Methods: D’Agostino 2008 and Wilson 2012 modeled primary (PE) and recurrent event (RE) risks, respectively. Both studies considered fatal and non-fatal aggregate CVD events and estimated a Cox Proportional Hazards (CPH) multivariate risk function. In the deterministic prediction, the means of the risk factors were used to predict the population’s risk directly from the functions. In the stochastic prediction, individual patient profiles (n= 10,000) were generated using Monte Carlo simulation. Individual risks were then estimated from the functions and averaged to compute the population’s risk. Multinomomial distributions were assumed for discrete variables (e.g. diabetes, number of vascular beds) and normal or log-normal distributions were assumed for continuous variables depending on skewness (e.g. age, total cholesterol). Probability distributions were parameterized based on the risk factors descriptives reported in the original references. Simulations were performed with and without considering dependence of risk factors. Results: Due to the nonlinearity of the CPH function, the stochastic prediction yielded 23% (PE) and 17% (RE) higher risks than the deterministic approach (14% and 10%, respectively, if age was kept constant). Differences between prediction approaches are even higher if the estimated correlation structure of risk factors is accounted for. Conclusions: When compared to the stochastic prediction, the deterministic approach leads to lower estimates of CVD risks. Therefore, economic models using this approach might underestimate treatment effect. PRM89 Are Cycles Needed in Markov Models? – The Continuous Model as a Simpler Approach Tichy E Evidera, Budapest, Hungary . Objectives: To present an alternative implementation for the conventional Markov models with area under the curve (AUC) approach: the continuous model (CM). To present how the CM avoids the need of determining cycles in theory and to compare the traditional and the CM approach in terms of results and complexity in an oncology model example. Methods: The AUC model assumes that the survival function is known at any timepoint not only at the beginning and end of model cycle. The CM calculates the model outcomes for the whole timehorizon by using the values of the survival function in every timepoint, instead of the discrete timepoints defined by cycle length. The CM approach overcomes the issue of the artificial characterization of time using cycles, that is often criticized in Markov models. Using CM can also lead to more precise estimates. A simple oncology AUC model with three health states (progression free survival, progression and death) and four-weekly cycles was built and converted to a CM model in Excel®, using user defined Visual Basic functions. Results, generalizability and user friendliness were compared. Results: The results of the two models were similar: for health outcomes differences were around 1%, for costs and incremental cost-effectiveness ratios around 0.5%. Calculations were done in a single cell/outcome instead of a column of 100-200 cells depending on cycle length and time horizon, giving less scope for bugs and facilitating easier debugging. As a result the implementation of the CM model was faster and technical validation easier. Conclusions: The CM approach requires more technical background from the developer; custom functions have to be built even for point estimates. However, results of a CM, requires smaller spreadsheet space, and provides more transparency and easier debugging, while providing similar or potentially more precise estimates compared to the AUC model results. PRM90 A Comparison of Modelling Techniques: Patient Simulation VERSUS Markov Modelling in Ophthalmology Claxton L 1, Malcolm W A 2, Hodgson R 1 1York Health Economics Consortium, York, UK, 2Novartis UK, Frimley, UK . . . . Objectives: Markov models are a currently popular means of estimating the costeffectiveness of interventions; however they are associated with certain limitations which may make them ill-suited to inform some health care decisions. Patient simulation models offer an alternative methodology which may overcome some of these A559 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 limitations potentially providing more accurate estimates of the cost-effectiveness. This study aims to explore the strengths and limitations associated with simulation modelling, and the appropriateness of this methodology in ophthalmology. Methods: An ophthalmology model that was previously developed using a Markov structure was adapted to a patient simulation model using the same cost and quality of life inputs, and clinical inputs adapted to the appropriate format. The deterministic and probabilistic results of each model were then compared to each other, with the costs being broken down by health state to identify key areas of differences. Each model’s results were verified against real-world observational data. Results: Results suggest that cost and quality of life outcomes are similar when unilateral disease is considered, with differences in quality of life and costs being seen when bilateral disease is incorporated. The key area of differences in costs is those relating to blindness, as this is a function of both eyes which is not captured well in a Markov model. One-off costs when a patient becomes blind are also not well captured in a Markov model. Conclusions: Our results suggest that the ability of simulation models to more accurately represent the real-world patient pathway allows a number of aspects of disease progression to be modelled, which cannot be easily done using a Markov structure. In particular, the benefits of a simulation approach can be demonstrated in the modelling of quality of life as a function of visual acuity in both eyes, and capturing the costs relating to blindness, resulting in differences in estimated cost effectiveness. PRM91 The Oncotyrol Prostate Cancer Outcome and Policy Model - How Latent Prevalence Affects the Benefit-Harm Balance of Screening Mühlberger N 1, Heijnsdijk E A M 2, Kurzthaler C 1, Iskandar R 1, Krahn M D 3, Bremner K 4, Oberaigner W 5, Klocker H 6, Horninger W 6, Conrads-Frank A 7, Sroczynski G 8, Siebert U9 1UMIT - University for Health Sciences, Medical Informatics and Technology / Oncotyrol - Center for Personalized Cancer Medicine, Hall i. T. / Innsbruck, Austria, 2Department of Public Health, Erasmus Medical Center, Rotterdam, The Netherlands, 3Toronto Health Economics and Technology Assessment (THETA) Collaborative, Toronto, ON, Canada, 4University Health Network, Toronto, ON, Canada, 5Cancer Registry of Tyrol, TILAK GmbH, Innsbruck, Austria, 6Department of Urology, Innsbruck Medical University, Innsbruck, Austria, 7UMIT - University for Health Sciences, Medical Informatics and Technology, Hall i. T., Austria, 8UMIT - University for Health Sciences, Medical Informatics and Technology/ ONCOTYROL - Center for Personalized Cancer Medicine, Hall in Tyrol/ Innsbruck, Austria, 9Medical Informatics and Technology, and Director of the Division for Health Technology Assessment and Bioinformatics, ONCOTYROL, Hall i. T, Austria . . . . . . . . . . . . . . Objectives: The ONCOTYROL Prostate Cancer Outcome and Policy (PCOP) Model is a state-transition micro-simulation model designed to evaluate prostate cancer (PCa) screening. We used the model to investigate how the benefits-harm balance of PCa screening is affected by the size of the latent prevalence pool. For this purpose, we recalibrated the natural history and detection component of the original PCOP model adopted from an earlier version of the Erasmus MISCAN model to match the higher prevalence observed by autopsy studies. The benefits and harms of screening predicted by the recalibrated model were then compared with predictions from the original model. Methods: For recalibration, we reprogrammed the natural history and detection component of the PCOP model as a deterministic state-transition model with stage- and grade-specific cancer states in the statistical software package R. All parameters were implemented as functions or variables and calibrated simultaneously in a single run using the ‘nlminb’ optimization algorithm available in R to minimize the deviation of model predictions from observed data. Calibration targets were observed data from autopsy studies, cancer registries and the European trial (ERSPC). Both the recalibrated and original models were identical except for calibrated parameters. Results: In total, we calibrated 46 parameters. Observed data could not be sufficiently fitted using the original set of parameters. Additional parameters, allowing for an interruption of disease progression in the stage- and grade-specific health states, and an effect modifier allowing for lower screening sensitivities in older men had to be implemented. Recalibration to higher prevalence demonstrated a considerable increase of overdiagnosis and decline of screening sensitivity, which significantly worsened the benefit-harm balance of screening regarding QALYs. Conclusions: Benefit-harm predictions of models, which use calibration to simulate PCa progression in the unobservable latent phase, can be significantly affected by the assumptions on latent cancer prevalence. PRM92 A Systematic Search and Methodological Review of Economic Models of Analgesics for Chronic Pain Sullivan W 1, Gladwell D 2, Fagnani F 3, LopezBastida J 4, Phillips C 5, Hirst M 6, Dunlop W 6 1Bresmed, Sheffield, UK, 2BresMed, Sheffield, UK, 3Cemka Eval, Bourg La Reine, France, 4University Castilla-La Mancha, Ciudad Real, Spain, 5Swansea University, Swansea, UK, 6Mundipharma International Limited, Cambridge, UK . . . . . . . Objectives: Economic modelling of analgesics for chronic pain (including the chronic pain sub-populations of musculoskeletal, neuropathic and malignant pain) is well established but characterised by methodological heterogeneity. The methods used to model pain vary substantially between and within the sub-populations and this variability inhibits comparability across the evidence base. This research aims to facilitate increased consensus for future development of pain models, by summarising existing published economic models and identifying key model characteristics. Methods: A systematic search and methodological review of published economic models of therapies in chronic pain was performed. MEDLINE, EMBASE, Health Technology Assessment, NHS Economic Evaluation Database and EconLit were accessed in April 2014 and, from studies that met the inclusion criteria, key methodological data were extracted and analysed. Results: Thirty-four studies were included in the final review. From these, eighteen original model structures were identified. Outcomes considered alongside analgesic effect varied substantially across studies. Assumptions were used to model multiple treatment lines in eleven model structures. Only three models used a time horizon greater than one year. The inclusion of the costs of adverse events was common (eleven models) but less prevalent in neuropathic pain models than for other chronic pain states. Finally the majority of models (eleven) used a Markov structure but four of ten neuropathic pain models used decision trees. Conclusions: Some methodological similarities can be identified when considering economic modelling within sub-populations in particular neuropathic pain. However, there is scope for further consensus in the key design attributes of pain models, in particular the choice of secondary outcomes. Further research is required to identify the strengths, weaknesses and complexities of the key modelling choices. FUNDING STATEMENT: This research was funded by Mundipharma International Limited. PRM93 Validation of A Global Economic Model to Evaluate The CostEffectiveness Of Targeted Treatments Using Companion Diagnostics In Advanced/Metastatic Cancer Treatment Using Kras Testing For Cetuximab Therapy In Metastatic Colorectal Cancer Mathurin K , Beauchemin C , Lachaine J University of Montreal, Montreal, QC, Canada . . . Objectives: A global economic model to evaluate the cost-effectiveness of targeted treatment using companion diagnostics in advanced/metastatic cancer treatment has been recently developed. Targeted treatment with cetuximab guided with KRAS testing in metastatic colorectal cancer (mCRC) is well implanted and has been subject of different economic evaluations. The objective of this study was to use data on cetuximab guided with KRAS testing in mCRC to validate the global economic model in advanced/metastatic cancer. Methods: Survival data and incidence of adverse events (infections, rash, and pain) were obtained from the CO. 17 study, which is a pivotal trial on the efficacy of cetuximab used according to the KRAS status vs. best supportive care (BSC). Prevalence of KRAS mutation, specificity/sensitivity and cost of KRAS testing, and the cost of cetuximab were obtained from the literature. Other parameters for the cost-effectiveness analysis were intrinsic components of the global model. Analyses were conducted from a Ministry of Health (MoH) and a societal perspective. Results: The incremental cost-utility ratio (ICUR) of cetuximab treatment guided with KRAS testing compared to no testing and BSC to all patients from a MoH perspective was estimated at $192,814/QALY and at $297,198/ QALY from a societal perspective. According to the deterministic analysis results, the ICUR of cetuximab with KRAS testing compared to no testing and BSC varied between $149,845/QALY and $387,971/QALY from a MoH perspective and between $232,417/QALY and $492,354/QALY from a societal perspective. Parameters that have the greatest impact on base-case ICURs were the cost of cetuximab and the specificity of KRAS testing. Conclusions: Validation using cetuximab with KRAS testing suggests that the global economic model is robust, as results obtained were in the range than those reported in the literature. Although further validations will be performed, the global model appears to produce quick, but accurate estimates. PRM94 Systematic Overview On Value-Of Information Analyses In Cancer Research Rochau U 1, Schnell-Inderst P 2, Burger E A 3, Baldauf M 4, Kühne F 4, Jahn B 4, Siebert U 5 - University for Health Sciences, Medical Informatics and Technology/ ONCOTYROL Center for Personalized Cancer Medicine, Hall in Tyrol/ Innsbruck, Austria, 2UMIT - University for Health Sciences, Medical Informatics and Technology, Hall i. T.; Innsbruck, Austria, 3University of Oslo, Oslo, Norway, 4UMIT - University for Health Sciences, Medical Informatics and Technology, Hall in Tyrol, Austria, 5Medical Informatics and Technology, and Director of the Division for Health Technology Assessment and Bioinformatics, ONCOTYROL, Hall i. T, Austria . . . . . . . . 1UMIT Objectives: Value of information (VoI) analysis helps to set priorities for future research by quantifying the value of collecting additional information. Determining whether further evidence is required is especially important in oncology, where novel, but expensive therapies with substantial uncertainty are emerging every day. We aim to provide an overview of published VoI studies within cancer medicine and summarize their methodological characteristics. Methods: We performed a systematic literature review using Pubmed to identify VoI analyses applied within cancer research. Studies exploring topics merely related to methodology were excluded. The search terms were refined following the recent publication by Tuffaha et al. 2014. Screening of articles was performed by two independent reviewers. We extracted characteristics, such as health care context, disease, and type of VoI analyses. Results: We identified a total of 96 references. Overall, 35 articles were included for analysis. Nearly one-half (16) of the studies were conducted in the health care context of UK, eight in the US, five in Finland, three in the Netherlands, and one in Canada, France and Denmark. Disease areas included breast cancer (26%), non-small-cell lung cancer (14%), colorectal cancer (14%), esophageal cancer (9%) non-Hodgkin lymphoma (6%) and prostate cancer (6%). Nearly all of the studies conducted expected value of perfect information (EVPI) analysis and about onehalf expected value of partial information (EVPPI). One study reported on expected value of individualized care, and only three studies did sample size and trial cost estimations. Conclusions: VoI analyses are becoming a more commonly applied method, following standard calculations of cost-effectiveness, to assess the value of further research. Simpler techniques, such as EVPI and EVPPI were reported more frequently compared to the computationally more demanding calculations of EVSI and ENBS, which often require complex statistical methods and estimates of a study’s costs need to be derived. PRM95 Clinical Trial Simulation Considering Quality Of Life Outcomes Mohseninejad L , Heeg B , Majer I M Pharmerit International, RotterdamThe Netherlands . . . . Objectives: Prior to the actual implementation, trial simulations are often performed to optimize registration study design and hence to maximize the probability of marketing approval. Almost exclusively, trials focus on clinical outcomes however reimbursement submissions require health economic evidence, in particular, information on patients’ quality of life (QoL) and estimates of quality-adjusted life A560 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 years (QALYs). Thus, considerations on QoL outcomes in the clinical trial design phase may lead to better optimized reimbursement submissions. The objective of this study was to develop a trial simulation model that is capable of addressing complex research questions, provides flexibility to test various assumptions, and predicts expected QALY outcomes. Methods: A patient-level simulation model was developed using hypothetical data in oncology. The model considered two treatments reflecting the common design of a pivotal trial. Individual survival times and time to progression data were simulated. Hazard ratios were used to include treatment effects. Using the simulated individual level data, a multistate life table model was constructed with three health states: pre-progression (with and without adverse events), post-progression, and dead. Utility and disutility values derived from literature were attached to the number of patients in each health state at a given point in time. Differences between the treatment arms were derived in terms of survival, QALYs, and the uncertainty around those (e.g. probability distribution, P-value). Results: The trial simulation model assessed various patient number scenarios to obtain the smallest sample size that provided a statistically significant minimum clinically meaningful QALY difference between the treatments. Simulations were performed (e.g. testing the effect of different survival profile scenarios, utility values) to assess the robustness of the results. Conclusions: The presented trial simulation model provided a flexible tool to inform clinical trial design considering QoL outcomes. The model can be also useful for manufacturers for pricing or investment decisions. PRM96 Systematic Review Of Mathematical Models Predicting Relative Effectiveness Hummel N , Panayidou K , Gsteiger S , Egger M , Kilcher G University of Bern, Bern, Switzerland . . . . . Objectives: To identify and assess mathematical models predicting the relative effectiveness of drug treatments in “real world” populations, based on data from randomized control trials and other sources of evidence. Methods: Systematic review of mathematical modelling studies addressing the step from relative efficacy to relative effectiveness. We identified eligible studies through electronic and manual searches of MEDLINE and EMBASE databases, selected websites and reference lists of relevant papers. Two reviewers screened the articles independently and extracted study characteristics such as model type, disease area, validation and software used via an extraction form. Results: Eight papers met the inclusion criteria covering four broad modelling approaches: multi-state models, simulationbased approaches, mechanistic models, and classical regression based models. The multi-state models were the predominant class of models. These models are defined as time-dependent stochastic processes with discrete event space. Most examples belonged to the special case of Markov multi-state models. Multi-state models were applied at the level of population groups or at the individual patient level. The other approaches we identified were less frequent. Discrete event simulation was used in one paper. This approach is entirely based on simulations. One article described a mechanistic model based on ordinary differential equations, which are typically derived from biological knowledge and first principles. Finally, more classical regression techniques from survival analysis were used in two papers. Six articles included models built for cardiovascular indications, the remaining ones covered oncology and neurosciences. Internal or external model validation was presented in six papers, while two papers considered only sensitivity analysis to evaluate the model performance. Conclusions: This review shows the range of models currently used for predicting the relative effectiveness of drug interventions in real world patient populations. They complement the available tools for evidence synthesis in comparative effectiveness research. PRM97 Cost-Effectiveness Analysis Of An Antimicrobial Transparent Dressing For Protecting Central Vascular Accesses In Critically Ill Patients Versus Standard Transparent Dressings In France: A Comparison Of Two Modeling Approaches: Decision-Tree Versus NonHomogeneous Markov Model M 1, Motrunich A 2, Maunoury F2 Palka-Santini 13M Germany, Neuss, Germany, 2Statésia, Le Mans, France . . . Objectives: To perform cost-effectiveness analysis (CEA) for routine use of a transparent dressing integrating a chlorhexidine gluconate (CHG) -containing gel pad versus standard transparent dressings, with a classical decision tree model and a Non-Homogeneous Markov Model (NHMM) previously developed. Methods: Clinical efficacy data was extracted from a multicentre randomized controlled trial (RCT) with 1,879 patients and economical data obtained from micro and macrocosting published studies. The baseline method is a NHMM previously developed in Microsoft Excel® with VBA using the same data sources. The decision tree was built with the TreeAge Pro® software 2013. One-way deterministic (DSA) and probabilistic sensitivity (PSA) analyses were conducted on key clinical and economic parameters. Results: Based on the decision-tree model, the CHG-dressing is a dominant strategy compared to standard dressings. The intervention prevents 13.5 infections per 1,000 patients and saves € 157 per patient. These results are robust across a range of values for several parameters in DSA. The PSA with the NHMM resulted in 11.8 infections avoided per 1,000 patients (95%CI: [3.85; 19.64)]) and a mean extra cost of € 141 per patient (95%CI: [€ -975; € 1,258]) when using antimicrobial dressing. Effectiveness as calculated by both models is similar while cost estimations diverge. Conclusions: Decision-tree and the NHMM are structurally different and even though their outcomes cannot be directly compared, they were coherent. The decision-tree model indicates that CHG-dressings are cost-saving and a dominant preventative strategy for CRBSIs. The Markov model supports cost-effectiveness compared to standard dressing. The main disadvantages of the decision-tree are the inability to integrate changes among health states during the ICU stay and to simulate possible observable trajectories in the patient history. The structure of the nonhomogeneous Markov model does not allow DSA for the incidence of the disease. PRM98 HEALTH ECONOMIC EVALUATION OF DIAGNOSTIC AND PROGNOSTIC PREDICTION MODELS. A SYSTEMATIC REVIEW Van Giessen A1, Wilcher B 2, Peters J 2, Hyde C 2, Moons K G 1, de Wit G A 3, Koffijberg H 1 Medical Center Utrecht, Utrecht, The Netherlands,, 2Exeter University, Exeter, UK, Institute for Public Health and the Environment, Bilthoven, The Netherlands . . . . . . . . 1University 3National Objectives: The aim of this study is to provide an overview of the quality of health economic evaluations (HEEs) of prediction models, the evidence used, and the challenges. Methods: The databases Medline, Embase, Econlit, and the NHS Economic Evaluations Database were systematically searched for HEEs of diagnostic and prognostic risk prediction models. The included HEEs were evaluated on their methodological quality using the Drummond checklist. Furthermore, an item list was developed incorporating descriptive items on the HEE, specific items on the HEE of prediction models, and statistical characteristics of the prediction model that could be incorporated into the evaluation. Results: The database search resulted in 791 unique papers, from which 653 were excluded based on abstract. After assessing full texts, 17 HEEs (all cost-utility studies) were included. A prediction model was compared to current practice in 11 HEEs and to an extended prediction model in 6 HEEs. On a 35-point scale the quality score ranged from 17 to 32 (median 25). In 7 papers there was no overlap between authors of the initial prediction model paper and those of the corresponding HEE. In 5 papers individuals were classified based on a single (set of) threshold(s); based on guidelines in 4 papers and once on expert opinion. In 8 papers the classification threshold was optimized in the CEA itself. A probabilistic sensitivity analysis was not included in 7 papers and uncertainty around predicted risks was only taken into account once. Conclusions: In most papers limited (prediction model) details were available. Potentially due to this lack of evidence and a lack of specific guidelines on HEE of prediction models, a large variety in the quality and methodology was observed. This variation may complicate the validation and interpretation of HEE results and thereby the decision making on implementation of prediction models in practice. PRM99 MIGRATION OF HEALTH ECONOMICS MODELS TO WEB AND MOBILE ENVIRONMENTS. WHY SHOULD MODELS GO WEB? Topachevskyi O 1, Volovyk A 2 Health Outcomes, Brussels, Belgium, 2Hashtago, Kiev, Ukraine . . 1Digital Objectives: To understand the key functional differences between conventional cost-effectiveness Excel and web based model types. Methods: An online survey consisting 18 end users and 5 model owners (n= 23) was conducted. Respondents were asked to rate key criteria of both model types on a scale from 0 to 10. Model types were compared with the following 13 criterias: model execution speed and size, general functionality support, accessibility, usability, model management and versioning, ease of localization, ease of model core modification, sharing, review process, usage analytics, integration with other content. No weighting to the scoring across criteria was applied. Results: Results of the survey indicate that web based models outperform standalone models in 10 of the 13 criteria assessed. Model review process, ease of model core modification and execution speed was rated higher for conventional standalone Excel models. 80% of model owners and 78% of model users assigned higher overall score for web based models compared to Excel models. Conclusions: Web based models offer advantages primarily related to model usage and lifecycle management. These models can be viewed on any hardware device or browser, thus overcoming the limitations of Excel models. The use of latest web technologies such as JavaScript, HTML5 and CSS3 improve user experience in model adaptation and presentation to end audience. Usage analytics, smart versioning, web sharing and automatic updates are the functional advantages that can not be achieved with conventional Excel models due to technical limitations. PRM100 Validating A Model To Predict Disease Progression Outcomes In Patients With COPD Risebrough N A 1, Briggs A 2, Baker T M 3, Exuzides A 4, Colby C 4, Rutten van-Molken M 5, Gonzalez McQuire S 6, Lomas D 7, Muellerova H 6, Tal-Singer R 8, Ismaila A 9 1ICON plc (formerly Oxford Outcomes), Toronto, ON, Canada, 2University of Glasgow, Glasgow, UK, 3ICON plc (formerly Oxford Outcomes), Morristown, NJ, USA, 4ICON plc, San Francisco, CA, USA, 5Erasmus University Rotterdam, Rotterdam, The Netherlands, 6GlaxoSmithKline R&D, Uxbridge, UK, 7University College London, London, UK, 8GlaxoSmithKline R&D, King of Prussia, PA, USA, 9GlaxoSmithKline, Research Triangle Park, NC, USA . . . . . . . . . . . . . Objectives: To validate a model for quantifying the COPD disease progression against both the data used to generate the model (internal validation) and clinical trial data not used in the model’s development (external validation). Methods: A model representing causal relationships between central disease attributes (lung function, exacerbations, symptoms and exercise capacity) and final outcomes (survival, quality of life, cost) was developed based on the Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) study dataset. Model predicted annual outcomes were compared to the corresponding annual observed data from ECLIPSE (n= 2,164) and TOwards a Revolution in COPD Health (TORCH) (n= 6,108) trials based on fitting the model baseline parameters to reflect each specific study population. Results: The model accurately predicted the ECLIPSE outcomes in at least two of the three annual time points within the 95% confidence interval (CI) of the observed data for survival, FEV1% predicted, and annual exacerbations (per patient per year [PPPY]. The model predicted 9.0 metres annual decline in Six Minute Walk Distance compared to ECLIPSE observed data of 5.7 metres decline. The model accurately predicted the TORCH placebo outcomes in at least two of the three annual time points within the 95%CI of the observed data for FEV decline and annual exacerbations PPPY. The model over predicted survival by 8% (absolute) compared to TORCH observed data at year 3. Conclusions: As expected, the model more accurately predicted the ECLIPSE observed outcomes in the internal validation exercise, than TORCH outcomes in the external validation. A561 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PRM101 Application Of A Model Of Decision Based On Fuzzy Logic To Pharmacoeconomics: Ranibizumab VERSUS Aflibercert In AMD Alonso Herreros J M 1, González-Cuello A 2 LOS ARCOS MAR MENOR, SAN JAVIER (MURCIA), Spain, 2Murcia University, MURCIA, Spain . . . 1HOSPITAL Objectives: The term “fuzzy logic” was introduced in 1965 by LAZadeh. Compared to traditional logic, fuzzy logic variables may have a truth value in degree. Fuzzy logic has been applied to many fields, from economic analysis, to artificial intelligence. However it has not been applied so far to pharmacoeconomics. We present a model of pharmacoeconomic decision based on fuzzy logic (Fuzzy Economic Review 2001; 6 (2): 51-73) and applied to the selection of ranibizumab-aflibercept in treating AMD. Methods: According to a decision analysis model based on fuzzy logic four fuzzy variables that affect the choice of treatment are defined: treatment success (expressed as a probability), cost of success, cost of failure (expressed as inverses), and other conditions about the cost (negotiation, handling of drugs...). Based on the value of these fuzzy variables, three linguistic variables (High, Medium, Low) are defined to expressing convenience of choice. The combination of the three possible values for each of the variables gives us 81 possible decision rules, so that the (HHHH) would be the most favorable option and (LLLL) the more unfavorable. So a new fuzzy variable called “ranking” is established for classifying these options with 7 possible values (Very-unfavorable, unfavorable, slightly-unfavorable, neutral, slightly-favorable, favorable, very-favorable). The value of the fuzzy variables for ranibizumab and aflibercept were established based on pivotal clinical trials at 52 weeks cited by the EMEA. Results: The matrices obtained for ranibizumab was (0.29,3. 55 10-4, -1.36 10-4 0.7), and aflibercept (0.269,7. 4 10-4 -2.59 10-4 0.3). These matrices correspond to decision rules (HLLM) and (HMML) and correspond to a ranking of “neutral” and “slightly-favorable”. Conclusions: It possible to apply methods of “fuzzy logic” to pharmacoeconomic studies to select the most favorable treatment. According to model, AMD treatment, with aflibercept would be a slightly more favorable option than ranibizumab. PRM102 Development Of An Influenza Outbreak Forecasting Model Using Time Series Analysis Methods Smolen H J Medical Decision Modeling Inc., Indianapolis, IN, USA . . Objectives: To use historical influenza incidence time series data to develop a predictive model using time series analysis methods to forecast expected number of reported influenza cases. BACKGROUND: Influenza is a common disease associated with high mortality. Low vaccination rates motivate health officials to predict outbreaks and intervene accordingly. A predictive model would facilitate in deciding whether an apparent excess of cases represents an outbreak or a random variation. Methods: Google Flu Trend project data from 2003 to 2014 was used to construct this predictive model. The influenza time series data clearly had a seasonal variation to it so a seasonally fit model using seasonal indicators, a seasonally fit model using trigonometric functions, and a multiplicative seasonal autoregressive integrated moving average (SARIMA) model were considered. Fifty-two weeks of data from the time series were withheld from the model fitting process so as to evaluate the predictive capability of the selected model using mean absolute percentage error (MAPE). The Akaike’s Information Criterion (AIC) goodness of fit measure was used to select the model that fit the data the best (lower the better). Results: The SARIMA model provided the best fit for the data with an AIC of 6361.7. The seasonally fit model using seasonal indicators had an AIC of 8473.9 and the seasonally fit model using trigonometric functions had an AIC of 8438.2. The SARIMA model MAPE for the predicted 52 weeks was 87.5%. The forecasted values were within the 95% confidence band of the actual ending 52 week data, though at the high end of the band. Conclusions: The SARIMA model was an appropriate predictor for flu cases in 2013-4. The data used to construct the model included flu epidemics so removing these time periods would result in a model more appropriate for non-epidemic periods. PRM103 Development Of A Global Economic Model To Evaluate The CostEffectiveness Of Targeted Treatments Using Companion Diagnostics In Advanced/Metastatic Cancer Treatment Mathurin K , Beauchemin C , Lachaine J University of Montreal, Montreal, QC, Canada . . . Objectives: With the development of high priced new targeted treatment for cancer, there is a need to know as soon as possible if these treatments are likely to be cost-effective. The objective of this study was to develop a model with global parameters to estimate the cost-effectiveness of targeted treatments using companion diagnostics in advanced/metastatic cancer treatment. Methods: The model was developed to take into account parameters usually considered in conventional economic models in cancer (treatment costs, costs of cancer care, target population characteristics, survival data, utilities, disutilities and costs associated with adverse events (AEs), etc.), and also parameters specific to the companion diagnostic itself (mutation prevalence, test specificity and sensitivity, and cost). The model had to allow performing cost-utility analyses from both a Health Ministry and a societal perspective and for most common cancers (lung, breast, colorectal, prostate, cervical/endometrial, bladder, and non-Hodgkin’s lymphoma). Results: The global model comprises a decision tree and a lifetime Markov model. The decision tree takes into account the sensitivity and specificity and cost of the companion diagnosis, and the prevalence of the biomarker/mutation in the eligible population. The Markov model with monthly cycles includes the following 3 health states: progression-free, progressive disease and death. Intrisic parameters of the model comprise the mean characteristics of the target population, utilities associated with health states, disutilities and costs associated with AEs, and costs associated with drug administration, cancer care, end-of-life care, follow-up visits, productivity losses, and informal care. Specific parameters to be entered by users are the prevalence of the mutation, treatment costs, specificity/sensitivity and cost of the test, survival data and the incidence of AEs. Conclusions: The proposed global model for the economic evaluation of targeted treatments using companion diagnostics in advanced/metastatic cancer treatment can, with minimal input, quickly generate cost-effectiveness analyses of targeted cancer treatment. PRM104 A Web Based Optimisation Model For A Portfolio Of Preventative Interventions Utilizing Multi Criteria Decision Analyses (Mcda) Framework Topachevskyi O 1, Volovyk A 2 1Digital Health Outcomes, Brussels, Belgium, 2Hashtago, Kiev, Ukraine . . Objectives: To inform decision makers who seek extension of Universal Mass Vaccination (UMV) about the most optimal allocation of funds across multiple preventative interventions or vaccines. To account for decision makers preferences using MCDA. Methods: A multi cohort markov model was developed to assess clinical and economic consequences of vaccine preventable diseases in Japan. Disease incidence rates, direct medical costs and QoL data were obtained from local sources. Payer perspective only was considered. Optimization module utilizing linear programming was developed to maximize outcome of interest which serve as an objective function subject to budget and intervention coverage constraints. A working version of the model can be found at http: //www. digitalho. com/models/a/portfolio/index. html. The model was initially developed in Excel and then automatically transformed into a JavaScript application to allow for an online access. One way sensitivity analyses was conducted to parametric unceranity. Results: Model results indicate that the optimal mix of interventions depends primarily on the objective function. Various single objective functions or a combination of multiple weighted objectives lead to different mix of interventions. When prevention of death is as an objective function then pneumococcal and rotavirus vaccines are chosen. Conclusions: The proposed web based model is a complementary addition to the conventional cost-effectiveness assessment for preventative interventions. This model helps to understand sequence of introduction of prevantative interventions and expected health and economic outcomes over time. The use of MCDA framework helps users to define specfic health objectives to be used in optimisation module. The web based modeling solution provides a widespread access to an easy to use tool that can by used by authorities, academia and non-modeling professionals. PRM105 Calibration And Statistical Modeling To Inform A Micro-Simulation Model For Early HTA Bongers M L 1, De Ruysscher D 2, Oberije C 3, Lambin P 2, Uyl-de Groot C A 4, Coupe V M 1 University Medical Center, Amsterdam, The Netherlands, 2University Hospitals Leuven/KU Leuven, Leuven, Belgium, 3MAASTRO Clinic, Maastricht, The Netherlands, 4Erasmus University Rotterdam, Rotterdam, The Netherlands . . . . . . . . . 1VU Objectives: For the evaluation of the potential cost-effectiveness of an early experimental therapy, we calibrated an existing micro-simulation model for radiotherapy planning in lung cancer using pilot data. Methods: We used an externally validated micro-simulation model, build using Real World Evidence data. The model contained four clinical states from alive to death, with intermediate states ‘local recurrence’ and ‘metastasis’, with 5 transitions. Based on individual and time-dependent hazard rates, patients move through the model according to their combination of patient characteristics and random variation. For the experimental dosis-escalation therapy we had limited pilot study data, which included overall survival and a number of baseline characteristics. The distribution of patient features in the cohort of the micro-simulation model was adjusted so that the simulated patients had the same baseline characteristics as the patients that received experimental therapy. Alternative radiotherapy strategies affected 5 transitions in the model, quantified by 5 hazard ratios (HRs). Subsequently, HRs for experimental radiotherapy compared to current radiotherapy were calibrated until they were able to satisfactorily reproduce the survival curve of the pilot data. The best fitting sets of HRs were selected based on the least Sum of Squared Errors (SSE) of the model predictions and the survival curve of the experimental therapy on three time points. Results: The best fitting set HRs resulted in a SSE of 0,005 based on prediction errors at 1,2 and 3-year survival. Although 33 out of 1000 sets produced predictions with less than 5% prediction error, hazard ratios varied strongly within and over the different sets. Conclusions: By using calibration, we obtained a micro-simulation model that is suitable for the evaluation of new treatments in the absence of empirical data. The model will be used for cost-effectiveness analyses, where the variation in hazard ratios within sets will be evaluated in scenario analyses. PRM106 How Does Uncertainty Around Costs And Effects Relate To Uncertainty Around Cost-Effectiveness? Jain M 1, Bhattacharyya S 1, Gupta S 1, Sonathi V 1, Mahon R 2, Malakar H 1, Vudumala U 1, Gunda P 1, Kumar P 1, Partha G 1, Thomas S K 3 Healthcare Pvt. Ltd., Hyderabad, India, 2Novartis Ireland Limited, Dublin, Ireland, 3Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA . . . . . . . . . . . . 1Novartis Objectives: HTAs require information on costs and outcomes as well as the uncertainty around them for making reimbursement decisions. Uncertainty around costs and effects (outcome uncertainty) can be substantial and increasingly so at more distal time points. However, the uncertainty surrounding the decision to adopt or reject a technology based on cost-effectiveness (decision uncertainty) evolves over time in a different manner. In this analysis, we intend to illustrate that increased outcome uncertainty need not result in increased decision uncertainty and that both may evolve over time differently. Methods: A previously published lifetime Markov model, built from UK health care perspective, was used in the analysis. The model compared the cost-effectiveness (CE) of A562 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Zidovudine + Lamivudine combination therapy vs Zidovudine monotherapy, to treat HIV infection. Based on probabilistic simulations, cumulative incremental net monetary benefits (CINMB) at a CE threshold of £20,000/QALY and probabilities of being cost-effective at various time-horizons (1-20 years) were estimated. Further, for each time-horizon, a CINMB frequency distribution was plotted and summary statistics were estimated. Results: For the combination therapy, while the outcome uncertainty increased over time, the decision uncertainty decreased. 95% confidence interval for expected CINMB was narrowest at year 1 (-1,771£ to -1,755£) and widest at year 7 (2,101£ to 2,209£); simultaneously the probability of being cost effective increased from 5% to 80% during this time. Outcome uncertainty, measured as the standard deviation of CINMB values stabilized after 5 years while probability of the combination therapy being cost effective continued to increase, indicating that decision uncertainty does not vary in tandem with outcome uncertainty. Conclusions: The above analysis shows that higher outcome uncertainty does not necessarily lead to higher decision uncertainty. CINMB could be a useful tool to observe the relationships between outcome uncertainty, decision uncertainty and time. PRM107 Development Of A Model To Assess The Cost-Effectiveness Of Therapies For Patients With Type 2 Diabetes Mellitus (T2DM) Following A Reference Model Framework Aguiar-Ibáñez R 1, Palencia R 2, Kandaswamy P 3, Flavin J 4, Gauthier A 5, Davies M J 6 1Amaris Consulting UK, London, UK, 2Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany, 3Boehringer Ingelheim UK, Bracknell, UK, 4Boehringer Ingelheim Canada Ltd, Burlingon, ON, Canada, 5Amaris, London, UK, 6University of Leicester, Leicester, UK . . . . . . . Objectives: To describe the practical approach implemented to construct a global cost-effectiveness model for T2DM therapies following a framework proposed for the development of reference models to inform public funding decisions. Methods: 1) A systematic review of published models was conducted to conceptualise the model in terms of natural history and relevant effects to include. 2) Clinical and health economic experts were selected to provide feedback during the model conceptualisation (to identify the appropriate modelling technique), the model implementation and the assessment of the results. 3) The model was built and populated based on the systematic identification of best available data, a network meta-analyses, a review of previous T2DM submissions to health authorities and other published information. The model incorporated several structures for uncertain areas, such as: treatment patterns; type and timing of adverse events; their impact in the occurrence of long-term complications; and the impact of weight changes on relevant endpoints. 4) The model was then validated based on outputs’ accuracy, feedback from country affiliates and consistency with the CORE model results. 5) The critical feedback received by HTA bodies has also been used to refine the model and improve its credibility accordingly. Results: Experts’ input proved invaluable at each developmental stage. One challenge related to the comparability with other published T2DM models, which were not fully transparent regarding assumptions. This framework resulted in a flexible model, accurate and stable, and easily adaptable to different health care systems. Country adaptations have contributed to the identification of aspects that require relevant structural changes and their rationale. Conclusions: The followed framework enhanced the transparency of the model and the accuracy of the results. Using a reference model across different countries, with adaptations made in consistency with this model, should help ensure consistent and comparable evaluations of the model across different countries. PRM108 Assessing The Relationship Between Individual Attributes Identified In Review Of Multi-Criteria Decision Analysis (MCDA) Of Rare Diseases And Annual Treatment Costs In Rare Endocrine Disorders Schey C 1, Irwin J 2, Teneishvili M 2, Krabbe P F M 3, Connolly M 4 of Groningen, St Prex, Switzerland, 2Shire Pharmaceuticals, Maidenhead, UK, 3University of Groningen, University Medical Center Groningen, Groningen, The Netherlands, 4University of Groningen, Groningen, The Netherlands . . . . . . . 1University Objectives: Payers have a perception that orphan products are extremely expensive. The current health technology assessment (HTA) systems might be too restrictive for orphan drugs, therefore potentially denying patients access to life-saving medicines. While price is important, it should be considered in relation to a broader range of product attributes, such as unmet need and disease severity that are not considered in cost-effectiveness analysis used by many HTA agencies. To overcome these challenges multi-criteria decision analysis (MCDA) has been proposed as an alternative to evaluate technologies. The aim of this study was to identify criteria reported in the literature, and to assess their impact on the total “score” for each product in relation to price. Methods: A systematic literature review was conducted to identify the most frequently cited attributes in MCDA. From the leading attributes identified, we reviewed and plotted the relationship between single attributes and the average annual treatment costs for several drugs used in the treatment of endocrine-related rare diseases. Annual treatment cost was based on UK prices for the average daily dose per patient. Results: The three most frequently mentioned attributes were ‘disease severity’, ‘treatment impact on condition’, and ‘level of research undertaken to support use of the product’. Disease severity was not shown to influence product price. Similarly, orphan drugs are not necessarily more expensive than products without orphan drug status. There is little discernible relationship between treatment ‘convenience’ and average annual treatment cost. A trend was observed between the market size and the average annual treatment cost. Conclusions: If society is concerned about equity and equal access to medicines for all patients, MCDA may offer a viable alternative to inform in reimbursement decisions for orphan drugs. The analysis can be used to inform investigations on the application of MCDAs in rare diseases. PRM109 Visualizing Methods For Discrete-Event-Simulations Using The Example Of A Breast Cancer Decision-Analytic Model Jahn B 1, Rochau U 2, Shterjovska J 1, Kurzthaler C 3, Kluibenschädl M 1, Urach C 4, Einzinger P 4, Piringer H 5, Popper N 4, Siebert U 6 1UMIT - University for Health Sciences, Medical Informatics and Technology, Hall in Tyrol, Austria, 2UMIT - University for Health Sciences, Medical Informatics and Technology/ ONCOTYROL Center for Personalized Cancer Medicine, Hall in Tyrol/ Innsbruck, Austria, 3UMIT - University for Health Sciences, Medical Informatics and Technology / Oncotyrol - Center for Personalized Cancer Medicine, Hall i. T. / Innsbruck, Austria, 4Vienna University of Technology, dwh Simulation Services, Wien, Austria, 5VRVis Zentrum für Virtual Reality und Visualisierung Forschungs-GmbH, Vienna, Austria, 6Medical Informatics and Technology, and Director of the Division for Health Technology Assessment and Bioinformatics, ONCOTYROL, Hall i. T, Austria . . . . . . . . . . Objectives: Discrete-Event-Simulation (DES) is a commonly used modeling tool to analyze the comparative effectiveness of alternative health technologies and to optimize resource allocation in health care settings. DES models are often rather complex and visualization is very important to improve transparency and acceptability. This study aims to illustrate and contrast alternative visualization techniques on a decision-analytic model for breast cancer. Methods: DES visualization methods and their applications in health care, engineering, and operations research were sought from a wide variety of sources, including literature databases (e.g., PubMed) and webpages of simulation conference (e.g., Winter Simulation Conference), academic societies etc. Based on this review, alternative visualization techniques for the conceptual model were selected, applied on a real world modeling example and compared. Results: In health care, the recently published ISPOR-SMDM Modeling Good Research Practice guidelines recommend flow diagrams or state charts to represent the key elements of a model, including the possible pathways, and the presence of queues and decision points. For flow charts, we found an international standard (ISO 5807). The application of standards like this could support harmonization of process-oriented models. In general, flow charts may lack the information of health states and transitions between health states that are relevant for clinicians to review the model. The semantic for state charts invented by Harel provides a further development of the bubble diagrams of State-Transition (Markov) Models (e.g. one state containing other states, one state detects changes in another). In state charts, health states could explicitly be named but treatment processes and resources use are less explicit. For DES software implementation, state charts seem to be less intuitive. For both methods, the application of visualization standards and guidelines was not always straight forward for our breast cancer model. Conclusions: In the case example there was no superior visualization technique. PRM110 Microsimulation Model For The Assessment Of Personalized Cancer Care: The Mapcca Model Framework Van der Meijde E 1, van den Eertwegh A J 1, Fijneman R J 1, Meijer G A 1, Linn S C 2, Coupe V M 1 University Medical Center, Amsterdam, The Netherlands, 2Netherlands Cancer Institute, Amsterdam, The Netherlands . . . . . . . . . . . 1VU Objectives: Most cancer care models are based on observed clinical events such as recurrence-free and overall survival. Times at which events are recorded depend not only on effectiveness of treatment, but also on timing of examinations and types of tests performed. Should these change, observation times would change as well. Construct a microsimulation model that describes the cancer disease process using a description of underlying tumor growth as well as its interaction with diagnostics, treatments and surveillance. The aim is to arrive at a framework that allows for exploration of the impact of simultaneously altering two or more aspects of the care process. Methods: The framework consists of two components; the disease model and the clinical management module. The disease model consists of atumor level, describing the growth and metastasis of the tumor, and a patient level, describing clinical observed states, such as recurrence and death, either from the disease or other causes. The clinical management module consists of the care patients receive, i.e. the diagnostic process, treatment and surveillance. This module interacts with the disease process, influencing the rate of transitioning between tumor growth states at the tumor level, and the rate of detecting a recurrence at the patient level. Results: A simulation study was performed to examine the feasibility of applying the framework to melanoma progression. Results demonstrated stage specific recurrence rates similar to those found in literature. Conclusions: The proposed microsimulation model framework allows for generating individual patient histories by simulating underlying tumor growth in interaction with clinical management. Our modeling approach allows for the exploration of the potential of drugs intervening in different parts of the tumor growth pathway. In addition, the approach allows for the evaluation of changing diagnostic patterns. PRM111 Methodological Evaluation Of The Impact Of Survival Costs In Oncology Modelling Taylor M 1, Filby A 1, Proudfoot C 2 1York Health Economics Consortium, York, UK, 2Sanofi, Guildford, UK . . . Objectives: Economic evaluations typically include all costs relevant to a disease, not only drug-related costs. This is particularly relevant to oncology modelling, as costs are assigned to each health state in the model, and, therefore, extending survival also increases costs. Because patients often incur higher health care costs in the post-progressed state of disease where costs of disease management are high, extending survival and increasing a patient’s time in the post-progressed stage can be particularly costly. Empirical analyses of the implications of such methods have not yet been extensively investigated by assessing different scenarios such as baseline severity and prognosis. The objective of this research was to investigate the methodology used in oncology modelling, and to determine the effect that this has on predicted cost-effectiveness. Methods: We developed a flexible three-state A563 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 economic model with ten key parameters to calculate the ICERs associated with various combinations of inputs. Published HTAs were reviewed to determine the model inputs. Extensive scenario and multiway sensitivity analyses were carried out to document informative patterns and relationships between parameters that affected the results. Results: Results showed that cancer sub-types with higher post-progression costs reduced a treatments likelihood of being cost-effective. The results also highlighted specific thresholds at which various cancer-specific case studies or combinations of inputs, including drug price, resulted in the drug being deemed not cost-effective using a threshold of £20,000 per QALY. Conclusions: The impact of post-progression costs can vary dependent on how these costs are modelled and also dependent on several factors, namely the ratios between health state utilities, ‘background’ costs, drug costs and the relative time spent in the stable and progressed disease states. It is demonstrated that, for many oncology treatments whose primary aim is to extend survival, this impact can be prohibitive to an intervention’s probability of being cost-effective. PRM112 Dose-Response Network Meta-Analysis To Address Dose Heterogeneity In A Cost-Effectiveness Analysis In Acute Migraine Reason T 1, Dias S 2, Welton N 2 Health, London, UK, 2University of Bristol, Bristol, UK . . . 1IMS Objectives: Network Meta Analyses (NMAs) are often used to parameterise efficacy in decision models for economic evaluation. A common source of heterogeneity in NMA arises from the fact that treatments may be given at different doses. This variation may manifest as unexplained heterogeneity in standard NMA models and propagates through to the decision analysis. We aim to explore how dose-response NMA can be used to inform cost-effectiveness analysis using a cost-utility analysis of treatments for acute migraine. Methods: We conducted four NMAs with different assumptions around dose-response to inform an economic evaluation in acute migraine. Separate 1-level NMAs were conducted where interventions were ‘lumped’ at the ‘dose’, ‘treatment’ and ‘class’ levels and a multi-level NMA was conducted, assuming monotonic dose-response. All NMAs were used to inform effect sizes in an economic model; the model structure, costing methods and utility inputs from the NICE Headaches guideline were adopted. The NMA models were compared in terms of heterogeneity and Deviance Information Criteria (DIC). We report the results of the economic analyses using cost-effectiveness acceptability frontiers (CEAFs). Results: Dose-response parameterisation lead to NMA models with lower heterogeneity and better fit. Different dose-response parameterisations substantially changed the resource allocation decision, particularly at lower willingness to pay thresholds. For the more complex NMA model, it is unclear from a decision making perspective which effect size estimates should be selected as inputs to the decision model and we show that careful consideration should be given to the relevance of individual doses and confounding bias. Conclusions: Dose-response NMA provides a useful and arguably more appropriate method for conducting NMA for decision analysis. Careful consideration should be given to how to treat doses of the same intervention in NMA since different parameterisations may lead to biased effect sizes and sub-optimal conclusions around cost-effectiveness. PRM113 Impact Of International And Therapeutic Referencing On Prices And Launch Optimization Hakim P 1, Weiss J 1, Degun R 1, Chalmers M 1, Kjeldgaard-Pedersen J 2, Sleeper M 3 Consulting, Inc, London, UK, 2Data2Impact, Copenhagen, Denmark, 3Asera Consulting, Surrey, UK . . . . . . 1Navigant Objectives: Majority of established pharmaceutical markets use pricing rules that reference products both across (international reference pricing; IRP) and within (therapeutic reference pricing; TRP) country-lines. Some markets, such as the UK, use no such rules and freely price therapies. IRP and TRP are used as effective measures to control price of pharmaceutical products. However, understanding the impact on a global level is considered highly complex, due to each market using different legislations and mechanisms. There is a growing need to better understand the impact of IRP and TRP on the decay of pharmaceutical drug prices and their effects on launch prices and sequencing across multiple markets. Methods: We utilized a model to simulate the impact of IRP and TRP, as well as parallel trade, to quantitatively assess impact on drug prices across 40 markets (EU28, Switzerland, Russia, Iceland, Norway, Turkey, Israel, Brazil, Japan, South Korea, Australia, Canada and USA). The model uses simulating annealing parameters to also yield optimized launch sequences in all markets. Results: Outputs of the model include impact to volume, price and revenue as well as parallel trade. Using a dynamic launch map, the model provides optimized price and launch sequence of a given pharmaceutical product. Within a set window, the launch and reimbursement dates were optimized based on a divergent set of rules and country baskets, often differing from manufacturers expected launch prices and sequences. While manufacturers are often able to secure higher price for therapies in free-price markets, the model clearly demonstrates the spill-over impact of referencing (both formally and informally) across countries and within therapeutic groups. Conclusions: Previously, launch sequences were optimized based on the implications on major markets. However, an expansive model looking at large number of markets that employ varying IRP and TRP rules will assist manufacturers in identifying an optimized price and launch sequence strategy. PRM114 Modeling Disease Progression In Alzheimer’s Dementia To Inform HTA (CEA) Green C , Zhang S Exeter University, Exeter, UK . . Objectives: Alzheimer’s dementia (AD) poses a significant challenge to health care systems around the world. Whilst treatment options are currently limited, with no effective disease modifying treatment, new advances in diagnosis and manage- ment of AD and promising advances in health technologies have the potential to significantly impact on the burden of the disease. However, alongside treatment advances it is important to improve the evaluation framework if we are to capture the potential benefits to people with AD. Methods to model disease progression over time, for use in comparative and cost-effectiveness analyses (CEA), is a priority area for further research. The objective in this research is to develop a new framework for modeling AD progression over time using the three main symptom domains of cognitive function, behaviour and mood, and functioning. Methods: Development of a descriptive system, comprising a set of health states for AD, using the three symptom domains. Statistical modeling of disease progression through states over time, using US data from the National Alzheimer’s Coordinating Center (NACC) (n= 3009). The model is tested in a decision-analytic context, using time to progression and a cost-per-QALY framework. Results: A 20-state disease progression pathway has been developed using multi-variate health states described using the three symptom domains. Transition probabilities and hazard rates have been estimated to model progression over time through the multi-variate descriptive system. In a baseline model over a 5-year timeframe, using mild-to-moderate AD starting states, 78% of people progressed to health states considered severe on at least one of the symptoms (46% severe for cognition). In a HTA context simulating a treatment with a modest effect the modeling framework predicted significant QALY differences between control and treatment over 5-years. Conclusions: This new modeling framework shows promise and presents a broader opportunity to capture the impacts of treatment over time using a range of symptom domains. PRM115 Use Of External Data To Guide Long-Term Survival Extrapolations Of Trial Data For Chronic Lymphocytic Leukemia Hawe E 1, Pearson I 1, Wolowacz S 1, Haiderali A 2 1RTI Health Solutions, Manchester, UK, 2GlaxoSmithKline, Collegeville, PA, USA . . . . Objectives: The National Institute for Health and Care Excellence recommends utilising external data to evaluate the validity of extrapolation beyond trial followup. The objective was to demonstrate the use of external data to guide long-term survival predictions where only short-term trial data are available. Methods: Four-year patient-level data for chlorambucil and ofatumumab+chlorambucil from the COMPLEMENT-1 trial were available; the survival rate was over 70% at four years. Published 18-year Kaplan-Meier data for chlorambucil from the C9011 study, which compared fludarabine and chlorambucil, were used to simulate a patientlevel dataset. These data were not used directly as outcomes have improved substantially since patients were enrolled (1990-1994) and the chlorambucil dosing regimens differed. However the curve was used to guide the extrapolation of the COMPLEMENT-1 data using a three-stage approach: 1) Survival functions were fitted for the COMPLEMENT-1 and C9011 chlorambucil arms with an indicator for study. 2) The average treatment effect (chlorambucil versus ofatumumab+chlorambucil) was estimated from survival analysis for both arms of COMPLEMENT-1 and the C9011 chlorambucil arm, with indicators for treatment and study. 3) For ofatumumab+chlorambucil, long-term survival was estimated by applying the treatment effect from stage 2 to the function for chlorambucil from stage 1. Exponential, gamma, Weibull, log-normal, log-logistic and Gompertz functions were fitted. Results: Weibull, Gompertz, and gamma functions provided reasonable fits during trial follow-up and plausible extrapolations (assessed by Akaike information criterion, Bayesian information criterion, graphical diagnostics, and expert clinical opinion). Survival predictions were markedly lower than conventional functions fitted to COMPLEMENT-1 data only (e.g. Weibull predictions for chlorambucil at 20 years were 8% vs 26%, respectively). Conclusions: Using long-term external data to guide the extrapolation provided plausible predictions for chlorambucil upon which alternative scenarios for the continuation of treatment effect observed in COMPLEMENT-1 could be explored. PRM116 Modelling Evolving Cancer Risk During Epidemiological Transition Using Economic Data Hughes M , Seesaghur A , De Silva D Decision Resources, Burlington, MA, USA . . . Objectives: Epidemiological projections sizing patient populations are fundamental to budget impact analyses and market forecasting. When disease risk evolves over time, as in the case of epidemiological transition, using historical estimates in epidemiological projection becomes unjustified. Incorporation of additional variables that model evolving risk may allow for more reliable forecasts. The hypothesis that gross domestic product per capita (GDP) is correlated with disease risk was tested for a variety of cancers using global epidemiological and economic datasets. Methods: Age-standardized incidence for 18 cancer sites across 188 countries was retrieved from the International Agency for Research against Cancer for the years 1993-1997. Corresponding country-specific GDP data was collected from the World Bank. For each site, correlation between GDP and incidence was measured using R2and linear co-efficient values. Results: Risk is strongly correlated with GDP for all of the sites studied, with the exception of the stomach (R 2= 0; p> 0.05). Correlation was strongest for those sites associated with diet/lifestyle factors prevalent in higher-income countries, namely: colorectal (R2= 0.61; p< 0.001), breast (R2= 0.61; p< 0.001) and lung (R2= 0.42; p< 0.001). Strong association was also seen for prostate (R2=0.47; p<0.001), although this may be explained by better case-detection in higher-income countries. Conversely, those cancer sites associated with infection showed a negative correlation, namely: cervix (R2= 0.33; p< 0.01), liver (R2= 0.09; p< 0.001) and oesophagus (R2= 0.06; p< 0.01). Conclusions: GDP is strongly associated with cancer risk and varies by organ site in a manner concordant with the evolving exposures to known pathogens that characterize epidemiological transition. On the assumption that such correlation is a marker for various causal relationships, and that robust economic methods underlie GDP forecasts, it is reasonable to conclude that these correlations could be used to make epidemiological projections more accurate than projections assuming constant disease risk. A564 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PRM117 Cost-Effectiveness Modeling Of Antimicrobial Dressings For Preventing Catheter-Related Bloodstream Infection: Homogeneous VERSUS Non-Homogeneous Markov Approaches Maunoury F 1, Motrunich A 1, Palka-Santini M 2 1Statésia, Le Mans, France, 23M Germany, Neuss, Germany . . . Objectives: To compare homogeneous (HMM) versus non-homogeneous Markov models (NHMM) for cost-effectiveness analysis (CEA) of routine use of transparent dressings containing a chlorhexidine gluconate gel pad versus standard transparent dressings. The antimicrobial dressing protects central vascular accesses reducing the risk of catheter-related bloodstream infections (CRBSIs) in intensive care units (ICU). The impact of the modeling approach on the decision of adopting antimicrobial dressings for critically-ill patients is discussed. Methods: Comparative clinical efficacy data from a multicentre randomized controlled trial (RCT) enrolling 1,879 patients and economical data from micro and macro-costing published studies were combined. The HMM and NHMM models were built separately using the same sources. The statistical unit was the ICU patient and the ICU perspective was chosen. Probabilistic sensitivity analyses (PSA) were conducted for both models for comparing the robustness of the CEA results. Results: The difference in clinical outcomes between each dressing strategies was statistically significant with both models while cost differences were not. The PSA with the NHMM resulted in 11.8 infections avoided per 1,000 patients (95%CI: [3.85; 19.64]) and a mean extra cost of € 141 per patient (95%CI: [€ -975; € 1,258]) when using antimicrobial dressing. The PSA with the HMM resulted in 6.45 infections avoided per 1,000 patients (95%CI: [0.15; 12.75]) and the mean extra cost of € 252 per patient (95%CI: [€ -924; € 1,428]). Conclusions: The antimicrobial dressings are consistently more efficacious in preventing CRBSIs whatever the model used. The HMM is less sensitive to simulate the real life of the ICU patients. Regardless the model approach chosen the antimicrobial strategy is more efficacious than the comparator, but its probability of being cost-effective is comparatively reduced with the HMM. Time dependent approach (NHMM) seems to be better adapted to model rare events as CRBSIs. PRM118 Development Of A Model To Predict Disease Progression In Autosomal Dominant Polycystic Kidney Disease (ADPKD) McEwan P 1, Bennett Wilton H 2, Robinson P 3, Hadimeri H 4, Ong A 5, Ørskov B 6, Peces R 7, Sandford R 8, Scolari F 9, Walz G 10, Woon C 11, O’Reilly K 3 1Swansea Centre for Health Economics, Swansea, UK, 2Health Economics and Outcomes Research Ltd, Cardiff, UK, 3Otsuka Pharmaceutical Europe Ltd, Wexham, UK, 4Department of Nephrology, Kärnsjukhuset, Skövde, Sweden, 5Academic Nephrology Unit, Department of Infection and Immunity, The University of Sheffield Medical School, UK, 6Department of Medicine, Renal Division, Copenhagen University Hospital, Roskilde, Denmark, 7Hospital Universitario La Paz, Madrid, Spain, 8Academic Laboratory of Medical Genetics, Addenbrooke’s Treatment Centre, Cambridge, UK, 9Department of Nephrology, University of Brescia, Italy, 10Department of Nephrology, University Medical Centre Freiburg, Zentrale Klinische Forschung, Freiburg, Germany, 11McCann Complete Medical, Macclesfield, UK . . . . . . . . . . . . Objectives: Autosomal dominant polycystic kidney disease (ADPKD) is a major cause of end-stage renal disease (ESRD) affecting approximately 4 per 10,000 people in Europe. There is a paucity of research regarding the natural history of ADPKD progression. This study aimed to utilise the results of a systematic literature review characterising predictors of ADPKD progression to construct a natural history disease model for ADPKD. Methods: An individual patient-level lifetime simulation was developed in Microsoft Excel, driven by baseline and time-dependent age, estimated glomerular filtration rate (eGFR) and total kidney volume (TKV). Rates of progression were informed by a large naturalistic study. Dialysis modality, transplant status and disease-specific mortality were also modelled. Relevant ADPKD complications were stratified by chronic kidney disease stages. Modification of disease progression rate was investigated in order to assess the potential of the model for evaluating treatment interventions. Results: On visual inspection, modelled and published eGFR trajectories for the general ADPKD patient were consistent (median age at ESRD of approximately 55 years). When patients are stratified by baseline TKV the model predicts variable rates of progression to ESRD, aligning with the assertion that the baseline TKV is the most important prognostic indicator for ADPKD progression. Modification of the risk equations to incorporate the impact of an intervention has shown promise to estimate important outcomes such as delay to ESRD. Conclusions: The model has demonstrated both face and predictive validity and is capable of predicting outcomes consistent with those reported in the ADPKD literature. It represents the first model capable of informing on important clinical outcomes relevant to both clinicians and patients, such as time to ESRD, with the potential to evaluate the long-term impact of treatment interventions on ADPKD progression. PRM119 Forecasting Cancer Incidence Using Gross Domestic Product Ayodele O A 1, Isherwood A 2, Hughes M 2 1Decision Resources Group, Burlington, MA, USA, 2Decision Resources Group, London, UK . . . . Objectives: To validate the use of change in Gross Domestic Product per capita (GDP) as a proxy for forecasting change in cancer risk. Methods: Using data from the International Agency for Research on Cancer (IARC) and GDP data from the World Bank, the correlation between the incidence of five cancers (breast, prostate, lung, liver, and colorectal) in 147 countries for 1993-1997 (midyear 1995), and the corresponding GDP was calculated. We used this correlation to retrospectively forecast the 2005 incidence for five cancers in four cities (Cali, Colombia; Delhi, India; Shanghai, China; St. Petersburg, Russia). The GDP-based forecasts were compared to a control assuming no change in incidence from the IARC incidence data for 1998-2002 (midyear 2000). The two forecast estimates with the actual reported IARC 2003-2007 incidence data were compared. Results: Overall, the GDP-based method correctly forecasts the directional change in incidence in 75% (95% confi- dence interval 67-83%) of instances compared with the control method (25%, 95%CI 17-33%), and among 83% (95%CI 76-91%) of positively-GDP-associated cancers compared with 17% (95%CI 9-24%) for the control method. There was no significant difference between the two methods for negatively-GDP-associated cancers. In terms of the relative magnitude of change compared to the actual incidences, there was no significant difference between the GDP-based forecast incidences and the control (mean magnitude difference 3.8%, 95% CI -3.5% - 11.0%). Conclusions: During epidemiological transition, cancer incidence is unlikely to remain static and so developing a proxy variable to evaluate risk over time is important. Using GDP as a proxy variable is preferable to the alternative of projecting historical values forward unchanged in terms of directional effect. However our study found no significant difference in terms of relative magnitude of the change over time. This latter result may be due to a small sample size of registries, indicating a need for further research. PRM120 Estimating Crossover Bias In A Randomized Clinical Trial Of Ovarian Cancer Treatment Kuehne F 1, Rochau U 2, Paracha N 3, Sabate E 3, Siebert U 4 of Public Health and Health Technology Assessment, UMIT - University for Health Sciences, Hall, Austria, 2UMIT - University for Health Sciences, Medical Informatics and Technology/ ONCOTYROL - Center for Personalized Cancer Medicine, Hall in Tyrol/ Innsbruck, Austria, 3F. Hoffman-La Roche, Basel, Switzerland, 4Medical Informatics and Technology, and Director of the Division for Health Technology Assessment and Bioinformatics, ONCOTYROL, Hall i. T, Austria . . . . . 1Department Objectives: Bevacizumab (BEV) has proven to be efficacious to delay ovarian cancer progression and in controlling ascites. Significant overall survival (OS) benefits were reported in GOG218 patients with ascites at baseline, however, not in the ITT population. This raises questions about the potential role of “progression-driven cross-over” in confounding OS trial results. The objective of this study was to use decision-analytic modeling to estimate the potential impact of this “cross-over bias” in a RCT. Methods: A decision-analytic Markov model with 6 mutually exclusive health states was built simulating OS as reported in GOG218. In the base case analysis, a 49 month time horizon was used (trial length). Input parameters were based on trial data and published literature. The model was calibrated to estimate the amount of patients developing ascites and switching. The model was then used estimating the unbiased treatment effect as it would have been observed without switching. Results: The model provided similar survival curves as reported. When switching treatment was disabled, the relative risk reduction for OS due to BEV compared to placebo changed from 11% to 32% over the 49 month time horizon. The results were sensitive to changes in the input parameter of switching, developing severe ascites and mortality after progression. Conclusions: Decision-analytic methods are suitable for describing causal relations and the impact of a potential bias. We estimated that about two thirds of the true treatment effect could not be captured in the biased data analysis. However, a formal adjustment of the original RCT data was not feasible due to lack of clinical trial data describing switching patterns. As crossover bias cannot always be avoided by trial design due to ethical reasons, we recommend collecting sufficient confounding variables determining cross-over so that specific statistical methods for causal-inference can be applied to control the cross-over bias. PRM121 The Impact Of The New Drug Co-Payment Scheme On Economic Evaluations In Spain Martin Saborido C , Zuluaga Sanchez S RTI-HS, Manchester, UK . . Objectives: To examine the new co-payment scheme for prescribed medications introduced in Spain in 2012 and to explore differences in patient and National Health Service (NHS) drug cost co-payments between the old and the new schemes, using an example in a modelled patient population with coronary heart disease (CHD). Methods: The new legislation was researched and the new co-payment scheme was summarised in a flowchart. A published economic evaluation of drugeluting versus bare metal stents for high-risk patients with CHD was used to calculate co-payments for the total cost of a prescribed drug (clopidogrel) from patients and from the NHS. The patient contribution was estimated from the income and expected work status of the model’s population. Results: In the new co-payment scheme, pharmacy-dispensed drugs are divided into three categories, using the Anatomical, Therapeutic, and Chemical classification system: 1) reimbursed with reduced contribution (4.26 € per prescription in 2014), 2) reimbursed without reduced contribution (ranging from 40% to 60%, depending on declared incomes for active workers, and from 10% to 60% for retired people) and 3) not reimbursed. Monthly limits for retired people (in 2014) range from 8.26 € to 62 € , with a monthly limit of 0.426 € for long-term, chronic conditions. The current, monthly, over-the-counter price of clopidogrel is 22.92 € . Under the old co-payment scheme, the model estimated that the average NHS payment was 19.62 € per patient (86% of the cost). Under the new scheme, this amount was estimated at 16.55 € (72%). The NHS contribution decreased by 14%. Conclusions: The new scheme results in a significant reduction to drug-related NHS co-payment contributions. This reduction could lead to significant changes in incremental cost-effectiveness ratio estimates. It is recommended that this adjustment be made in economic evaluations developed or adapted for Spain. PRM123 Stratified Cost-Effectiveness Analysis To Guide Genetic Screening For Cancer Risk Folse H J , Dinh T Evidera, San Francisco, CA, USA . . . Objectives: Genetic screening identifies candidates for intensified cancer screening and prevention. Due to the high cost of genetic testing, it is impor- A565 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 tant to identify patients who are most likely to benefit. Doing so using clinical trials is prohibitively expensive; thus a mathematical modeling approach is required. Methods: We developed a framework for stratified cost-effectiveness analysis using individual-based discrete-event simulations, consisting of a natural history component that captures mutation distribution, correlations between mutation and other risk factors (e.g. family history), and cancer incidence, progression and mortality, and a health care process component that captures interactions between the patients and the health care system, through genetic testing, screening, diagnosis and treatment, and their costs. The genetic screening strategy consists of 3 steps: a benefit- risk assessment step, in which patients are assessed for risk of carrying mutations and potential benefits from genetic testing, a genetic testing step, in which qualified patients within an optimal risk bracket are given the appropriate tests and an intervention step, in which patients are given care based on the results from the genetic tests. Results: We use the following approach to explore and identify optimal strategies for 3 genetic screening applications: Dinh et al. demonstrated that primary screening for Lynch syndrome in patients at least 25 years old and with a risk of at least 5% was cost effective. Folse et al. showed that single-nucleotide polymorphism (SNP) screening for breast cancer risk for recommending patients to MRI screening was most cost-effective in women age 40 with a lifetime risk of 16 to 28%. Green et al showed that the same genetic test for recommending patients to chemoprevention was most costeffective for women age 50-59 with a 5-year risk of 1.2-1.66%. Conclusions: As more genetic tests becomes available, this method can be used to identify screening strategies that maximize cost-effectiveness. PRM124 Discrete Event Simulation For The Cost-Effectiveness Evaluation Of Pet-Ct Scans In The Diagnosis Of Conn’s Disease In Hypertensive Patients Maruszczak M 1, Stewart G 1, Kusel J 1, Brown M J 2 1Costello Medical Consulting Ltd., Cambridge, UK, 2University of Cambridge, Cambridge, UK . . . . . Objectives: To develop a flexible and computationally efficient discrete event simulation (DES) model which could be employed in a cost-effectiveness analysis comparing the use of PET-CT scans versus current diagnostic procedures for Conn’s disease in hypertensive patients. Methods: Visual Basic was used for the model simulation with Microsoft Excel constituting the front-end software. In order to ensure a high level of flexibility, individual patients could be assigned a number of personal traits and the clinical, cost and utility inputs were easily adjustable. Individual diagnostic procedures were programmed in separate modules with the aim of simplifying potential modifications to the diagnostic pathway. Results: A DES was constructed to evaluate the cost-effectiveness of new treatments based on the experience of patients assigned to intervention and comparator arms. Patients were considered individually in each arm, using the same background mortality. Time, gender and event dependent risk equations enabled efficient modelling of endogenous heterogeneity of the population. Continuous time accounting allowed for the modelling of competing adverse events and provided a realistic representation of patients’ experience. Preliminary results indicate that the use of PET-CT scans for the screening of Conn’s syndrome could be cost-effective. Conclusions: The newly developed model is the first formal attempt to evaluate the cost-effectiveness of this alternative screening technique for hypertensive patients who are suspected of suffering from Conn’s disease. The model will be further developed to include probabilistic sensitivity analysis and bootstrapping in order to evaluate the robustness of the potential results. Evolutionary algorithms will be incorporated to define the most optimal solution from the continuous spectrum of potential screening strategies. As the model will utilise actual patient level data, it could be used by the decision maker to determine the most cost-effective diagnostic strategy. PRM125 Modelling Long-Term Changes In Opioid Induced Constipation (OIC) Altincatal A 1, Lawson R 2, King F 3, Marsh K 4 . . . . 1Evidera, Lexington, MA, USA, 2AstraZeneca, R&D | Global Medicines Development|Payer and Real Pricing and Reimbursement, Gaithersburg, 3AstraZeneca, Global World Evidence, Cheshire, UK, MD, USA, 4Evidera, London, UK Objectives: Patients’ experience of OIC may be unstable, with periods of constipation and non-constipation, an observation supported by physician reports. There is, however, a lack of quantitative evidence of this experience. Such evidence would be valuable to inform development of economic models for OIC treatments. The objective of this abstract is to fill this gap utilizing data from two pivotal Naloxegol studies, KODIAC 4 and 5, which demonstrated significant improvements in SBM frequency response compared to placebo over 12 weeks. Methods: 892 non-cancer pain patients with OIC were randomized to Naloxegol 25 mg or placebo in two pivotal studies. A 4-week rolling determination of OIC and non-OIC status at weeks 4 through 12 was used in time-to-event analyses. Patients were considered OIC if they reported < 3 SBMs for > = 2 out of the 4 weeks and non-OIC if reported > = 3 SBMs for > = 3 out of 4 weeks. Those with non-OIC status at week 4 were selected as the baseline and first observed OIC status was considered an event. Parametric analyses with Exponential, Weibull, Gompertz, Log-normal, Loglogistic, and Generalized Gamma distributions were conducted. Results: Based on the parametric time-to-event analysis results, the Log-normal distribution was selected as the best fit and provided plausible long term projections. Naloxegol had a noticeable separation for extending the time to first OIC event when compared to placebo over the projected long-term follow-up. Conclusions: This research demonstrates that the natural fluctuation between OIC and non-OIC is substantial and requires integration into an economic model. Even in the absence of treatment, a substantial proportion of patients become non-OIC, and a significant proportion of these remain in non-OIC subsequently. Nevertheless, a treatment effect for Naloxegol was observed over and above this ‘background’ placebo variation in the experience of OIC. PRM126 Companion Diagnostics-Targeted Therapies Pairings Model-Based Economic Evaluation: Reflection On A General Modeling Framework And Key Methodological Points Marty R 1, Roze S 1, Tisseau A 2, Borget I 3, Chouaid C 4 1HEVA HEOR, Lyon, France, 2Merck Serono, on belhalf of the LEEM Biomarker Working group, Lyon, Paris, France, 3Institut Gustave Roussy, Villejuif, France, 4Santé publique au cabinet, Creteil, France . . . . . BACKGROUND: Companion diagnostics (CD) testing aims to stratify the patient population. It conditions the choice of best available therapeutic options, limiting targeted therapy (TT) to subgroups most likely to benefit and triggers potential cost savings. Objectives: To provide a general framework and a list of key methodological points to be addressed while conducting a model-based economic evaluation of CD-TT pairings, especially in oncology. Methods: Based on a health economic literature review and a clinical expert panel with examples drawn from cases of CD testing selection biomarker predictive of the response level towards an anti-cancer TT. Results: As CD and TT have embedded values, it is important to assess them concomittantly within a shared modeling framework. We propose a decision tree to model the patient population stratification and to incorporate impacts of analytical and clinical validity. The former refers here to the inability of the CD to accurately and reliably inform the biomarker resulting in true (false) positive/negative cases, whereas the latter relates to the penetrance, i.e. the strength of association between the biomarker and clinical phenotypes (treatment effect). Such parameters are crucial especially in cases multiple distinct lab-tests (commercial vs. home-brew, technics, amount of informations provided regarding the biomarker). Each patient sub-group outcomes are required to be modeled (costs and health effects). A Markov state-transition model either based on treatment pathway and/or disease staging represents both adequate approaches to simulate the clinical outcomes, incorporating specific efficacy parameters per sub group depending on their biomarker expression levels. In instances, time spent until the CD result delivery exceeds a clinical significant threshold, testing delay shall be modelled such as all parameters driving loss of opporunity. Conclusions: Beyond reasonable simple binary-type of selection biomarker, more complex types of biomarkers and CD technologies (full sequences) has risen additional complexity and poses new methodological challenges. PRM127 Patient Preferences And Hiv Drugs: What About Uncertainty? Broekhuizen H 1, IJzerman M J 2, Hauber A B 3, Groothuis-Oudshoorn C G M 1 1University of Twente, Enschede, The Netherlands, 2University of Twente and MIRA institute for Biomedical Technology & Technical Medicine, Enschede, The Netherlands, 3RTI Health Solutions, Research Triangle Park, NC, USA . . . . . . . . Objectives: Quantitative patient preferences are increasingly considered for health care policy decisions. The objective of this study is to develop a methodology to combine patient preferences with clinical evidence in a multi-criteria framework that takes into account uncertainty in both preferences and clinical evidence. The methodology will be illustrated with a case on antiretroviral treatments. Methods: Treatments under consideration are eight highly active antiretroviral therapies (HAART) recommended for treatment-naïve patients by the National Institute of Health. The treatments are compared on the probabilities of virologic failure, hypersensitivity reaction, bone damage, and kidney damage; and on the treatability of bone/kidney damage. Preferences from 147 patients were elicited with a discrete choice method in an earlier study. Preferences were assumed to be distributed with a multivariate normal distribution. Treatment performances as identified from clinical trials were assumed to be distributed with beta distributions. The probability distributions around preferences and clinical performances were combined with a Monte Carlo simulation method to estimate the joint probability distribution around each treatment’s patient-weighted utility. Results: The three treatments with the highest mean patient-weighted utility were dolutegravir+abacavir/lamivudine (-0.4, 95% CI: -1.3 to 0.5), raltegravir+tenofovir/emtricitabine (-0.5, 95% CI: -1.6 to 0.7) and darunavir/ritonavir+tenofovir/emtricitabine (-0.6, 95% CI: -2.0 to 0.8). There was considerable overlap between the probability distributions of patient-weighed utilities (probability of first rank reversal: 49%; probability of any rank reversal: > 99%). When ignoring uncertainty around patient preferences, the probability of a first rank reversal dropped to 12%, and that of any rank reversal dropped to 88%. Conclusions: A probabilistic multi-criteria methodology was developed that explicitly combines patient preferences and clinical evidence. The individual or joint impact of uncertainty in these on the treatments’ patient-weighted utilities is assessed. Although limited by the small number of attributes, the illustrative case suggests the choice of HAART is highly sensitive to patient preferences. PRM128 MODELLING HEALTH-RELATED QUALITY OF LIFE (HRQOL) LONGITUDINALLY. A BAYESIAN MIXED BETA REGRESSION APPROACH Gheorghe M , Brouwer W , van Baal P Erasmus University, Rotterdam, The Netherlands . . . Objectives: Cross-sectional studies showed that, for modelling health-related quality of life (HRQoL), beta regression is superior in terms of fit and predictive accuracy to other commonly used methods based on normality distribution assumption. Although, longitudinal HRQoL measurements are widely used in clinical trials, not much is known about beta regression suitability in this context. This is mainly due to software unavailability with classical estimation methods. This study proposes to model the longitudinal HRQoL outcome using a mixed beta regression estimated by Bayesian Markov chain Monte Carlo (MCMC) methods implemented in WinBUGS. Compared to the classical approach, not only the Bayesian estimation is considerably easier to implement but has other advantages; for example, the possibility of including informative priors, enabling analysts to incorporate multiple sources of evidence in a single model. Methods: We used a 16-year longitudinal follow-up for modelling the relationship between SF-6D HRQoL and variables age, gender and mortality risk by means of a mixed beta regression. Besides modelling the mean A566 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 parameter, we also modeled the precision parameter using a regression structure. Regression coefficients and predictive accuracy from this model estimated using the Bayesian approach with vague priors were compared to those from a linear mixed effects model estimated classically. Results: Our results indicated that beta distribution fitted the SF-6D outcome better than normal distribution. Furthermore, compared to the linear mixed effects model, the mixed beta regression model was superior in terms of predictive model accuracy. We found that mortality risk had a significant effect on both mean and precision parameters: we observed lower mean HRQoL for higher mortality risk and higher variation of health utilities for higher mortality risk. Conclusions: Mixed beta regression offers a superior approach for modelling the HRQoL outcome longitudinally. Furthermore, such a model can be easily implemented using freely available Bayesian software. PRM129 Effect Of Sample Size And Data Maturity On Parametric Survival Modeling Projections In Advanced Cancer Graham C N , Davis K L , Goyal R K RTI Health Solutions, Research Triangle Park, NC, USA . . . . . . Objectives: Parametric survival modeling (PSM) is often used in cost-effectiveness analyses of oncology treatments to aid in lifetime projections due to right censoring of data. We sought to better understand the effect of sample sizes and data maturity (follow-up time) on PSM projections to aid in the design of clinical trials and the interpretation of cost-effectiveness models. Methods: We modeled overall survival (OS) for advanced colorectal cancer patients treated with first-line chemotherapy and/or a biologic using SEER-Medicare data (2004-2010). Survival was estimated using Kaplan-Meier (KM) and PSM methods. From the full cohort, we randomly drew patients to match typical sample sizes from Phase II and III clinical trials (n= 50, 100, 200, and 400). Additionally, arbitrary data cutoffs were created to proxy clinical trial follow-up times (t= 3, 6, 9, 12, 24, and 36 months). Using PSM methods mean survival from the full cohort was compared with survival from the combinations of sample sizes and follow-up times. Results: Using the KM method, 6% of patients were alive at the end of the follow-up period (6.5 years). Mean OS from the full cohort was estimated to be 21.9 months using the PSM method (best fit Weibull curve). OS estimates for the sample size and follow-up time combinations ranged from 5.9-28.0 months. Minimum and maximum survival projections represented a 73% underestimation and 28% overestimation of survival compared with the full cohort projection, respectively. Projection accuracy was improved when t≥ 6 months and n≥ 200. Conclusions: Both sample size and data maturity have a profound effect on survival projections. Care should be taken when interpreting projections in cost-effectiveness models, especially when sample size is low and follow-up time short. In addition to power calculations, clinical trial design should account for these issues. Additional analyses in other cancer types may provide further guidance for optimum trial design. PRM130 Understanding Real Life Treatment Patterns Among Patients With Hypertension: A Markov Model Letierce A 1, Jouaville S L 1, Boutmy E 2 strategic Data, Boulogne-Billancourt, France, 2Cegedim Strategic Data, BoulogneBillancourt, France . . . . 1Cegedim Objectives: Approximately 65% of patients diagnosed with hypertension are not well controlled and two third of patients need to be treated by two or more drugs for achieving target blood pressure. The aim of this study was to describe antihypertensive treatment patterns in real life settings by using Markov chain model. Methods: Data concerning prescriptions of patients with diagnosed hypertension were obtained from a large primary care survey conducted in 2012 using CSD Longitudinal Patient Database. Patients selected were treated with sartans, prescribed alone or in combination with amlodipine and/or hydrochlorothiazide (HCT), either in free associations or in fixed combinations. A Markov chain model M1 with 4-states (A: single sartan, B: sartan+ amlodipine, C: sartan + HCT, D: sartan + amlodipine + HCT) was proposed to model transitions from one treatment to another over time. A second chain, M2, with 6 states was also studied, in which dual-therapies were divided into two states depending on whether the combination was free or fixed. Age and sex were included as covariates. R packages and MSM DIAGRAM were used. Results: 11,976 patients were selected, 49% were men, aged 69 years on average (SD = 11), suffering from hypertension for 7 years on average, 35% had diabetes, 5% renal failure and 31% a previous cardiovascular event. At the time of selection, the distribution between the states was as follows: A: N = 5187 (43.3%), B: N = 1060 (8.9%), C: N = 5120 (42.8%), D: N = 609 (5%). At the end of one year, > 93% of patients had remained in the same state of M1. The M2 model showed substantial transitions from free to fixed associations. Age and sex did not change the coefficients. Conclusions: The Markov chains used to visually describe the evolution of treatment regimens are useful for analyzing large longitudinal databases of prescriptions. PRM131 The Proportional Odds Model Is More Efficient Than The Multinomial Logistic Model For Network Meta-Analyses Of Ordered Outcomes Bouwmeester W 1, van Beurden-Tan C 1, Bennison C 2, Heeg B 1 1Pharmerit International, Rotterdam, The Netherlands, 2Pharmerit Ltd, York, UK . . . . Objectives: Network meta-analysis (NMA) techniques have been developed to study relative treatment effects for several outcome types (e.g. time-to-event outcomes). No literature exists comparing models of NMA for ordered categorical data, though models are available with different characteristics. This study compared the proportional odds (PO) and multinomial logistic (ML) model for NMA in ordered categorical datasets based on model fit and qualitative characteristics. Methods: To contrast model performance, two extreme datasets were simulated, one which exactly satisfied the PO assumption (POA dataset), and one which did not (nPOA dataset). The models were also tested in a clinical dataset including ordered response categories for four different treatments in psoriasis patients. Both fixed and random effects models were studied. Results: In the POA dataset, the PO fixed effects model had the lowest residual deviance (54.8 versus 58.9 for the ML model) and uncertainty of treatment effects (49% lower standard error (SE)). In the nPOA dataset, the predictions of the PO model were biased, and the ML model had the lowest residual deviance (52.7 versus 271.0 for the PO model). Visual inspection indicated a partial violation of the PO assumption in the psoriasis data. Analyses of the psoriasis data, showed that the PO fixed effects model had the lowest residual deviance (18.1 versus 20.9) and uncertainty (62% lower SE). However, PO model predictions were biased for treatment responses which violated the PO assumption. Conclusions: Statistical selection of NMA models for ordered outcomes should be based on the PO assumption and deviance measures. If data satisfies the PO assumption, the PO model differentiated treatment effects better as a result of lower uncertainty. In terms of flexibility, the PO model can handle data from studies that use different cut-offs for response categories and the ML model can be applied to datasets violating the PO assumption. PRM132 Evaluating The Effect Of Immunotherapy In Advanced Non-SmallCell Lung Cancer Patients Using Two Components Mixture Model Sánchez L 1, Muchene L 2, Luaces P 1, Viada C 1, Rodríguez P C 1, Frias A 1, Shkedy Z 2, Lage A 1 of Molecular Immunology, Havana, Cuba, 2Hasselt University, Diepenbeek, Belgium . . . . . . . . . 1Center Objectives: The aim of the study was to assess the effect of an immunotherapy for the treatment of advanced non–small-cell lung cancer (NSCLC). Methods: Data from a phase III, multicenter, randomized, open-label trial evaluating the efficacy of one EGF-based cancer vaccine (CIMAvaxEGF) as switch maintenance in patients with advanced NSCLC, were used. Survival analysis using Kaplan Meier estimates was performed. Weighted log-rank was conducted to assess the later effect of the immunotherapy. Additionally, a finite mixture model to the primary endpoint (Overall Survival, OS) was fitted. Weibull distribution was assumed for the overall survival and a mixture model consists of one, two or three components was fitted.. All analysis was conducted using the NLMIXED procedure in SAS. The Akaike Information Criterion (AIC) was used for model selection. Results: Intention-to-treat (ITT) analysis showed 1.44 months of OS benefit for vaccinated patients with confirmed delayed-separation phenomena (OS: Vaccine arm, 10.37 months vs. Control arm, 8.93 months; p= 0.043). The mixture model with the best goodness to fit to the data consists of two components (AIC= 3097.7). The two mixture components represent short-term and long-term survival subpopulations. The proportions of the subpopulations are estimated to be equal to 0.89 and 0.11 respectively. The median OS was estimated to be equal to 9.59 and 60.32 for short- and long- term survival populations, respectively. 23.26% and 76.74% from the patients who were classified into long-term survival subpopulation were from the control and vaccinated group, respectively. From them, 7 (70%) patients in the control group and 25 (75%) patients in the treated group were still alive at the end of the study. Conclusions: The results confirm that the vaccination with CIMAvaxEGF prolongs the survival of the advanced NSCLC patients. Mixture models allow assessing the efficacy/effectiveness of vaccines and biological products in the presence of heterogeneous populations. PRM133 Predictive Modelling: Predicting Hospitalisation And Estimating The Cost And Risk To The Third Party Funder Celliers R E Agility Global Health Solutions/University of Pretoria, Centurion, South Africa PREDICTIVE MODELLING: PREDICTING HOSPITALISATION AND ESTIMATING THE COST AND RISK TO THE THIRD PARTY FUNDER R CELLIERS (MsC. (Mathematical Statistics) University of Pretoria) . . Objectives: In the third party funder environment most analyses focus on retrospective claims analyses; the aim of predictive modelling is to estimate future claims or current risk, based on the probability of a hospital event using historical data. Methods: A logistic regression approach is followed where the likelihood of a hospitalisation event is established and mapped to a cost estimate. The modelling process involved establishing a development and validation sample, identifying the predictor variables, building and lastly validating the sample. During the model building process the development and validation population consisted of 149 416 and 47623 beneficiaries respectively; where the data was obtained from a third party funder consisting of 3 years of data. During the building process the dependant variable (Y= Logg (odds)) takes on the value 1 or 0 depending on whether or not a hospital event occurred. To calculate the cost per beneficiary a weighted probability was multiplied by the average cost of a hospital authorisation. Beneficiaries were classified as high risk if log (odds) > = 0.7. Results: The final model is: Log (odds) = -0.11X1+0.00576X2+0.409X3+0.179X4+0.614X5 where X1-X5 denotes the predictor variables. X1 denotes an indicator variable for gender, X2 the predictor variable for age, X3 the HIV indicator variable, X4 the diabetes indicator variable and X5 the chronic indicator variable. The strongest predictors were the chronic indicator variable and age. The validation process resulted in 79% of the beneficiaries being correctly classified and the cost estimation resulted in totals within 3-5% of the actual values. Conclusions: The proposed model predicts hospitalisation efficiently at a beneficiary level and can be implemented to monitor risk and the associated cost (hospital or total) for individuals, employer group or the third party funder. Third party risk management and cost estimation are other applications of the model. PRM134 Joint Modelling Of The Change In Tumor Size And Overall Survival; A Parametric Model Considering Patient Heterogeneity Not Observed At Baseline Majer I M , Heeg B Pharmerit International, Rotterdam, The Netherlands . . . A567 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: In economic evaluations in oncology, survival data is typically extrapolated without taking into account prognostic factors. If individual level trial data are available, patient and disease characteristics observed at baseline are considered. However, survival models typically disregard information that are not known at baseline, e.g. response to treatment, but that may be valuable for the prognosis of patients and hence for decision making. In this study we present a parametric survival model that included response to treatment over time. Methods: Data from 99 patients with late-stage soft tissue sarcoma from a clinical trial was used. Survival information and the percentage change in the sum of the longest diameters of target lesions (i.e. the basis for response evaluation) measured repeatedly during follow-up were utilized. A joint model was estimated linking a random effects sub-model for the change of tumor size with a Weibull sub-model for the survival outcome. The association between change of tumor size over time and overall survival was assessed. Several different functional forms were explored to model the tumor size data and the best fitting model was selected. Results: The median follow-up time in the trial was 1.6 years; 63 patients died. On average, 4.8 measurements on tumor size were available per patient. A flexible cubic B-spline sub-model provided the repeatedly measured tumor size change data the best model fit. The association between tumor growth and overall survival was marginally statistically significant with a P value of less than 0.10. Conclusions: The presented joint model demonstrated that response to treatment over time may be important to consider when building survival models for health economic evaluations in oncology. The model explicitly incorporated the heterogeneity of patients not observed at baseline providing a clinically relevant survival model. Individual survival predictions can be prepared using patient-specific history of tumor growth. RESEARCH ON Methods – Patient-Reported Outcomes Studies PRM135 Assessment Of The Huntington Quality Of Life Instrument (H-QOL-I) Cross-Cultural Validity Clay E 1, Belhadj A 2, Squitieri F 3, Dorey J 4, Auquier P 5, Zielonka D 6, Trigo P 7, Toumi M 8 Ceutical, Les Berges du Lac, Tunisia, 3Neurogenetics and Rare Disease Centre, IRCCS Neuromed, POZZILLI, Italy, 4Creativ-Ceutical USA, Chicago, IL, USA, 5Université de la Méditerranée, Marseille, France, 6Poznan University of Medical Sciences, Poznan, Poland, 7Hospital Ramón y Cajal, Madrid, Spain, 8University of Marseille, Marseille, France . . . . . . . . 1Creativ-Ceutical, Paris, France, 2Creativ Objectives: The Huntington Quality of Life Instrument (H-QoL-I) is the first selfreported specific instrument developed to assess the health-related QoL (HRQoL) of patients with Huntington’s disease (HD). It includes three subscales: motor (4 Likert-type items), psychology (4 Likert-type items) and socializing (3 Likerttype items). The aim of the study was to assess whether patients from different countries respond differently to the H-QoL-I instrument. Methods: Data were from the European study of HD burden (EURO-HDB) survey and included data across 6 countries: France, Germany, Italy, Spain, Poland and the USA. The Differential Item Functioning (DIF) method was adopted to examine whether patients from different countries with the same characteristics had different probability of giving a certain response on H-QoL-I. An item was considered as displaying a DIF if the p-value associated with the two-degree-of-freedom Chi-squared test comparing the two ordinal logistic regressions (with/without country effect and an interaction term between the total rest score and country) was lower than 0.01 and the Zumbo-Thomas effect size was higher than 0.130. Zumbo-Thomas effect size measure and associated p-value were calculated for each item and for all pairs of countries (i.e. 15 combinations). Results: The study included 633 patients (176 French, 124 Italian, 44 German, 60 Polish, 59 Spanish and 170 American). No DIF was detected across all combinations of countries for all items. Sixteen pairwise Zumbo-Thomas effect size measures referring to 7 items were found significant but were lower than 0.130. Conclusions: This study did not detect any variation across the studied countries in the assessment of HRQoL of HD patients using the H-QoL-I instrument. These results support the crosscultural validity of the H-QoL-I. PRM136 Assessment Of The Huntington Clinical Self-Reported Instrument (H-CSRI) Cross-Cultural Validity Dorey J 1, Belhadj A 2, Squitieri F 3, Clay E 4, Auquier P 5, Zielonka D 6, Trigo P 7, Toumi M 8 USA, Chicago, IL, USA, 2Creativ Ceutical, Les Berges du Lac, Tunisia, 3Neurogenetics and Rare Disease Centre, IRCCS Neuromed, POZZILLI, Italy, 4Creativ-Ceutical, 5 Paris, France, Université de la Méditerranée, Marseille, France, 6Poznan University of Medical Sciences, Poznan, Poland, 7Hospital Ramón y Cajal, Madrid, Spain, 8University of Marseille, Marseille, France . . . . . . . . 1Creativ-Ceutical Objectives: The H-CSRI is the first clinimetric self-reported instrument for patients with Huntington’s disease (HD). It includes three subscales: motor (13 Likert-type items in 4 dimensions), functional (7 Yes/No questions) and behavioural (13 Likerttype items in 4 dimensions). The aim of the study was to assess whether patients from different countries respond differently to H-CSRI. Methods: Data were from the European study of HD burden (EURO-HDB) survey and included data across 6 countries: France, Germany, Italy, Spain, Poland and the USA. The Differential Item Functioning (DIF) method was adopted to examine whether patients from different countries with the same characteristics had different probability of giving a certain response on H-CSRI. An item was considered as displaying a DIF if the p-value associated with the two-degree-of-freedom Chi-squared test comparing the two ordinal logistic regressions (with/without country effect and an interaction term between the total rest score and country) was lower than 0.01 and the ZumboThomas effect size was higher than 0.130. Zumbo-Thomas effect size measure and associated p-value were calculated for each item and for all pairs of countries (i.e. 15 combinations). Results: The study included 633 patients (176 French, 124 Italian, 44 German, 60 Polish, 59 Spanish and 170 American). Almost all the items (24 of 26) didn’t show any cross/cultural difference. The two items show- ing DIF were related to the dimension “precise movement” and were detected in the Spain-Italy comparison (Δ R2 [left hand] = 0.2165, Δ R2 [right hand] = 0.1618) and in the Spain-France comparison (Δ R2 [left hand] = 0.1571, Δ R2 [right hand] = 0.1578). Conclusions: Globally, these data support the H-CSRI cross-cultural validity. Further analyses should be conducted to confirm if those particular items need to be revised in the Spanish version. PRM137 Severity And Functional Disability Of Patients With Occupational Contact Dermatitis: Validation Of The German Version Of The Occupational Contact Dermatitis Disease Severity Index (ODDI) Apfelbacher C 1, Popielnicki A 2, Bauer A 3, Diepgen T L 4, Elsner P 5, Dawsey R 6, Mahler V 7, Molin S 8, Schmitt J 9, Weisshaar E 10, Ofenloch R F 10 - University of Regensburg, Regensburg, Germany, 2TransPerfect, Boston, MA, USA, 3University Hospital Carl Gustav Carus, Technical University Dresden, Dresden, Germany, 4University Heidelberg, Heidelberg, Germany, 5University Hospital Jena, Jena, Germany, 6TransPerfect, Atlanta, GA, USA, 7University Hospital of Erlangen, Friedrich-Alexander-University Erlangen-Nuremberg, Erlangen, Germany, 8Ludwig-Maximilians-University Munich, München, Germany, 9Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany, 10University Hospital Heidelberg, Heidelberg, Germany . . . . . . . . . . . . . 1UR Objectives: The Occupational Contact Dermatitis Disease Severity Index (ODDI) was designed in Australia to measure severity and functional disability in patients with occupational contact dermatitis (OCD) of the hands. The psychometric properties of the German version of the ODDI are unclear. Our objective was to investigate the validity and reliability of the German ODDI version. Methods: The ODDI was translated and linguistically validated into German for Germany, following industry standard procedures of concept definition, dual forward translation, back translation and reconciliation, and clinician review. Once the German version was available, data was drawn from the baseline assessment (T0) and first follow-up (T1) of the German chronic hand eczema (CHE) registry (carpe). Spearman correlations of the ODDI with reference measures were computed to assess validity. Cronbach’s alpha was calculated as a measure of internal consistency and the intraclass correlation coefficient (ICC) to assess retest-reliability. Smallest real difference (SRD) and minimal clinical important difference (MCID) were calculated to assess sensitivity to change. Physician Global Assessment (PGA) was used as an anchor for the MCID. Results: 422 patients (54.5% female, mean age: 45.1 years) were included for analysis. Cronbach’s α was found to be 0.73. The ICC was 0.79. Correlations of the ODDI total and the Dermatology Life Quality Index (rho= 0.36) as well as the PGA (rho= 0.48) and patient-assessed disease severity (rho= 0.40) were of moderate strength. The MCID (1.29) was found to be smaller than the SRD (1.87). Conclusions: The German ODDI version is reliable and valid to measure functional impairment and disease severity in patients suffering from OCD. The MCID falls within the range of measurement error and should not be used. PRM138 Mapping Fact-P To EQ-5D In Metastatic Castration-Resistant Prostate Cancer (MCRPC): Performance Of A Previously Developed Algorithm When Applied On A Sample With A Different Disease Stage Ivanescu C 1, Longworth L 2, Skaltsa K 3, Holmstrom S 4 Consulting, Hoofddorp, The Netherlands, 2Brunel University, Uxbridge, UK, 3Quintiles Consulting, Barcelona, Spain, 4HEOR, Astellas Pharma Global Development, Leiden, The Netherlands . . . . 1Quintiles Objectives: To evaluate the predictive performance of a previously published mapping algorithm for converting the prostate cancer specific instrument FACT-P (Functional Assessment of Cancer Therapy–Prostate) to EQ-5D utility values (UK tariff) on a sample with a different disease stage than the one on which the model was generated [Skaltsa et. al. ViH 2014]. Methods: We applied a previously developed algorithm to the data obtained from a randomized, double-blind, placebo-controlled phase 3 trial in asymptomatic/mildly symptomatic chemo-naïve mCRPC patients. The trial collected EQ-5D and FACT-P data at baseline and until treatment discontinuation. The mapping model was developed on mCRPC patients in a post-chemo setting, included the FACT-P subscale scores and baseline variables and used separate algorithms for patients with good and poor health defined as a FACT-P score exceeding or not 76. Model performance was assessed by mean absolute error (MAE) and root mean squared error (RMSE). Results: The testing dataset contained 1,669 patients with baseline and ≥ 1 post-baseline scores. The average baseline EQ-5D utility and FACT-P total score were 0.844 and 119.5 respectively. Percentage of perfect health was 37% across all visits (ceiling effect). The average (across all visits) observed and predicted EQ-5D utility index value was 0.823 and 0.842, respectively. The model yields accurate predictions (MAE= 0.107; RMSE= 0.150) comparable to the ones obtained on the development sample (MAE= 0.117; RMSE= 0.162). The model predicts well for milder health states, but overpredicts for the more severe ones (EQ-5D utility≤ 0.5: MAE= 0.436, RMSE= 0.258; EQ-5D utility > 0.5: MAE= 0.096, RMSE= 0.125). Conclusions: Although external validation is recommended using similar samples, our findings show that the algorithm developed in the post-chemo setting performed well in a pre-chemo setting in mCRPC patients, although overpredicts for severe states. This model seems suitable for predicting utility values for economic evaluation when a preference-based measure is absent in chemo-naïve mCRPC populations. PRM139 How Do Individuals Complete The Choice Tasks In A Discrete Choice Experiment? Veldwijk J 1, Determann D 2, Lambooij M S 1, van Til J A 3, Korfage I J 4, de Bekker-Grob E 5, de Wit G A 1 Institute for Public Health and the Environment, Bilthoven, The Netherlands, 2National institute for public health and the environment, Bilthoven, The Netherlands, 3University of Twente, Enschede, The Netherlands, 4University Medical Center Rotterdam, Rotterdam, The Netherlands, 5Erasmus medical center, Rotterdam, The Netherlands . . . 1National . . . . . . . . A568 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To explore how participants evaluate and complete the choice tasks in Discrete Choice Experiments (DCE), with special attention to the impact of educational level and health literacy. Methods: Two existing DCE questionnaires on rotavirus vaccination and prostate cancer screening served as a case for the current study. In total, 70 participants were sampled based on educational level (35 per case study). During structured interviews, participants completed five choice tasks aloud. Interviewers monitored how participants read the choice tasks, how they interpreted the included risk attributes and what decision strategy they used to make their decision and if the monotonicity and continuity axioms hold. Results: The majority of the participants read all the attributes within each choice task. Nearly all participants chose the scenario with the optimal attribute levels (monotonicity axiom). In accordance with the continuity axiom, most participants mentioned three or more attributes when motivating their decisions. Overall, higher educated and literate participants more often included three or more attributes when motivating their decision and used trading between attributes more often as a decision strategy. Conclusions: The majority of the participants complete a DCE as presumed by its underlying methodology. However, the assumptions did not hold for a subset of lower educated and less literate participants. Based on participants’ age, educational level and health literacy additional measures should be undertaken to enhance participants’ understanding of the attributes, the attribute levels and the choice tasks in a DCE. PRM140 The Measurement Of Utilities In Asthma Patients: A Preliminary Study Yong Y V , Shafie A A Universiti Sains Malaysia, Penang, Malaysia . . . . Objectives: To assess the feasibility of a computer-based Standard Gamble (SG) visual prop whilst measuring utilities of different asthma health states at the same time. Methods: Twenty adult asthma patients literate in either Malay or English language were conveniently sampled from a public hospital in Penang, Malaysia. They were interviewed by two trained interviewers using a bilingual script. Each patient was requested to value the given health states using Visual Analogue Scale (VAS) prior to SG exercise. There were three chronic health states (C1-C3) for 10 years, three temporary states (T1-T3) for 3 months, and two anchor states (healthy and dead). During the SG exercise, the visual prop was fully operated by the interviewers. The probability of being in a worse state was changed in a ‘ping-pong’ fashion until the indifference point was reached. Results: All patients understood the SG exercise and rated SG easier than VAS. Around 85% (n= 17) completed SG within 30 minutes. There was 90% (n=18) who ranked T3 as the worst temporary health state during VAS. Two patients provided logical inconsistency data in SG. The preferences by SG were higher than VAS. Preferences were also higher in temporary states measured by chained SG than other states by conventional SG. The mean utilities for C1= 0.56 (SD 0.38), C2= 0.47 (SD 0.33), C3= 0.53 (SD 0.38), T1= 0.65 (SD 0.31), T2= 0.53 (SD 0.35), and T3= 0.38 (SD 0.38). Conclusions: The SG methods including the props are feasible for utilities measurement in asthma, based on the agreements achieved with other studies on the pattern of utilities measured in this preliminary study. PRM141 DISCRETE-CHOICE EXPERIMENT VERSUS RATING SCALE EXERCISE TO EVALUATE THE RELATIVE IMPORTANCE OF ATTRIBUTES: A STUDY OF THE MAASTRICHT ISPOR STUDENT CHAPTER Wijnen B F M 1, van der Putten I M 1, Groothuis S 1, de Kinderen R J A 1, Noben C Y G 1, Paulus A T G 1, Ramaekers B L T 2, Vogel G C W M 2, Hiligsmann M 1 1Maastricht University, Maastricht, The Netherlands, 2Maastricht University Medical Center, Maastricht, The Netherlands . . . . . . . . . . . . . . . . . . . . . . . Objectives: Eliciting preferences has become increasingly important in health care. Several methods are available to evaluate the relative importance of different aspects of health and health care. In this study, we aim to examine the difference between a discrete-choice experiment (DCE) and a rating scale exercise (RSE) to determine the most important attributes of undergraduate students when selecting a study specialization. Methods: First-year health sciences students were asked to complete a questionnaire that included a DCE and a RSE. Six attributes were identified in focus groups: “possible acquainted masters”, “job opportunity”, “scope of specialization”, “quality of education”, “hours self-study” and “personal interest”. Fourteen unlabeled choice tasks were constructed using a statistically efficient design and a mixed multinomial logistic regression analysis was used for data analysis. In the RSE, attributes were rated on perceived importance using a 7-point Likert scale. Two versions of the questionnaire were distributed in which the RSE was put before and after the DCE. Results: A total of 254 students filled out the questionnaire. In the DCE, three attributes were statistically significant of which “personal interest” was the most important attribute followed by “job opportunity” and “quality of education”. In the RSE, all attributes except “hours of self-study” were rated 4 or higher. The RSE scores of the attributes with a relatively low importance in the DCE were significantly lower in the questionnaire version that started with DCE than in the other version. Results of the DCE did not significantly differ between the two questionnaire versions. Conclusions: The DCE had a differentiating effect on the relative importance of attributes whereas in the RSE attributes were rated more equally and were, except for one, all considered important. Forcing respondents to make first trade-offs between attributes (using a DCE) leads to lower RSE scores for less important attributes afterwards. PRM142 THE DEVELOPMENT AND PRELIMINARY VALIDATION OF THE MANCHESTER SLEEP SYMPTOMS INDEX (MSSI) FOR PEOPLE WITH CHRONIC OBSTRUCTIVE PULMONARY DISEASE (COPD) Garrow A P 1, Yorke J 1, Khan N 2, Tyson S 1, Singh D 3, Vestbo J 4 1University of Manchester, Manchester, UK, 2The Medicines Evaluation Unit, Manchester, UK, 3University of Manchester, Medicines Evaluation Unit, Manchester, UK, 4The University of Manchester, Manchester, UK . . . . . . . Objectives: In COPD, disturbed sleep is related to exacerbation frequency, poor quality of life and early mortality. We developed the Manchester Sleep Symptoms Index (MSSI) to assess night-time symptoms and disturbed sleep in COPD. Methods: Identification of potential items was guided by interviews and focus groups involving COPD patients and age-matched controls. Hierarchical methods and Rasch analysis informed item deletion and development of a unidimensional scale. Internal consistency and test-retest reliability were assessed. Concurrent validity was examined using Pearson’s correlation with the St George’s Respiratory Questionnaire (SGRQ), modified MRC Dyspnoea Scale, FACIT fatigue scale, MOS Sleep Scale, HADS Anxiety and Depression scores. Results: Qualitative data from 36 patients with COPD and 10 age-matched controls informed an initial list of 22 items. The cross-sectional study included 203 COPD patients (GOLD: I: 14% II: 41% III: 25% IV: 7%; male: 63%, mean age: 64.7; SD: 7.5 years) and 50 non-COPD controls. 12 items were removed during hierarchical methods and a further two following Rasch analysis. The final MSSI contains 8 sleep-related items that are specific to COPD patients: breathlessness, chest tightness, cough and sputum production. The index has good internal consistency (Cronbach’s alpha 0.87), test-retest repeatability (intra-class coefficient 0.77) and validity. Total MSSI scores significantly correlated with the SGRQ (r= 0.64); MRC Dyspnoea scale (r=0.46), FACIT-F (r=-0.61); MOS problems index 2: (r=0.62); MOS Sleep adequacy (r=0.40); MOS Sleep disturbance (r=0.53) HADS anxiety (r=0.54) and depression (0.48). There was good overall fit to the Rasch model (Chi-squared: 29.2 df: 16 p=003.) and distribution of item scores. Conclusions: The MSSI is a reliable, valid, uni-dimensional self-reported outcome measure of sleep and night-time symptoms for people with COPD. It is simple and quick to use making it suitable for research and practice. Further work is needed to determine the minimal clinical important difference and cross cultural validity. PRM143 Psychometric Evaluation Of The Patient’s Knee Implant Performance (PKIP) Questionnaire For The Assessment Of Primary Total Knee Arthroplasty Coles T M 1, Dwyer K A 2, Mordin M 3, Williams V 1, Clatworthy M 4, Yates P 5, Hamilton W 6 1RTI Health Solutions, Research Triangle Park, NC, USA, 2DePuy Orthopaedics, Inc, Warsaw, IN, USA, 3RTI Health Solutions, Ann Arbor, MI, USA, 4Ascot Hospital, Auckland, New Zealand, 5Murdoch Orthopaedic Clinic, Murdoch, New Zealand, 6Anderson Orthopaedic Research Institute, Alexandria, VA, USA . . . . . . . . . Objectives: The objective of this study was to evaluate the psychometric properties of a new patient-reported measure of knee implant functional performance associated with physical activities prior to and following primary total knee arthroplasty (TKA). The Patient’s Knee Implant Performance questionnaire (PKIP) was developed to assess factors that lead to patient dissatisfaction and describe unmet needs in knee functional performance. Methods: The psychometric analysis sample (n=764) was based on a multicenter, prospective, noncomparative longitudinal study of patients with osteoarthritis undergoing TKA at 22 international sites. The PKIP and additional patient-reported outcomes and clinical measures were collected preoperatively, postoperatively at less than 1 year, at a minimum of 1 year, and at 2 years. The PKIP structure and its reliability, construct validity, discriminating ability, and responsiveness were assessed. Results: Based on inter-item correlations, factor analyses, and results of previous qualitative research, the PKIP was scored as four subscales (Stability, Confidence, Satisfaction, and Activity Modification) and an Overall PKIP score. The Overall PKIP score met reliability standards (internal consistency: alpha = 0.78 at minimum 1 year; test-retest: intraclass correlation coefficient = 0.77). Correlations between the PKIP and other available measures provided evidence of construct validity. For example, the PKIP correlated 0.19 and 0.50 with the American Knee Society Score preoperatively and at less than 1 year, respectively, and correlated 0.69 and 0.77, with the Knee Injury and Osteoarthritis Outcome Score Quality of Life subscale. The PKIP was capable of discriminating between groups of patients with better or worse knee functioning as defined by clinician-rated measures; hypothesis tests were in the predicted direction and mostly statistically significant. The effect size for the Overall PKIP score was 2.38, indicating that the PKIP was highly responsive. Conclusions: The reliability, validity, and responsiveness of the PKIP support its use among patients undergoing primary TKA. PRM144 Validity And Responsiveness Of The Bristol Rheumatoid Arthritis Fatigue Multidimensional Questionnaire (BRAF-MDQ) In A Randomized Controlled Clinical Trial Kirwan J 1, Coteur G 2, Dures E 1, Nicklin J 1, Bryson J 3, Hewlett S 1 Royal Infirmary, Bristol, UK, 2UCB Pharma, Brussels, Belgium, 3UCB Pharma, Raleigh, NC, USA . . . . . . 1Bristol Objectives: To evaluate the validity of the BRAF-MDQ in a new group of patients in a clinical trial setting, to confirm its internal factor (domain) structure and to document its sensitivity to change. Methods: Pooled data from a randomized controlled trial (NCT01242488) in patients with moderate to severe RA were collected at baseline (BL), Wk10 and Wk12. Spearman’s correlation coefficients, Bland-Altman plots and confirmatory factor analysis tested construct validity, reproducibility and internal factor structure of the BRAF-MDQ. Responsiveness was assessed amongst clinical responders at Wk12 by effect sizes of changes from BL in BRAF-MDQ. Results: There were 219 patients (mean age: 55.5 years; disease duration: 12.6 years; BL DAS28: 5.77). The proportion of missing item answers was very low (0–3%). BRAF-MDQ scores correlated with patient global, pain and HAQ at BL (r= 0.49, 0.46 and 0.58) and at Wk12 (r= 0.63, 0.65 and 0.64). Changes in the Physical and Living domains were more closely related to changes in patient and physician global scores and DAS scores than did changes in Cognition and Emotional domains. Reproducibility was high (r> 0.87 for all total and domain scores; narrow BlandAltman limits of agreement), as was internal consistency (Cronbach’sa: 0.97 for total scores; > 0.82 for each domain at BL). The Butler comparative fit index (CFI; 0.92) indicated that the established structure within the BRAF-MDQ accounts well for data variation. Effect sizes for BRAF-MDQ in clinical responders at Wk12 were very high A569 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 (> 0.7) for the Physical domain, high (> 0.5) for the Total score and the Living domain and moderate (>0.3) for the Cognition and Emotional domains. Conclusions: The BRAF-MDQ was completed well by participants, related to appropriate measures of disease severity, retained its factor structure, gave reproducible results and was responsive to clinical change, confirming its validity as a measure of RA fatigue. PRM145 Cdad-Daysyms™: A New Patient-Reported Outcome Tool For Clostridium Difficile-Associated Diarrhoea Kleinman L 1, Talbot G H 2, Schüler R 3, Broderick K 4, Revicki D 5, Nord C E 6 LLC, Anna Maria, FL, USA, 3Actelion Pharmaceuticals Ltd, Allschwil, Switzerland, 4Cubist Pharmaceuticals, Lexington, MA, USA, 5Evidera, Bethesda, MD, USA, 6Karolinska Institutet, Karolinska University Hospital, Stockholm, Sweden . . . . . . . . 1Evidera, Seattle, WA, USA, 2Talbot Advisors Objectives: Patient-reported outcome (PRO) measures provide relevant information on how patients function and feel about their health. Such data are useful in clinical practice and registrational trials. Despite the importance of symptom assessment in CDAD, there is no validated PRO for Clostridium difficile-associated diarrhoea (CDAD). A qualitative research study was conducted to develop a CDAD PRO according to US FDA PRO guidelines. Methods: Content development comprised 2 study phases, with input from an advisory group of 6 CDAD experts in Europe and North America. Phase I elicited patients’ experiences of CDAD symptoms in open-ended discussions during telephone interviews. Supplementary interviews obtained nurses’ observations. A draft PRO was developed following demonstration of concept saturation. Readability and translatability were assessed. Phase II involved 2 rounds of patient interviews, with revision of the draft PRO after each round. All patients were ≥ 18 years old, with confirmed CDAD. IRB approval and participant informed consent were obtained. Results: Phase I interviews included 18 patients and 6 nurses in the United States; 16 additional patients were interviewed in Phase II. Patients were representative of the general CDAD population, and diverse in age, gender, and disease severity. Concept saturation was reached in Phase I for spontaneously reported CDAD symptoms. Items were organised in a draft conceptual framework with 5 hypothesised domains: diarrhoea, abdominal discomfort, tiredness, lightheadedness, and other symptoms. Phase II demonstrated initial content validity of the 13-item draft daily diary (CDAD-DaySyms™). Participants reported the questions were clear, relevant, and comprehensive; were able to use the instructions to complete the diary correctly; and considered the 24-hour recall period appropriate. Conclusions: The CDAD-DaySyms™ captures symptoms relevant to CDAD patients, demonstrating initial content validity. To allow its use in clinical practice and CDAD clinical studies, final content and psychometric validity are being evaluated in 2 ongoing international clinical trials. PRM146 Development And Content Validity Testing Of A Treatment Acceptance Measure For Use In Hypercholesterolemia Patients Receiving Treatment Via Subcutaneous Injection Tatlock S 1, Grant L 1, Arbuckle R 1, Khan I 2, Manvelian G 3, Sanchez R 3 1Adelphi Values Ltd, Bollington, UK, 2Sanofi, Bridgewater, NJ, USA, 3Regeneron Pharmaceuticals, Tarrytown, NY, USA . . . . . . Objectives: In phase II studies, alirocumab, a PCSK9 inhibitor administered via subcutaneous injection, demonstrated significant reduction in LDL-C levels. The objective of this study was to conduct qualitative research to explore the concept of treatment acceptance in patients receiving a subcutaneous injection to support the development and testing of a patient-reported outcome measure, the Injection-Treatment Acceptance Questionnaire (I-TAQ). Methods: A literature review led to the targeting of ‘treatment acceptance’ as the measurement concept in preference to treatment satisfaction. Concepts generated from the literature and instrument review informed the drafting of 17 items in the I-TAQ; item wording was adapted from three validated instruments. Qualitative interviews were conducted among 29 US-English speaking patients participating in alirocumab’s phase III program who self-administered the treatment via autoinjector (n=19) or pre-filled syringe (n=10). First, concept elicitation (CE) questioning was used to elicit concepts relating to patients’ treatment experiences and acceptance. The I-TAQ was then cognitively debriefed using “think-aloud” methods. Verbatim transcripts were analyzed using thematic analysis and Atlas. ti. Revisions were considered after each round of interviews. Results: Qualitative analysis of CE data indicated treatment acceptance to be high, with the following concepts identified as relevant: perceived efficacy, side effects, self-efficacy, convenience and overall acceptance. Ten (34%) patients reported an initial fear of needles, which subsided with no impact on discontinuation. Pain was not considered relevant by patients, suggesting no pain associated with the injection. Five items were added following round 1 interviews, three were retained after round 2 testing and two were added at finalization, forming the conceptually comprehensive 22-item I-TAQ. Patients demonstrated good understanding of item wording, instructions, response scales and recall period. Conclusions: Successive rounds of interviews resulted in a treatment acceptance measure with strong content validity. Next steps are to psychometrically validate the I-TAQ in a population with experience of taking alirocumab. PRM147 The Influence Of Gene Expression Profiling (GEP) On Decisional Conflict In Chemotherapy Treatment Decision-Making For EarlyStage Breast Cancer (BRCA) Marshall D 1, MacDonald K 1, Deal K 2, Trudeau M 3, Leighl N 4, Bombard Y 5 of Calgary, Calgary, AB, Canada, 2McMaster University, Hamilton, ON, Canada, 3Sunnybrook Health Sciences Centre, Toronto, ON, Canada, 4Ontario Cancer Institute, Toronto, ON, 5 Canada, Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Toronto, ON, Canada . . . . . chemotherapy treatment decision-making. Methods: We embedded the validated Decisional Conflict Scale (DCS) into our discrete choice experiment survey examining preferences for chemotherapy treatment in early BrCa. Of the 1004 general population participants, 200 completed the DCS before (DCS-1; no GEP test score in scenario) and after (DCS-2; GEP test score added to scenario) the discrete choice experiment. The 16-item DCS was scored from 0-100 with five subscores. Mean total and subscores, standard deviations and change in scores were calculated, with significance based on matched pairs t-tests (p< 0.05). We anticipated GEP would decrease decisional conflict in individuals unsure of their chemotherapy treatment decision. Results: As anticipated, total score and all subscores (uncertainty, informed, values clarity, support, and effective decision) decreased significantly (all p< 0.05) in the group of respondents (n= 33) who indicated uncertainty about taking chemotherapy in DCS-1 but changed to no chemotherapy after receiving a GEP test score in DCS-2. In the group of respondents (n= 25) who indicated they would undergo chemotherapy in DCS-1 but changed to unsure in DCS-2, their effective decision subscore increase significantly (24.5 to 34.5, p< 0.05). In the overall sample (n= 200), total decisional conflict decreased from DCS-1 to DCS-2 by 0.5 (p= 0.3) and all subscores had non-significant decreases with the exception of effective decision, which had a non-significant increase. Conclusions: GEP influences chemotherapy treatment decisional conflict in individuals who are initially unsure in their treatment decision-making. However, we do not observe this effect in individuals who do not change their chemotherapy treatment decisions. PRM148 Patient Reported Utilities In First-Line Advanced Or Metastatic Melanoma: Analysis Of Trial CA184-024 Porter J 1, Lee D 1, Hertel N 2, Hatswell A J 1 1BresMed, Sheffield, UK, 2Bristol Myers Squibb, Uxbridge, UK . . . . . Objectives: In oncology, the impact of interventions on health-related quality of life (HRQL) is traditionally modelled based on disease progression status. The aim of this analysis was to assess if more meaningful patterns exist in HRQL data, based on other clinically important events that should be considered in modelling utility. Methods: HRQL data from the CA184-024 trial of ipilimumab plus dacarbazine in previously untreated patients with unresectable malignant melanoma were analysed. European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core-30 (EORTC QLQ-C30) responses were mapped to a generic, preference-based measure (EORTC-8D) by means of a published and validated mapping algorithm. The utility observations available for each patient were used to examine the relationships between HRQL and a number of disease- and time-based variables, including treatment effect, progression status and time to death, via a mixed effects regression model. Results: Progression status was found not to be significantly predictive of utility (p=0.29). Of the variables considered, the strongest relationship was with time to death, the mixed effects model for which was significantly predictive of utility (p≤0.001). HRQL dropped as patients approached death; patients treated with ipilimumab had a utility of 0.86 if time to death was more than 1 year, which reduced to 0.61 during the final month of life. The ipilimumab treatment variable was associated with a small negative coefficient (-0.02), accounting for the adverse event profile of the drug when added to dacarbazine (p=0.06). Conclusions: Analysis of the CA184-024 HRQL data showed that time to death rather than progression status was significantly predictive of utility. Hence, modellers should carefully examine primary data to determine if a time to event approach or a progression based approach is appropriate to model utility best reflecting the pathology of the disease. PRM149 RASCH FIRST? FACTOR FIRST? Chen W H , McLeod L D , Coles T M RTI Health Solutions, Research Triangle Park, NC, USA . . . . . . Objectives: Rasch modelling theory and its extensions have become popular tools in assessing psychometric properties of patient-reportedoutcome (PRO) instruments. Since the Rasch model assumes a unidimensional structure, it is important to assert that this assumption has not been violated. However, there has been much debate about using factor analysis as the first step to assess dimensionality or using the Rasch model directly to identify items not fitting the unidimensional model. This study uses simulated data to compare the two techniques to examine the unidimensional assumptions. Methods: Simulated data that represent a typical PRO instrument are generated based on the following variables: sample size (200 or 400), number of factors (1,2, or 3), and correlation among factors (0.4 or 0.7). Each simulated PRO assessment contains 15 items with 5 response categories. Exploratory factor analysis is conducted, and the number of factors proposed by the results are noted. A Rasch model theory analysis is also conducted, and the number of mis-fit items is noted. The unidimensional test associated with Rasch model is also conducted. Results: Preliminary results suggest that when there is one dominant factor, the Rasch results identify the simulated factor and it becomes the underlying trait. Items not belonging to the dominant factor are flagged as mis-fit items or as forming secondary factors based on the unidimensional test. In situations where there is no dominant factor, factor analysis is able to identify separate factors, whereas the results are inconsistent using Rasch model. Conclusions: Rasch first or factor first? The preliminary findings suggest that, when there is prior knowledge of a dominant factor, Rasch modelling can be conducted first and will result in a unidimensional measure. When there is no prior knowledge of a dominant factor, then factor analysis should be conducted first to examine the dimensionality. . 1University Objectives: Individuals with BrCa have high decisional conflict with respect to treatment decisions. GEP of tumours informs risk prediction, potentially affecting decisions about adjuvant chemotherapy in early BrCa, where only 15% will experience recurrence. We aimed to examine whether GEP reduces decisional conflict in PRM150 Predicting Suicidal Behavior In Veterans And Active Military Personnel: Possibilities For Electronic Deployment To Discover A Predictive Assessment Sage C C 1, Platko J V 1, Nokela M 2 Market Access Services, San Diego, CA, USA, 2Covance Market Access Services, Solna, Sweden . . 1Covance . . . A570 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Currently no measure can identify, with a high degree of positive predictive power, suicidal behavior. Because suicide occurs at a low base-rate, studies of instruments designed to predict this outcome often lack an adequate sample size to prove the tool’s predictive ability. Our aim is to identify an assessment with the most promise of predicting suicidal behavior in veteran or military patients and present options to overcome previous research hurdles. Methods: Two systematic reviews, one performed for the US Department of Veterans [1], and the other as part of NICE guidance development [2] provided the background for our analysis. These summaries were reviewed to identify the most predictive assessments yet developed. Sensitivity, specificity, positive and negative predictive power were tabulated for all reported instruments. Study limitations were recorded along with these results. Results: One instrument showed promise of meeting the objective. The Affective States Questionnaire (ASQ) was able to predict suicide within 3 months with a sensitivity of 60%, specificity of 74% and positive predictive power of 32% [3]. The study that produced those results was performed on an inpatient and outpatient veteran population. However, the sample was small with only 283 patients and the risk of bias was unclear. A larger study may provide the needed evidence to make the Affective States Questionnaire a useful screening tool. We propose the Affective States Questionnaire be transferred to electronic administration and provided as part of routine admissions at VA facilities. Deploying the tool electronically could provide the large sample sizes required to detect effects on this low base-rate outcome. Conclusions: The Affective States Questionnaire shows promise of becoming an appropriate screening tool for suicide in a military population. Electronic capture may allow for large scale deployment, therefore gaining sufficient sample to determine applicability as a screening tool. PRM151 Re-Validation of the Self-Injection Assessment Questionnaire© (Siaqv2.0©) in Rheumatoid Arthritis Patients on Certolizumab Pegol Treatment Coteur G UCB Pharma, Brussels, Belgium . Objectives: To evaluate psychometric properties of the revised Self-Injection Assessment Questionnaire in rheumatoid arthritis patients (pts) receiving certolizumab pegol. Methods: In the study (NCT00674362), pts with low to moderate rheumatoid arthritis (RA) received certolizumab pegol (CZP; 400mg at Weeks [Wks] 0, 2, 4, then 200mg every other wk). In the open-label extension (OLE; NCT00843778), pts could self-administer CZP using a pre-filled syringe. Pts in OLE completed the revised Self-Injection Assessment Questionnaire (SIAQv2.0©) at Wks 0, 2, 4, 6, 8, 10, 12. Domain scores were calculated per authors’ recommendations, and internal consistency was assessed using the Cronbach’s alpha statistics. Floor and ceiling effects were reported as % pts with the worst/best domain score. Construct validity was assessed by confirmatory factor analysis fitting the current conceptual framework of the questionnaire and by calculating the Bentler’s Comparative Fit Index (CFI) and the root mean square error of approximation (RMSEA). Results: 86 pts (mean age: 50.8 years; disease duration: 4.6 years) entered the OLE and completed the SIAQ at least once. At first self-injection visit, DAS28 (ESR) was 4.0 and HAQ-DI 0.9. The internal consistency of all domains was >0.8 at any visit. Floor effect was <5% at any visit; ceiling effect was ≤ 13% for Self-Confidence, Ease of Use (EU) and Satisfaction domains, but reached 40% for Feeling and Injection-Site Reactions domains. The ceiling effect of the EU domain was lower (10%) than the original validation (22%). The conceptual framework structure was supported by the confirmatory factor analysis with CFI values of 0.75–0.86 and RMSEA values of 0.10–0.13, which, given the limited sample size, would indicate reasonable goodness of fit. Conclusions: Modifications brought to the SIAQv2.0© appeared to remediate the acquiescence bias issue that was noted during the validation of SIAQv1. The appropriateness of the internal consistency reliability and construct validity of the SIAQ v2.0© were confirmed. PRM152 Physicians’ Preferences for Bone Metastases Treatments in Turkey González J M 1, Gatta F 2, Arellano J 3, Qian Y 3, Ertugrul G 4, Hauber A B 1, Posner J 1, Oksuzoglu B 5 1RTI Health Solutions, Research Triangle Park, NC, USA, 2Amgen (Europe) GmbH, Zug, Switzerland, 3Amgen Inc., Thousand Oaks, CA, USA, 4Amgen Turkey, Istanbul, Turkey, 5Ankara Oncology Training and Research Hospital, Ankara, Turkey . . . . . . . . . . Objectives: To evaluate Turkish physicians’ preferences when selecting between the different bone-targeted agents (BTAs) available for preventing skeletal-related events (SREs) in patients with bone metastases from advanced solid tumors. Methods: Physicians from several centres, currently treating patients with bone metastases from solid tumours were recruited by phone or personal invitation and then engaged in a face-to-face interview where they completed a web-enabled discrete-choice experiment survey. Each survey included 10 choices between pairs of hypothetical treatment profiles for the two putative patient profiles. The hypothetical treatment profiles included five attributes within a pre-defined range (based on prescribing information): time until first SRE (10, 18 and 28 months); time until worsening of pain (3, 6 and 10 months); annual risk of osteonecrosis of the jaw (ONJ; 0, 1 and 5%); annual risk of renal impairment (0,4 and 10%); and mode of administration (oral tablet, subcutaneous injection, 15-minute or 120-minute intravenous infusion). Choice questions were based on an experimental design with known statistical properties. A main-effects random parameters logit model was estimated. Results: A total of 105 physicians agreed to participate in the face-to-face interview and accessed the online survey. Of these, 104 physicians were eligible and consented to participate and 99 were included in the analysis. Estimated preference weights for all applicable attributes were consistent with the natural ordering of the categories. Risk of renal impairment and months until first SRE were the most important attributes influencing physicians’ decisions, with better clinical outcomes preferred to worse outcomes. Preventing pain progression was the third most important attribute followed by mode of administration. Annual risk of ONJ was the least important attribute. Conclusions: When making treatment decisions regarding choice of BTA for patients with bone metastases, the main treatment goals for Turkish physicians are reducing risk of renal impairment and delaying first SRE. PRM153 Validity of the Eq-5d-5l in Stroke Patients Golicki D 1, Niewada M 1, Buczek J 2, Karlinska A 2, Kobayashi A 2, Janssen M F 3, Pickard A S 4 1Department of Experimental and Clinical Pharmacology, Medical University of Warsaw, Warsaw, Poland, 22nd Department of Neurology, Institute of Psychiatry and Neurology, Warsaw, Poland, 3Department of Medical Psychology and Psychotherapy, Erasmus MC, Erasmus University, Rotterdam, The Netherlands, 4Department of Pharmacy Systems, Outcomes, and Policy, College of Pharmacy, University of Illinois at Chicago, Chicago, IL, USA . . . . . . . . . Objectives: To assess EQ-5D-5L validity in patients with acute stroke, in comparison to EQ-5D-3L, EQ VAS, modified Rankin Scale (mRS) and Barthel Index (BI). Methods: Cross-sectional study of 408 patients (51.5% males; mean age 69 years), after median 8 days from stroke onset. We assessed: construct validity in terms of known-groups validity, convergent validity of EQ-5D-5L dimensions with other stroke outcome measures, and criterion-related validity in terms of concurrent validity, with mRS as a gold standard. Results: A total of 2.9% EQ-5D-5L and 3.7% EQ-5D-3L questionnaires had at least one missing answer, indicating good feasibility of both instruments in patients with stroke. The proportion of patients reporting ‘no problems’ was 38.2% for BI, 6.1% for EQ-5D-3L, 5.6% for EQ-5D-5L, 5.0% for mRS and 2.5% for EQ VAS. Results of the known-groups validity tests confirmed prior hypotheses: health state utilities were lower in females, patients with high mRS score, low BI or VAS score, patients with subarachnoid hemorrhage or intracerebral hemorrhage, and when proxy respondent was used. Convergence of EQ-5D-5L dimensions with mRS, BI and EQ VAS was improved or at least the same as EQ-5D-3L dimensions. For predicting outcome in patients with stroke, the sum of mRS related EQ-5D-5L dimensions (Mobility, Self-care, Usual activities), gave 1% of false positive and 0% of false negative results. Conclusions: Results support the validity of the EQ-5D-5L descriptive system as a generic measure assessed by self-report and proxy in patients with acute stroke, demonstrating some psychometric advantages in comparison to EQ-5D-3L and substantially lower ceiling effect in comparison to Barthel Index. PRM154 An Evaluation of the Performance of Eq-5d: A Review of Reviews of Psychometric Properties Longworth L 1, Singh J 1, Brazier J 2 1Brunel University, Uxbridge, UK, 2University of Sheffield, Sheffield, UK . . . Objectives: EQ-5D has been widely used to measure health status in a variety of conditions and the amount of evidence of its performance has increased over recent years. The aim of this study was to consolidate this evidence by reviewing papers reporting systematic reviews of the psychometric properties (validity and reliability) and/or responsiveness of EQ-5D. Methods: Medline and Embase were searched for systematic reviews of the performance of EQ-5D. Supplementary searches were carried out in Cochrane Library, Web of Science, reference lists of included studies, the EuroQol database and hand searching of EuroQol Scientific Plenary Proceedings. In addition the website of the Oxford Patient Reported Outcome Measures (PROMs) Group was searched for reports. Data were extracted using a template designed specifically for the study. Results: 25 reviews were identified in this study and a further 18 were identified from the Oxford PROMs group website. The majority of studies focussed on adults. Overall there was evidence of good to fair performance of EQ-5D in depression, diabetes (type 2), rheumatoid arthritis, skin conditions, cancer, cardiovascular disease, asthma, personality disorder and urinary incontinence. Evidence was mixed in COPD, dementia, schizophrenia and vision disorders, and poor for hearing disorders. The was little evidence for liver transplantation, venous leg ulcers, haemophilia, bipolar disorder and low back pain; although limited, the evidence showed positive results for liver transplantation, haemophilia and leg ulcers. No evidence was identified for, among others, skin cancer and systemic lupus erythematous. Conclusions: This study has provided a comprehensive overview of the evidence of the performance of EQ-5D. Most evidence suggests good psychometric properties of EQ-5D; however there are particular concerns about its ability to capture the impact of dementia, schizophrenia, visual impairment and hearing disorders. Further research is encouraged in conditions where data or reviews of psychometric properties of EQ-5D are lacking. PRM155 Health-Related Quality of Life in Italian Patients With Moderate and Severe Crohn’s Disease: Interim Results from the Sole Study Lazzaro C 1, Cappello M 2, Cortelezzi C 3, Costantino G 4, Fiorino G 5, Mastronardi M 6, Giannotta M 7, Galletti B 8, Cicala M 9, Vadalà di Prampero S 10, Gualberti G 11, Caprioli F 12, Gasbarrini A 13, Meregaglia M 14 1Studio di Economia Sanitaria, Milan, Italy, 2Università di Palermo, Palermo, Italy, 3AOU di Circolo - Fondazione Macchi, Varese, Italy, 4Università di Messina, Messina, Italy, 5Istituto Clinico Humanitas, Rozzano (MI), Italy, 6IRCCS S. De Bellis, Castellana Grotte, Italy, 7Azienda Ospedaliero Universitaria di Careggi, Firenze, Italy, 8Ospedale S. Salvatore, L’Aquila, Italy, 9Università Campus Bio Medico, Roma, Italy, 10Azienda Ospedaliero-Universitaria S. Maria della Misericordia di Udine, Udine, Italy, 11AbbVie, Campoverde di Aprilia (LT), Italy, 12Università degli Studi di Milano, Milano, Italy, 13Università Cattolica del Sacro Cuore, Rome, Italy, 14Bocconi University, Milan, Italy . . . . . . . . . . . . . . Objectives: to investigate health-related quality of life (HRQoL) in Italian patients with moderate and severe Crohn’s disease (CD) (Harvey Bradshaw Index> 8). Methods: EuroQoL 5-dimension 3-level (EQ-5D-3L) questionnaire and visual analogue scale (VAS) were administered to 540 consecutive patients with moderate and severe CD who referred to a convenience sample of 38 Italian inflammatory bowel disease centres (21 teaching-hospitals; 4 research hospitals; 8 self-governing hospitals; 3 Local Health Authority hospitals; 2 private hospitals) participating in the ongoing Survey on Quality Of Life in Crohn’s Patients (SOLE) A571 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 during 2012. As Italian EQ-5D-3L tariffs were not available at the time of the research, UK ones were used according to other comparable studies. Results concerning the first out of the 4 visits planned in SOLE study were reported as mean, standard deviation (SD), median and range. Correlation between EQ-5D-3L questionnaire and VAS scores was investigated via Kendall’s tau-b (Ktau-b). Results: 531 patients (mean age: 41.3; SD: 13.8; median: 41; range: 18—84) responded to EQ-5D-3L questionnaire (98.3%) and 536 to VAS (99.6%). The most frequently marked levels for EQ-5D-3L questionnaire were: 1 for mobility (66.2%) and self-care (85.2%); 2 for usual activities (51.8%), pain/discomfort (70.7%), and anxiety/depression (53.4%). Overall scores for EQ-5D-3L questionnaire and VAS were 0.7 (SD: 0.3; median: 0.8; range: -0.2—1) and 54.3 (SD: 20.8; median: 55; range: 0—100), respectively. Six out of 531 EQ-5D-3L questionnaire responders (1.1%) valued their health state potentially worse than death (utility< 0). VAS scores of 10 or multiple of 10 were marked by 298 out of 536 patients (55.6%). Ktau-b between EQ-5D-3L questionnaire and VAS scores reached 0.45 (p< 0.001). No significant differences were detected when EQ-5D-3L questionnaire and VAS scores were stratified according to hospital institutional features. Conclusions: Italian patients with moderate and severe CD report a remarkable reduction in HRQoL. PRM156 Current Sample Size Practices in the Psychometric Evaluation of Patient-Reported Outcomes for Use in Clinical Trials Coles T M , Chen W H , Nelson L M , Williams V , McLeod L D RTI Health Solutions, Research Triangle Park, NC, USA . . . . . . . . . Objectives: Sample size (N) affects the robustness of psychometric results, but for evaluations of patient-reported outcome (PRO) measures, N is often a compromise between timelines and resources. Currently, there are no psychometric N guidelines or requirements for the development of PROs for use in clinical trials. The objective of this study is to review current N practices by conducting a systematic literature review of the psychometric methods and N choices made in the evaluation of PRO measures (for use in clinical trials) over the past 10 years. Methods: This systematic literature review included abstracts that described the psychometric evaluation of PRO measures that were likely developed for use in clinical trials. The review included English-language journal abstracts published in the past 10 years and identified in PubMed. Characteristics of each study were tabulated including the number of items and dimensions in each PRO of interest, the psychometric methods employed (e.g., internal consistency, test-retest reliability, factor analysis, responsiveness), and N. Results: The literature search yielded 252 abstracts describing studies conducted mostly in Europe, Canada, and the United States. Preliminary results indicate that Ns ranged from approximately 40 to 4,000. The most frequently reported psychometric method was Cronbach’s alpha to quantify internal consistency. Approximately thirty percent of studies employed methods that demand the largest Ns such as item response theory (IRT) and factor analysis (FA). Ns for studies using IRT or FA analysis ranged from approximately 100 to 4,000 participants. Conclusions: A wide range of Ns were employed for the psychometric evaluation of PROs developed for use in clinical trials. Ideally, researchers should consider the complexity of the PRO measure, its intended use, and the purpose of the evaluation when deciding on a study N. Additional studies should work toward developing best practices for PRO N guidelines in clinical trials. PRM157 Re Introduction of the Ranking Task in Eq-5d Valuation. Improved Data Quality and Reduced Level of Inconsistencies? Ramos-Goñi J M 1, Rand-Hendriksen K 2, Igarashi A 3, Ikeda S 4, Pinto-Prades J L 5 1The EuroQol Group Foundation, Rotterdam, The Netherlands, 2Faculty of Medicine, University of Oslo, Oslo, Norway, 3University of Tokyo, Graduate School of Pharmaceutical Sciences, Tokyo, Japan, 4International University of Health and Welfare, Otawara City, Tochigi, Japan, 5Glasgow Caledonian University, Newton Mearns, UK . . . . . . . Objectives: The EuroQol group observed problems related to the quality of valuation data collected using the 5L protocol. The objective of this study was to test if reintroducing a ranking task prior to TTO could help to improve the quality of the data. Methods: Respondents were assigned to three different study branches in Spain, while just two branches were used in Japan. The control group (Arm 1, both countries) was the 5L protocol. The second branch (Arm 2, just Spain) was the EQ-VT preceded by a ranking task, in which the respondents were asked to rank the target health states using physical cards. The last branch (Arm 3, both countries) was with ranking prior to the EQ-VT, but unlike the second branch, the ranked states remained visible through the Composite TTO (C-TTO) tasks, and the C-TTO states order were determined by the rank order. A sample of respondents was randomly selected from Canary Islands and Tokyo population. All interviews were performed face to face. We compared the number of C-TTO-based inconsistencies. Results: The final sample size was 932 respondents (141 and 191 for branches 1 and 3 respectively in Japan and 196,205 and 199 for branches 1, 2 and 3 respectively in Spain). The percentages of inconsistencies on C-TTO responses (state 55555 not valued as the worst state) were 13.27%, 10.73% and 13.57% for branches 1, 2, 3 respectively in Spain, and 20.6% and 10.5% for branches 1 and 3 in Japan. Grouping the data, the inconsistencies were 16.3%; 10.7% and 12.1%. However, none of the observed differences were statistically significant. Conclusions: Differences in proportions observed in the Spanish data are smaller than in Japanese data. Given that Japanese team did all control arm interviews prior to ranking task interviews could play in favor of ranking arm (learning effects). PRM158 Are “Lively” and “Full of Pep” Similar Or Different Concepts? Challenges in Translating These Terms in Seven Languages Gil-Campos I 1, Anfray C 1, Acquadro C 2 1Mapi, Lyon, France, 2Mapi Research Trust, Lyon, France . . . Objectives: Feeling “lively” or “full of pep” are common items used in PRO measures. The Profile of Mood States (POMS) identifies and assesses six fluctuating mood or affective states: 1) Tension-Anxiety, 2) Vigour-Activity, 3) DepressionDejection, 4) Fatigue-Inertia, 5) Anger-Hostility, and 6) Confusion-Bewilderment. “Lively” (item 7) and “full of pep” (item 56) belong to the Vigor-Activity domain. The objectives of this study were to evaluate the difference in concept (if any) between these items and to translate them in 7 languages (Bulgarian, Estonian, French, German, Latvian, Portuguese and Slovakian). Methods: The POMS was translated following the Mapi methodology in compliance with the ISPOR guidelines. The process consisted of: 1) definition of each concept behind each item, 2) two forward translations by native translators, reconciliation of the translations, 3) one back-translation by an English-speaker fluent in the target language, 4) final reconciliation, and 5) cognitive interviews with 35 native-speaking healthy subjects (n= 5 per country). Results: The first step of the translation process, i.e., definition of concepts, helped to clarify the conceptual differences between “lively” and “full of pep”. “Lively” had to be considered more as an action or behavior-based descriptor and “full of pep” as descriptive of the internal state. “Lively” was considered as the manifestation of feeling “full of pep”: With this definition in mind, each local team could find appropriate equivalents in each language. “Lively” was translated with literal wording in all languages. As for “full of pep”, translations varied from “enthusiastic” (Bulgarian, Portuguese) to “full of vitality” (Estonian, Slovakian) or “peppy” (German), “full of spirit” in French and “full of energy “in Latvian. The translations were well understood. Conclusions: “Lively” and “full of pep” are two different concepts which could be easily translated with the help of the developers who provided clear definitions of the intent of each item. PRM159 A Reliability Generalisation of the Eortc Qlq-Br23 Smith A B 1, Taylor M 2, Cocks K 1, Parry D 3 1University of York, York, UK, 2York Health Economics Consortium, York, UK, 3AstraZeneca UK, Macclesfield, UK . . . . . Objectives: The collection of patient-reported outcome (PRO) data is becoming more routine in oncology clinical trials and in clinical practice. The European Organisation for Research and Treatment of Cancer (EORTC) has developed and validated a number of PRO instruments, including the QLQ-BR23, a breast cancer specific measure. The objective of this study was to evaluate the internal reliability of the EORTC QLQ-BR23 through reliability generalisation, a meta-analytic technique for pooling Cronbach’s alpha coefficients. Methods: A systematic literature review was undertaken of articles reporting the internal reliability of the EORTC QLQ-BR23 to identify potential studies for inclusion. Internal reliability was assessed for each of the five domains of the EORTC QLQ-BR23: Body (BI), Sexual Functioning (SF), Arm Symptoms (AS), Breast Symptoms (BS) and Systemic Therapy Side Effects (ST). Reliability generalisation was undertaken using fixed and random effects models for each domain. A value for Cronbach’s alpha coefficient above 0.70 was considered to indicate reliability. Results: A total of 15 studies were found resulting in 18 datasets. Of the 3102 patients who had been recruited into studies 2888 had completed the BR-23 (93%). Almost half the studies had used European language versions of the instrument (7/15,47%). The unweighted average (fixed effects) Cronbach’s alpha coefficient ranged from 0.69 (breast symptoms) to 0.86 (sexual functioning). There was statistically significant heterogeneity present across the 5 domains under the fixed effects model justifying the use of random effects. All Cronbach’s alpha > 0.7 under the random effects model, e.g. BI: 0.84 (95%CI: 0.80-0.87), SF: 0.87 (95%CI: 0.84-0.89), AS: 0.71 (95%CI: 0.66-0.76), BS: 0.70 (95%CI: 0.65-0.74) and ST: 0.71 (95%CI: 0.66-0.75). Conclusions: The results suggest that the 5 domains of the EORTC QLQ-B23 are reliable and may be used in clinical trials and practice to capture PRO data to help inform trial, as well as individual clinical decision-making processes. PRM160 Improving Performance in Diabetes Care: Benefits of Information Technology Enabled Diabetes Management Saric T 1, Lazic G 2, Poljicanin T 3, Prenda Trupec T 4 savjetovanje, Zagreb, Croatia, 2Private practice family medicine Karlovac, Karlovac, Croatia, 3Croatian National Institute Of Public Health, Zagreb, Croatia, 4Croatian Health Insurance Fund, Zagreb, Croatia . . . . 1Promeritus Objectives: Empirical data suggest that compliance with diabetes clinical practice recommendations is inadequate in primary care and that a large proportion of patients with diabetes remain at high risk. Changing physician practice patterns with use of electronic integrated system would provide a suite of technologies to support a full range of diabetes management activities. The objectives of this study are to determine the financial and clinical benefits of implementing information technology enabled diabetes management systems. Methods: The simulations were performed using the CORE model – widely validated and broadly used to enable a reliable estimation of costs and clinical effects associated with diabetes. Several estimates of care process improvements were derived, representing different percentage of patients covered by designed management activities and reaching target HbA1c. The primary outcome was medical cost savings and secondary measures include reduction of cardiovascular, cerebrovascular, neuropathy, nephropathy, and retinopathy clinical outcomes. Results: Gradual, sustained, and statistically significant improvements in a lifelong perspective performance measures were observed. According to simulation, as percentage of patients covered by information technology enabled diabetes management is growing, health care expenditures are reduced. With 10% of patients covered 6% of total patient costs were saved in comparison to 23% of total patient costs saved with 100% of patients covered. Conclusions: Implementation of information technology enabled diabetes management has demonstrated significant potential for improving processes of care, preventing the development of diabetic complications, and generating cost savings. Moreover, this improves the synthesis of information, the delivery of knowledge, and the efficiency of communication, allowing for coordination of care across delivery teams. A572 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PRM161 Assessing the Methodological Value of Digital Real-Time Collection of Qualitative Content in Supporting In-Depth Qualitative Interviews Exploring the Symptoms and Impacts of Gout on Health-Related Quality of Life Rudell K 1, Tatlock S 2, Panter C 2, Arbuckle R 2, Symonds T 1 Ltd, Tadworth, UK, 2Adelphi Values Ltd, Bollington, UK . . . . . 1Pfizer Objectives: Gout is the most common arthritic condition, but research of Gout patient experiences is limited. Qualitative interviews are a valid and well-accepted means of gaining in-depth insight into the patient experience, but are typically conducted in artificial environments relying on patient recall. Novel digital methods of collecting qualitative data through real-time data capture (RTDC) have recently emerged and have stronger ecological validity. The objective of this research was to assess the added methodological value of analyzing audio/visual data recorded and submitted by gout patients through a mobile phone application while experiencing flares, in addition to traditional patient interviews. Methods: Concept elicitation interviews were conducted with 20 American gout patients using open-ended exploratory questions to facilitate spontaneous elicitation of content. Following interviews, 50% of the sample took part in a RTDC exercise by submitting selfrecorded videos and images in response to six tasks issued on a mobile phone application over seven days. All data were subject to thematic analysis using Atlas. ti. Interviews and RTDC data were compared in terms of conceptual coverage and insights. Results: Qualitative analysis demonstrated both forms of data collection led to the identification of the same symptoms and impacts with no additional concepts identified in either form of data. Symptoms and impacts of the disease during flares were high. RTDC data provided additional insight into the severity of symptoms and level of impact burden, specifically with regards to images of the level of swelling experienced during a flare and sleep disturbances through videos recorded by patients at the time of disturbance. Conclusions: Whilst traditional patient interviews remain the gold standard in exploring the symptoms and impacts of a condition, RTDC was found to provide additional valuable insights in this sample of gout patients, which can inform future measurement strategies and enhance the field of patient-centered research. PRM162 What Is the Appropriate Comparator Health State To Use in Time Trade-off Studies? Shah K Office of Health Economics, London, UK . Objectives: Studies to produce utility values for the EQ-5D-5L instrument are ongoing internationally. These include the valuation of 10 EQ-5D-5L health states using the time trade-off (TTO) method. In some of the studies carried out to date, relatively low mean TTO values for mild health states have been observed. It is hypothesised that this is because the health states under evaluation are being compared to “full health”, whereas in previous studies they were compared to 11111 (the “best” health state in the descriptive system). The objective is to assess differences in TTO valuations using two different comparators (full health and 11111). Methods: Preferences for EQ-5D-5L health states were elicited from a broadly representative sample of the UK general public. TTO data were collected using computer-assisted personal interviews, carried out in respondents’ homes. Respondents were randomly allocated to one of two arms: in arm 1 the comparator health state was full health; in arm 2 the comparator health state was 11111. After completing 10 TTO valuations, respondents were asked follow-up questions which sought to examine their interpretations of the term “full health”. Results: 450 interviews were completed in mid-2014. Health state 11111 was almost always given a value of 1; yet the majority of respondents who self-reported as being in 11111 did not consider themselves to be in “best imaginable health”. Preliminary analyses suggest that the use of 11111 (rather than full health) as the comparator does not increase the average values elicited for mild health states. A sizeable minority of respondents did not agree that 11111 and full health are equivalent. Vision and spirituality were mentioned by respondents as examples of important aspects of health not covered by 11111. Conclusions: The low observed values for mild EQ-5D-5L health states cannot be explained by the choice of comparator health state alone. the Total SGRQ score (r= 0.73) and Symptoms (r= 0.62); Activity (r= 0.55), Impact (r= 0.72) subscales; the modified MRC Dyspnoea scale (r=0.54), FACIT-F (r= -0.66); HADS Anxiety (r=0.53) and Depression (r= 0.54). The MEMSI also demonstrated good overall fit to the Rasch model (Chi-squared 26.6; df 20; p=0.26) and distribution of item scores. Conclusions: The final MEMSI contains 10-item and is a reliable, valid, unidimensional self-reported outcome measure of early morning symptoms for people with COPD. It is quick and simple making it suitable for use in research and practice. Further work is underway to determine the minimal clinical important difference. PRM164 An Electronic Version of the Pdq-39: Acceptability To Respondents and Assessment of Alternative Response formats Morley D , Dummett S , Kelly L , Dawson J , Jenkinson C University of Oxford, Headington, UK . . . . . Objectives: Firstly, to migrate the paper-based version of the PDQ-39 to a computer based platform and assess its usability and acceptability to respondents. Secondly, to investigate the impact of implementing non-response options on response rates and data completeness. Methods: Six people with Parkinson’s (PwP) participated in cognitive interviews in order to assess the usability and acceptability of the electronic version of the PDQ-39, the ePDQ. This was followed by an online survey of 129 PwP, randomly assigned to one of two groups; one required to provide a response to every item and one with the option to skip any item they did not wish to answer. Results: Cognitive interviews indicated that the ePDQ is acceptable to PwP, with positive feedback regarding layout, features and functionality. 125 PwP fully completed the ePDQ. Following randomization 60 participants completed the forced response ePDQ and 65 completed the non-forced version. Response rates of 98.4% were achieved for the forced response ePDQ and 95.6% for the non-forced. Missing value analyses calculated levels of missing data at below 5% in the non-forced sample. Conclusions: The ePDQ is user-friendly and acceptable to respondents. Additionally, there appears little difference when implementing non-response options on response rates and data completeness. PRM165 Efficacy of Virtual Reality Exposure Therapy in the Management of Symptoms Associated With Post Traumatic Stress Disorder Jiandani N , Nair S R , Shukla H Capita India Pvt. Ltd., Mumbai, India . . . . Objectives: Management of symptoms and illness for post traumatic stress disorder (PTSD) requires effective and prolonged psychiatric support. Re-experiencing traumatic memories and events via virtual reality exposure therapy (VRET) can lead to reduction of illness and symptoms associated with PTSD. This systematic literature review aims to demonstrate the value of VRET in the management of illness and symptoms associated with PTSD. Methods: Studies published in the English language for PTSD and VRET were retrieved from Embase, PubMed, and Cochrane databases using relevant search strategies. Two researchers are independently reviewing studies as per the Cochrane methodology for systematic literature reviews. We considered VRET as a tool to deliver therapy programs via a virtual platform to patients suffering from PTSD. The main outcome will be improvement in symptoms such as anxiety, various phobias and depression developed as a result of PTSD. Outcomes will be measured as change in baseline characteristics in patients using VRET, by using questionnaires specific to the symptoms being measured such as Beck’s Anxiety Inventory, Beck’s Depression inventory-II (BDI-II), ClinicianAdministered PTSD Scale (CAPS), and patient reported outcomes. Results: In total, 2574 potentially relevant studies were retrieved from the databases and are being screened for inclusion in the review. The detailed results from the systematic review will be presented in the poster. Conclusions: The evidence from this systematic literature review will hopefully suggest the role of VRET as a promising new tool for managing PTSD from a psychotherapeutic perspective. PRM166 Quality of Life Elements in Schizophrenia for Patients and Carers offer Challenges To and Opportunities for Intervention Floyd D 1, Gemmell E 2, Brown J 1 . . . 1PHMR Associates, London, UK, 2PHMR Associates, Newcastle PRM163 The Development and Preliminary Validation of the Manchester Early Morning Symptoms Index (Memsi) for People With Chronic Obstructive Pulmonary Disease (Copd) Garrow A P 1, Yorke J 1, Khan N 2, Tyson S 1, Singh D 3, Vestbo J 4 1University of Manchester, Manchester, UK, 2The Medicines Evaluation Unit, Manchester, UK, 3University of Manchester, Medicines Evaluation Unit, Manchester, UK, 4The University of Manchester, Manchester, UK . . . . . . . Objectives: Early morning symptoms (EMS) of COPD are associated with poor health, impaired daily activities and increased exacerbation risk. We describe the development and preliminary validation of the Manchester Early Morning Symptom Index (MEMSI) to quantify EMS in COPD. Methods: Potential items were identified from interviews and focus groups with COPD and non-COPD participants. Cognitive de-briefing elucidated the relevance and understanding of the potential items followed by a cross-sectional study to finalise the items for inclusion. The draft scale was completed twice, one week apart to assess test-retest reliability. The St George’s Respiratory Questionnaire (SGRQ), modified MRC Dyspnoea Scale, FACIT fatigue scale and Hospital Anxiety and Depression Scale (HADS) evaluated convergent validity. Hierarchical methods and Rasch analysis informed item reduction and assessed uni-dimensionality. Results: 36 patients with COPD initially identified 22 items. The cross sectional study included 203 COPD patients (GOLD: I: 14% II: 41% III: 25% IV: 7%, male: 63%, mean age 64.7, SD: 7.5 years) and 50 age and gender matched non-COPD controls. 12 items were removed during hierarchical methods. The MEMSI demonstrated excellent internal consistency (Cronbach’s alpha 0.9) and good testretest repeatability (r=0.82). The scale also showed moderate-good correlation with upon Tyne, UK Objectives: Schizophrenia is responsible for high levels of individual morbidity: acute schizophrenia had the highest disability score of any condition in the 2010 Global Burden of Disease study. Consequently, quality of life (QoL) overall and related to health is a major issue for patients, as well as their families/carers. Methods: We undertook a pragmatic literature search of publications relating to QoL and schizophrenia over the past 7 years to identify main themes and trends. Results: Patients: Patient-reported QoL is often over-estimated compared to objective QoL as measured by health care professions, particularly during psychotic episodes. Factors that appear to be adversely associated with QoL include symptoms, treatment side effects, physical mobility, lack of relationships, daily activity, housing, social stigma and self-stigma. In some cases, there is a complex cause and effect relationship. Families/carers: Those looking after patients with schizophrenia suffer impaired QoL and this can have an impact on the whole household and its social milieu. Poor carer QoL can ultimately impair levels of care and increase risk of mistreatment, which in turn risks a relapse of symptoms and ultimately hospitalisation. On the other hand, the experience of caring for a patient with schizophrenia may lead to personal emotional/psychological growth. Disease insight and education for some schizophrenia sufferers can lead to diminished QoL and an increased risk of suicide. On the other hand, disease understanding can help improve treatment adherence. Conclusions: QoL in schizophrenia is complicated and must be fully understood to help develop effective programmes to improve QoL for sufferers and their families/carers. Programmes that do not reflect this complexity may increase risk of symptom recurrence and even suicide. Thus, when devising interventional programmes to supplement drug treatment for management of schizophrenia, a range of patient- and carer-related factors must considered. A573 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PRM167 Comparability of Interview and Self-Administration of the Functional Assessment of Chronic Illness Therapy-Tuberculosis (Facit-Tb) Instrument in Iraqi Pulmonary Tuberculosis Patients Dujaili J A 1, Syed Sulaiman S A 1, Hassali M A 2, Blebil A Q 3, Awaisu A 4, Bredle J M 5 Sains Malaysia, Minden, Malaysia, 2Universiti Sains Malaysia, Penang, Malaysia, 3UCSI University, Kuala Lumpur, Malaysia, 4Qatar University, Doha, Qatar, 5FACITtrans, Elmhurst, IL, USA . . . . . . . . . . . 1Universiti Objectives: To investigate the extent to which two different modes of administration (interview by a trained interviewer versus self-administration) yielded a comparable estimate of health-related quality of life (HRQL) in pulmonary tuberculosis (PTB) patients. Methods: The study was conducted between September 1st 2012 and July 31st 2013, among consecutive PTB patients treated at Thoracic and Respiratory Disease Specialist Centre in Baghdad, Iraq. The mode of administration of the Functional Assessment of Chronic Illness Therapy-Tuberculosis (FACIT-TB); a new tuberculosis (TB) -specific instrument, at baseline was registered in 305 subjects. Results: Although the FACIT-TB was designed for self-administration, most patients in our sample (N = 193,63.278%) requested some help from the interviewer to fill out the questionnaire. Mann Whitney U test showed that those patients capable of self-administration were younger (38.16 ± 12.93 versus 43.58 ± 16.41 years, P = 0.005) and required less time to complete the questionnaire compared to those who interviewed by a trained interviewer (14.64 ± 3.24 versus 17.22 ± 2.61 minutes, P < 0.001), while Chi-Square statistics showed that this group of patients had a higher education level (P < 0.001). No differences in gender were observed. HRQL score across all domains for those who interviewed by a trained investigator was slightly lower than those who answered the questionnaire by self-administration. However, the results did not reach statistical significance (P> 0.05). Conclusions: Technical equivalence has been demonstrated in the sample of PTB patients in Iraq. FACIT-TB instrument is flexible and it is able to accommodate the needs of patients with diverse social, educational, and functional skills. Technical equivalence across different modes of administration of questionnaire permits unbiased assessment of the impact of the disease and its treatments on patients’ HRQL. PRM168 Health Related Quality of Life in Cancer Patients: Evaluation With A Self-Administered Ipad Application Yuce D , Hayran M , Kilickap S , Erman M , Celik I Hacettepe University, Ankara, Turkey . . . . . Objectives: Cancer has the second highest mortality rates after cardiovascular diseases in the world. Advances in treatment options caused significant enhancements in survival of cancer patients. However, the major parameter affecting treatment success and treatment adherence in these patients is the quality of life (QoL). We aimed to develop a self-administered iPad application for evaluation of QoL in a short questionnaire. Methods: As part of routine practice in Preventive Oncology Department EORTC QLQ-C30 was administered to 1549 cancer patients treated in Hacettepe University Oncology Hospital. We determined 10 questions that explains most of the variation in QoL using factor analysis, and designed a new application for iPad, where patients can record responses themselves. Results: The 10 factors that described by the factor analysis had the power of 74,2% explaining QoL variances. The reliability analysis of these factors showed a Cronbach alpha coefficient of 0,75. The new self-administered iPad application was tested in a pilot study that conducted in 127 patients taking chemotherapy regimens in the outpatient setting. The validity and reliability analyses revealed that the new application can be effectively used in Turkish cancer patients. Conclusions: Our results revealed that our software application will be useful and efficient for monitoring of the changes in QoL during their treatment course. Furthermore, this kind of mobile applications may be practical for health professionals in daily routine clinical assessments of patients. Also, the audio and visual enhancements in electronic applications provide increased accessibility for the cancer patients. PRM169 The Selection of Appropriate Health State Utility Values (Hsuvs) for Health Technology Assessment (Hta): Lessons To Be Learnt Patel A , King D , Greenall G , Mitchell S A , Purushotham S , Hudson P M Abacus International, Bicester, UK . . . . . . . . Objectives: Incorporation of health-related quality of life (HRQoL) evidence into economic models is a requirement of many countries for the purposes of Health Technology Assessment (HTA), and therefore appropriate health state utility values (HSUVs) are often sought. The objective of this review was to: (i) identify and summarise the principal limitations of HSUVs used in recent submissions appraised by the National Institute for Health and Care Excellence (NICE) and (ii) produce a categorical checklist that can be used by manufacturers to reduce uncertainty when selecting HSUVs for HTA. Methods: Evidence appraisal documents for the 50 most recently published technologies assessed by NICE were retrieved in June 2014. Economic models were assessed and utility inputs reviewed. Critiques of the utilities reported by the evidence review group or final appraisal committee were extracted, reviewed and categorised. Results: Of the appraisals reviewed (43 single technology appraisals (TAs) and 7 multiple TAs), utility inputs were either sourced from the literature (n= 27), published mapping algorithms (n= 11), de novo mapping algorithms (n= 2), or derived from clinical trials (n= 10). The concerns expressed by review groups can be categorised into four categories: (i) generalisability – relevance of HSUVs to UK clinical practice, deviation from NICE scope, and the use of other countries’ valuations for health states; (ii) HSUV selection – inadequate justification of HSUVs, and lack of consideration for covariates and disutilities; (iii) mapping algorithms – use of non-validated or non-peer-reviewed publications, incomplete reporting of key model information, and ambiguity regarding selection and justification of mapping function; (iv) risk of bias – sample size, instrument response rates, and general study quality identified as factors affecting HSUV validity. Conclusions: The selection of appropriate HSUVs is critical to reduce uncertainty in economic models. A checklist based on critiques of recent HTAs will be a useful tool for manufacturers when selecting relevant HRQoL parameters. PRM170 Translation and Linguistic Validation of the Electronic Colombia Suicide Severity Rating Scale in Asia-Pac Anderson H 1, Gordon-Stables R 1, Wild D 1, Riale K 2 1ICON plc, Oxford, UK, 2ERT, Pittsburgh, PA, USA . . . . Objectives: The Columbia Suicide Severity Rating Scale (C-SSRS) has been developed into an electronic self-rated version (the e-CSSRS) in order to facilitate compliance with regulatory requirements for prospective monitoring of suicidal ideation and behaviours. The e-CSSRS v2.0 for IVRS has been translated and linguistically validated for use in over 60 countries. Whilst translating and linguistically validating the scale it was noted that particularly in the Asia-Pacific region there were some challenging issues around the concept of suicide from a linguistic and cultural perspective. This study aimed to determine what these issues were and highlight how they were resolved. Methods: Eighteen reports were reviewed from the AsiaPacific region. The languages were: China-Mandarin, India-English, India-Gujarati, India-Hindi, India-Kannada, India-Malayalam, India-Marathi, India-Tamil, IndiaTelugu, Korea-Korean, Malaysia-English, Malaysia-Malay, Malaysia-Mandarin, Philippines-English, Philippines-Tagalog, Singapore-English, Singapore-Malay, and Singapore-Mandarin. Each report was reviewed for challenges relating to translation and cultural adaptation. Results: Across all reports specific homonymic confusions were encountered within two target languages: in Indian Hindi the same word is used for “pill” and “bullet”; in Singapore Malay the words for “end” and “saving” sound similar, occasioning confusion in prompts about attempts to end life. Translations were adapted to avoid these confusions. In two cases the use of a gun in suicide attempts was found to be rare or unknown: in Singapore English the relevant prompt was clarified to minimise confusion; for Malayalam for Kerala, where shooting is rare and hanging is common, “getting a gun” was changed to “getting a rope” wherever it appeared and for similar reasons “collecting pills” was changed to “collecting a poison.” Conclusions: The e-CSSRS v2.0 IVRS is now available in for use in over 60 countries. Some issues were identified relating to translation and the methods of suicide in a small number of languages but these were resolved throughout the linguistic validation process. PRM171 From Clinically Relevant Outcome Measures To Quality of Life in Epilepsy de Kinderen R J A 1, Wijnen B F M 1, Postulart D 2, Majoie M H 2, Aldenkamp A P 2, Evers S M 1 University, Maastricht, The Netherlands, 2Epilepsy Center Kempenhaeghe, Heeze, The Netherlands . . . . . . . . . . . . . 1Maastricht Objectives: Utilities can be easily derived using generic quality of life (QoL) instruments. However, problems in collecting utility scores may occur because clinical evaluations still favor disease specific instruments over generic quality of life instruments for reasons of sensitivity and reliability. Especially in case of the fluctuating nature of seizures in epilepsy, generic QoL-instruments are often find to be unsuitable for outcome research. A proposed method to bridge the gap between clinically relevant outcome measures and QoL is to derive utility scores for epilepsy health states. The aim of this study is to develop a scoring algorithm to transform epilepsy health states into utility scores. Methods: The proposed scoring algorithm was based on valuations of health states generated by the Time Trade-Off (TTO) method. The TTO was based on clinically important attributes (seizure frequency, seizure severity and treatment related side-effects). A full factorial design was used which resulted in 78 scenario’s. Besides standard demographics, every participant was asked to value 10 or 11 different health states. A multilevel regression analysis was performed to account for the nested structure of the data. The TTO was conducted using online survey software. Results: In total 531 subjects of the general population, with an average age of 42 years, have participated in the TTO study. Preliminary results show that the best health state (no seizures and no side-effects) is estimated at 0.89 utility and the worst state (seizures twice a day, many side-effects, type 5 seizure) is estimated at 0.16. Conclusions: This study provides a scoring algorithm for transforming clinically relevant outcome measures of epilepsy into utility estimates which can be incorporated into economic evaluations. Although seizure frequency is the most commonly reported primary outcome measure in epilepsy research, this study suggest that the impact of seizure severity alone should not be underestimated. PRM172 An Analysis of the Health Technology Assessment Recommendation and Guidance on Use of Eq-5d-5l in Cost-Effectiveness Modeling Ovcinnikova O 1, Firth Z A 2, Dimova M 2 . . . . 1MAPI, London, UK, 2Mapi, London, UK Objectives: Several national health technology assessment (HTA) bodies including the National Institute for Health and Care Excellence (NICE) for England and Wales recommend EQ-5D as the preferred health-related quality of life (HRQL) measure for use in cost-effectiveness analyses. This study aims to evaluate the recommendation and impact of using EQ-5D-5L versus EQ-5D-3L for cost-effectiveness modeling in HTA submission guidelines and subsequent reimbursement decisions. Methods: A targeted review of national HTA submission modeling guidelines since the introduction of the EQ-5D-5L measure was carried out for the UK and France. The costeffectiveness modeling guidelines were assessed for the requirements for EQ-5D at either level. The impact of guideline requirements on submissions were analysed by examining the presence of EQ-5D-5L in the 20 most recent technical appraisal manufacturer submissions for each HTA body. Results: While all HTA bodies recommended the use of EQ-5D as a generic measure of HRQL, only NICE specifically recommended EQ-5D-5L. However, it was not a requirement. Of the latest published guidance only NICE and the Scottish Medicines Consortium had one EQ-5D-5L submission each. Despite evidence suggesting increased sensitivity and reduced ceiling A574 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 effects associated with EQ-5D-5L, this measure is still not a key requirement for costeffectiveness modeling in selected HTA submissions. Conclusions: EQ-5D-3L has the potential to distort the true cost-effectiveness in conditions that are insensitive to the measure. Awareness and requirements for the use of EQ-5D-5L should increase amongst manufacturers and HTA bodies to ensure submissions present accurate cost-effectiveness models. Further research into the evaluation of costeffectiveness results between these two measures following the introduction of validated independent value sets for EQ-5D-5L is encouraged. PRM173 Modification of Patient Reported Outcomes Measures of Compliance, Gastrointestinal Symptoms, Palatability and Treatment Satisfaction for Patients Needing Iron Chelation therapy Lasch K 1, Côté I 2, Roma T 2, Srivastava B 2, Horodniceanu E G 3, Dhatt H 3, Carter J A 3, Bal V 2 International, Cambridge, NJ, USA, 2Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA, 3Pharmerit International, Bethesda, MD, USA . . . . . . . . . . 1Pharmerit Objectives: In patients requiring frequent blood transfusions due to transfusion dependent anemia (TDA) (e.g., sickle cell disease [SCD]) and myelodysplastic syndrome (MDS), life expectancy is directly related to the quality of chelation therapy, and poor adherence to treatment increases the risk of complications and shortens survival1. Improved palatability ratings and gastrointestinal (GI) tolerability could have positive impacts on adherence with iron chelation therapy (ICT) 2. Therefore, patient-reported outcomes (PROs) measuring compliance, GI symptoms, and palatability were developed and the Satisfaction with Iron Chelation Therapy (SICT) questionnaire was modified, as electronic PROs (ePROs) specific to a new formulation of ICT. Methods: Eleven patients with TDA or MDS provided informed consent and were included in this qualitative study. Two sets of face-to-face cognitive interviews were conducted iteratively; modifications to items were debriefed in the second set. Interviews began open-endedly to elicit patients’ spontaneous experiences with ICT. Interviews were audio recorded and transcribed. An item tracking matrix documented the changes made for each item. Data analysis used ATLAS. ti software. This study was conducted according to best practices for development3 and modification4 of PROs in an ePRO format. Results: Patients were 73% (n= 8) male and 27% (n= 3) female with a mean age of 43 (range 14-81 years); 45% (n= 5) had SCD, and 27% (n= 3) had MDS. Patient spontaneous reports and cognitive debriefing responses confirmed concepts in the draft PROs and helped eliminate irrelevant items. Changes made after cognitive debriefing ensured the comprehensibility, lack of redundancy, and appropriate instructions and response options. The resultant PROs included the Compliance (2 items), GI Symptom Diary (6 items), Palatability (4 items), and modified SICT (13 items). Conclusions: Results support the content validity of PRO measures of compliance, GI symptoms, palatability, and satisfaction with ICT. These measures require psychometric validation of validity, reliability, and responsiveness before recommending their use in future clinical research. PRM174 The Importance of Migration Assessments: Ecoa Translations and Linguistic Validation Sweeney E , Kelley T TransPerfect, New York, NY, USA . . Objectives: As the use of clinical outcomes assessments (COAs) in global studies continues to increase, early collaboration between eCOA and linguistic validation providers becomes critical to the success of global initiatives. Early involvement of a linguistic validation partner in the eCOA migration process offers insight into migration issues that may not be present in the English version of the instrument, but if not identified, can lead to study delays as well as increased costs for the sponsor. Methods: An examination was conducted of previous linguistic validation projects that included either newly developed eCOA instruments or pen/paper to eCOA migration. A comparison of various eCOA platforms, the corresponding issues, and details relating to migration solutions were assessed. Results: Frequently, a line of text that is present in an eCOA platform (proprietary software, Excel, etc.) is coded such that it populates into the eCOA device in multiple locations. These segments are sometimes referred to as “computed text” and are commonly used for response options that repeat for multiple items of a questionnaire. While this may work adequately in English, many languages require a variance in the translation used based on the context of the item and/or response choice. While a pen/ paper version may allow for these variances, the initial eCOA programming may not. The variance is only determined further into the linguistic validation process and presents challenges if the device requires re-programming. Conclusions: A migration assessment, separate from equivalency testing, allows for the eCOA and linguistic validation providers to assess the initial setup of the eCOA software. This assessment can determine whether system modifications are necessary to allow for the translations to properly be mapped and displayed. This additional step will also prevent study delays as well as quality issues as it allows issues to be addressed early and avoid later difficulties. feedback on a 4 point scale for applicability, comprehensibility and appropriateness of the response options of each item. They were also encouraged to leave free-text feedback. Results: A total of 218 subjects provided feedback, 61.47% female. Mean age was 55.78 years, with patients from 18-80 years, 40.37% between 55-65 years. The most represented conditions were Parkinson’s (27), Fibromyalgia (23), Multiple Sclerosis (20) and Rheumatoid Arthritis (19). The mean number of quantitative assessments per item was 55.74, while the mean number of comments was 11.64. Considering a cut-off of 2 points in the 0-3 scale, 18 items had low applicability, just one item had low comprehensibility and all items had appropriate response options. Conclusions: An impressive amount of feedback was obtained in just a week using the Open Research Exchange platform. The lowest scores were in applicability. However, as the items are destined to an adaptive test, some of them are very extreme (destined to severe cases). Therefore applicability was considered a secondary concern. With this caveat, quantitative feedback was very positive, and added to the great amount and detail of qualitative feedback suggest some changes that we think will greatly improve the instrument. PRM176 Quality of Life in Patients Undergoing Hemodialysis: A Vision of the Influence of Time Maia R C F 1, Monteiro W M S 1, Silva M G C 2, Almeida R F C 1 1Secretaria de Saúde do Estado do Ceará, FORTALEZA, Brazil, 2UECE, Fortaleza, Brazil . . . Castejón N 1, Harrington M 2, Campillo-Álvarez Á 1, Rebollo P 1 1LASER ANALYTICA, Oviedo, Spain, 2PatientsLikeMe, Cambridge, MA, USA . . . . Objectives: CAT-Health is a generic health related quality of life computer adaptive test developed and validated in Spain. Based on its 96 item pool, a new instrument is being developed in English. The objective of the present study was to obtain substantial feedback from patients on the items, through Open Research Exchange platform, that centralizes PROM research and is integrated with PatientsLikeMe, an online community. Methods: The item pool was split in four sets of 24 items each. 750 patients with various conditions received an e-mail that included link to a survey open for 7 days. They were asked to answer one of the sets and give . . . . . . . . Objectives: To measure and compare the quality of life in patients undergoing hemodialysis with respect to time. Methods: An observational, prospective study conducted in 2012/2013 consisting of 50 patients with Terminal Chronic Kidney Disease, undergoing HD in a public clinical care within the metropolitan area of Fortaleza. We used the KDQOL instrument early in therapy and one year after its beginning in order to observe changes in quality of life over time. Statistical analysis included frequency distribution and measures of central tendency. Results: The general dimensions of this instrument indicate that the dimensions “physical function”, “social function” and “emotional function” had the lowest scores early in therapy. It is observed that these same dimensions showed significant improvements in patients after one year of HD. Only “general health” was not significant. At baseline, specific dimensions of the instrument as “professional role”, “sexual function” and “disease overload” have the lowest scores and remain the same after one year of the begin of therapy, showing that the time difference has no statistical significance. Conclusions: The quality of life of patients undergoing hemodialysis after a year of therapy is superior to the quality of life showed at the beginning of treatment although this difference is not statistically significant. It is important to search for alternatives that can positively influence the quality of life of these patients. PRM177 Methods Used To Measure Patient Preferences In Psoriasis Treatments – An Overview With Regards To The German Iqwig And G-Ba Gutknecht M , Herrlein O , Augustin M University Medical Center Hamburg-Eppendorf, Hamburg, Germany . . . Objectives: According to the methodology of the German Institute for Quality and Efficiency in Healthcare (IQWiG), the benefit assessment of a new treatment intervention refers to patient-reported outcomes exclusively. Given the existence of different patient-relevant outcomes the Federal Joint Committee (G-BA) can engage the IQWiG to aggregate these to one comprehensive benefit measurement. So far, the IQWiG uses no standardized method for this. In its current method paper, among the benefit QALY, the IQWiG refers to methods of multi-criteria decision making or preference evaluation like analytic hierarchy process and conjoint analysis. For psoriasis, one of the most frequent chronic skin diseases worldwide, a variety of treatment interventions is available. In the light of recent discussions, the objective was to give an overview of methods which have been used to date to prioritize and weight patient-relevant outcomes in psoriasis treatment. Methods: The present review is based on a systematic literature research until 31 December 2013 in the databases PubMed, Embase, Ovid Medliner, Cochrane Library, EconLit and CINAHL using the keywords “psoriasis” and “preferences”. Results: The search resulted in 288 hits without duplicates. 16 articles met the predefined inclusion criteria. In addition to methods to calculate QALY like time-trade off and standard gamble, conjoint analysis, willingness-to-pay and other preference methods were used. In view of the method paper of IQWiG, no study exists in the field of psoriasis where the analytic hierarchy process was used. Conclusions: The results of the presented review show, that the analytic hierarchy process was not used in psoriasis studies, so far. The use of this method in future studies might provide new essential knowledge in the evaluation of patient preferences in psoriasis treatment. PRM178 Impact of Osteoporotic Fractures on Quality of Life – Design of A Mapping Study of Qualiost To Eq-5d Hansen L 1, Vestergaard P 1, Petersen K D 2 University, Aalborg East, Denmark, 2Aalborg University, Aalborg, Denmark . PRM175 Patient Driven Questionnaire Development, Item Feedback from Users of A Patient Network . . . . 1Aalborg Objectives: The QUALIOST (quality in life questionnaire in osteoporosis) is one of the most commonly used osteoporosis-specific health-related quality of life (HRQoL) questionnaires and it is often used in clinical studies to document the longitudinal impact of osteoporosis and related fractures. It is unknown whether the QUALIOST is better at estimating HRQoL for osteoporotic fracture patients compared to the commonly used EQ-5D questionnaire. Preference scores have not yet been developed for the QUALIOST and, thus, cost-utility analyses are difficult to perform. The purpose of the present paper is to describe the first steps of a future study on mapping from the QUALIOST to the EQ-5D questionnaire. Methods: A questionnaire, containing both EQ-5D-5L and QUALIOST, is distributed to patients in an orthopaedic outpatient clinic in Denmark and the aim is to include 150 patients. Patients above 50 years of age and with a recent fracture (less than 2 weeks old) are invited to participate in the study. A575 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 The patients are asked to complete the questionnaire at initial contact with the outpatient clinic and at four follow-up time points; one week, and one, three, and twelve months post-fracture). The last inclusion date will be December 31st2014. Results: The hypothesis is that the QUALIOST will provide a better estimate of the impact of osteoporosis-related fractures on HRQoL as it contains disease-relevant aspects, which may not be sufficiently covered in the generic EQ-5D-5L questionnaire. We will develop a mapping algorithm to predict EQ-5D derived utilities for Danish fracture patients from the QUALIOST, which can be used in future studies, where utilities then may be estimated from QUALIOST results. Conclusions: Mapping from QUALIOST scores to EQ-5D-5L derived utilities will enable estimation of preference-based HRQoL utilities for patients with osteoporotic fractures. PRM179 Review of Patient-Reported Outcomes in Diabetic Macular Edema Ishii K , Narimatsu A , Adachi K Bayer Yakuhin, Ltd., Tokyo, Japan . . . Objectives: Diabetic macular edema (DME) is a serious condition occurring in patients with diabetic retinopathy (DR) which is a common complication of diabetes. Understanding not only objective measurements such as visual acuity, but also the patients’ perspective is important in gaining a comprehensive understanding of the impact of DME and its treatment on their functioning and well-being. The aim of this study is to understand possible measurements to investigate the impact of DME on health-related quality of life (HRQoL), by identifying currently available measurements, their ability to differentiate between DR and DME, and possible reasons for difficulties in measuring HRQoL specifically for DME. Methods: We performed a literature review on articles describing instruments of patient-reported outcomes (PRO) for DME. We summarized the current evidence on the usefulness of the instruments and whether they differentiate between DR and DME. In addition, we investigated possible hurdles in measuring HRQoL for DME, given the clinical understandings of the systemic aspects of DME. Results: We identified seven condition-specific or vision-specific measurements and seven general measurements including generic HRQoL and utilities for possible instruments for DME. However, realistic use for DME appears questionable even for vision-related measurements. The possible reasons could be that: 1) DME is not the primary disease, 2) many of DME patients have comorbidities such as DR, cataracts, and DR-induced glaucoma, and 3) those comorbidities itself may lead to decreased visual acuity with decreased HRQoL. Conclusions: Evidence is limited when measuring the impacts of DME on diabetic patients largely due to the lack of an effective PRO instrument for DME. The major difficulties in developing such instruments is that DME patients have complex health statuses, which provides for multiple reasons in decrease of QoL which may not be directly due to their DME. PRM180 Development of the Behavior Rating Inventory of Executive Function - Preschool Version (Brief-P) in 10 Languages Vasarri S 1, Isquith P K 2 . . . 1Mapi, Lyon, France, 2Dartmouth University, Norwich, VT, USA Objectives: The Behavior Rating Inventory of Executive Function Preschool Version (BRIEF-P) was designed to provide a better understanding of preschool children’s selfcontrol and problem-solving skills. It is composed of 63 items organized in five clinical scales which measure five aspects of executive functioning (Inhibit, Shift, Emotional Control, Working Memory, Plan/Organize). It is used to evaluate children aged 2-5 years old with a wide spectrum of developmental and acquired neurological conditions, such as learning disabilities, Tourette’s disorder, traumatic brain injury, attention-deficit/ hyperactivity disorder or pervasive developmental disorders/autism. The objective of this study was to develop the BRIEF-P in 10 languages (Afrikaans, Bulgarian, Estonian, Greek, Japanese, Lithuanian, Romanian, Serbian, Slovak, Ukrainian). Methods: The following methodology was used: (1) Clarification of concepts with the developers; (2) Forward and backward translation steps in each target country. Results: The translation process did not reveal any cultural issues since most of the concepts assessed were cross-culturally relevant. The main difficulties consisted in finding conceptual equivalents of the original items with strong idiomatic content or containing words pertaining to the same semantic field. For instance, the most problematic items were items 10, 23 and 50. Item 23 (Is fidgety, restless, or squirmy) raised difficulties since it was not always possible to find direct equivalents for all adjectives in all languages. As for item 50 [Acts overwhelmed or overstimulated in crowded, busy situations (such as lots of noise, activity, or people)], discussions focused on how to convey the differences between “overwhelmed” and “overstimulated”, as well as between “busy” and “crowded.” In all cases, issues were solved with repeated discussions with the developers. Conclusions: The cross-cultural adaptation of the BRIEF-P into 10 languages required an international collaboration. The involvement of the developers during the whole process enabled the production of conceptually equivalent and culturally appropriate measures. PRM181 Challenges In Recruiting Patients For The Linguistic Validation Of Pro Instruments Developed For Rare Diseases: A Case Study With Alagille Syndrome Kennedy C 1, Abetz-Webb L 2, Lambe J 3 1Lumena, San Diego, CA, USA, 2Patient-Centred Outcomes Assessments LTD, Macclesfield, Cheshire, UK, 3Mapi, Lyon, France . . . Objectives: Observer and Patient Itch-reported Outcome Instruments (ItchRO) are electronic morning and evening diaries developed in US English to assess itchrelated symptom severity (rubbing, scratching, skin damage, sleep disturbances or irritability) in pediatric patients suffering from rare cholestatic liver diseases such as Alagille Syndrome (ALGS). Itching is a key symptom in ALGS and significantly impacts on a child’s daily life. This research presents the recruitment challenges faced during the linguistic validation of the ItchRO into French, German, Polish and Spanish in ALGS patients. Methods: Native translators performed two forward and two back translations of the English ItchRO. Cognitive interviews with five children with ALGS (patient report) and five parents (observer report) were to be performed. Key inclusion criteria were: children aged 5-9 years with cholestatic liver disease (ALGS) or progressive familial intrahepatic cholestasis (PFIC), who were also experiencing itching. Results: Recruitment was challenging and threatened to delay the clinical trial programme. Thus, recruitment was extended to include children who were 3-17 years of age, and in Spain and Germany, patients with alternative pathologies that manifested primarily with generalized pruritus were included. The final sample included: 15 with ALGS (France n= 5; Poland n= 5; Germany n= 3; Spain n= 2), 2 with PFIC (Spain), 2 with atopic dermatitis (Germany) and 1 with contact dermatitis (Spain). The ItchRO translations were well understood; there were no apparent differences in translation interpretations by disease state. Conclusions: Recruitment in linguistic validation projects in rare diseases can be a hurdle, yet patient understanding, rather than those of ‘lay people, ’ is preferable in paediatrics to ensure adequate interpretation, which can vary substantially across children. As a result patients with similar symptom and impact characteristics could be recruited if recruiting the primary population is likely to delay clinical trial research in rare diseases, where there is an urgent need for treatment. PRM182 Responsiveness and Minimal Clinically Important Difference of A Specific Health Related Quality of Life (Hrqol) Questionnaire for Home Enteral Nutrition (Hen) Patients: Nutriqol® Questionnaire Cuerda C 1, Virgili N 2, Irles J A 3, Cuesta F 4, Apezetxea A 5, Casanueva F 6, Carrillo L 7, Layola M 8, Lizán L 9 1Hospital Universitario Gregorio Marañón, Madrid, Spain, 2Hospital Bellvitge, Barcelona, Spain, 3Hospital Universitario Nuestra Señora de Valme, Sevilla, Spain, 4Hospital San Carlos, Madrid, Spain, 5Hospital Basurto, Bilbao, Spain, 6Hospital Universitario Santiago de Compostela, Santiago de Compostela, Spain, 7Centro de Salud Victoria de Acentejo, Santa Cruz de Tenerife, Spain, 8Nestle health science, Barcelona, Spain, 9Outcomes’10, Castellon, Spain . . . . . . . . . . Objectives: To determine NutriQoL® responsiveness and minimal clinically important difference (MCID) in patients receiving HEN. Methods: NutriQoL®, a specific questionnaire, developed and validated in Spain, for the assessment of HRQoL in patients receiving HEN regardless of the underlying condition was administered to a prospective cohort from 4 Spanish hospitals. It includes 17 pairs of items of HEN-related HRQoL, scoring from -51 (worst HRQoL) to 51 (best HRQoL). NutriQoL® was completed three times within 1-month (±15 days) intervals (visit 1/visit 2/visit 3). Responsiveness was assessed by estimating the effect size and the standardized response mean between visits 1 and 3. For MCID calculation an anchor-based approach was performed. Interquartile range of the change in NutriQoL® from patients, who reported changes in their health-status (worse or better) between visits 2 and 3, was used. Results: A total of 86 subjects who presented clinical changes between visits [63% male; mean (SD) age 61 (13)] participated. Cancer was the main diagnosis leading to HEN prescription (66.3%). NutriQoL® scores were 16.98 (14.57), 16.63 (14.86) and 18.92 (15.25) for visit 1, 2 and 3, respectively. Up to 78.79% of patients reported improvements in their HRQoL since the introduction of HEN. The estimated effect size and the standardized mean response between visit 1 and 3 were 0.23 and 0.24. Regarding MCID, NutriQoL® total scores difference between visit 2 and 3 in those patients that perceived changes in their health status was between -3.75 and 4.25 (interquartile range). Conclusions: NutriQoL® responsiveness is moderate. This may be due to the limited temporal period between visits since most of patients referred improvements in their condition. A difference of ±4 points on NutriQoL® total score regarding a previous administration demonstrates a clinical change that affects patients HRQoL. PRM183 Measuring Change in Quality of Life: Can We Distinguish Recall Bias and Scale Recalibration? Blome C , Augustin M University Medical Center Hamburg-Eppendorf, Hamburg, Germany . . Objectives: Treatment effects on health-related quality of life (QoL) often differ depending on whether they are measured prospectively (before and after treatment) or retrospectively (after treatment only). It is not clear which of either evaluations is more valid: Prospective evaluations may be biased by scale recalibration (a changed understanding of the response scale), and retrospective evaluations may be biased by recall bias (a wrong assessment of former QoL). Methods: Based on an analysis of literature, we present an overview on (a) possible biases in prospective and retrospective measurement of QoL, (b) how these biases are named and defined in literature, and (c) current approaches to distinguish scale recalibration and recall bias. Results: The definitions of different biases are inconsistent. Many authors do not clearly distinguish measurement bias from true change. Furthermore, some consider only scale recalibration or only recall bias. There are different approaches for distinguishing scale recalibration and recall bias. We argue that these make too extensive assumptions to be valid. Conclusions: Much of the current discussion on the validity of prospective and retrospective QoL measurement suffers from unclear definitions, especially of “response shift” and “recall bias”, or from neglecting one of the possible biases. We suggest more elaborate definitions for different types of bias, and recommend taking both kinds of bias into consideration when measuring change in QoL. Due to a lack of valid methods, there is not enough evidence on the extent of these biases yet; therefore the best approach for outcomes studies might be to include both prospective and retrospective assessments. In the long run, valid methods need to be developed to determine the most valid method of QoL assessment. PRM184 Methods To Elicit Patient Preferences: A Case Study In Metastatic Breast Cancer Copher R 1, DiBonaventura M 2, Basurto E 2, Faria C 1, Lorenzo R 2 1Eisai, Inc., Woodcliff Lake, NJ, USA, 2Kantar Health, New York, NY, USA . . . . . A576 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Patient preferences have implications for treatment decision making, treatment adherence and follow-up care. This study aimed to highlight, using metastatic breast cancer (mBC) as an example, a method to elicit preferences and, of particular novelty, examine individual differences of those preferences. Methods: Using mixed methods, a qualitative study (n= 10) of patients with mBC informed the development of the preference survey (a cross-sectional Internet survey administered to women with mBC). Survey participants (N= 181) completed a conjoint exercise that included a series of choice questions. Each choice question included a pair of hypothetical treatments that were presented in terms of eight safety attributes, single attributes for effectiveness, dosing regimen, and quality of life. Survey choice data were analyzed using hierarchical Bayesian logistic regression models. Predicted values from this model were then analyzed to understand individual differences in patient preference. Results: Qualitative interviews identified the most relevant side effects to include in the choice task (e.g., alopecia, nausea/vomiting, etc.) and reinforced the importance of quality of life when making treatment decisions. In the survey data, treatment effectiveness was most strongly associated with treatment preference, followed by alopecia, fatigue, neutropenia, and quality of life. Predicted values from the choice model enabled preference comparisons across treatment experience subgroups (e.g., 6+ rounds of chemotherapy vs. less). Preference strength for individual attributes, e.g., side effects was correlated with various demographic and health history variables, though only modest associations were detected (Pearson rs< 0.25). Conclusions: Understanding patient preferences provides opportunities for improved care and outcomes. Combining qualitative and quantitative methods in this study allowed for specificity of preferences and generalizability (albeit limited). Patient preferences derived across the sample informed predicted values from the choice models that can also be used for comparing preferences across subsamples and identifying factors that may be associated with certain preferences. RESEARCH on Methods – Statistical Methods PRM186 Real-World VersUS Randomised Controlled Trial Data: A Case Study On The Cost-Effectiveness Of Laparoscopic Surgery For Chronic Reflux Faria R , Liu S , Epstein D , Manca A University of York, York, UK . . . . Objectives: Real-world (observational) data has the potential to address the limitations of randomised controlled trials (RCTs) but presents its own challenges given the increased risk of bias. We compared the costs and quality-adjusted life years (QALYs) of patients following random vs preference-based allocation and assessed the performance of different methods to address these challenges. Methods: The REFLUX study was a pragmatic trial in patients with chronic reflux comparing laparoscopic fundoplication (surgery) with medical management (MM) over 5 years of follow-up. The trial included randomised and non-randomised preference-based allocation. We compared the cost-effectiveness of surgery in the RCT vs preference cohorts as unadjusted raw differences, applying methods to handle biases from selection and confounding (regression adjustment, propensity score matching and instrument variable analysis) and explored the impact of receiving the preferred treatment on the results. Results: The preference surgery group accrued greater costs and QALYs than the randomised surgery group (£3,524 vs £2,852; 3.723 vs 3.612 QALYs). The preference MM group had lower costs but slightly better QALYs than the randomised MM group (£861 vs £1,415; 3.541 vs 3.411 QALYs). The incremental cost-effectiveness ratio (ICER) for the preference cohorts was similar to that obtained in the RCT using the different methods at around £8,000 per QALY gained (vs £7,149 in the RCT and £14,632 unadjusted raw differences). Receiving the preferred treatment was significantly associated with lower costs and better QALYs after adjusting for prognostic variables. Conclusions: Real-word data can be used in cost-effectiveness analysis to complement RCT evidence. However, more research is needed on how to choose the most appropriate method to adjust for selection bias and how to account for patient preferences when making recommendations on value for money. PRM187 Integrating Health Psychometrics With Health Economics: Can the ‘Mapping’ Toolbox Be Extended Using Ordinal Structural Equation Models? Patton T , Manca A University of York, York, UK . . Objectives: Mapping enables the prediction of health-state utility values via health outcome measures in trial data using algorithms linking those measures available to preference-based measures (PBMs). However, the unusual distributional features of PBMs mean that there is no consensus around the most appropriate statistical methodology for obtaining mapping algorithms. Existing studies have shown that structural equation modelling (SEM) developments open up a range of opportunities for effectively analysing PBMs. This study draws upon some of the methodological advances around SEMs from other fields in a case study. Progress towards developing a mapping algorithm for prediction of EQ-5D scores using a disease-specific measure, the Seattle Angina Questionnaire (SAQ), is discussed. Methods: EQ-5D and SAQ data were obtained from a completed RCT, the RITA-3 trial. Several psychometric analysis approaches were considered to decide which subscales and items of the SAQ were likely to be relevant to the model. Several SEMs were specified and the first looked at the EQ-5D index values and SAQ subscales as continuous responses in a linear regression model. External information about the reliability of the diseasespecific measure was incorporated into this model. The subsequent models looked at the specification of outcomes at the item level with an ordinal logistic approach. Models were compared in terms of their ability to fit the observed data and their predictive performance. Results: Only three of the five subscales of the SAQ were selected for the development of mapping algorithms. The linear regression model had best model fit. However, the SEM that incorporated item level data for ordinal responses to each instrument showed better predictive performance, especially at the lower end of the EQ-5D distribution. Conclusions: This study demonstrates the flexibility of SEMs for the development of mapping algorithms. The models developed showed comparable predictive performance to existing models in the published literature. PRM188 Evaluation of Methods for the Inclusion of Real World Evidence in Network Meta-Analysis – A Case Study in Multiple Sclerosis Jenkins D 1, Czachorowski M 2, Bujkiewicz S 1, Dequen P 1, Jonsson P 2, Abrams K R 1 1University of Leicester, Leicester, UK, 2National Institute for Health and Care Excellence (NICE), Manchester, UK . . . . . . . Objectives: Network Meta-Analysis (NMA) is becoming a key component of submissions to reimbursement agencies world-wide, especially when there is limited head-to-head evidence for multiple technologies. However, almost all NMAs only consider Randomised Controlled Trials (RCTs) even though there may be considerable Real World Evidence (RWE) available – for example observational studies or registry-based studies. Evaluation of methods to enable the inclusion of RWE, especially in the light of the changing nature of RCTs from both a regulatory and reimbursement perspective, is considered here. Methods: RCTs and RWE studies were searched for using standard filters and databases up to, and including, the regulatory approval of Fingolimod by the European Medicines Agency (EMA) in 2011 for Multiple Sclerosis (MS). A number of NMAs were then conducted and which included; only RCTs, both RCTs and RWE (accepted at face-value), both RCTs and RWE but including an additional level in the NMA hierarchical model to represent the different study designs, and finally both RCTs and RWE but adjusting the RWE for potential biases. Results: Identification of RWE in addition to RCTs in this MS example significantly increased the number of studies (and comparisons) that were potentially included in the NMA. Whilst the inclusion of the additional RWE led to a reduction in the level of uncertainty surrounding most effect estimates, this depended on the method of inclusion adopted for the RWE, and the extent to which biases were adjusted for. Conclusions: This initial evaluation of methods for the inclusion of RWE in NMAs indicates that methods of adjustment for the potential biases in RWE can have a significant impact on the level of uncertainty. Consequently further work investigating both empirical evidence for such biases and methods of elicitation from experts on the extent of biases associated with individual RWE studies is warranted. PRM189 Frequentist Approach for Detecting Heterogeneity in MetaAnalysis Pair-Wise Comparisons: Enhanced Q-Test Use By Using I2 and H2 Statistics Laliman V 1, Roïz J 2 1Ensai, Bruz, France, 2Creativ-Ceutical, London, UK . . Objectives: In meta-analysis, model selection is an important criterion which needs to be tested and validated by strong statistical evidence. The Cohran’s Q-test allows in theory to decide between random-effect and fixed-effect models. Due to the highly conservative nature of this test, three statistics have been built to estimate the heterogeneity between studies to lead the model decision: the I2, the H2 and the R2. We conducted a review of the Q-test utility in diverse scenarios with a comparison of three different methods to estimate the heterogeneity between studies. Methods: Based on the global formulation of the Cochran’s Q-test, we proposed to analyse jointly the first error species and the second error species in different scenarios based on the number of studies included in each meta-analysis. The goal was to determine the reliability of the Q-test in extreme situations but also to give some benchmark for the reliability of this test. We use simulation methods to analyse the three different methods for calculating the between-study variance compared to the real value of heterogeneity. We also compared different arbitrary levels for model selection using these statistics in different scenarios. Results: The Cochran’s Q-test is too conservative with a large number of studies and concludes to the presence of heterogeneity whatever the situation is when the number of studies is higher than 18. In comparison, the different statistics have an average value conversely linked with the number of studies in case of non-heterogeneity: the higher the number of studies, the lower the statistics’ average values. Conclusions: The I2 and H2statistics can eventually enhance the use of Cochran’s Q-test by solving conservative issue associated with this test. The model selection can, eventually, be led by benchmark of these statistics jointly with the Cochran’s Q-test. PRM190 The Use and Acceptance of Novel Statistical Analyses To Support Technology Submissions To Hta Authorities Batson S , Mitchell S A , King D Abacus International, Bicester, UK . . . . Objectives: Indirect comparisons are increasingly accepted to model the clinical- and cost-effectiveness of treatments. The purpose of this study was to (i) assess the literature reporting on the use of novel statistical methods [simulated treatment comparison (STC), and matching-adjusted indirect comparison (MAIC)]; and (ii) assess technology appraisals (TAs) submitted to the National Institute of Health and Carel Excellence (NICE) to determine whether these techniques have been accepted by reimbursement authorities. Methods: Embase, Medline and the Cochrane Library were interrogated to identify publications reporting on the use of MAIC or STC. NICE TAs published from 2011-2014 which reported MAIC or STC analyses were identified and the critique by the appraisal group was summarised. Results: Six publications reported on the use of MAIC in six indications. Results from these analyses concluded that MAIC offered several advantages over conventional meta-analysis methods that rely on aggregate data. Findings from the review of NICE TAs indicated that these novel statistical techniques have not been A577 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 widely used in manufacturers’ submissions to date. Of the most recent 60 NICE TAs, analyses employing MAIC methodology have been presented in two submissions and a STC analysis in a single HTA, all in the oncology setting. In all cases the review group identified limitations with the statistical methodology presented, although their use as exploratory analyses supporting results from conventional meta-analyses was highlighted in one submission. In particular the use of nonrandomised data from single treatment arms was highlighted as a potential weakness of STC. Conclusions: In spite of the increasing published evidence base reporting on MAIC in a range of indications, both MAIC and STC have not been widely used in manufacturer’s submission to NICE. Assessment bodies critiquing the technology submissions remain to be convinced of the appropriateness of these novel techniques for the robust assessment of relative efficacy. PRM191 Meta-Analysis in Open Bugs: How To Assess the Convergence of Mcmc Chain? Laliman V 1, Roïz J 2 1Ensai, Bruz, France, 2Creativ-Ceutical, London, UK . . Objectives: Meta-analysis is often conducted in OpenBUGS. This software, like all BUGS projects, is based on MCMC simulations by using Gibbs sampling. One of the main issues in the use of Markov chains in a continuous space is the chain convergence. If the chain does not converge, transient states will be accounted for in our posterior distributions. Since these states are not bound to the empirical data but only with the chain’s starting point, the estimated parameters of the posterior distribution will be biased. To help assessing the convergence of MCMC chain, several methods exist. Methods: Based on the literature, we run several simulation scenarios in order to test built-in OpenBUGS graphical methods and to assess the power of the “thin” approach, a fixed-step jumping-data method, for convergence. Then, we focus on the existing diagnoses, their supplementary assumptions and their associated computation costs. To help perform these diagnoses directly on BUGS objects, we present the R-package coda. Results: The use of jumping-data method leads to loss of power and a poorer estimation of posterior distribution even in case of high autocorrelation. Consequently, the use of the thin method is not recommended to obtain a quicker convergence and better posterior distribution estimation. We have also seen that although autocorrelogram and trace can be useful to assess convergence, they can lead to misinterpretation in case of extremely low number of studies and conclude to convergence. Alternatively, using the Geweke diagnosis seems, in terms of computation cost and assumptions, recommended for two main advantages: it gives a measure of trust of being in a stationary process and very low computation cost. Conclusions: We presented methods to assess convergence of MCMC chains and argued on their pros and cons. The Geweke diagnose was found to provide best trade-off between computational cost and interpretability. PRM192 Adjusting for Treatment Crossover in A Trametinib Metastatic Melanoma Rct: Identifying the Appropriate Method Bell H 1, Latimer N 1, Amonkar M 2, Casey M 2 of Sheffield, Sheffield, UK, 2GlaxoSmithKline, Collegeville, PA, USA . . . . 1University Objectives: Treatment crossover refers to the situation in randomised controlled trials (RCTs) where patients randomised to the control group switch onto the experimental treatment. This leads to biased estimates of treatment effects if not appropriately controlled for. Several crossover adjustment methods are available, but previous research has shown that the optimal adjustment method depends upon the characteristics of the trial. This study applies crossover adjustment methods to an RCT comparing trametinib to chemotherapy in patients with BRAF V600E/K mutation-positive advanced or metastatic melanoma (NCT01245062), and investigates which adjustment method best fits this case study. Methods: The crossover adjustment methods applied include the Rank Preserving Structural Failure Time Model (RPSFTM), Iterative Parameter Estimation (IPE) algorithm, Inverse Probability of Censoring Weights (IPCW) and a two-stage accelerated failure time model estimation procedure. Suitability of each method is compared by assessing the plausibility of the underlying assumptions of the models in this case study and analysing output and performance indicators associated with each method. Results: In the primary efficacy population (patients without history of brain metastases) 67.4% of chemotherapy patients switched onto trametinib. The intention to treat (ITT) hazard ratio (HR) for overall survival (OS) was 0.72 (95% CI 0.52-1.01). Point-estimates of the adjusted HRs produced by the most plausible applications of the RPSFTM, IPE, IPCW and two-stage methods ranged between 0.43 and 0.49, consistently favouring trametinib. Results were sensitive to the technique used to apply each method. Key issues included recensoring, the active nature of the comparator, and the choice of covariates included in the analyses. Conclusions: Each of the crossover adjustment methods result in a lower HR than the ITT analysis. However, results are uncertain and sensitive to key assumptions. It is important to carefully analyse trial characteristics and model output when identifying which applications of adjustment methods are most plausible. PRM193 Assessing Balance in Baseline Characteristics Using Different Propensity Adjusted Methods for Bipolar I Mixed Disorder Patients Initiating Asenapine Versus Other Oral Atypical Antipsychotics Rouleau A 1, Guiraud-Diawara A 2, Landsman-Blumberg P 3, Lokhandwala T 3, StafkeyMailey D 3, Verpillat P 1 1Lundbeck SAS, Global Epidemiology, Issy-les-Moulineaux, France, 2Lundbeck SAS, Global Analytics, Issy les Moulineaux, France, 3Xcenda, Palm Harbor, FL, USA . . . . . . Objectives: Asenapine (ASE), an oral Atypical Antipsychotic (AA), was initially used for more severe bipolar I mixed disorder. Different propensity score (PS) methods were investigated to achieve balanced baseline characteristics between ASE and four oral AA cohorts for eventual outcomes analyses. Methods: Adults with ≥ 1 asenapine, aripiprazole, olanzapine, quetiapine, or risperidone prescription fill (Aug 2009 to Dec 2010) and diagnosis of bipolar I mixed disorder (ICD-9-CM: 296.6x) from MarketScan® claims databases, yielded 230 ASE, 2726 aripiprazole, 984 olanzapine, 3056 quetiapine, and 1623 risperidone patients. PS were derived using logistic regression models for ASE and each AA with baseline demographic and clinical characteristics as covariates. PS, inverse probability treatment weight (IPTW: 1/ PS ASE; 1/ (1-PS) AA), and standard mortality ratio weight (SMR: 1 ASE; PS/ (1-PS) AA) distributions were evaluated. ASE-AA un-weighted, IPTW, and SMR baseline characteristics were compared using standardized differences, chi-squares, and t-tests. Results: Un-weighted asenapine patients had pre-index greater bipolar I episodes rates, psychiatric drug use, dyslipidemia and obesity (all comparators). PS distributions for asenapine-olanzapine overlapped to some degree while PS of asenapine and the other comparators overlapped little to not at all. Comparing IPTW baseline characteristics, asenapine more resembled the AA cohorts. Demographic imbalance increased between asenapine and each AA. IPTW improved clinical characteristic balance for asenapine versus olanzapine and risperidone, but only slightly improved imbalance versus aripiprazole and quetiapine. However some clinical characteristics not previously balanced in the un-weighted analyses for asenapine versus each AA were now imbalanced. Applying SMR, AA cohorts more resembled the asenapine cohort and all baseline demographic and clinical characteristics were finally balanced. Conclusions: SMR, a less common PS method, resulted in balanced baseline characteristics. SMR should be considered when IPTW leaves imbalance and the cohort of primary interest differs significantly from the broader underlying population to which it’s being compared. PRM194 Task-Based Versus Case-Based Analysis of Time Outcomes in Multi-Country Time and Motion (T&M) Studies: Methodological Considerations and Application De Cock E 1, Kritikou P 2, Tao S 3, Ngoh C A 4 1United BioSource Corporation, Barcelona, Spain, 2United BioSource Corporation, London, UK, 3UBC: An Express Scripts Company, Dorval, QC, Canada, 4F. Hoffmann-La Roche Ltd, Basel, Switzerland . . . . . Objectives: A challenge in multi-centre Time and Motion (T&M) studies is performing inferential statistics, in light of hierarchical data. Our objective was to investigate two approaches to analyze the data. Methods: Task-based approach analysed tasks independently, mean times were summed, and 95% confidence intervals (CIs) were computed based on Variance Sum Law I (assuming time is independent among all tasks). Case-based approach involved imputation for missing time; all tasks per observation were summed, and a single time variable was analysed. Both approaches were applied to three countries participating in a multi-country T&M study comparing intravenous [IV] and subcutaneous [SC] administration processes. Absolute and relative differences in country means (case-based minus task-based) and the difference in CI range were computed using a random intercept model, to account for centre clustering. Results: Mean times were similar for both approaches. For IV process, absolute (relative) differences in time were -0.03min (-0.1%) in France, -0.77min (-2.3%) in Italy, and -0.07min (-0.3%) in Russia. For SC process, results were 0.30min (2.1%) in France, 0.90min (4.5%) in Italy, and 0.01min (0.1%) in Russia. The differences in CI range between both approaches were noticeable: 0.51min (5%) in France, 25.04min (57%) in Italy, and 10.06min (46%) in Russia for IV and 4.38min (40%) in France, 0.88min (5%) in Italy, and 8.19min (56%) in Russia for SC. Conclusions: The choice of task-based or case-based approach did not impact mean process time; however, since task-based approach assumed independence of task times, it resulted in much narrower CI range. On the other hand, case-based approach eliminates the underestimation of variations, thus may therefore be a more optimal choice to analyse time outcomes for complex processes. With only a single time variable being analysed, it also allows pooling of data across countries, therefore providing more power to generate reliable CIs. PRM195 Graphical Interactive Meta-Analysis Module for Facilitating Evidence-Based Decision Making in Health Care Spata E 1, Bujkiewicz S 2 1Leicester University, Leicester, UK, 2University of Leicester, Leicester, UK . . Objectives: In health technology assessment (HTA) decisions about reimbursement of new health technologies are largely based on effectiveness estimates obtained from pre-prepared meta-analysis of evidence from randomised controlled trials. However, there is not always a consensus amongst the decisionmakers about the inclusion criteria of studies into the meta-analysis. Therefore an approach that allows stakeholders to manipulate the content of the meta-analysis, thus facilitating a critical sensitivity analysis in real time during the decisionmaking process, would be valuable from the point of view of the transparency of the HTA submissions. A Graphical-User-Interface (GUI) was designed to facilitate such a transparent decision-making process. Methods: The GUI was designed using freely available software packages which included WinBUGS for development of meta-analysis and meta-regression models and R which was used to design GUI, to link data with statistical models in WinBUGS and to extract the results. R was also used to develop graphical tools for presentation of results (forest and bubble plots) and for visual assessment of publication bias (funnel plots). Software was designed for an illustrative example in rheumatoid arthritis where effectiveness of TNF-alpha inhibitors was measured on different scales (DAS28, HAQ, ACR, and EULAR). Results: R-based Transparent Interactive Decision Interrogator (R-TIDI) was developed, which is a user-friendly tool with “point and click” options that allows users to choose an outcome measure and run randomeffects or fixed-effects meta-analysis and meta-regression models. Users are not required to have knowledge of statistical software or programming skills since the use of WinBUGS and R is entirely “behind the scenes”. R-TIDI enables users to interactively include/exclude studies from the meta-analysis allowing for conducting sensitivity analyses in real time. Conclusions: R-TIDI is a useful tool for non-statistical decision-makers. It allows users to run sensitivity meta-analyses A578 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 in a user-friendly environment during a decision-making setting avoiding disadvantages of pre-prepared analyses. PRM196 An Evaluation and Comparison of Methods Used in Survival Analysis To Fit Distributional Curves To Kaplan-Meier Data Haines P Curo, Marlow, UK and we show for different scenarios how close the estimates come. For selected treatment comparisons we present effect-sizes with confidence intervals. We apply also the Bayesian extension and discuss its advantages. Conclusions: Based on our research we give recommendations of when the Lumley-method should be best applied, and discuss limitations.1 Lumley T: Network meta-analysis for indirect treatment comparisons. Stat. Med. 2002; 21: 2313-2324. . Objectives: HTA bodies increasingly require accurate survival estimates in order to provide reliable recommendations. It is argued that access to individual patient data (IPD) can improve their accuracy. This paper aims to assess to what degree extracting IPD from published Kaplan-Meier curves helps improve extrapolated survival estimates. Some methods currently used for HTA submissions fit a survival curve directly to a published Kaplan-Meier curve, but does this lack accuracy? Methods: Two methods used to extract the IPD from Kaplan-Meier curves reviewed in this paper are by Guyot et al (Guyot, Ades, Ouwens, & Welton, 2012) and Hoyle and Henley (Hoyle & Henley, 2011) which were compared against the outcomes of the standard ‘least squares’ method. Comparisons were made for two situations: 1) when numbers at risk are available at different time points throughout the KaplanMeier curve and 2) when numbers at risk are only available at the start. Results: The three methods resulted in the long-normal distribution showing the best fit, with all containing the true mean and median within their confidence intervals. However, the Hoyle and Henley method estimates a mean marginally closer to the true mean than the other methods in both situations. When many numbers at risk are provided, the Hoyle and Henley method gives narrower confidence intervals. Both extraction methods slightly outperformed the least squares method. The three methods give median estimates and resulting confidence intervals which are statistically equivalent to that of the IPD, except for the Guyot method when numbers at risk are not available. Conclusions: In conclusion, extraction methods can give marginally better results than the Least Squares method. However, these results may not be applicable to other examples. In addition, the extra time taken to run extraction methods could be too large to account for the small improvement in accuracy of results. PRM197 Multi-Level Network Meta-Analysis To Account for Dose-Response and Class Effects Reason T 1, Dias S 2, Welton N 2 Health, London, UK, 2University of Bristol, Bristol, UK . . Iakovlieva L , Kyrychenko O , Gerasymovà O , Kuznetsov I National University of Pharmacy, Kharkiv, Ukraine . . . . Objectives: Evaluation and comparison of oral antidiabetic drugs (OAD) consumption at state level are an important element at control of 2ndtype DM patients treatment quality. The objective of the study is to determine the volumes and structure of OAD consumption in Ukraine and to compare it with consumption in the other countries. Methods: ATC/DDD-methodology with application of DDD/1000/day (DID). The evaluation is based on consumption volume, provided by “PharmXplorer/ Pharmstandard” analytical system of market research. Results: In year 2008, OAD consumption in Ukraine was 5.78 DID and increased to 11.13 DID in year 2013. In year 2011, OAD consumption was 54.28 DID in France, 44.58 - in Germany (S. Pichetti, C. Sermet, S. van der Erf, 2013), 33.25 - in Estonia, 29.87 - in Latvia (Baltic Statistics on Medicines 2010–2012), showing that OAD consumption in Ukraine was very low. Structure of OAD consumption in Ukraine shows that 98.95% of the total consumption volume is distributed to 2 groups: sulfonylureas (73.84%) and biguanides (25%) and only 1,05% to gliptines, glitazones, glucosidases and glinides. The total share of preparations of sulfonylureas and biguanides group in the total consumption structure in France and Germany was 71.8% and 80.1% respectively of the total consumption OAD. In year 2013, OAD of the II generation - gliclazide (3.01 DID) and glibenclamide (2,09 DID) had the highest consumption level in sulfonylureas group, preparations of III generation - glimepiride (2,09 DID) had lesser consumption rate. Out of 61 OAD trade names (TN), presented in the pharmaceutical market of Ukraine, 8 TNs took 89.78% of the total consumption volume. Conclusions: Very low rate of OAD consumption in Ukraine shows the necessity of its increase. Analysis of OAD consumption structure evidences the application of relatively cheap and long-used medical preparations for treatment of 2nd type DM, which is largely due to financial capacities of payers. . 1IMS Objectives: A frequent challenge in Network Meta-Analysis (NMA) arises from the fact that several interventions may belong to the same class and be given at multiple doses. Models have been proposed for NMA accounting for dose-response, but these models do not also consider class effects at the same time, which are important from a decision making perspective. We aim to develop a framework that extends dose-response NMA methods to account for dose and class effects simultaneously, and explore the ability of these models to explain heterogeneity, improve model fit and increase precision of the estimated treatment effects. Methods: Using clinical trial data of treatments for acute migraine obtained from Cochrane reviews, we developed multi-level NMA models to simultaneously account for dose-response and class-effects, in particular defining a ‘dose’, ‘treatment’ and ‘class’ hierarchy within the NMA models. We explored a non-parametric “random walk” model constrained to be monotonically increasing with dose. Multi-level NMA models were compared to 1-level (standard) NMA models where interventions were ‘lumped’ at each level separately. Results: The model that explicitly included monotonic doseresponse and class effects showed the best fit and least heterogeneity, and produced more precise measures of treatment effect than all 1-level models. NMA models that made less plausible assumptions around dose-response had poorer fit than models with monotonic dose-response. Conclusions: We have developed a framework for simultaneously estimating treatment effects at the ‘dose’, ‘treatment’ and class level within the same NMA model. The framework can help decision makers identify the most appropriate class, drug, and dose, however, results of dose-response models are not straightforward to interpret or implement from a decision making perspective. Careful consideration should be given to dose-response and similarity of interventions when conducting NMA. PRM198 The Lumley-Method, A Recommended Network Meta-Analysis for Indirect Comparisons, Summarized for Practitioners Petto H 1, Kadziola Z 1, Belger M A 2 1Eli Lilly Regional Operations GmbH, Vienna, Austria, 2Eli Lilly and Company, Windlesham, UK . PRM199 Analysis of Volume and Structure of Oral Antidiabetic Drugs Consumption in Ukraine . . . Objectives: In recent years we have seen a growth in the use of network metaanalysis as part of the evidence base for Health Technology Assessments, with the Lumley method, published in 20021being a key reference when considering both indirect and direct comparisons. Unfortunately the program-code included in the manuscript cannot easily be run, and the given examples cannot be replicated, even with corrected code. To give practitioners helpful insight into the method, we start from individual patient data of head to head trials and show how from subsequent data-aggregation the Lumley-model (a random-effects model) can be derived. Methods: We give more details than in the article of how the proposed variance function aggregates study-heterogeneities and of how effect-sizes and confidence intervals can be derived from the parameter- and variance-estimates. We discuss why dependencies coming from the network-structure should be incorporated into confidence-interval calculations and of how the model can be extended with an in the article suggested Bayesian approach for modeling the random-effects parameters. Results: We include an example of how the Lumley-method can be applied in practice. We present based on the program-example in the article a corrected R-version and a translation into SAS. For both we show how aggregated study-data should be structured and dummy-coded before running the program. The Lumley-method was applied to simulated data with known model-parameters PRM200 Development of A Web-Based Tool To Elicit the Opinion of Regionally Dispersed Health Care Professionals Responsible for Medical Device Vigilance Pibouleau L 1, Galtier T 1, Sailly A C 2, Maison P 2, Katsahian S 3 1INSERM, CRC, Paris, France, 2Agence Nationale de Sécurité des Médicaments, Saint-Denis, France, 3Hôpital Européen Georges Pompidou, Paris, France . . . . . . Objectives: In the context of uncertainty due to the lack of sound data, expert opinion is considered as a legitimate source of information for decision-makers. The use of experts’ opinion requires to quantifying their uncertainty about a specific event by eliciting a probability distribution of the event. The objectives of this study were to develop a web-based tool enabling users to remotely elicit the opinion of a group of geographically dispersed experts and to evaluate the measurement properties of this tool. Methods: The web-based tool allowed first to elicit univariate probability distributions separately from each expert and secondly to calculate an aggregated distribution. The elicitation method was the four-interval method that was judged to be more appropriate for non-statistician experts due to its clarity of use. As recommended to limit biases, the elicitation questionnaire included a training exercise and a graphical feedback so that the experts could validate their distributions. A pilot survey was conducted among all the French regional medical device vigilance correspondents (n= 24) about the risk of failure (%) of an implantable medical device. Results: Twenty-two correspondents (92%) completed the survey. An aggregated distribution was calculated from the elicited individual distributions and a beta distribution was fitted reflecting the group uncertainty about the risk of failure. Feasibility was judged in view of the users’ feedback and time to completion. Validity and reliability were assessed using data on comprehensiveness, internal coherence and test-retest reliability. Conclusions: The proposed web-based tool was feasible, valid and reliable. It should be useful in making expert elicitation easier and more practical. PRM201 Knowledge on Medication Taking Behaviour, Balanced Diet and Physical Activity - A Survey Among the Adolescents sandeep Chowdary K , Nallani V R R Chalapathi Institute of Pharmaceutical Science, Guntur, India . . . . Objectives: The main objective of the study is to promote awareness and assess the knowledge of adolescents on medication use, importance of balanced diet and physical activity. Methods: The study was conducted among the adolescents aged from16-18 years in Regions of Guntur. The volunteers are allowed to fill their informed consent to be a part of the study. The questionnaire was distributed to all the volunteers included in the study, which includes questions on their medication taking behaviour, dietary habits and physical activity. The response was then analyzed to assess the knowledge on medication use, balanced diet and physical activity. Results: Among the 165 individuals on assessment of their medication taking behavior78% of them do not follow their prescription, 61% of the individuals do not have any idea on their medication use and a majority of 74% are not aware of the unwanted effects caused by the medication. On assessment of their dietary habits and physical activity, 62% of the individuals include meal rich in fat, 42% of the individuals skip their breakfast every day and 41% of them will not include leafy vegetables as part of their regular meal. 65% of individuals do not perform a regular physical activity Conclusions: It is the responsibility of the pharmacist A579 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 to promote awareness on the medications, balanced diet and physical activity to improve the quality of life of an individual. or to refine the decision formulation if uncertainty is high, potentially leading to improved decision-making. PRM202 Simulating Individual Patient Level Data To Address Treatment Switching When Only Summary Data Are Available PRM205 Systematic Review and Critical Appraisal of the Statistical Methods Used in Published Studies To Indirectly Compare Novel Anticoagulants (Noacs) With Warfarin for the Prevention of Stroke in Patients With Atrial Fibrillation (Af) Boucher R , Abrams K , Lambert P C University of Leicester, Leicester, UK . . . . Objectives: Treatment switching commonly occurs in the pivotal HTA evidence for advanced or metastatic cancer treatments submitted to reimbursement agencies. Simple approaches, such as Intention-to-treat (ITT) analysis, have typically been used to analyse data with treatment switching, despite simulation studies showing these to drastically underestimate the underlying treatment effect. With more manufacturers conducting indirect comparisons (ICs) to compare treatments, summary data are being used more in analysis. The method outlined addresses treatment switching when only summary data are available to ensure appropriate estimates for the treatment effect are achieved when the data is then used in an IC. Methods: Using digitised survival curves, multiple datasets that are representative of the original individual patient data (IPD) are simulated. Treatment switching information is estimated from reported information on progression-free survival, and then established methods which adjust appropriately for treatment switching used to analyse the simulated data. This approach is applied to an example from a technology appraisal (TA) submitted to National Institute for Health and Care Excellence (NICE), and the ITT hazard ratio and median survival obtained and compared with those reported, before analysis using a Rank Preserving Structural Failure Time Model (RPSFTM). Results: Averaging over 2000 datasets, the replicated summary statistics were similar to those reported. Both median survival times were within 1 month of those stated in the TA and the hazard ratio less than 0.05 different. Subsequent analysis using an RPSFTM shows the new treatment to be more effective, and inappropriately adjusting for crossover to have underestimated the treatment effect. Conclusions: Adjusting summary data is important as otherwise, subsequent analysis conducted will give inappropriate results. The simulated data approach well represents the original IPD, giving on average similar results to those reported. Hence, the further analysis to address treatment switching issues gives more appropriate treatment effect estimates. PRM203 Modeling the Effect of Combining Alogliptin With Dual therapy in Type 2 Diabetes Johnson K I 1, Palin H J 1, Sibbring G C 1, Selby R 2 Clarity, Macclesfield, UK, 2Takeda UK, High Wycombe, UK . . . . . . . 1Complete Objectives: To estimate the impact of combining the dipeptidyl peptidase-4 (DPP4) inhibitor, alogliptin, with metformin and sulfonylurea (alogliptin triple therapy) to achieve glycemic control in patients with type 2 diabetes. Methods: Since no clinical trial of alogliptin triple therapy has been conducted, the effect of adding alogliptin to dual therapy (metformin+sulfonylurea) was modeled using novel additive effect methodology, utilizing data from a previous mixed treatment comparison (MTC). The following assumptions were made: the efficacy of triple therapy can be estimated as a function of its constituent parts, and the efficacies of the constituent parts are equivalent. Pooled data for the absolute change from baseline in glycosylated hemoglobin (HbA1c) from trials of sitagliptin, linagliptin, and vildagliptin triple therapy, and for their constituent parts, informed the model. A weighting factor, β coefficient, derived from DPP-4 mono, dual, and triple therapy trials, was used to estimate the effect size for triple therapy using the sum of the constituent parts. The estimated mean β value was validated against the observed effect size of alog liptin+pioglitazone+metformin, using the pooled effect from the MTC. Results: An estimated mean β coefficient value of 0.83 represented the DPP-4 inhibitor class. Validation of the approach resulted in a similar β coefficient for pioglitazone triple therapy (0.82). Absolute change in HbA1c from baseline for alogliptin triple therapy was estimated as -0.77% (95% CI -1.16, -0.39). Similar values were observed in the MTC for sitagliptin -0.94% (95% CI -7.34,5. 40), linagliptin -0.65 (95% CI -6.87,5. 60), and vildagliptin -0.80% (95% CI -7.00,5. 43). Conclusions: The wide confidence interval is consistent with expectations in the literature and is a limitation of the method employed, in that it requires the variance of the individual studies to be summated. Nevertheless, the method demonstrates the value of modeling when clinical trial evidence is not available. PRM204 Uncertainty and Probabilistic Methods in Multi-Criteria Decision Analysis O Meachair S , Walsh C Trinity College Dublin, Dublin, Ireland . . Objectives: Multi-Criteria Decision Analysis (MCDA) is a collection of techniques for choosing optimal decisions when two or more criteria need to be taken into account in the decision process. Most MCDA techniques require the specification of a number of parameters; criteria weights, utility functions or indifference thresholds. We wish to account for the uncertainty in these parameters which may arise due to the fuzzy nature of the Decision Maker’s preferences, conflicting opinions between a group of decision makers or population group, or the abstract nature of the parameters. Methods: We implement some MCDA models from a Bayesian perspective where parameters come from posterior probability distributions representing the combination of available knowledge on the parameters. Such knowledge can come from empirical data, expert elicitation, survey data, decision-making committees, or some combination of these. Results: Depending on the method used, the end result is either a benefit function which quantifies the uncertainty in the benefit score for each action, or a rankogram which depicts the uncertainty in the ranking of actions. Conclusions: Knowledge about this uncertainty allows decision makers to make more informed decisions. A decision action may be clear when uncertainty is sufficiently low, or it may be necessary to request more information Daacke I M Boehringer Ingelheim UK, Bracknell, UK . . Introduction: The three main novel anticoagulants (NOACs) currently licensed in Europe, apixaban, dabigatran and rivaroxaban, have all been directly compared against warfarin in randomised controlled trials. However, none of the three drugs have been directly compared against each other. Thus, there has been an increase in the number of meta-analyses and indirect comparisons published comparing the relative efficacy and safety of these novel anticoagulants against each other via warfarin as a common comparator. Objectives: Systematically review all meta-analyses and indirect comparisons evaluating the NOACs against warfarin for the prevention of stroke in patients with AF and critically appraise the statistical methods used to do so. Methods: Systematic searches of EMBASE, MedLine, EBM Reviews, EconLIT as well as manual searches of ClincalTrials. gov, the Cochrane Library, CADTH, NICE, NHSEED and HTA were conducted. Data was abstracted from any citation applying statistical methods to compare the efficacy and safety of NOACs for the prevention of AF-related stroke. Information regarding the statistical approach; model assumptions; data presentation; interpretation of the evidence; and discussions of internal and external validity was used to quality rate each study. Results: Bucher’s method of adjusted indirect comparison was most widely used. There were generally three main model assumptions required: the similarity, homogeneity and consistency assumptions, each being investigated with varying scrutiny in the studies reviewed. According to the quality assessment, the indirect comparison conducted by Wells and colleagues (2012) is of the highest relative quality. Conclusions: The limited number of RCTs available comparing the NOACs to standard therapy, creates considerable uncertainty surrounding the comparative efficacy and safety of these anticoagulants. In order to establish which individual NOAC is most likely to benefit a given patient population, indirect comparisons and meta-analyses are increasingly used. However, the quality of indirect comparison studies are variable and results should be interpreted with care. PRM206 Methodological Assessment of Matching-Adjusted Indirect Comparisons: Case Study Application To Attention Deficit/ Hyperactivity Disorder (Adhd) Shafrin J 1, Sikirica V 2, Shrestha A 1, Henkhaus L E 3, Erder M H 2, Chandra A 4 Health Economics, Los Angeles, CA, USA, 2Shire Development, LLC, Wayne, PA, USA, 3University of Southern California, Los Angeles, CA, USA, 4Harvard University, Cambridge, MA, USA . . . . . . . . 1Precision Objectives: Matching-adjusted indirect comparison (MAIC) is a novel comparative effectiveness approach to address biases that can appear in traditional indirect comparison (IC) methods when patient characteristics differ across trials. We examined three unanswered MAIC methodological questions and applied the proposed solutions to a comparison of ADHD treatments. Methods: Using individual patient data from two randomized controlled trials (RCTs) comparing guanfacine (GXR) vs placebo and published summary statistics from four RCTs comparing atomoxetine (ATX) vs placebo, MAIC was used to reweight the GXR data so that observable GXR patient characteristics matched those of ATX patients. Change in ADHD-RS-IV scores was the primary endpoint. Comparative efficacy results were evaluated for their sensitivity to changes in the following three MAIC specifications: variable selection using regression-based methods, statistical moments matched (i.e., mean vs mean and variance), and matching on placebo-arm outcomes. Results: Both treatments decreased ADHD-RS-IV scores relative to placebo (-17.9 GXR vs -10.7 placebo; -14.6 ATX vs -5.8 placebo). In the baseline MAIC specification adjusting for patient baseline characteristics and placebo arm outcomes, GXR produced larger decreases in ADHD-RS-IV scores than ATX (Δ: -3.9, p< 0.004). The results were insensitive to adding variables to the matching algorithm (Δ : -3.8, p< 0.023), or matching only covariate means rather than both means and variances (Δ : -3.6, p< 0.006). Applying MAIC without matching placebo arm outcomes indicated a slightly greater decrease in ADHD-RS-IV scores for ATX, but there was no statistically significant difference between GXR and ATX (Δ : 0.6, p< 0.649). Conclusions: In this study, MAIC results were insensitive to variable selection via regression and the statistical moments matched, but matching the placebo arms altered the results. Matching placebo arm outcomes is valid when unobserved trial-specific factors have a differential impact on a trial’s treatment and control arm outcomes; this was likely the case in this GXR-ATX study. PRM207 Proposed Checklist for Non-Statisticians To Assess the Quality of A Network Meta-Analysis in the Context of A Nice Submission Pacou M 1, Gauthier A 2, Taieb V 2, Bec M 2, Belhadi D 2, Guillemot J 2 . . . . . . 1Amaris, Paris, France, 2Amaris, London, UK Objectives: To develop a checklist to assess the quality of a network meta-analysis (NMA) in the context of a submission to NICE. This checklist is intended to be comprehensible and easy-to-use by non-statisticians to assess whether an NMA is suitable for a submission to NICE and/or to populate cost-effectiveness models within the context of the NICE requirements. Methods: An ad-hoc search of the literature was conducted to identify existing checklists. Items from these checklists were extracted and critically reviewed. Recommendations from NICE as well as existing NICE submissions and corresponding comments from the evidence review groups (ERG) were used to develop the checklist. Our checklist was validated by health economists and pharmacists not trained in NMA on the basis of a NICE submission A580 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 in type 2 diabetes mellitus. Comprehension of the checklist was assessed based on qualitative interviews of each analyst. A measure of the inter-analysts agreement was estimated to ensure the reliability of the checklist. Results: Checklists identified from the literature included the checklist developed by the NICE Decision Support Unit and the one developed by the ISPOR task force. These two checklists were developed for analysts who conduct NMAs as well as analysts who critically review NMAs. However, they seem to lack clarity for non-statisticians. We developed a new checklist, which included the following items: definition of the study question (list of comparators, study population), methods (study selection, data extraction, statistical model, selection of fixed versus random effects model, assumptions for the base case, heterogeneity and inconsistency assessment and sensitivity analyses), reporting of results (network and source data, median or mean and 95% credibility interval) and interpretation of results. Conclusions: Our checklist can be used by analysts not trained in statistics to prepare or review NMAs to be submitted to NICE and/or to populate cost-effectiveness models. PRM208 Methodology For Selecting Expert Groups For The Purpose Of Decision-Making Tasks Ivlev I 1, Bartak M 2, Kneppo P 1 University in Prague, Kladno, Czech Republic, 2Jan Evangelista Purkyne University, Usti nad Labem, Czech Republic . . . 1Czech Technical Objectives: This work aims to develop a methodology for determining the qualitative composition of an expert group for the purpose of participation in decisionmaking in health care technology. Its goal is also to evaluate the methodology based on an example of the selection of large medical equipment. Methods: The complex weighting factor is a comprehensive evaluation of an expert. It is based on the expert’s overall work experience, experience in solving tasks, level of education and scientific record, interest in solving the particular task, current position, and awareness of how to solve the task. Also taken into account are the relevance of the expert’s knowledge and the overall self-evaluation concerning his or her total competence in solving the task. For the purpose of validating the methodology, 96 potential experts were interviewed. These subjects included managers from relevant departments in hospitals and hospital staff members who were from 72 health facilities in the Czech Republic. Results: Unlike the other models, the calculation model that was selected is able to eliminate errors in estimating the proportionality of extreme values and to reduce the impact of uncertainty in the experts’ overall self-evaluations concerning their total competence to the combined ratio. Based on this model, a methodology for selecting experts was developed. A statistically significant correlation was found between the complex weighting factor and the following characteristics: the expert’s experience in dealing with similar tasks (r= 0.512, p< 0.001), the expert’s theoretical background (awareness) and the relevance of the expert’s knowledge (r= 0.44, p< 0.001), the expert´s current position (r= 0.319, p= 0.002), and the level of his or her education and scientific record (r=0.28, p= 0.007). Conclusions: This methodology will be especially useful in scientific and technological forecasting, medical and managerial decision-making, quality assessment, and operational research. PRM210 Modelling Longitudinal Trajectories of Patient-Reported Outcomes To Evaluate Treatment Effect Jen M H Eli Lilly and Company, Surrey, UK . . Objectives: To evaluate treatment effect on longitudinal patient-reported outcomes using appropriate analytical strategy. Methods: This was an ad-hoc analysis of longitudinal patient-reported outcomes using a two stages simulated data in which the true model is known, to explore and to evaluate the capability of the group-based trajectory method to identify the distinctive features of a highly irregular but still continuous population distribution of trajectories. Firstly, we created six different types of underlying trajectory in which the true model is known and added in level-one between occasion random noises. Then we added a level-two, between- patients variation (of the random intercept form) with differing variability to each of the six distinctive trends. This simulation allows us to examine how the software implementation identifies different group trajectories as well as their level-one and -two variances. It was recognized that a priori assignment of distinct longitudinal trajectories may not be appropriate and that no ability to calibrate the precision of individual classifications exists if ex-ante rules are used. Thus, latent group-based trajectory model, a method to map the developmental course of symptoms and assess heterogeneity in response to clinical interventions, was used to identify patient groups with varied response. Results: The fitted trajectories closely approximate the true shapes and there is also a close correspondence for the percentage of places attributed to each group. Even the size of the level 1 random term is correctly estimated. The semi-parametric group-based trajectory method has demonstrated unequivocally its capability to capture the unobserved subgroups in the presence of considerable level-1 random variation. Conclusions: Patients in many disease areas experience changes in QoL in different ways. Identification of those groups is essential for appropriate evaluation of therapy treatment effects and identification of factors contributing to those groupings. PRM211 The Randomized Blind Start Trial: Evaluation of A New Study Design for Assessing Clinical Outcomes in Rare and Heterogeneous Patient Populations Signorovitch J 1, Ayyagari R 2, Kakkis E 3 1Analysis Group, Inc., Boston, MA, USA, 2Analysis Group Inc., Boston, MA, USA, 3Ultragenyx Pharmaceutical Inc., Novato, CA, USA . . . Objectives: Clinical development of therapies for rare diseases can benefit from improvements to conventional trial designs. This study evaluated a new trial design, the randomized Blind Start. Methods: The Blind Start design randomizes patients to ≥ 3 groups which initiate double-blind active therapy at different times from baseline, preceded by 0,1, 2 or more intervals of placebo. Analytical and simulationbased investigations were conducted to compare the statistical power, required assumptions, practical considerations and economic features of the Blind Start to conventional single-arm and randomized designs. Results: Given the same number of patients, the randomized Blind Start provides equivalent statistical power for detecting changes pre- vs. post-treatment compared to a conventional single-arm design. However, by concealing treatment initiation times, the Blind Start enables more objective assessments of outcomes that are effort-based, patient-reported or subjectively assessed by investigators. In addition, compared to a conventional 2-arm randomized trial, analysis of parallel treatment and placebo groups embedded within a Blind Start design provides greater power to detect treatment effects over any fixed time interval. For example, with N= 16 and a treatment effect equal to 1 standard deviation of the outcome measure, a 4-arm Blind Start design provides 85% power in a pre-post analysis and 79% power in an analysis of embedded parallel groups. In contrast, a conventional 2-arm randomized trial provides 52% power in this scenario. Benefits of the Blind Start design come at the expense of 1) more patient-time in the trial and 2) lack of stringent control over patient status upon active treatment initiation. Conclusions: The randomized Blind Start design can improve precision for treatment effect estimation vs. parallel-group designs and reduce risk of bias vs. single-arm designs. Endpoint choice and statistical analysis strategies for the Blind Start design can maximize the assessment of treatment effects on multiple outcomes. RESEARCH on Methods – Study Design PRM212 The Quality of Search Methodology and Search Reporting in Published Systematic Reviews of Economic Evaluations: Search Sources Wood H , Arber M , Glanville J York Health Economics Consortium, York, UK . . . Objectives: The economic evaluation of health care interventions is now an accepted element of health care decision-making and priority-setting. As the number of published economic evaluations has grown, so has the number of systematic reviews of economic evaluations. However, the quality of search methodology used in recent reviews has not been widely investigated. This study sought to identify which search resources are being used to identify studies in recent, published systematic reviews of economic evaluations, and to investigate whether choice of resources reflects current recommendations for the conduct of such reviews. Methods: A search to identify systematic reviews of economic evaluations published since January 2013 was undertaken in MEDLINE. Two reviewers extracted the following information from reviews which met the inclusion criteria: general medical literature databases searched, specialist economic databases searched, health technology assessment sources searched, supplementary search techniques used. Results were compared against the search resources recommended by NICE when searching for economic evidence for single technology appraisals, and the summary of current best evidence provided in Sure Info (http://vortal.htai.org/?q= node/336). Results: Sixty-five systematic reviews met the inclusion criteria; 23 of these could not be accessed in full text, data was extracted from 42 reviews. Five reviews (12%) met or exceeded the search resources recommended by NICE (MEDLINE, Embase, NHS EED, EconLit). Nine reviews (21%) searched at least four of the six types of resource recommended by Sure Info (specialist economic databases, general databases, HTA databases, webpages of HTA agencies, grey literature, collections of utility studies). None of the reviews searched all six. Although all reviews explicitly described the resources searched, reporting frequently contained errors or lack of clarity in the names of databases and interfaces. Conclusions: The information resources used to identify evidence for the majority of recently published systematic reviews of economic evaluations do not conform to current recommendations. PRM214 Identifying Psoriasis and Psoriatic Arthritis Patients in Retrospective Databases When Diagnosis Code Is Not Available: A Validation Study Comparing Medication/Prescriber Visit Based Algorithms To Diagnosis Codes Dobson-Belaire W 1, Borrelli R 1, Goodfield J 1, Liu F F 2, Khan Z M 3 1IMS Brogan, Mississauga, ON, Canada, 2Celgene Inc., Mississauga, ON, Canada, 3Celgene Corporation, Summit, NJ, USA . . . . . . . Objectives: Retrospective database studies rely on the ability to accurately identify patient cohorts of interest within health care databases. Diagnosis code-based algorithms are the primary method of identifying patient cohorts; however, many databases lack reliable diagnosis code information. Our aim was to develop precise algorithms based on medication claims/prescriber visit (MC/PV) to identify psoriasis (PsO) patients or psoriatic patients with arthritic conditions (PsO-AC), a proxy for psoriatic arthritis, in databases lacking diagnosis codes. Methods: Algorithms were developed using medications with narrow indication profiles in combination with prescriber specialty to define PsO and PsO-AC. For the study period of July 1, 2009 to June 30, 2013, algorithms were validated using the PharMetrics Plus™ (PharMetrics) database, which contains both adjudicated medication claims and diagnosis codes. Positive predictive value (PPV), negative predictive value (NPV), sensitivity and specificity of algorithms developed for PsO and PsO-AC were assessed using diagnosis code as the reference standard. Results: In the PharMetrics database, 183,328 patients were identified by diagnosis code or medication claim for validation. The highest PPVs for PsO (85%) and PsO-AC (65%) occurred when a predictive algorithm of ≥ 2 MC/PVs was compared to the reference standard of ≥ 1 diagnosis code. The majority of PsO-AC false positives had a diagnosis of PsO and pain or joint symptoms. NPV and specificity were also high (99 – 100%), while sensitivity was low (≤ 30%). Reducing the number of MC/PVs or increasing diagnosis claims decreased A581 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 the algorithms’ PPV. Conclusions: We have demonstrated that MC/PV algorithms can be used to identify PsO patients with a high degree of accuracy, while PsO-AC accuracy requires further investigation. Such methods allow researchers to conduct retrospective studies in databases where diagnosis codes are absent. getting a positive final test result (i.e. wrongly continued). The developed software (i.e. R-codes) can easily be applied to other cases. PRM215 Assessing the Effectiveness of Counter Matching for Improving the Efficiency of the Nested Case-Control Design in Observational Studies Jaksa A , Westbrook L , Daniel K , Ho Y S Context Matters, Inc., New York, NY, USA Kiri V 1, MacKenzie G 2 & JK Consulting Ltd, Guildford, UK, 2University of Limerick, Limerick, Ireland . . 1FV Objectives: The nested case-control (NCC) design offers a simple method for avoiding unreasonable assumptions in the evaluation of time-dependent treatment effect. Its results are easy to interpret. Its strength rests largely on the appropriateness of the controls which are matched to the cases- suggesting a matching strategy that ensures maximum number of discordant case-control pairs may be more efficient since effect estimation is based entirely on the off-diagonal data of the resulting 2x2 tables in the conditional logistic regression. In theory, the more off-diagonal pairs that are generated by the random sampling scheme, the more improvement we can expect on efficiency. The objective of this study is to assess the efficiency of counter-matching on treatment compared with the classical matching approach in the NCC design based on results from the cohort design using simulated data. Methods: In each simulation of 1000 patients at 100 replications per run, we assumed an underlying event hazard of Weibull distribution using inputted values for the scale and shape parameters, treatment, age and sex as the factors. Each run involved distinct treatment prevalence of between 10%-50% with event rate varying from common to very rare. We compared the proportion of matched pairs (at 1 control per case) used in the analyses of the resulting data between the classical and counter matching strategies. Results: The counter-matched strategy was more efficient than the classical approach. The proportion of matched pairs used was on average over ten times more and it also gave mean effect estimates which were more consistent with the full cohort values, particularly for low treatment exposure and rare events. Conclusions: Our results suggest counter matching is more efficient and more accurate estimates than classical matching in nested casecontrol design. These benefits may be particularly important for studies involving rare events or low treatment exposure. PRM216 Assessment of the Methodological Quality of Randomized Controlled Trials Published in “Russian Allergology Journal” in 2009-2013 Rakina E 1, Dombrovskiy V S 2, Rebrova O 3 1Autonomous Non-profit Organization “National Centre for Health Technology Assessment”, Moscow, Russia, 2The Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia, 3Pirogov Russian National Research Medical University, Moscow, Russia . . . . Objectives: To assess the methodological quality of randomized controlled trials (RCTs) published in “Russian Allergology Journal” (RAJ) in 2009-2013. Methods: Retrospective analysis of 96 original publications was carried out. For 8 RCT the risks of biases were assessed using the methodology of the Cochrane Collaboration. Accuracy of statistical analysis was assessed in accordance with established in 2009 journal’s requirements, made in accordance with best international practice. Results: 96 articles were analysed, 8 (8%) of them were identified as RCTs. All the RCTs have a high risk of the biases and major mistakes in the statistical analysis. Conclusions: The methodological quality of RCTs is insufficient and needs to be improved. We consider that the most important role should play improvement of trials’ planning. Collaboration with the experts in clinical trials’ methodology is strongly recommended. The analysis empowers researchers to consider existing experience and to improve methodological quality of RCTs, their relevance to international standards. PRM217 Simulation of An Additional Go/No-Go Efficacy Interim Analysis in A Head-To-Head Rct Van Montfort K Nyenrode Business University, Breukelen, The Netherlands . Objectives: A head-to-head randomized clinical trial (RCT) to evaluate a new drug is financially risky, because a positive outcome is uncertain. We simulate and evaluate a head-to-head RCT design incorporating an additional go/no-go efficacy interim analysis to show the consequences of this additional interim analysis. Methods: We simulate the endpoint event-free-survival (EFS) of patients in a head-to-head RCT. The decision rule of the additional interim efficacy analysis (i.e. stop or continue) depends on the number of patients (i.e. 300,400 or 500 patients) and the significance level (i.e. α = 0.05, α = 0.10 or α = 0.20) of the interim analysis. The RCT without an interim-analysis has significance level α = 0.05, power= 0.86 and sample size 800. Results: Each combination of sample size and significance level, which is called a scenario, is investigated by simulating 2,000 trials of 800 patients. Per simulated scenario we report, among others, the “Probability of positive final analysis test result GIVEN negative interim analysis test result” (= wrongly stopped) and the “Probability of negative final analysis test result GIVEN positive interim analysis test result” (= wrongly continued). The results of the first scenario, i.e. an interim analysis at 300 included patients and significance level α = 0.05, are as follows. If the actual improvement of the EFS hazard rate is 0%, the abovementioned probabilities are 3.65% and 7.85%. An actual improvement effect of 5% changes the probability values to 20.25% and 12.80, while an actual improvement effect of 10% causes the probability values 28.50% and 5.55%. Conclusions: The simulated probabilities were mainly influenced by the actual EFS improvement. The smaller the actual outcome improvement, the greater the probability of continuing the trial up to 800 included patients without PRM218 Negative Reimbursement Consequences From Trial Design Choices . . . . . Objectives: Health Technology Assessments (HTA) use clinical trial data to determine comparative efficacy and cost-effectiveness; thus study design plays a roll in market access. The objective is to determine how often reimbursement decisions cite trial design defects. Methods: We analyzed 1,702 HTAs from CADTH, G-BA, HAS, NICE, PBAC, and SMC. We examined the clinical assessment rationale for the decision and the reimbursement decisions. An explicit trial design defect was defined as a clinical assessment of “inappropriate comparator” or “inappropriate patient population.” Clinical assessments of lower, uncertain, or unknown efficacy or a clinical determination of insufficient or lack of evidence were defined as potential trial design defects. Results: Reviews that cited trial defects resulted in significantly more negative reimbursement decisions (6.6%) than positive reimbursement decisions (0.4%; p< .001). This pattern held true for each individual agency examined. G-BA was the agency most likely to cite an explicit trial deficit (39%), while HAS was the least likely (0.6%). In addition, significantly more reviews that cited a potential trial defect resulted in negative reimbursement decisions (44%) than positive reimbursement decisions (5.5%; p< .001). This also held true for each agency. Again, G-BA was most likely to note a potential trial defect in reviews (46%) while SMC was least likely (6.4%). Among disease conditions with more than 10 reviews, explicitly cited trial defects were mostly frequently seen in Cystic Fibrosis and Parkinson’s Disease (15% and 13%, respectively). Potential trial defects were most frequently cited in Atrial Fibrillation and Depression reviews (42% and 39% respectively). Conclusions: Explicit and potential trial design issues have negative consequences for reimbursement outcomes. Negative decisions are more likely than positive decisions to cite trial design issues. G-BA is more inclined than other agencies to cite these trial design issues when issuing their reimbursement decisions. Manufacturers should consider market-access outcomes when designing clinical trials. PRM219 Evidence Resulting From Chart Review Methodology Applied To Named Patient Programme Participation And Compassionate Medication Use: Peri-Approval Approximation Of Post-Market Practice Patterns And Costs Stein D 1, Jean-Mary J 2, Goldwin A E 2, Lau M R 3, Manson S 3 Express Scripts Company, Dorval, QC, Canada, 2United BioSource Corporation, London, UK, 3GlaxoSmithKline Oncology, Uxbridge, UK . . . . . . . 1UBC: An Objectives: Compassionate use programmes provide peri-approval drug access based on unsolicited physician requests for patients with unmet need. Practice pattern evaluations in this context, using chart review methodology, permits the collection of pre-approval data outside of clinical trial settings that can approximate real-world post-market use. Data can be used to inform important economic evaluations, value dossiers, and drug safety assessments. Methods: Study design and operational considerations related to chart review studies of compassionate use populations have been summarized by evaluating three multi-national case studies. Results: The source populations of patients were drawn from compassionate use programmes providing oral anti-cancer therapies. These were initiated pre-approval following positive clinical trial findings. Countries included Australia, Belgium, Greece, Ireland, Israel, Italy, The Netherlands, New Zealand, Spain, Switzerland and the United Kingdom. Data including patient characteristics, patterns of care and drug dosing, duration of treatment and reasons for treatment discontinuation, overall survival, clinical benefit, progression free survival and adverse and serious adverse events are being collected to inform health economic and other burden of illness assessments. Site and patient selection was performed using compassionate use enrollment data, facilitating an efficient study start-up. Data were cleaned at point of data entry and via an electronic query process in real-time resulting in tailored international datasets. Patients provided consent for their medical data to be used prior to initiating compassionate use therapy, and additional ethical approval for the chart review was sought on a local and/ or national level. Chart review study design requires a balance between scientific and operational rigor and practicality and feasibility. Conclusions: Peri-approval chart review studies of patients in compassionate use programmes offer an important opportunity to characterize patterns of use and associated treatment costs as well as the clinical impact of investigational medications in non-trial settings to inform clinical, health economic and market access decisions. PRM220 Immature Survival Data From Early Trial Termination – Theory and Hta Practice Pruefert A 1, Skaltsa K 2, Maervoet J 1, Van Engen A 1 1Quintiles Consulting, Hoofddorp, The Netherlands, 2Quintiles Consulting, Barcelona, Spain . . . . Objectives: Scientific research suggests that randomized controlled trials terminated early for benefit considerations systematically overestimate treatment effects of the primary outcome. This study assessed whether Health Technology Assessment (HTA) agencies accept the increased uncertainty around overall survival (OS) estimates in oncology trials arising from early termination. Methods: Public scientific databases were searched to identify scientific articles and pivotal trials involving early trial termination. A selection of related HTA appraisals, published between January 2011 and February 2014, were analysed. Current scientific evidence on the impact of early stopping on outcome estimates was compared to the conclusions made by 11 HTA agencies. Results: Twelve scientific articles, 12 pivotal trials, and 31 related HTA appraisals were selected for in-depth analysis. The scientific literature suggests that more stringent significance levels in the repeated interim A582 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 analyses reduce the probability of finding significant results due to chance while large numbers of outcome events reduce the overestimation of treatment effects. Our analysis finds that a statistically significant gain in OS is an important decision driver for even the most critical HTA agencies, although the treatment effect may still be questioned when the trial is unblinded early. HTA agencies appreciate to receive the latest available information (UK, Australia and Germany) and may reject the use of oncology drugs when there is too much uncertainty around OS estimates to justify the proposed price. It is generally useful to continue data collection and follow-up patients should HTA agencies still request more reliable OS estimates for modeling purposes (UK and Australia) or long-term risk-benefit evaluation (France). Conclusions: Payers are aware of the overestimation of effect size due to early trial termination and may reject drugs for high uncertainty around OS estimates. For adequate responses to requests for more reliable data, it is advised to continue data collection and follow-up patients. PRM221 The Management of Irritable Bowel Syndrome (Ibs) in England: A Real World Study in Primary Care Clinical Practice Caldwell I 1, Collins J 2, Rance M 3, Dew R M 4 1Swan Lane Medical Centre, Bolton, UK, 2NIHR Clinical Research Network: Greater Manchester, Manchester, UK, 3Almirall, Uxbridge, UK, 4pH Associates Ltd, Marlow, UK . . . . . Objectives: IBS is often a diagnosis of exclusion, with poor diagnosis coding in primary care and identification of eligible research participants challenging. We present the methodology of an on-going multi-centre, observational, retrospective research study, designed to overcome the challenges of IBS patient identification. Methods: FARSITE, a software tool for identification of research participants developed by the Greater Manchester Comprehensive Local Research Network and North West eHealth, was used to screen anonymised primary care records for potentially eligible patients. Ethical approval reference 13/LO/0692. Search criteria: patients aged 18-60; combination READ code symptoms indicative of IBS and prescription of IBS medications 01/01/2009–31/12/2011. GPs at 8 participating practices in Salford & Greater Manchester reviewed clinical records of the FARSITE-generated list of patients to apply full eligibility criteria for final patient selection. Inclusion criteria: medical diagnosis of IBS or meeting ROME III criteria; provision of consent. Exclusion Criteria: diagnosis excluding IBS; IBS symptoms secondary to other condition; IBS medications only for non-GI symptoms. Results: FARSITE identified 1089 (1.3%) patients, of which 297 (27.3%) were eligible. 97 patients consented to participation (79% female). Main reasons for non-eligibility were not meeting ROME III criteria or IBS excluded by medical opinion. Patients were most commonly coded as irritable colon (37%), difficulty defecating (21%), abdominal pain (18%), diarrhoea symptoms (14%). Four (4%) patients had a READ code specific for IBS. The median (IQR) time from 1st presentation with abdominal symptoms to study eligibility was 3.98 (0.00-9.04) years. Conclusions: Identification of patients with IBS using READ codes is sub-optimal in primary care. A combination search of READ codes with symptom and prescription data via FARSITE has enabled potential participants to be identified with a reasonable screening failure rate. FARSITE is a valuable research tool aiding study feasibility by reducing the need for manual patient identification. PRM222 The Effect of A Likely Overemphasis on Ficiency-Related Test Attributes on Acmg Recommendations and Access To Newborn Screening (Nbs) Rittenhouse B MCPHS University, Boston, MA, USA . Objectives: Patient access to NBS has been greatly influenced by the 2006 American College of Medical Genetics (ACMG) recommended expansion of NBS. ACMG relied largely on a stakeholder survey on 19 attributes of 84 rare conditions. The percentage of respondents agreeing to an attribute’s presence for a condition, along with its weight, determined attribute score. Sums of scores determined the entry point to an algorithm for final recommendations. This research examines 6 attributes that appear to be associated with the same concept and asks whether these are really one (over-weighted) concept. Methods: The ACMG report provided attribute scores. Six questions addressed test efficiency (simplicity, high throughput, cost < $1/condition, multiple analytes/test run, other conditions identified/analyte, multiple conditions detected/test). We examined correlations between the 6 answers for a given condition across conditions and associations with recommendations. Results: After eliminating conditions with missing data, 78 remained. Pairwise correlations between the 6 answers were high (mean= .85; range, .72-.96). Of those conditions (37) scoring at least 75% of the possible points on one question (“high throughput”), 79% were recommended as Core conditions to be screened and only 8% were Not Recommended. The mean total scores for the 6 similar questions was 339 (500 possible). Of those (19) scoring 25% or fewer of the possible points for that one question, only 3% were Core, 72% Not Recommended (mean score:89). Conclusions: The high correlations support the idea that the 6 similar questions were answered as if they were the same concept, weighting the common general attribute very highly. A more systematic approach, say MCDA, would likely have eliminated some of these questions with significant consequences for ACMG recommendations. PRM223 Workflow Mapping for Paediatric Vaccination Process in the United Kingdom (Uk): A Precursor of A Time and Motion (T&M) Study Mokiou S 1, De Cock E 2, Standaert B 3 1UBC: An Express Scripts Company, London, UK, 2United BioSource Corporation, Barcelona, Spain, 3GlaxoSmithKline Vaccines, Wavre, Belgium . . . Objectives: Time and Motion (T&M) methodology allows quantifying timerelated outcomes for a health care delivery process by disaggregating the process in its constituent parts to measure task durations. The design of a T&M study requires early process mapping to define the time outcomes to be measured. The mapping of paediatric vaccination process in the United Kingdom (UK), as a precursor of a real-world study, is described. Methods: A targeted review of publicly available information was conducted to gain comprehensive understanding of the paediatric vaccination process in the UK. A survey was designed eliciting the chronology of vaccination process prior to and on vaccination day, including estimates of active health care professional involvement. Face-to-face interviews with a nurse were conducted at three general practitioner surgeries routinely performing vaccinations. A subsequent follow-up call with each nurse was also arranged. Descriptive statistics were generated and preliminary cost calculations made. Results: Paediatric vaccination process can be broken down in 6 and 8 clearly discernible steps prior to and on vaccination day, respectively. Activities prior to vaccination day include, among others, inventory, ordering, cold-chain management and are typically for multiple subjects. Mean time for those activities, recalculated per single vaccination visit, was 6.7 minutes, of which 61% dedicated to administrative duties. Activities on vaccination day include, among others, room preparation, consultation, vaccine administration. Estimated time per single visit totaled 25.4 minutes. Estimated total cost per single vaccine administration, with nurse salary cost from PSSRU, was £10.4. Costs may vary substantially depending on the level of “on-costs” to nurse’s gross salary. Conclusions: The detailed mapping of paediatric vaccination process in the UK identified clearly discernible tasks, time estimates, factors impacting variability of time outcomes, and early cost estimates. This forms the basis of a real-world T&M study aiming to generate robust time and cost outcomes. PRM224 Comparative Effectiveness Research of Medical Devices – New Methods Needed? Hunger T 1, Schnell-Inderst P 2, Arvandi M 1, Conrads-Frank A 1, Siebert U 1 - University for Health Sciences, Medical Informatics and Technology, Hall i. T., Austria, 2UMIT - University for Health Sciences, Medical Informatics and Technology, Hall i. T.; Innsbruck, Austria . . . . . 1UMIT Objectives: Guidelines for Health Technology Assessment (HTA) and Comparative Effectiveness Research (CER) largely focus on pharmaceuticals and only few explicitly consider other health care technologies. CER of medical devices (MD) faces some challenges that raise questions about how adequate current CER methods account for the specific features of MD and how well MD fit in the paradigm of drug HTA. Our aim was to identify challenges and gaps in methodology related to specific issues of MD. Our comprehensive framework for the evaluation of clinical effectiveness of MD includes recommendations for generation of primary data and analyzing and synthesizing data in systematic reviews of CER of MD. Methods: We performed a targeted literature review for CER methods and specific features of MD. An electronic database search was combined with systematic screening of tables of content of selected journals in the fields of epidemiology, HTA, statistics, and evidence-based medicine, which have a strong focus on methods. Additionally, we screened the reference lists of the most relevant papers. Results: More than 200 publications about the general evaluation of MD and about specific CER methods were included. The MD’s physical mechanism of action, the dynamic development and regulatory evidence requirements are the driving features that suggest the increased use of certain methods for the evidence generation, finding of information for HTA, data analysis and synthesis, and interpretation of results. Rather than following the paradigms of drug evaluation, MD resemble more the notion of complex interventions. Our methodological framework is compatible with the EUnetHTA core model and integrates existing recommendations for other complex interventions. The consideration of observational data, operator characteristics, active control trials, and decision-analytic modeling are of special importance, as well as the application of Bayesian methods. Conclusions: The assessment of the clinical effectiveness of MD does require specific, although not necessarily new methods. PRM225 (Cost-) Effectiveness of A Multi-Component Intervention for Adults With Epilepsy: Study Protocol of A Dutch Randomized Controlled Trial Wijnen B 1, Leenen L A M 2, de Kinderen R J A 1, Majoie M H 2, van Heugten C M 3, Evers S M 1 1Maastricht University, Maastricht, The Netherlands, 2Epilepsy Center Kempenhaeghe, Heeze, The Netherlands, 3MHENS, School for Mental Health and Neuroscience, Maastricht, The Netherlands . . . . . . . . . . . . . Objectives: Poor adherence to anti-epileptic drugs has been shown to be the most important cause of poorly controlled epilepsy. Furthermore, it is emphasized that an increase in quality of life among patients with epilepsy could be reached by counseling and treatments aimed at increasing their self-efficacy and thus stimulate self-management. However, there is a need for evidence on the effectiveness of such programs, especially within epilepsy care. Therefore, we have developed a multi-component intervention (MCI) which combines a self-management/education program with e-Health interventions. Hence the overall objective of this study is to assess the (cost-) effectiveness of a MCI aiming to improve self-efficacy in people with epilepsy compared to care as usual. Methods: A randomized controlled trial in 2 parallel groups will be conducted to compare the MCI intervention with a waiting list control condition in epilepsy patients. One hundred eligible epilepsy patients will be recruited from the Kempenhaeghe epilepsy center and allocated to intervention or control group. Patients in the intervention group will receive an education program of six meetings including e-Health intervention and will be followed for 12 months. Patients in the control group will be followed for 6 months after which they will be offered to participate in the MCI. The study will consist of three parts: 1) a clinical effectiveness study, 2) a cost-effectiveness study, and 3) a process evaluation. The primary outcome will be self-efficacy. Outcome assessments will be done using questionnaires at baseline and after 3, 6, 9, and 12 months. Results: N/A. Conclusions: This study will determine the (cost-) effectiveness of an MCI intervention to improve the self-efficacy of epilepsy in adult patients. The MCI is designed to stimulate self-management skills and awareness of epilepsy patients in combination with A583 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 the use of e-health interventions.*Both B. F. M. Wijnen and L. A. M. Leenen contributed equally to this work. PRM226 Implementation of International Chart Review Studies: An Assessment of Ethics and Regulatory Considerations Jean-Mary J 1, Stein D 2, Yeomans K 2, Payne K A 3 BioSource Corporation, London, UK, 2UBC: An Express Scripts Company, Dorval, QC, Canada, 3United BioSource Corporation, Dorval, QC, Canada . . . . . 1United Objectives: In the absence of secondary sources of health care data, chart review studies can result in patient level data repositories including patient characteristics, care patterns, treatment effectiveness and clinical and safety outcomes. Data can be used to populate economic evaluations, and value dossiers, and inform drug safety assessments. For successful implementation, however, knowledge of country-specific ethics and regulatory approval processes is paramount. Methods: Operational, ethics and regulatory issues and considerations as well as strategies for study success have been summarized in the context of eleven recent multi-national chart review case studies. Results: Two of 11 studies also collected data prospectively; two studies were categorized as post authorization safety studies and three studies were conducted in peri-approval compassionate use program populations. The majority of studies (9) were oncology focused, with two studies focused on infectious diseases and opioid-induced constipation. Sample sizes varied from 20 to 500 patients, the number of countries from 1 to 8, and the number of sites from 4 to 61. All studies included at least one European country. Across studies, key operational considerations that impacted the ethical/regulatory approval process were ambiguous/amorphous multinational regulatory requirements/guidelines; commercial availability or non-availability of the sponsor product at the time of chart abstraction; data collection method(s) (i.e., retrospective vs. hybrid chart review plus prospective data collection); country variation in informed consent requirements and definitions of personal data; and multinational contractual requirements with the participating sites. Conclusions: International chart review studies are an effective methodology to resolve data gaps not solved by existing secondary health care data sources resulting in tailored, patient-level datasets. Current knowledge of the highly variable and evolving global regulatory requirements, as well as the development of a risk management plan informed by methodological and operational lessons learned at study-outset will facilitate risk mitigation and allow researchers to overcome key challenges. PRM227 Cost Per Patient in Non Interventional Studies and Added Value of Direct To Patient Contact Service Fournie X Mapi, Lyon, France . Objectives: In addition to study outcome concerns arising from patients lost to follow-up (LFU) in pharmacoepidemiology and pharmacovigilance studies, the financial impact of LFU can be significant. Our objectives were to estimate cost per patient in Non-interventional studies, to identify variables that may affect this patient cost, to estimate cost of patient lost to follow-up (LFU), and financial benefits that can be expected from LFU minimization through Direct to Patient Contact service (DPC). Methods: Analysis of 2013 proposals and budgets submitted to study sponsors. Selection criteria: non interventional, prospective, longitudinal patient follow-up, full CRO services. Analysis were performed according to patient sample size, study duration, disease category, and different hypothesis for LFU rates. Results: 1) 20 studies (Domestic, Regional or Global) met all inclusion criteria; 2) Annual cost per patient -ranging from € 1,068 to € 4,370- decreases as the study duration increases (set-up cost is more diluted in the patient annual cost). But the longer the study is the more expensive the overall cost per patient; 3) Mean annual patient cost significantly differs according to rarity of disease/population; rarity is an important criterion that greatly impacts overall and annual patient cost, especially for study lasting more than 1 year. Below 1 year, the cost per patient remains quite similar between types of diseases/populations; 4) Cost are more significant in rare diseases studies, therefore DPC can provide the best overall cost savings in these populations; and 50 The cost savings are depended on the expected rate of patient LFU-with/ without DPC service and the planned patient sample size. Conclusions: Return On Investment plays an important role for Sponsors to determine if DPC is valuable in a study. The financial investment may be beneficial regardless of the cost to insure completion of the patients, thus meeting the scientific study objectives. But it could generate cost savings as well. PRM228 Retrospective Chart Review Studies: Strategies To Ensure Robust Data Quality Stein D 1, Bassel M 1, Payne K A 2 Express Scripts Company, Dorval, QC, Canada, 2United BioSource Corporation, Dorval, QC, Canada . . . . 1UBC: An Objectives: Retrospective chart review studies can result in robust naturalistic data to inform evaluations of treatment patterns, resource utilization, costs of care, clinical outcomes and safety. Data quality control is challenging both as a result of poor quality documentation in the usual care medical chart, or as a result of data abstraction and data entry processes. Methods: Ten chart review case studies conducted in the United States, Canada and Europe were evaluated to provide recommendations for improving chart review data quality control mechanisms. Results: All 10 studies used electronic data capture (EDC) systems. Common lessons learned across the studies were that the case report forms (CRFs) should only include necessary data points required to fulfill the analysis. Direct chart-to-EDC data entry and remote real-time data quality control is recommended to reduce additional transcription errors that may occur if using paper CRFs. It is important to ensure the EDC system includes a cohort-control platform that enables selection of patient cohorts (i.e., random selection) and tracking of eligibility screening to reduce selection bias risk. Automated edit checks of primary data endpoints should be programmed into the EDC system prompting data abstractors to revise erroneous data and/or confirm data outside of expected ranges at entry. To confirm abstracted data reflect source documents (patient medical charts), a second abstractor at the site can re-abstract pre-defined critical study variables from patient medical charts for cross-referencing for data discrepancies. Site training must be effective to ensure compliance with chart abstraction and data quality requirements. Conclusions: Given the frequent incomplete or poor quality medical chart information and the potential for human error in data abstraction and entry processes, data quality control methods are paramount. Approaches to protocol, CRF and study training materials design can positively impact data quality. RESEARCH on Methods – Conceptual Papers PRM229 Research Prioritization In An Mcda Context: Existing Methods New Results Janssen M P 1, Koffijberg H2 1University Medical Center Utrecht, Utrecht, The Netherlands, 2University Medical Center, Utrecht, The Netherlands . . Objectives: Health technology assessment typically involves consideration of multiple conflicting criteria. Therefore, trade-offs are required between different objectives such as maximizing health, restricting budget impact, increasing health equity and maximizing safety. Methods such as multiple decision criteria analysis (MCDA) are therefore increasingly being used to reflect such trade-offs in a transparent and consistent manner. Although MCDA can be combined with cost-effectiveness analysis it may, however, invalidate results from Value of Information (VOI) analysis when it also includes other health-related or cost-related objectives. Methods: In two case studies we first applied VOI methods directly and only to cost-effectiveness estimates, and then also applied these methods separately to all relevant decision criteria. In a simulation study on two drugs we calculated the expected value of perfect information (EVPI) with drug selection concerning a trade-off between cost-effectiveness and drug safety. In a clinical study on the primary prevention of cardiovascular disease using improved versus standard risk prediction we calculated the EVPI with selection of the best risk prediction strategy concerning a trade-off between cost-effectiveness and budget impact. Results: In our simulation study we found EVPI estimates per patient based only on cost-effectiveness were up to € -586 lower and € +459 higher compared to EVPI estimates also acknowledging the safety criterion, depending on its weight. In our clinical study, the EVPI estimates based only on cost-effectiveness were consistently lower, up to € -540 per patient, compared to EVPI estimates also acknowledging the budget impact criterion. Conclusions: When decisions are based not only on cost-effectiveness but on other criteria as well, some of which also relate to costs or health effects, standard VOI estimates are no longer valid. However, separate application of VOI methods to each of the relevant decision criteria is straightforward and can facilitate transparent research prioritization in a complex MCDA context. PRM230 A Statistical Modeling Framework To Characterize the Impact of Progression on Survival in Oncology Ishak K J Evidera, St-Laurent, QC, Canada . . The benefits and value of new cancer treatments often focus on the overall survival (OS) gains that patients may derive. Trials are typically not long enough to allow detailed understanding of OS, and potential benefits must be inferred from benefits on progression-free-survival (PFS). This raises questions such as whether early or later progression impacts survival, whether the increase in mortality following progression is sustained or gradually diffused, and whether a benefit observed on PFS implies a benefit in OS. Answering these questions requires an analytical framework in which progression and survival can be analyzed together and parameterized to address key questions. We propose a statistical modeling framework based on Cox regression and time-dependent predictors and effects. A simple formulation of this model would include a time-dependent indicator for progression, whose coefficient would measure the increase in risk of death following the event. This is very limiting, however; it assumes that the timing of progression does not matter and that the increase in risk of death is sustained indefinitely. A more flexible formulation can be built using two descriptors of event: the timing of progression (TP) and time since progression (TSP). These can be continuous measures or categorized (e.g., early vs. late TP), as appropriate. The coefficient for TP reveals whether later progression is associated with higher/lower subsequent mortality, while the coefficient of TSP reflects whether and for how long the increase/decrease in mortality is sustained and whether it ever returns to the level of patients who had not progressed. The impact of treatment can be captured on each of these parameters separately. The proposed framework will be illustrated with an example, and extension of the approach to other applications (e.g., measuring the impact of a stroke on survival) will be discussed. PRM231 Towards Integration of Research Evidence on Patient Preferences in Coverage Decisions and Clinical Practice Guidelines: A Proposal for A Taxonomy of Preference-Related Terms Utens C M 1, Joore M A 1, van der Weijden T 2, Dirksen C D 2 of Clinical Epidemiology and Medical Technology Assessment, Maastricht University Medical Centre, Maastricht, The Netherlands, 2Department of Family Medicine, CAPHRI School for Public Health and Primary Care’, Maastricht University, the Netherlands, Maastricht, The Netherlands . . . . . . . 1Department Despite the availability of a large body of research evidence on patient preferences for health outcomes and/or health care services, its use in health care policy decisions is limited. This contrasts with the current increasing attention for patient-cen- A584 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 tred care and integration of the patient perspective in health care policy decisions. A major challenge for the integration of evidence on patient preference is that research on patient preferences is performed by various disciplines (e.g. psychology and economics) that do not share a common language. It has been recommended to perform conceptual and taxonomic work on the definition and conceptualisation of ‘preference’ and related terms. The aim of this study was to develop a taxonomy of preference-related terms. The taxonomy was developed in three steps: 1) the identification of preference-related terms; 2) providing all identified terms with a definition from the dictionary; and 3) the identification of dominant theories or models from (health) economics and psychology that deal with the referencerelated terms. The proposed taxonomy consists of several building blocks that hold all identified preference-related terms and demonstrate the relation between terms. The building blocks are centred around a factual event. Ex ante to this factual event lies building block 1, “decision making” holding terms like “choice” and “decision”. Ex post lie building block 2 “evaluation process” and building block 3 “outcome of evaluation process”. Building block 3 holds terms like “utility”, “quality of life” and “satisfaction”. Building blocks 1-3 are influenced by building block 4 “the value system”. This value system is divided in cognition, affect and conation and holds terms like “beliefs”, “expectation”, “attitudes”, “desires” and “intention”. In this taxonomy, preferences can be considered as a part of the value system. The proposed taxonomy is a first step towards conceptual clarity to facilitate the integration of research evidence in health care policy decisions. PRM232 When It May Not Be Necessary To Model Overall Survival for Economic Evaluations of Anti-Cancer Drugs Hoyle M 1, Hamilton W 1, Rudin C 2 of Exeter, Exeter, UK, 2Royal Devon & Exeter Hospital, Exeter, UK . . . 1University Overall survival (OS) is traditionally modelled in economic evaluations of anti-cancer drugs. However, OS is commonly associated with problems such as immaturity of the data, or confounding due to treatment switching or use of inappropriate treatments after progression. Fortunately, analysis of historical trials reveals that there is good evidence across a range of cancers that the mean time in post-progression survival (PPS) is equal between treatment arms, i.e. Δ PPS = 0. Therefore, we recommend that the default position is to assume equal mean times post-progression. If there is no a priori biological reason to suppose that the PPS times are likely to differ between treatments (e.g. due to differences in cross-resistance or long term toxicities between treatments), our recommendation is that it should be assumed that the mean time in progressive disease is equal between treatment arms if any of the following apply: OS is very immature; treatments post-progression are substantially imbalanced between treatment arms; in particular, treatment switching has occurred at progression; treatments post-progression are different to those routinely given in clinical practice; only single arm trials are available. If none of the above apply, or if there are a priorireasons to suggest that ΔPPS differs from 0, then the recommendation is to model OS and PFS in the traditional way. For chronic cancers, it is recommended that analyses should either assume equal times post initial treatment or equal time post progression. The assumption that Δ PPS = 0 substantially simplifies the economic analysis because cost-effectiveness becomes insensitive to OS. The methodology has been endorsed twice by NICE appraisal committees in assessments of drugs for chronic myeloid leukaemia. The cost-effectiveness of several drugs recently assessed by NICE are re-calculated using the methods proposed. Next, we give simplified formulae for the maximum drug price acceptable for reimbursement under the methodology. PRM234 Feasibility of Conducting Retrospective Studies Using Hashtags and Social Media Data From Facebook and Twitter Volovyk A 1, Topachevskyi O 2 . Cost-effectiveness models often require the consideration of a sequence of treatments. This enables the downstream implications of a treatment to be captured, and alternative sequences to be compared. However, when many treatments are available, the number of feasible sequences can be large. Also, if the objective is to maximise net benefit for a given ICER threshold, then a comparative analysis to identify the optimal sequence may not be possible. This is further compounded when using individual patient simulation (IPS), because of the increased computational burden compared with cohort approaches. The aim of this study was to undertake a systematic review of optimisation methods that are applicable to a treatment sequencing IPS model. 28 key papers were identified across a range of academic subjects. Metaheuristics including simulated annealing, tabu search and genetic algorithms have been applied to simulation-optimisation problems and a bespoke review framework was applied to determine their appropriateness. Based on the review, a framework for the economic evaluation of treatment sequences was developed. The framework considers the requirements of a cost-effectiveness model to efficiently evaluate sequences, the application of the reviewed metaheuristics to determine the optimal sequence, and the consideration of these results within a decision-making context. This will be applied to a case study in rheumatoid arthritis. Alternative metaheuristic algorithms will be applied in an attempt to estimate a (near) optimal treatment sequence. Preliminary results of these experiments will be available in time for the November 2014 ISPOR conference. If these methods prove successful and feasible, then the framework may have potential applicability to sequencing models in many diseases. Whether there is the capability for it to be applicable within the current process for decision-making organisations such as NICE remains an open question, however, identifying an optimal sequence in a decision problem is of interest to decision makers. . 1Hashtago, Kiev, Ukraine, 2Digital PRM236 Novel Indirect Comparison Methodology for Estimating Time-Dependent Response To Antimuscarinics for the Treatment of Oab Snedecor S J , Sudharshan L Pharmerit International, Bethesda, MD, USA . . . Background: Common indirect treatment comparison (ITC) methodology in overactive bladder involves combining absolute reduction in urge urinary incontinence (UUI) episodes at study endpoint (e.g., week 12) to estimate the overall treatment effect. Trials with differing endpoints must assume equivalence to be included in the network. Further, analyses of endpoint data are not sufficient to predict efficacy at intermediate time points (e.g. 4 or 6 weeks). We developed and tested an alternate methodology to utilize available intermediate time points into an ITC of published studies of fesoterodine and tolterodine. Methodology: Study-level mean UUI reduction over time can be represented as the percent reduction from baseline, which can be modeled as a monotonically-increasing function with a theoretical maximum of 100%. This function is expressed with two parameters: %red = bi*time/ (ci + time), where bi is the maximum possible reduction for treatment i, and ci is the time required to reach half the maximum reduction. The inverse %red is a linear function of 1/time that can be used within a Bayesian ITC framework to generate a placebo-adjusted indirect comparison of efficacy. Conclusions: The endpoint results obtained from the alternate methodology were comparable to those obtained from an endpoint ITC. This novel methodology has the additional advantage of utilizing all available time point data within a single analysis, which can then be used to generate efficacy estimates at intermediate time points, which may be utilized within economic models. Limitations include unavailability of uncertainty estimates of the %red variable and difficulty of estimating combinations of parameters within functional constraints. Finally, our alternate methodology may be used for any longitudinal data exhibiting a monotonic increase or decrease and may be expanded to include a network with multiple treatments. Health Outcomes, Brussels, Belgium Various online services such as Socialbakers, Keyhole, Gnip offer tools to analyze, fetch and collect data from social media. This data is often presented in a form of interactive web based dashboards, displaying various trends: number of posts, mentions, shares, likes over time. Facebook and Twitter have an API to access data on social media profiles of real people. Users profiles usually have data on age, sex, employment and relationships status, specific group membership, etc. We conducted a simple feasibility study using Facebook API in diabetes area using profiles of people posting hashtags as a primary source of data. We then expanded the sample by adding people who liked, shared and reposted messages containing diabetes relaed hashtags #Diabetes, #dedoc, #ourD. We applied exclusion criteria to derive a sample consisting of patients only, hence targeting specific group of people. Our assumption was that people who interact with posts containing specific hashtag have diabetes. We used descriptive statistics to characterize obtained sample (n= 17296) by calculating mean age, age distribution histogram, proportion of males and females and other descriptive metrics. We also calculated conditional probabilities of being in multiple disease area Facebook groups such as obesity groups or groups of people with increased risk of cardiovascular disease. Future area of research will be concentrated on aspects of in-degree centrality in network of diabetic people, hypothesis testing between two different groups, analyses of changes in positive/negative posting trends following drug launch, locating agents and influencers in the network and conducting prospective studies in social media using hashtags. Social Media data can be a valuable addition to a real life post launch data. Evidence on changes in positive/negative postings can be used as an additional piece of information in Phase IV studies or risk-sharing agreements. PRM237 Bayesian Models for Cost-Effectiveness Analysis in the Presence of Structural Zero Costs PRM235 A Framework for the Economic Evaluation of Sequential Therapies for Chronic Conditions Moran P , Harrington P , Ryan M Health Information and Quality Authority, Dublin, Ireland Tosh J , Stevenson M , Strong M , Akehurst R University of Sheffield, Sheffield, UK . . . . Baio G University College London, London, UK . Bayesian modelling for cost-effectiveness data has received much attention in both the health economics and the statistical literature, in recent years. Costeffectiveness data are characterised by a relatively complex structure of relationships linking a suitable measure of clinical benefit (\eg QALYs) and the associated costs. Simplifying assumptions, such as (bivariate) normality of the underlying distributions are usually not granted, particularly for the cost variable, which is characterised by markedly skewed distributions. In addition, individual-level datasets are often characterised by the presence of structural zeros in the cost variable. Hurdle models can be used to account for the presence of excess zeros in a distribution and have been applied in the context of cost data. We extend their application to costeffectiveness data, defining a full Bayesian specification which consists of a pattern model for the individual probability of null costs, a marginal model for the costs and a conditional model for the measure of effectiveness (given the observed costs). The model is presented using a working example to describe its main features. In addition, we present a R package (BCEs0) that directly implements this framework and can be used to run a full Bayesian cost-effectiveness analysis of individual data in the presence of structural zero costs for some subjects. PRM238 Effective Prioritisation of National Health Technology Assessments . . . Prioritisation of assessment topics is an essential activity within HTA. Failure to successfully identify technologies that are likely to have the greatest impact on the health system carries an opportunity cost that is measured in poorer decision A585 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 making, reduced patient benefits and less efficient use of public resources. Within individual agencies, prioritisation is also a business function that must balance the need to plan for and manage the allocation of resources with the need to provide expeditious advice to decision makers and adapt quickly to changing circumstances. This research describes a transparent and responsive framework for selecting health technologies to assess, minimising the potential for important technologies to be missed and providing a useful resource for HTA agencies facing similar issues. Topics are identified through a mix of routine horizon scanning, a formally convened advisory group consisting of the major decision makers from within the publicly funded health system and informal business intelligence gathering. Screening is carried out to eliminate technologies that are clearly unsuitable and provisionally grade all remaining candidates according to three principal criteria; 1) clinical impact (patient population, potential incremental effect and availability of alternatives); 2) economic impact (incremental costs and potential disruptive effect on how services are currently organised) and 3) policy impact (link to decision-making and factors that make it likely to feature on the national health care agenda). The screening process feeds into an in-depth expert group discussion, which also considers operational issues such as the extent of the advice required to inform the decision, data availability and costs associated with the assessment. We also describe a software visualisation tool developed to facilitate the prioritisation process, as well as measures for quality assurance and ongoing performance evaluation. PRM239 Goal Attainment Scaling – A Useful Individualized Clinical Outcome Measure Jones M , Kharawala S , Langham J , Gandhi P Bridge Medical, London, UK . . . . Goal Attainment Scales (GAS) capture outcomes relevant to individual patients and provide “real-world” outcome measurement. This abstract will describe the background to their use, their operationalization, strengths and limitations. Traditional outcome measures assess a standardised set of questions regardless of their relevance to each patient. GAS overcomes these weaknesses because it is an individualised assessment based on achievement of goals which are personal to each patient. Despite its widely cited use in academic literature and good psychometric properties it is rarely used in drug intervention studies. Operationalisation of GAS varies but follows these basic steps: 1) The patient’s specific problem areas are assessed and goals for each defined; 2) A GAS for each goal is created and an “expected outcome” for each agreed; 3) Goal attainment levels are defined for each point on, typically, a 5-point scale (expected outcomes are usually scored 0; baseline is often -2, but may be -1 or 0 depending on potential for deterioration). Each level must be carefully described in a way that is relevant, observable, measurable and consistent with study design. Published standardised goals are available; and 4) A standardized statistical formula provides overall goal attainment. Benefits include: ease of use; relevant goals; no redundant items; assessment of multiple domains; provides quantifiable and applicable outcomes across different conditions and severities; potentially more sensitive measure than traditional scales. Limitations include: potential bias; appropriate goal selection and outcome prediction; observable changes may differ from pre-defined outcomes; time consuming; may require independent GAS assessors for blinded trials; may require “control” goals not affected by treatment; statistical issues around single overall score. In capturing those outcomes relevant to each individual patient, GAS has potential use in supporting product labeling claims and value assessment of a medicine by HTA and payers. PRM240 Avoiding and Identifying Errors and Other Threats To the Credibility of Health Economic Models Tappenden P , Chilcott J University of Sheffield, Sheffield, UK . . Health economic models have become the primary vehicle for undertaking economic evaluation and are used in various health care jurisdictions across the world to inform decisions about the use of new and existing health technologies. Models are required because a single source of evidence, such as a randomised controlled trial, is rarely sufficient to provide all relevant information about the expected costs and health consequences of all competing decision alternatives. Whilst models are used to synthesise all relevant evidence, they also contain assumptions, abstractions and simplifications. By their very nature, all models are therefore “wrong.” Whilst the presence of imperfect evidence provides the impetus for developing models, it is also the reason why we can never fully validate them. As such, the interpretation of estimates of the cost-effectiveness of health technologies requires careful judgements about the degree of confidence that can be placed in the models from which they are drawn. The presence of a single error or inappropriate judgement within a model may lead to inappropriate decisions, an inefficient allocation of health care resources and ultimately suboptimal outcomes for patients. This study sets out a taxonomy of threats to the credibility of health economic models. The taxonomy segregates threats to model credibility into three broad categories (1) unequivocal errors, (2) violations and (3) matters of judgement, and maps these across the main elements of the model development process. These three categories of threats to model credibility are defined according to the existence of criteria for judging correctness, the degree of force with which such criteria can be applied, and the means by which potential threats can be handled. A range of suggested processes and techniques for avoiding and identifying these threats is put forward with the intention of prospectively increasing the credibility of any given model. PRM241 Assessing Heterogeneity of Treatment Effect Using Real World Data Murray J F 1, Kadziola Z 2, Zagar A 1 1Eli Lilly and Company, Indianapolis, IN, USA, 2Eli Lilly Regional Operations GmbH, Vienna, Austria . . . . There is increasing scrutiny of pharmaceuticals on their value proposition as well as a growing demand for evidence on real world effectiveness once they are commercially available. There are many challenges in producing valid and reliable estimates of real world effectiveness. A major challenge is assessing a product’s effectiveness relative to why patients may respond differently to a treatment (i.e., identifying groups of patients exhibiting “Heterogeneity of Treatment Effect” (HTE) using subgroup identification methods). Assessing HTE is critical to understanding differences that may exist between the efficacy observed in randomized clinical trials and a product’s real world effectiveness. Understanding causes for HTE is required for correct attribution of any observed difference between efficacy and effectiveness to the product versus other sources (e.g., patient behavior); Not recognizing and accounting for HTE will confound assessment of a product’s performance, which ultimately affects its acceptance and use by payers, physicians, and patients. Failure to define and incorporate subgroups is a frequent criticism of systematic evidence reviews and comparative effectiveness research reports. However, the analytical methods for finding factors that define subgroups that explain HTE are challenging due to many known statistical issues (e.g., limited statistical power, multiplicity adjustments) Real world data exacerbates the analytical challenges due in part to biases (e.g., selection bias) and issues (e.g., data quality) inherent in the data. We will describe the data and bias challenges that create these analytical complexities for detecting the cause and magnitude of HTE when using real world data. We will present results from a simulation experiment that compared and validated several subgroup methods developed to address these data and analytical issues. We simulated 22 permutations of subgroups with known identification criteria and treatment effects to determine the performance of the methods. PRM242 Impacts of Epro Data Collection Mode Selection on Patient Inclusion Holzbaur E , Ross J , Wade M , Rothrock T Almac Clinical Technologies, Souderton, PA, USA . . . . Objectives: The Electronic Patient Reported Outcome (ePRO) data collection mode selected for trials is often based on efforts to minimize timelines, budgets, and patient burden. However, are sponsors inadvertently introducing bias into trial results in this selection process? This conceptual paper reviews common ePRO modes and explores patient groups that may be excluded. Methods: Common modes for ePRO data collection are reviewed. An assessment of potential patient groups that may be excluded is performed based on ePRO mode. Results: Common ePRO modes include telephone, web, and handheld device. As sponsors look to reduce costs, improve data quality, and reduce patient burden, industry has continued its shift towards patients using their own telephone, computer, tablet, or smartphone and away from sponsors provisioning these devices to patients. Choice of patient-provisioned device: Patients from certain geographic areas may be excluded where internet connections and cellular/mobile telephone reception is limited. Requiring patients to use their personal web/mobile device may exclude patient groups with certain economic, cultural, or demographic characteristics who live in rural or underdeveloped areas. Choice of sponsor-provisioned device: Logistics issues, i.e. shipment of devices including customs considerations, reliability of data transmission, storage and replacement of devices and cords, training, etc. Conclusions: The objective of clinical trials is to establish treatment effectiveness, generalizable to the overall patient population. ePRO mode selection may impact inclusion of individuals from certain economic, cultural, demographic, and geographic areas. Exclusion of these groups could impact results; therefore, it is important to understand the potential bias that can be introduced when selecting an ePRO mode. Proper planning should include assessment of patient population and inclusion of regions that would render generalizability. ePRO mode selection should be based on which method works best for the required regions to optimize inclusion, as well as the patient population’s characteristics to minimize burden. PRM243 Clinical Outcome Assessment (Coa) Instrument Scoring: the Validity and Precision of Unweighted Summary Scores VersUS Irt Weighted Scores, and the Added Value of Irt Standard Errors Coon C D 1, Lenderking W R 2 1Adelphi Values, Boston, MA, USA, 2Evidera, Lexington, MA, USA . . . . COA development experts in recent years have given thought to the psychometric evaluation of instruments and their ability to detect meaningful differences between patient groups. The scoring of the instruments, however, has received less attention, with various approaches sometimes suggested without a clear preference or justification. The score is ultimately used for evaluating patient outcomes and treatment efficacy and is what requires validation, so this seems like a significant omission. We examine the traditionally accepted unweighted summary score approach and compare it to the more complex IRT weighted scoring to evaluate if the gain in precision justifies the increased scoring complexity. Precision may differ depending on whether the score is close to the mean of the population or closer to the extreme ends of the distribution. Simulated data are used for this comparison to evaluate if the precision of the scores differs depending on the location of the score and if the instrument is used for group comparisons versus individual diagnosis. Additionally, we recognize that the reliability of a scale is likely to be variable across the range of its scores. With that in mind, we consider an approach to comparing mean scores between groups that incorporates the standard error of each individual IRT score into the model. By using the IRT standard errors, we can adjust for the different levels of uncertainty associated with ranges of scores along the scale, ultimately providing us greater confidence in the group comparison results. PRM244 Evaluation of Estimators of Treatment Effect in Observational Studies Faries D E 1, Lipkovich I 2, Kadziola Z 3 . . . . A586 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1Eli Lilly and Company, Indianapolis, IN, USA, 2Quintiles Innovation, Morrisville, NC, USA, 3Eli Lilly Regional Operations GmbH, Vienna, Austria Many methods are currently available to estimate treatment effects with observational data. We conducted simulations evaluating some well-established methods (regression, propensity weighting, stratification or matching on propensity) as well as some newer ideas (tree based methods, local control, entropy balancing, prognostic scoring) under several different scenarios including homogeneous and heterogeneous treatment effects. Mean square error, bias, coverage probability for the overall treatment effect, and prediction accuracy of personalized treatment contrast (for scenarios with heterogeneous treatment effect) were assessed. We will present some guidelines for estimating treatment effects with observational data and strategies that are appropriate with respect to (i) tree-structured treatment models (ii) polynomial outcome model with interactions (iii) presence of noise covariates. PRM245 Health Technology Assessment and Environmental Costs: Time for Health Care To Catch Up? Marsh K 1, Ganz M 2, Hsu J 3, Strandberg-Larsen M 4, Palomino Gonzalez R 4, Lund N4 1Evidera, London, UK, 2Evidera, Lexington, MA, USA, 3Harvard Medical School, Boston, MA, USA, 4Novo Nordisk A/S, Bagsværd, Denmark . . . . . Objectives: The United Nations recently published its most definitive report, calling for greater action on climate change. Historically, however, health technology assessment (HTA) has had a more narrow focus with emphasis on the health of patients and health inequalities. Recently some health care decision makers have extended the focus to include the environment, e.g. the Swedish Government is considering a Green Premium for generic drugs, and the UK NHS has CO2 emissions targets. We consider the case for incorporating environmental impacts into HTA, and the associated methodological challenges. Methods: We reviewed health care decisions where environmental impacts were considered - a summary will be provided in the paper. We then convened a workshop with key opinion leaders. Results: There are two lines of reasoning for incorporating environmental impacts into HTA: 1) Direct impact: changes in the environment could affect the health of individuals; and 2) Health decision makers’ objectives are broader and are informed by other policy goals, such as the CO2 targets adopted by the NHS in the UK. We also identified two types of methodological challenges for implementation. First, the nascent evidence base is insufficient to support the accurate comparison of the environmental impact of technologies. Second, uncertainty about how best to incorporate evidence into HTA. The cost-utility analysis approach favoured by many HTA agencies could capture some of the value of environmental impacts – in particular, those that generate health impacts. Both cost-benefit analysis and multi-criteria decision analysis have potential, having both previously been applied to evaluate both health and environmental interventions, though are less familiar to health care decision makers. Conclusion: Further work is needed to track decision makers’ demand for evidence on environmental impacts. Robust methods also are needed for capturing and incorporating environmental data as part of HTA as more decision makers begin incorporating environmental impacts. PRM246 Multiple Decision Criteria for Assessing An Incremental CostEffectiveness Ratio of Expensive Health Technologies Kamae I , Yamabe K , Sugimoto T The University of Tokyo, Graduate School of Public Policy, Tokyo, Japan . . Objectives: To develop a new method that naturally extends the UK NICE way of single-threshold for incremental cost-effectiveness ratio (ICER). It aims to provide multiple decision criteria for assessing an ICER of expensive health technologies such as molecular-targeted cancer drugs and regenerative medicine products. Methods: We took a theoretical approach, provided that the cost (C) -effectiveness (E) function, C = f (E), is known regarding the treatment alternatives for a disease area, given two points Pb and Ps plotted with a pair of C and E on the C-E plane, where Pb represents the best comparator, and Ps as the second best for a new technology X located at the point. Px (ex (effectiveness; known), cx (cost; assumed)). At first, given a single threshold of ICER, the “expensiveness” in the C-E function was defined using a tangent/derivative method (poster presentation PRM119 by Kamae I, et. al. in ISPOR Montreal 2014). Second, we estimated three benchmarks based on the C-E function: 1) the ICER of Pb to Ps, 2) the tangent at Pb, 3) the tangent at the point on the C-E curve which intersects with the vertical line at the point: (E, C) = (ex, 0). Then the magnitude relationship was examined between the three benchmarks and the ICER of the technology X defined by the slope of the line connecting Px with Pb. Results: Multiple decision criteria at six levels were identified and formulated as for acceptance of the “expensive” cost-effectiveness of a new health technology: 1) unconditional acceptance (simple dominance), 2) preferred (extended dominance), 3) less preferred, 4) minimally preferred, 5) not preferred, but negotiable, and 6) cannot accept. Example calculations clarified how the theory works in practical setting. Conclusions: Our approach offers multiple decision criteria to assess expensive health technologies as a natural extension beyond the NICE way of single-threshold assessment. PRM247 Efficacy, Effectiveness and the “Efficacy-To-Effectiveness Gap”: Review of the Current State of Play and Perspectives. First Results From the Imi Getreal Consortium Nordon C 1, Karcher H 2, Pichler F 3, Rossignol M 4, Abbe A 5, Abenhaim L 6 Lilly and Company, Windlesham, Surrey, UK, 4McGill University, Montreal, QC, Canada, 5SANOFI, Chilly-Mazarin, France, 6London School of Hygiene and Tropical Medicine, London, UK . . . . . . 1LASER Analytica, Paris, France, 2LASER Analytica, London, UK, 3Eli Background: The concept of “efficacy-effectiveness gap” (EEG) has gained awareness in the scientific community and started to erode the confidence in decisions taken on drugs: authorization, appraisal, and medical choice between alternatives. The Innovative Medication Initiative was launched the GetReal project to tackle this issue. It gathers representatives of more than 14 pharmaceutical companies, public institutions such as EMA, HAS, ZIN, NICE and Academic research teams and aims at (1) better understanding how evidence of efficacy and effectiveness should be considered and reconciled and (2) proposing operational solutions. Objective and method: We conducted a focused literature review to gain clarity and perspective on the concept of EEG: on which historical background it emerged, how it is understood and which solutions have been suggested to narrow it. Results: A disconnect between outcomes from clinical trials and information needed for clinical practice has been identified in the process of standardization of drugs assessment (Schwartz, 1967), evidence-based medicine (Feinstein, 1997), and knowledge dissemination (Lehman, 1995), and called the EEG. Several factors have been identified to explain it, including characteristics of real-life health care settings (physician and patient behaviours) and the weak generalizability of clinical trials due to their design. The need for a more systematic assessment of effectiveness is now widely acknowledged. Adaptive licencing was recently proposed to account of the sequential evidence generation on drugs outcome (Eichler, 2011). The EEG can be conceptualized as the interaction of drug effect and “real-life” contextual factors (Unutzer, 1999). Conclusions: Although the literature on the EEG is extensive, the contextual factors that actually impact drug’s outcome in real-life are still to be identified. Innovative and integrative study methods and designs are required to enable the EEG to be addressed adequately early on in the drug development process: this is the next step in the GetReal project. PRM248 Health Technology Assessments for Personalised Medicines: Are Current Methodologies Suitable for the Assessment of Personalised Therapies? Lauks S 1, Gee A 1, Wilson L E 2 Consulting, Reading, UK, 2Quintiles, Reading, UK . . . . 1Quintiles Objective: An increased drive towards personalised healthcare and medicine by policy-makers, alongside technological advances in medicines and diagnostics, is leading to more personalised medicines coming to market. Given that personalised medicines differ from traditional medicines in their development, use and cost, previously published articles have stated that current health technology assessments (HTA) methodologies are not designed to appropriately evaluate these technologies. This research was conducted to provide insights on methods for evaluating personalised medicines and what modifications to current HTA processes would be needed to ensure robust and timely assessment. Methods: Qualitative interviews were conducted with five experts in personalised medicine and market access across the UK, US and Germany to discuss the movement towards and benefits of personalised medicines as well as the key metrics on which they should be evaluated. These insights, supported with secondary research, were used to provide suggestions on the structure and methodology of personalised medicine assessments and how current assessment processes would need to be altered to accommodate these unique technologies. Results: The key areas where personalised medicines would need special consideration in HTAs identified were: - Study design: population size, geography, ethnicity - Companion diagnostics: cost, logistics - Unmet need: individualised view of perceived benefit - Cost effectiveness: costs and outcomes of therapy and companion diagnostic, reduction in overall health care costs Conculsion: The key areas identified are discussed in further detail, specifically, as to how they could be incorporated into current HTA models to effectively assess personalised medicines and how they would influence the decision-making process. PRM249 Challenges In Meeting Evidence Needs Of Payer, Physician, Patient And Industry Stakeholders For Novel Therapeutics Pauer L Santen, Inc., Emeryville, CA, USA . Objectives: Evidence development programs for novel therapeutics must simultaneously demonstrate safety, efficacy, clinical significance, economic value and effectiveness to meet diverse stakeholder requirements. This is further complicated by variable evidence needs across global markets and resource limitations. Understanding how to address all stakeholder perspectives for treatments in ophthalmology was accomplished through primary research. Methods: Conducting primary research through in-depth phone interviews and advisory board meetings with patient, payer and clinical stakeholders in three ophthalmic conditions: chronic, non-infectious posterior uveitis, wet age-related macular degeneration and glaucoma provided important insights. Patient research questions focused on identifying patient burden and unmet needs. Further research with physicians and payers was accomplished through evaluating diverse criteria applied to assessing clinical and economic evidence plans. Results: When therapeutic areas lack universally accepted clinical guidelines that can be relied upon to guide treatment decisions, payers rely heavily on clinicians to understand current standard of care and accepted endpoints. Clinician leaders, interestingly, were not in universal agreement. Moreover, physicians and payers differed on use of economic endpoints and appropriate therapeutic comparators. Physicians differed on the endpoints that would be most relevant for demonstrating treatment response and how to address the patient burden. Overall, clinicians were more willing to consider clinical trial endpoints that differed from those endpoints found in published data. Payers preferred to have consistent endpoints to facilitate indirect comparisons between treatments. Conclusions: The differing needs of payer and clinician stakeholders create additional barriers for development planning for novel therapeutics, particularly when published treatment guidelines are not available. Manufacturers must consider multi-stakeholder insights across global markets in clinical trial design development. 1Santen Pharmaceuticals, Emeryville, CA, USA. 2GfK Market Access, Wayland, MA, USA. A587 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PRM250 Pathways of Implementation of Multi-Criteria Decision Analysis Into Orphan Drug Approval Procedure for Drug Supply Programs in Russian Federation Serpik V G 1, Yagudina R I 2 Research Institute of Public Health, Moscow, Russia, 2First Moscow State Medical University named after I. M. Sechenov, Moscow, Russia . . . . 1National Background: While the orphan drug supply program is in progress, development of decision-making rules for approving orphan drug for supply program of Russian Federation becomes very actual. Real world data provides evidence, that routine approaches for approving such kind of drugs, e.i. pharmacoeconomic conclusions, are not applicable. Than the need in more appropriate approaches is existed. Multi-criteria decision analysis is one such approaches (MCDA). Objective: To evaluate prospective of implementation of MCDA in health care system of Russian Federation and to develop road map of MCDA in Russia. Methods: Literature review, cluster analysis, interviewing experts. Results: The first step (qualitive) to implement MCDA is to test various MCDA methods to find out optimal one for Russian Federation: it is expected to select the most relevant criteria from the wide range of them. First of all, MCDA is considered to be the instrument to improve the quality of discussion and its transparency, to underline different point of view and unmet needs. On the second stage it may be possible to use quantity MCDA assessment as a rule to approve orphan drugs for drug supply programs. Local recommendations for MCDA in Russian Federation has been published. Conclusion: Implementation of MCDA as assisting instrument for orphan drug approving for drug supply programs is likely to be a valuable approach, that may improve the quality, transparency of decision-making process and to provide social equity for accepting decisions. PRM251 Propensity Score Matching and Subclassification With Multi-Level Treatments Kadziola Z 1, Yang S 2, Imbens G W 3, Cui Z 4, Faries D E 4 1Eli Lilly Regional Operations GmbH, Vienna, Austria, 2Harvard School of Public Health, Boston, MA, USA, 3Graduate School of Business, Stanford University, and NBER, Stanford, CA, USA,, 4Eli Lilly and Company, Indianapolis, IN, USA . . . . . . . There is extensive literature on methods, such as propensity scoring, for estimating the causal effects for two treatments using real world data. Much less work has been done for the more general setting with three or more treatments. Whereas the literature has suggested that these propensity-based methods do not naturally extend to the multi-level treatment case, we show, using the concept of weak unconfoundedness, that adjusting for or matching on a scalar function of the covariates removes biases associated with observed covariates. We focused on subclassification and matching approaches as these have found to be effective for two treatments and are among the most popular methods in that setting. We apply the proposed methods to an analysis of the effectiveness of treatments for fibromyalgia from a prospective observational study. We also carried out a simulation study to assess the performance of those new methods relative to such approaches like: pairwise propensity score matching; matching on the Mahalanobis distance of all covariates; matching on the set of propensity scores (with the number of scores equal to the number of distinct treatment levels minus one (Rassen, 2013)); weighting on the inverse of the binary treatment propensity scores (McCaffrey, 2013). The simulations suggest that the proposed methods are simple and viable options for comparing the effectiveness of three or more treatments. RASSEN et al.: Matching by propensity score in cohort studies with three treatment groups. Epidemiology 24, 401–9. MCCAFFREY et al.: A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Stat. Med. 32,3388–414. PRM252 Getting To Reimbursement Faster: Combining Randomised, Pragmatic, and Observational Clinical Trial Data Alsop J Numerus Ltd, Wokingham, UK . Reimbursement authorities often require pharmaceutical companies to provide them with more than just placebo-controlled data from RCTs. Instead, they typically seek data from a wider “real-world” setting, where the focus is on generating evidence of comparative effectiveness. The natural temptation for many pharmaceutical companies is to provide this evidence from separate, post-market approval studies. However, this approach can be expensive and undoubtedly leads to delays in reimbursement. We propose that both the additional costs of evidence gathering and the delays between regulatory and reimbursement approvals could be reduced by combining the main design elements of randomised, pragmatic, and prospective observational studies into a single, integrated Phase 3/4 study. This single study approach would typically begin with a standard RCT phase where, for example, an initial cohort of patients would be randomised to receive either the investigational therapy or placebo. Either in parallel with or following this phase, a second patient cohort would be randomised under pragmatic clinical trial conditions with the aim of comparing the investigational therapy with placebo and a limited number of active comparator treatments. Lastly, a third (observational) cohort would be enrolled and allocated to a wider range of therapies, as per clinical practice. Data from the RCT cohort would be used to obtain limited regulatory approval. Following this, data from the pragmatic cohort, once available, would then be formally combined using standard statistical techniques with data from the RCT cohort in order to obtain a wider regulatory approval and possibly some form of conditional reimbursement. The pragmatic and observational cohorts would then provide the comparative effectiveness data to allow for reimbursement across different patient groups. We outline the strengths and weaknesses of this approach, and discuss its operational considerations. PRM253 An Epidemiologic Modeling Application To Pharmacoeconomics for Improved Health Care Planning Cid Ruzafa J , Cox A , Merinopoulou E , Baggaley R , Leighton P , Desai K Evidera, London, UK . . . . . . Epidemiologic and pharmacoeconomic models differ in terms of populations considered, mathematical techniques used, and questions addressed. A typical pharmacoeconomic model assesses chronic or acute conditions, uses Markov techniques, and considers a closed patient group receiving a defined therapy to assess incremental costs needed to achieve gains in quality adjusted life years. A typical epidemiologic model assesses vaccination or public health interventions for infectious disease using differential equations and considers open populations representing communities to estimate prevalence or numbers of disease cases averted. The manner of conducting sensitivity analyses also differs. In oncology, in which multiple lines of treatment are available, the epidemiologic approach has application to estimate the patient point prevalence or the number of patients who can start on a line of therapy over a certain time period, when this cannot be determined from clinical trials or registers (which usually focus on single lines of therapy or limited types of patients that are not representative of the overall patient population). The approach consists of conceptualizing an open population that incorporates incidence of the condition and the transition of patients through various lines of treatment until death, and uses systems of difference/differential equations. Parameterization is challenging if there are several prognostic factors to describe the patient population, multiple or complex treatment pathways, and a wide range of variability. Parameters are obtained from the published literature, analyses of database information, and/ or surveys to experts in the field. Steady state solutions of the model equations estimate point and period prevalence. This approach is applicable to gastrointestinal stromal tumours and multiple myeloma. Resulting estimates are important for budget impact analysis and health care services planning by reducing uncertainty associated with identifying the patient numbers eligible for a given treatment. Epidemiologic modelling permits a framework to estimate disease prevalence that is little used in pharmacoeconomics. PRM254 Non-Interventional Research Ethical Requirements in England and France: Shared Experience From A Binational Research Project Guillemot J , Boval M C , Gauthier A Amaris, London, UK . . . . Background: Ethical review for non-interventional research is progressively becoming part of research standards. This evolution ensures that participants in research are respectfully considered. In practice, information on ethical requirements for noninterventional research seems insufficient. Increasing and legitimate expectations from peer-reviewed journals regarding reviews by ethics committees sometimes challenge researchers. In this presentation, we share our experience of investigating ethical requirements for conducting a questionnaire-based research on physicians in France and England. Methods: This investigation consisted of a documentary analysis, including official guidance documents on ethical requirements, communications with institutions and publications reviews. Documents were identified using an ad hoc search on official websites. Publications were identified on PubMed. Findings: In England, the service of the National Research Ethics Service (NRES) serves as the ethics reviewer. It offers an informal preliminary review of the study protocol and estimates ethical risks associated with non-interventional research projects. Depending of the target population, the methods and the risk level associated with the research project, the NRES states whether a formal ethics application is necessary or not. In case of low risk projects the NRES supplies an email which can be used as a justification for peer-reviewed journals. In France, structures to support ethical reviews for non-interventional research are the result of an on-going reform. Comités de Protection de la Personne, or CPPs, fulfil the role of ethics reviewers although they were initially designed to collaborate for hospital-based research. Gaining ethical review in France was more complex due to the infrequent character of such request from the industry. Conclusion: This experience showed the increasing role of ethical requirements in non-interventional research. It is a domain in constant movement which calls for innovative approaches to compile and disseminate information regarding ethical requirements for non-interventional research across Europe and the world, especially regarding cross-national research projects. PRM255 Real World Studies, Challenges, Needs and Trends from the Industry Batrouni M , Comet D , Meunier J P Axonal, Paris, France . . . . Objectives: To understand key challenges, needs and trends for conducting real world studies (RWS). Methods: An online survey conducted in September 2013 within key players in the pharmaceutical and medical device industry in EU and US. 456 persons have been solicited through emails and phone calls, 107 have responded to the questionnaire. Respondents were mostly occupying senior positions in medical affairs, health economics and outcome research. Results: 27% RWS conducted are requested by Health Authorities, 73% on the industry initiative. 75% of those studies are subcontracted to a CRO. The main criteria of choice are the experience in RWS, particularly in the regulation process, the capacity to deliver on time and a flexible and adaptable structure. The RWS activity is expected to increase by 25 % in the next two years. Most of those studies have safety and effectiveness objectives and to a lesser extent drug utilization and health economics and the most common therapeutic areas are: oncology, cardiovascular and metabolic disorders. In addition, pharmaceutical companies are conducting more and more epidemiological studies to prepare dossiers for market access (disease understanding, unmet needs, population targeting). Conclusion: The pharmaceutical market is becoming global and is expanding into new countries and therapeutic areas. The result is an increase in the need for RWS where the regulatory agencies are asking for additional data A588 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 concerning the long term safety and effectiveness of the drugs when used on larger populations. Pharmaceutical companies face big challenges for the coming years especially in EU and there is an increase need for local regulatory knowledge. There’s still need to increase awareness for the importance of real world studies and the impact it has on the patient’s life. PRM256 Publication Manual of Budget Impact Analysis (Bia) by the Department of Science and Technology of the Ministry of Health (Decit) Koury C D N 1, Elias F T S 2 1FIPE -Fundação de Ensino e Pesquisas Econômicas, Brasilia, Brazil, 2Ministry of Health of Brazil, Brasilia, Brazil . . . . . . The epidemiological and economic methods applied to health technologies evaluations had a significant development in the last two decades. The need to balance the incorporation of new technologies in health care and limited financial resources promoted the construction and application of instruments supporting the decision making of health technology. The requirement Budget Impact Analysis formally stated in Law 12.401/2011 establishing the incorporation process technologies in SUS. In this context, in 2010/2011, the National Agency of Sanitary Surveillance (ANVISA) and DECIT, in partnership Institute for Health Technology Assessment (IATS) for drawing up of this guideline. In the first stage of development were used international recommendations of Canada, Australia, the UK and Poland, the recommendations of the International Society for PharmacoEconomics and Outcomes Research (ISPOR) and the methods used in studies of budgetary impact that had already been published. Afterwards, drafted a preliminary version of the Guideline and a standard tool - Excel worksheets - to estimate the uptake of monetary resources required for adoption of new technologies. Revisions were carried out by technicians DECIT and health agencies, and the proposal was submitted to the Working Group on Development of Methodology REBRATS, composed of experts and academic researchers from several Brazilian states. Were also carried out workshops for the application of spreadsheets. In 2012, the first edition of the Guidelines was published two thousand copies in Portuguese in order to provide best practice recommendations for studies of budget impact. DISEASE - SPECIFIC STUDIES RESPIRATORY-RELATED DISORDERS – Clinical Outcomes Studies PRS1 Prospective Study on Cost-Effectiveness of Nurse Interviw Introducing Retesting With in Vitro Diagnostics (IVD) To Parents of Children With Suspected Food Allergy in Finland Hermansson L L 1, Korhonen K 2, Silvan M 2, Rantanen S 2, Isoaho R 3, Savolainen J 3 1Thermo Fisher Scientific, Uppsala, Sweden, 2Härkätie Primary Care Center, Lieto, Finland, 3University of Turku, Turku, Finland . . . . . . . Objectives: Accordance to Finnish Allergy Program 2008-2018, to decrease food avoidance diets by 50%. Focus in algorithm with patient history +IVD in school children with suspected food allergy and reason for declining re-diagnosis. NICE clinical guideline (Food Allergy Diagnoses, 2011) suggested further work made on effect of diagnosing allergies in realistic population and cost effectiveness of retesting. Methods: Prospective trial with patients from Finnish primary care database (2885 school children). School kitchen had allergy restricted diets for 179 children. In the pilot phase, 179 families were contacted by letter. Of the 24 who were included in pilot, 17 were not allergic (70%). In this study families were interviewed by telephone. Of 156 families 107 agreed to participate in this study and 47 children will be diagnosed by component resolved diagnostics (CRD) and 60 with sIgE and CRD. Results: Prevalence of food avoidance diets: 6,2%. Reasons for declining re-testing: 23 were not allergic, 9 were busy, 9 have own physician, 3 did not believe allergy tests, 8 scared of needles, 7 already tested, 4 tested often due to health problems, 2 in pilot study and 7 did not recognize a benefit. Conclusions: Telephone consultation by nurse decreased special diets for 23 children (13%) and 39 (22%) had non-medical reason to decline retesting. Nurse consultation to introduce retesting with IVD can be considered as cost effective approach in decreasing food avoidance diets in children. PRS2 Effectiveness of Montelukast on Asthma Control in Infants: A Claims Data Study Belhassen M 1, Ginoux M 1, Laigle V 2, chanut-Vogel C 2, Lamezec L 2, de Blic J 3, Fauroux B 3, de Pouvourville G 4, Laforest L 1, Van Ganse E 1 1University of Lyon, Lyon, France, 2Laboratoires MSD France, Courbevoie, France, 3Pediatric Medicine Necker University, Paris, France, 4ESSEC Business School, Cergy, France . . . . . . . . . . Objectives: Montelukast 4mg (MTL-4) is an add-on therapy for asthmatic infants. Given the quality and exhaustivity of the data, French claims data (SNIIR-AM) is a relevant tool to investigate MTL-4 effectiveness in infants. The objective was to compare the effectiveness of MTL-4, associated or not with ICS, vs. ICS without MTL-4, on health outcomes of infants with mild to moderate uncontrolled asthma. Methods: Infants (6-24 months) receiving ≥ 2 consecutive dispensing of respiratory drugs from 2010 to 2011, and presenting an initial exacerbation within 6 months of the first dispensing were preselected. Asthma-related outcomes included hospitalizations, dispensing of oral corticosteroids, addition of short-acting beta agonists to existing respiratory therapy, switch to a higher ICS dosage, or nebulized CS. The studied groups were infants receiving MTL-4 +/- ICS (MTL-4 group) and infants receiving ICS without MTL-4 (ICS group). The two groups were matched, e. g. on initial therapy before initial exacerbation and past asthma related hospitalization. The two groups were compared, as to the occurrence of a new exacerbation and the total num- ber of exacerbations during the 6 month follow-up following initial exacerbation. We also compared health care utilization between both groups. Results: Among 115,489 infants (mean age: 13.9 months; 62.9% boys), 4,477 infants of the MTL-4 group were matched with 13,386 infants of the ICS group. In multivariate analysis, the risk of a new exacerbation was lower in infants of MTL-4 group compared to infants in ICS group (HR= 0.91, IC95% [0.87; 0.95]). The total number of exacerbations did not differ between the 2 groups during the 6-month follow-up (p= 0,8617), neither the cost of asthma management (344€ for MTL-4 group vs. 308€ for ICS group, p= 0.1410). Conclusions: MTL-4 and ICS appear to be comparable therapeutic strategies, with similar effects on exacerbation and equivalent costs. The SNIIR-AM allows conducting comparative effectiveness research. PRS3 Clinical Trial-Based Cost-Effectiveness Analysis of Indacaterol (ONBREZ® 150 MCG) Versus Tiotropium (SPIRIVA®) in the Treatment of Chronic Obstructive Pulmonary Disease (COPD) IN TURKEY Saylan M , Beykoz V , Keskinaslan A Novartis Pharma, Istanbul, Turkey . . . Objectives: COPD is a disease that is characterized by chronic and progressive restriction of the airflow. The cost of COPD medications can be reduced significantly by implementing a treatment algorithm that is consistent with the GOLD guidelines. Indacaterol and tiotrpium administered by inhalation are indicated for maintenance treatment of COPD in Turkey. We aimed to compare, from the perspective of the Turkish social security institution, the cost-effectiveness of indacaterol 150 mcg once daily and long-acting tiotropium 18 mcg once daily at months 3 and 6 in patients with moderate to severe COPD aged 30 years and above. Methods: From payer perspective, a cost-effectiveness analysis based on two separate clinical trials (INTENSITY-once daily indacaterol and tiotropium vs. placebo and INHANCEindacaterol vs tiotropium) was performed. The primary endpoints of the clinical trials (Trough FEV1, Transition Dyspnea Index [TDI] and Saint Georges Respiratory Questionnaire [SGRQ]) were included in the cost-effectiveness analysis. Incremental cost effectiveness ratio (ICER) of indacaterol vs. tiotropium for different treatment success criteria (week 12 FEV1 > 0.12L increase, ≥ 1 improvement in TDI score, ≥ 4 decrease in SGRQ score) were compared. Incremental cost effectiveness ratios were calculated over incremental differences versus placebo. Probabilistic sensitivity analysis was performed using the Bootstrap method. Results: FEV1success rates at month 3 for indacaterol and ipratropium were 26.5% and 24.3%, respectively. At month 3, ICERs of indacaterol versus ipratropium were -1002TL for FEV1, -434TL for TDI and -878TL for SGRQ. At month 6, FEV1 success rates were 54.8 and 47.4%, TDI success rates were 58.7% and 54.4% and SGRQ success rates were 81.8% and 77.1%, respectively. ICERs of indacaterol versus ipratropium at month 6 were -616TL for FEV1, -1049TL for TDI and -1014TL for SGRQ Conclusions: Based on this clinical trial-based analysis, indacaterol was cost effective treatment and cost reducing choice vs. tiotroprium in COPD treatment. PRS4 A Network Meta-Analysis Comparing the Efficacy And Safety of Ceftobiprole and Selected Comparators in the Treatment of Hospital-Acquired Pneumonia Pooley N 1, Chadda S 1, Madrigal A M 2, Kuessner D 3, Posthumus J 4 1PHMR Associates, Newcastle upon Tyne, UK, 2PHMR Associates, London, UK, 3Basilea Pharmaceutica Ltd, Basel, Switzerland, 4Basilea Pharmaceutica International Ltd., Basel, Switzerland . . . . . . Objectives: Hospital-acquired pneumonia (HAP) is a severe respiratory tract infection which develops more than 48h after hospital admission. Ceftobiprole, the active moiety of its prodrug ceftobiprole medocaril, is a new cephalosporin with bactericidal activity against a broad spectrum of pathogens including resistant bacteria such as methicillin-resistant S. aureus (MRSA), penicillin-resistant pneumococci andP. aeruginosa. Ceftobiprole was shown safe and effective for the treatment of HAP (excluding ventilator-associated pneumonia), when compared with linezolid plus ceftazidime in a large-scale phase-III clinical trial (NCT00210964). Methods: MEDLINE, EMBASE, Medline-In-Process and the Cochrane Library were searched for randomised controlled trials that included ceftobiprole and/or comparators ceftazidime, meropenem, imipenem/cilastatin, piperacillin/tazobactam, ciprofloxacin, levofloxacin, moxifloxacin and gentamicin as intervention in the treatment of HAP. The efficacy of ceftobiprole was compared to comparators using a random effects model implemented within a fully Bayesian framework. Primary outcome was clinical response after end of treatment in the clinically evaluable (CE) population. Results: Eleven studies (2413 patients) with HAP were included in the analysis, 1618 patients were eligible for analysis of clinical response in the CE population. The comparative efficacies (odds ratio, 95% credible intervals) of ceftobiprole to each comparator were 0.92,0. 092-8.8 (ceftazidime), 1.1, 0.054-19 (piperacillin/tazobactam), 1.9, 0.12-30 (meropenem), 0.83, 0.019-32 (levofloxacin), 0.96, 0.047-16 (imipenem/cilastatin), and 0.87, 0.025-22 (ciprofloxacin). No comparison was possible to gentamicin or moxifloxacin due to a lack of comparative studies against other comparators. No significant difference was seen between ceftobiprole and any comparator in clinical response or in any of the secondary outcomes, including mortality and adverse events Conclusions: The results of this multi-treatment comparison support the comparable efficacy and safety of ceftobiprole to relevant comparators in the treatment of HAP. This analysis was limited by the small number of available studies, and by the fact that among the drugs compared, only ceftobiprole provides coverage of MRSA. PRS5 Comparative Efficacy of Umeclidinium Bromide Versus Other LongActing Anticholinergic Monotherapies as Treatments for Copd Patients Ismaila A 1, Huisman E 2, Punekar Y S 3 Park, NC, USA, 2Mapi - HEOR & Strategic Market Access, Houten, The Netherlands, 3GlaxoSmithKline, Uxbridge, UK . . 1GlaxoSmithKline, Research Triangle . . A589 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To assess the relative efficacy of umeclidinium bromide 62.5 mcg OD (UMEC) versus tiotropium bromide 18 mcg OD (TIO), aclidinium bromide 400 mcg BID (AB) and glycopyrronium bromide 50 mcg OD (GLYCO). Methods: A systematic literature review was performed to identify RCTs ≥ 12 weeks duration comparing TIO, AB, GLYCO or UMEC to placebo in adult patients with COPD. Random effects meta-analyses were performed by pooling results of each treatment vs. placebo on change from baseline at 12 and 24 weeks in trough FEV1, SGRQ total score, TDI focal score and rescue medication use. The results were synthesized by using an indirect treatment comparison (ITC) within a frequentist framework based on the Bucher method. Scenario analyses were performed to evaluate the robustness of the results to variations in the included studies and assumptions. Results: At 12 weeks, ITC results show that treatment with UMEC resulted in a comparable but numerically higher change from baseline in trough FEV1 compared to TIO [18.06mL (95%CI: -19.11, 55.23, p=0.341)], AB [35.77mL (95%CI: -7.84, 79.38, p=0.108)] and GLYCO [27.86mL (95%CI: -8.74, 64.45, p= 0.136)]. At 24 weeks, UMEC resulted in comparable trough FEV1 values vs. TIO (p= 0.854), AB (p= 0.663) and GLYCO (p= 0.777). UMEC also resulted in comparable TDI focal scores and rescue medication use at both time points compared with TIO, AB and GLYCO. UMEC resulted in numerically lower (better) change from baseline at 12 weeks in SGRQ total score compared with TIO [-2.65 (95%CI: -7.09, 1.79, p= 0.242)], AB [-2.68 (95%CI: -7.12, 1.75, p= 0.235)] and GLYCO [-2.15 (95%CI: -6.60, 2. 31, p= 0.345)]. At 24 weeks there was no statistically significant difference in change from baseline in SGRQ total score between UMEC, TIO, AB and GLYCO. Conclusions: UMEC showed comparable efficacy to TIO, AB and GLYCO on trough FEV1, SGRQ, TDI and rescue medication use at 12 and 24 weeks. PRS6 Systematic Review of Observational Studies and Rcts of Omalizumab in Severe Persistent Allergic Asthma and MetaAnalysis Feasibility Assessment Bergrath E 1, Hwa Ong S 2, Bousquet J 3, Balwin M 4, Manga V 2, Rao S 2, Cope S 5 Pharma AG, Basel, Switzerland, 3University of Montpellier, Montpellier, France, 4Novartis Pharma AG (Employed by Novartis at time of systematic literature review), Basel, Switzerland, 5Mapi, Toronto, ON, Canada . . . . . . . 1Mapi, Inc., Boston, MA, USA, 2Novartis Objectives: To compare the effectiveness of omalizumab versus standard of care (SOC) based on randomized controlled trials (RCTs) compared with ‘real-world’, single cohort, observational studies that assess patients ‘before and after’ the use of omalizumab. Methods: A systematic literature review was conducted to identify RCTs and observational studies that assessed omalizumab in patients with severe persistent allergic asthma. Study and patient characteristics, outcome definitions, and differences in baseline risk and observed study effects were compared in terms of exacerbations and hospitalizations across the RCTs and observational studies. Results: 11 RCTs and 24 observational studies were identified. A wide range of clinically significant exacerbation rates was observed across RCTs in terms of baseline risk (SOC: 0.40–2.86) and the treatment effect (rate ratio [RR]: 0.39–0.75). This differed from observational studies in terms of baseline risk (before omalizumab: 3.48–6.00) and the treatment effect (RRs: 0.12–0.46). A limited range of severe exacerbation rates was observed in RCTs regarding baseline risk (SOC: 0.42–0.48) and the treatment effect (RR: 0.50–0.56). However, considerable differences were identified in observational studies in terms of baseline risk (before omalizumab: 2.20–4.50) and the treatment effect (RR: 0.05–0.39). In terms of hospitalization rates, a limited range was observed for RCTs with respect to baseline risk (SOC: 0.12–0.17) and the treatment effect (RR: 0.12–0.54). Again, a wider range was observed across the observational studies in terms of baseline risk (before omalizumab: 0.32–4.45) and the treatment effect (RR: 0.09–0.71). Conclusions: ‘Real-world’ evidence reinforces the efficacy of omalizumab in patients with severe allergic asthma derived from RCTs, although differences in potential treatment effect modifiers were identified. Patients in observational studies may represent a more severe population compared with those in RCTs. PRS7 Impact of Omalizumab on Poor Asthma Control Events and Medication Utilisation in Patients With Moderate or Severe Persistent Asthma Yu T C 1, Nazareth T 1, Turner S J 2, Raimundo K 3, Zhou H 4, Ortiz B 1, Li L 5 1Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA, 2Novartis Pharmaceuticals, East Hanover, NJ, USA, 3Genentech Inc., South San Francisco, CA, USA, 4KMK Consulting Inc., Florham Park, NJ, USA, 5Career International Inc., Shanghai, China . . . . . . . . . Objectives: Poor asthma control is associated with increased health care cost in patients with moderate or severe asthma. Here we evaluate the impact of omalizumab on poor asthma control events (PACE) and medication utilisation (MU) in a case-crossover study of US patients with moderate or severe persistent asthma. Methods: Truven MarketScan database was used to compare PACE (hospitalisation, ER visit, corticosteroid [CS] burst or ≥ 7 short-acting beta-agonist [SABA] fills) and MU for 1 year pre/post omalizumab exposure, during the period 1-January-2007 to 30-September-2012. Included in the analysis were patients aged ≥ 12 years who had 1 inpatient or 2 outpatient Asthma claims (ICD-9= 493. XX) and used omalizumab continuously for 1 year, with 2 years continuous coverage (1 year pre/post omalizumab index date). Patients were categorized as Moderate or Severe based on their most recent 8 weeks of continuous, NHLBI-guideline-recommended, therapy preceding omalizumab. Results: In total, 429 patients (mean age, 46.6 years; female, 59.0%; Moderate= 340, Severe= 89) were included in the analysis. Omalizumab was associated with reductions in proportions of All, Moderate, and Severe asthma patients with PACE (41.3%, 48.3%, 17.2%, respectively; all p< 0.05). Specifically, reductions in patients with ≥ 1 asthma-related hospitalisation, ≥ 1 asthma-related ER visit, ≥ 2 CS bursts, and ≥ 7 SABA fills in the Moderate group (Moderate: 55.9%, 77.8%, 53.8%, and 40.6%, respectively; all p≤ 0.0196) drove reductions in All patients (all p≤ 0.0159). Reductions in patients with ≥ 1 OCS fill and ≥ 1 SABA fill were observed in All and Moderate patients (20.3%–26.1%; all p< 0.0001); reductions in mean OCS fills and mean SABA fills were observed in All and Moderate patients (26.0%-42.5% p< 0.0001), while reductions in mean SABA fills were also observed in Severe patients (20.1%; p= 0.0328). Conclusions: Omalizumab initiation was associated with significant reduction in PACE and MU in patients with moderate or severe persistent asthma. PRS8 Indirect Comparison of Exacerbation Frequency Between Aclidinium and Tiotropium in Patients With Chronic Obstructive Pulmonary Disease Lee H , Choi S E , Bae E , Lim E A , Kim J , Park H Korea University, Yeongi-gun, South Korea . . . . . . . . Objectives: The purpose of this study is to compare the frequency of exacerbations between aclidinium and tiotropium in patients with chronic obstructive pulmonary disease (COPD). Methods: Through a systematic literature search in Medline (PubMed), we included randomized controlled trials that evaluated the exacerbation frequency of aclidinium 200µg and 400µg twice a day and tiotropium 18µg once a day regimens compared to placebo. Inclusion criteria were at least 12 weeks of treatment from January 1990 to January 2014, an age over 40, current or former smokers, and diagnosis with moderate to very severe COPD. The main outcome is the frequency of exacerbation. Indirect comparison analysis was performed to estimate the odds ratio of exacerbation between aclidinium and tiotropium. Results: After screening 278 full-text articles, we identified 19 clinical trials that total 19,741 COPD patients were participated: 3 trials of aclidinium 200µg and 400µg BID and 16 trials of tiotropium 18µg QD. tiotropium 18µg was associated with a significant reduction in exacerbation compared with placebo (OR 0.90; 95% CI 0.84 to 0.96). Other two anticholinergic agents showed comparable effects in reducing exacerbation compared with placebo: aclidinium 200µg (OR 0.73; 95% CI 0.53 to 1.01) and aclidinium 400µg (OR 0.72; 95% CI 0.52 to 1.00). Aclidinium 200µg (OR 0.84; 95% CI 0.603-1.167) and aclidinium 400µg (OR 0.83; 95% CI 0.592 -1.156) BID showed the similar frequency of exacerbation to tiotropium 18µg QD. Conclusions: Our study substantiates that tiotropium 18µg provides superior effects on lowering the risk of exacerbation compared with placebo but there was no significant difference in the frequency of exacerbations between aclidinium and tiotropium. PRS9 Treatment Plan Comparison: An Observational Study of the Marche Region Sciattella P 1, Marcellusi A 2, Mennini F S 3 1University of Rome “Tor Vergata” Italy, Rome, Italy, 2University of Rome, Rome, Italy, 3University of Rome “Tor Vergata”, Italy, Rome, Italy . . . . Objectives: To estimate the number of users of Theophylline (ATC: R03DA04) and Doxofylline (ATC: R03DA11) for the treatment of chronic asthma in adults, in the Marche Region. Moreover, we wanted to estimate the cost of the two treatments, taking into account the prescriptions of other drugs associated with them. Methods: The drug prescriptions were extracted from the Information System of the Pharmaceutical Prescriptions of the Marche Region (PHARM), containing all the recipes sent by pharmacies within the region and reimbursed by the National Health System. The number of prescriptions per year has been obtained by selecting all the recipes for each ATC code in the years 2008-2012, while the number of users has been estimated by identifying the subjects who received at least one prescription of the ATC codes of interest. The number of concomitant prescriptions was estimated by selecting all the recipes for potentially associated ATC, dispenced between 5 days before and 5 days following the prescription of ATC codes. The price of prescriptions has been calculated using the information “price” contained in the PHARM record. Results: For both drugs, the users are approximately 5,000 per year in the study period. Theophylline had a mean base price lower than Doxofylline (4.81€ vs 6.37€ per prescription); however, Theophylline was more associated than Doxofylline (34.4% vs 23.7%) with other drugs for the treatment of Asthma. Consequently, the total treatment cost for Theophylline was equal to 33.65€ vs a total cost for Doxofylline equal to 22.49€ (+ 49.6%). Conclusions: The PHARM allows the estimate of drugs’ utilization, taking into account the overall patient’s treatment plan. In our study, the prescription of the first ATC code is more associated with prescriptions of other drugs, and this implies an increasing in the cost of the treatment plan despite a lower average initial price. PRS10 A Database Study to Investigate the Incidence of Anaphylaxis and the Prescription Rate of Self-Injection Epinephrine in Japan Shima D 1, Ii Y 1, Yamamoto Y 2, Nagayasu S 2, Fujimoto Y 1 1Pfizer Japan Inc, Tokyo, Japan, 2MinaCare co. ltd, Tokyo, Japan . . . . . Objectives: A database research was conducted to investigate the incidence of anaphylaxis/shock using a Japanese health-claims database (HDB). In addition, the prescription rate of self-injection epinephrine was investigated among those patients with anaphylaxis for the management of future reactions. Methods: A Japanese HDB which contains approximately 1.8 million subjects covered by employment-based health insurance (MinaCare Co. Ltd) was used for this retrospective study. In order to identify actual anaphylaxis/shock precisely, diagnosis recorded in the claims based on ICD-10 code (T78.0, T78.2 and T88.6) was combined with claim records of medical practice and prescriptions. Specifically, prescription for epinephrine/adrenaline or oxygen inhalation therapy was required for “anaphylactic shock” and the use of an infusion therapy or venous catheter was required for detecting “anaphylaxis (except for anaphylactic shock)”. For this study, the data associated with events occurring in fiscal years 2010 to 2013 (2010/4/1 to 2013/3/31) were included. Results: Of approximately 2.9 million person-years of observations, 13.3 anaphylactic shock events per 100,000 person-years (crude rate) were identified. The rate was 42.9 per 100,000 person-years when non-shock anaphylaxis events were considered. The age-specific anaphylactic shock event rates (per 100,000 personyears) were: 27.6 (0-6 years), 12.8 (7-12 years), 11.0 (13-18 years), and 11.9 (> 18 years). Among the 389 anaphylactic shock events, etiologies of anaphylaxis were food 113 A590 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 (29.0%), venom 31 (8.0%), drug 28 (7.2%), other/unknown 217 (55.8%). The prescription rate of self-injection epinephrine immediately after the anaphylaxis was 10.9% (after the insurance started to cover the service in 2011/9). Conclusions: This is the first report on the incidence of anaphylaxis in general population in Japan; the rates were consistent with those reported in the West. The prescription rate of self-injection epinephrine was notably low even among those who experienced anaphylaxis, indicating the importance of increasing the awareness of the availability of life-saving anaphylaxis management. PRS11 Epidemiology and Severity of Chronic Obstructive Pulmonary Disease (COPD) in the United Kingdom (UK) Raluy-Callado M1, Lambrelli D1, MacLachlan S1, Merinopoulou E1, Hagan MA2, Khalid JM3 1Evidera, London, UK, 2Takeda Pharmaceuticals International, Inc., Deerfield, IL, USA, 3Takeda Development Centre Europe, Ltd., London, UK Objectives: In 2013, the Global Initiative for Chronic Obstructive Lung Disease (GOLD) recommendations on COPD management and prevention were updated to include additional therapeutic options for patients based on severity using a combined assessment of symptoms, degree of airflow limitation and number of exacerbations. The objective of the present study was to quantify the prevalence and incidence of COPD in the UK and estimate disease severity by 2013 GOLD categories A/B (low risk) and C/D (high risk). Methods: Patients with a diagnosis of COPD aged ≥ 40 years were identified in the population-based Clinical Practice Research Datalink. Point prevalence was calculated on December 31, 2013. Incidence was estimated using newly diagnosed patients between 2009-2013. Rates were standardised using 2011 UK population age and gender. % predicted FEV1, modified British Medical Research Council grade and exacerbations defined by Read codes and prescriptions were used to classify patients by GOLD categories. Patient characteristics were reported. Results: 49,286 prevalent patients were diagnosed with COPD with mean age of 70 years; 51% were male. Median time since diagnosis was 5 years. Overall prevalence was 33.0 per 1,000 people (95% CI: 32.7-33.4). Of these, 66.4% were classified as GOLD A/B and 33.6% as GOLD C/D. 27,224 newly diagnosed patients were identified with mean age of 67 years at first diagnosis; 53% were male. Incidence was 2.2 per 1000 person-years (95% CI: 2.1-2.2). Conclusions: A third of COPD patients in the UK are considered high-risk according to the 2013 GOLD categories. Classification of patients is key to identifying appropriate treatment options to reduce symptoms and the frequency of COPD exacerbations. PRS12 Incidence and Prevalence of COPD By Gold 2013 Classification in the Netherlands Van den Berg E J 1, Overbeek J A 1, Penning-van Beest F J A 1, Khalid J M 2, Dekhuijzen P N R 3, Herings R 1 1PHARMO Institute for Drug Outcomes Research, Utrecht, The Netherlands, 2Takeda Development Centre Europe, Ltd., London, UK, 3Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands . . . . . . . . . . . . . Objectives: To quantify the five-year incidence (2008-2012) and 2012 prevalence of COPD in The Netherlands by the Global initiative for chronic Obstructive Lung Disease (GOLD) 2013 combined assessment categories. Methods: Using the General Practitioners Database of the PHARMO Database Network, the five-year incidence (2008-2012) and prevalence at July 1, 2012 of COPD (ICPC code R95) by GOLD 2013 combined assessment categories among individuals ≥ 40 years of age was assessed. Based on degree of airflow limitation (using post-bronchodilator FEV1) and risk of exacerbations (based on medication or as recorded by the GP) patients were classified as low-risk COPD (FEV1 ≥ 50% and/or ≤ 1 exacerbations) or high-risk COPD (FEV1 < 50% and/or ≥ 2 exacerbations). Results: Using a source population of 813,800 individuals ≥ 40 years of age the five-year (2008-2012) incidence (95% CI) of COPD among patients ≥ 40 years of age was 0.50 (0.49-0.50) per 100 person years; this was 0.54 (0.53-0.55) among males and 0.45 (0.44-0.46) among females. The 2012 prevalence of COPD in a source population of 805,112 individuals ≥ 40 years of age was 3.7 (3.6-3.7) per 100 persons; this was 4.0 (3.9-4.1) among males and 3.4 (3.3-3.4) among females. Mean (± sd) age of incident and prevalent COPD patients was 65 ± 12 and 67 ± 12 years, respectively. The distribution of lowrisk COPD and high-risk COPD was 90% versus 10%. For patients treated by their GP this distribution of low-risk and high-risk COPD was similar, while patients treated by a specialist had a distribution of 82% versus 18%. Conclusions: This study describes the epidemiology of COPD in the Netherlands. Results on the distribution of low-risk and high-risk COPD depend on the population studied and the definitions used. Additional information on symptoms would allow a more detailed classification of patients. PRS13 Estimating Smoking Cessation Rates and Smoking Prevalences Using Public Data and a Published Dynamic Model Kumar R , Govindan B Vantage Research, Chennai, India . . al1 Objectives: Mendez et developed a dynamic forecasting model to predict the prevalence of smoking. They use initiation and prevalence rates (data from 1965-1993) and estimate cessation rates for that period. Further, they assume the persistence of the cessation rates and predict future prevalence of smoking. We re-created the Mendez et al model in order to estimate smoking cessation rates (n) and updated cessation rates for 2000-2012 using newly available data on smoking prevalence (R). Further, smoking prevalences for the decade 2012-2024 were predicted based on a couple of alternate hypotheses of smoking initiation Methods: We re-created the Mendez model in Excel, including the mathematical manipulations for estimates of mean rates of cessation from the years 1970-1994 and used the re-created Mendez model set-up and newly available data on smoking prevalence (R), initiation etc. to estimate cessation rates between 2001 and 2012. Further, we predicted smoking prevalences for 2012-2024 and explored a couple of different scenarios of smoking initiation. Birth and death rates of the general population are assumed to be the average of the previous decade. Relative Risk of mortality is assumed to be the same. Results: Smoking prevalence in 2024 is estimated to be 18.5%, even with alternate hypotheses around smoking initiation in the 2012-2022 decade. Conclusions: Aging of smoker population will continue to contribute to reducing prevalence of smokers. Existing dynamic forecasting models were re-created and used to estimate smoker prevalence using recent data. References 1. Mendez D , Warner K E , Courant P N , Has smoking cessation ceased? Expected trends in the prevalence of smoking in the United States. Am J Epidemiol. 1998 Aug 1;148(3):249–258. PRS14 Patients With COPD who Initiate Roflumilast in Sweden Khalid J M 1, Mushnikov V 2, Vattulainen P 2, Johansson G 3, Korhonen P 2, Hoti F 2 1Takeda Development Centre Europe, Ltd., London, UK, 2EPID Research, Espoo, Finland, 3Uppsala University, Uppsala, Sweden . . . . . . . Objectives: In Sweden, reimbursement for roflumilast is limited to eligible patients with severe to very severe chronic obstructive pulmonary disease (COPD) who are intolerant to inhaled corticosteroid therapy. Therefore, patients being treated with roflumilast in Sweden may differ by their characteristics to patients in other countries. Our aim was to describe demographic and disease characteristics of patients with COPD prescribed roflumilast in Sweden, at time of first prescription. Methods: Patients with diagnoses of COPD or chronic bronchitis (CB) who initiated use of roflumilast at age ≥ 40 during 2011 were identified from the Swedish Hospital Discharge Register and Swedish Prescribed Drug Register. Summary statistics were calculated for demographic and disease characteristics at time of first prescription. The Charlson comorbidity index (CCI) was used to score the presence of comorbidities. Results: 1,161 patients (42.8% male) with COPD/CB initiated roflumilast. Mean age at time of first prescription was 70.8 (SD±8.3) years. Mean time since first COPD diagnosis was 6.1 (SD±3.8) years. 640 (55.1%) patients were hospitalized at least once in the year prior; 99 (8.5%) had ≥ 5 hospitalisations in this period. 21.7% of patients had congestive heart failure, 13.5% myocardial infarction, 17.1% diabetes, and 9.1% a mood disorder indicated in their medical records. 37.2% of patients had a CCI score ≥ 3 at time of first roflumilast prescription. Conclusions: Patients who are prescribed roflumilast carry a very severe disease burden. Appropriate methodology should be used when making comparisons between patients who are exposed to roflumilast to those who are not, using real world data in Sweden. PRS15 The Prevalence of Tobacco Smoking in Patients With Diabetes in Hospital Pulau Pinang, Malaysia Albaroodi K A I 1, Syed Sulaiman S A 2, Shafie A A 3, Awaisu A 4, Lajis R 1 1University Sains Malaysia, Pulau Penang, Malaysia, 2Universiti Sains Malaysia, Minden, Malaysia, 3Universiti Sains Malaysia, Penang, Malaysia, 4Qatar University, Doha, Qatar . . . . . . . . . Widespread evidence has demonstrated the negative effects of tobacco smoking in patients with diabetes. Although many studies have explored the prevalence of tobacco smoking in the general population, data are lacking regarding its prevalence in a specific population with a chronic disease such as diabetes.Objectives: This study aims to determine the prevalence of tobacco smoking among patients with diabetes in Hospital Pulau Pinang, Malaysia. Methods: A cross-sectional survey was conducted to study diabetic patients who attended the endocrine clinic at Hospital Pulau Pinang in Malaysia from March to August 2012. All the diabetic patients who attended the endocrine clinic during that period were asked about their smoking status, and their medical records were reviewed. A total of 1,118 patients with diabetes were reviewed to determine the prevalence of tobacco smoking in diabetic patients at the endocrine clinic of Hospital Pulau Pinang. Results: The majority of the study population was male, with Malay and Chinese patients in almost equal proportions and a smaller proportion of Indian patients. Most of these patients had started smoking before they were diagnosed with diabetes. Among the 1,118 diabetic patients, only 108 patients smoked; therefore, the prevalence of tobacco smoking in our patients with diabetes was 9.66%. Conclusions: A low prevalence of tobacco smoking was estimated in this study. This prevalence is close to the corresponding value in the general population in Malaysia. PRS16 Current Annual Cost Calculation is the Best Predictor of Mortality at Three Years in COPD Dal negro RW Nationa center for Respiratory Pharmacoeconomics & Pharmacoepidemiology, Verona, Italy Objectives: Chronic Obstructive Pulmonary Disease (COPD) is a progressive condition which is characterized by a dramatic socio-economic impact. Sensitivity of clinical signs and lung function in predicting death is variable in different COPD phenotypes. Aim: To assess the predictive value of COPD annual cost on mortality. Methods: Gender; age; smoking habit, clinical data, and complete lung funtion were assessed in 275 consecutive COPD patients aged > 40y together with the annual cost calculated over the last twelve months. Statistics: t test for comparing means ± sd; linear regression for checking any relationship between each variable and mortality (p< 0.05 was accepted). Results: The whole mortality was 40.4% over three years (n= 12; 47, and 52 subjects, respectively). Subjects still survived after three years (n= 164) proved originally different from those (n= 111) who died in terms of mean age; FEV1 (in l); RV; TLCO/VA; 6’ walking test; BODE index, and Charlson index (all p<0.001), but not of FEV1% pred. and FEV1/FVC (p=ns). Mean total A591 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 COPD cost was 3,290.7 [95%CI: 2539.9; 4051.2], but € 1161.0 [95% CI: 968.4; 1,353.6] in survivors and 6,158.9 [95% CI: 5,508.0; 6,809.8] in those who died, respectively (p< 0.001). The hospitalization cost impacted for 78.2% of the total annual cost in subjects who died, the absolute value being sixfold higher than in survivors. All economic components of cost were discriminant at three years, independently of gender. Conclusions: 1) several clinical and lung function variables contribute to predict mortality in COPD; 2) identification of COPD phenotypes is crucial, but multiple indices are required; 3) total annual cost proved the most sensitive predictor of mortality at three years; 4) annual cost is much easier and cheaper to obtain over twelve months; 5) data are supporting the high predicting value and the convenience of COPD “economic phenotyping”. RESPIRATORY-RELATED DISORDERS – Cost Studies PRS17 Budgetary Implications of Introducing Fluticasone Furoate/ Vilanterol for COPD in the UK Doyle S , Armstrong S , Bowditch S GSK, Uxbridge, UK . . . Objectives: Fluticasone furoate/vilanterol (FF/VI) 92/22 mcg is a once-daily, fixeddose combination inhaled corticosteroid + bronchodilator licensed for the treatment of COPD in the UK. A budget impact model (BIM) was designed to explore the cost implications of prescribing FF/VI 92/22 in adult patients who continue to exacerbate or are currently receiving an off-license therapy versus alternative ICS/LABA therapies, in line with clinical guidelines. Methods: a one-year BIM was constructed to explore the financial outcomes of prescribing FF/VI 92/22 as an alternative treatment option to currently prescribed therapies based on market shares. The BIM is based on UK prescription analysis, epidemiological and resource data. The model explores three routes for progressing patients: 1) exacerbating patients currently on LAMA or LABA monotherapy progressing onto ICS/LABA combination therapy; 2) exacerbating patients currently on ICS/LABA combination therapy progressing onto triple therapy; 3) patients currently on off-license therapies moved onto licensed COPD products. The model does not explore differences in patient outcomes, efficacy or safety; it explores drug acquisition cost alone. Results: The BIM estimates that the average health economy (e. g. clinical commissioning group; local health board) in the UK has 3066 COPD patients of whom 2138 continue to exacerbate or are prescribed off licence therapies. In year 1, progressing exacerbating patients onto alternative ICS/LABA combinations or triple therapy (50% implementation rate), or patients on off-license therapies onto licensed therapies (100% implementation rate) results in a budget impact of £156,498 compared with a budget impact of -£51,907 if these patients are moved onto FF/VI 92/22. Conclusions: The introduction of FF/VI 92/22 in COPD has the potential to reduce the budget impact and total spend on combination ICS/LABA therapies by £208,405 in the average UK health economy compared to current patterns of prescribing. PRS18 Budgetary Implications of Introducing Fluticasone Furoate/ Vilanterol for Asthma in the UK Doyle S , Armstrong S , Bowditch S GSK, Uxbridge, UK . . . Objectives: Fluticasone furoate/vilanterol (FF/VI) is a once-daily, fixed-dose combination inhaled corticosteroid + bronchodilator licensed for the treatment of asthma in the UK. A budget impact model (BIM) was designed to explore the cost implications of initiating FF/VI in patients ≥ 12 years not adequately controlled on ICS monotherapy (and as needed SABA) versus alternative ICS/LABA therapies, in line with BTS-SIGN guidelines. Methods: A one-year BIM was constructed to explore the financial outcomes of prescribing FF/VI as a treatment option in these dynamic patients. The BIM is based on UK epidemiological and resource data. It compares a scenario in which all eligible patients are treated with currently available ICS/LABA products based on market shares with a scenario in which FF/VI is introduced for uncontrolled patients eligible for step up onto an ICS/LABA combination therapy in line with clinical guidelines. The model does not explore differences in patient outcomes, efficacy or safety; it explores drug acquisition cost alone. Results: The BIM estimates that the average health economy (e. g. clinical commissioning group; local health board) in the UK has 7326 patient on ICS monotherapy of whom 3736 are inadequately controlled and appropriate for step-up to an ICS/LABA combination therapy. In year 1, presuming a 50% implementation rate, stepping up these patients onto alternative ICS/LABA combinations at current usage rates results in a budget impact of £342,413 compared with a budget impact of £313,167 if these patients initiate FF/VI therapy. Conclusions: The introduction of FF/VI in asthma has the potential to reduce the budget impact and total spend on combination ICS/ LABA therapies by £29,246 in the average UK health economy compared to current patterns of prescribing. PRS19 The Budget Impact of Duoresp® Spiromax® (Budesonide + Formoterol Fumarate Dihydrate) Compared With Symbicort® Turbohaler® for the Management of Asthma and Chronic Obstructive Pulmonary Disease in the United Kingdom: Impact on Health Care Costs and Inhalation Technique Lewis A 1, Blackney M 1, Torvinen S 2, Holmes J 3, Osborne M 3, Dale J 3, Chandler S 3, Plich A 2 Inc., London, UK, 2Teva Pharmaceuticals Europe B. V, Amsterdam, The Netherlands, 3Teva UK Limited, Essex, UK . . . . . . . . 1Covance Objectives: DuoResp® Spiromax® (budesonide + formoterol fumarate dihydrate) is a fixed-dose combination (FDC) of inhaled corticosteroid (ICS) + long-acting beta agonist (LABA) in a novel dry powder inhaler (DPI). An economic model was developed to assess the budget impact of using DuoResp ® Spiromax® instead of Symbicort® Turbohaler® – a DPI delivering the same FDC – to manage adult patients with persistent asthma and chronic obstructive pulmonary disease (COPD) in the United Kingdom (UK). The potential cost benefit of improved inhalation technique, due to the innovative characteristics of the Spiromax® inhaler, was also investigated. Methods: The eligible adult patient population was based on current confirmed UK asthma and COPD diagnosis rates, with the proportion of patients receiving FDCs based on market research data. The costs of Symbicort® Turbohaler® and scheduled and unscheduled health care events were taken from publically available UK sources. Frequency of poor inhalation technique with Symbicort® Turbohaler® and the associated increased risk of unscheduled health care events were taken from a large (n= 1,664) cross-sectional, Italian observational study. The estimated reduction in the proportion of patients with poor inhalation technique with DuoResp® Spiromax® compared with Symbicort® Turbohaler® was based on a conservative assumption. Results: An estimated 409,445 adult patients used Symbicort® Turbohaler® annually in the UK and were therefore eligible for treatment with DuoResp® Spiromax®, with 178,108 of these exhibiting poor inhalation technique. Assuming a hypothetical uptake of DuoResp® Spiromax® reaching 25% in year 4 and 5 and its anticipated price, the model predicted drug cost savings totalling £36.09 million. Furthermore, 39,266 unscheduled health care events could be avoided due to the predicted improvement in inhalation technique with DuoResp® Spiromax® compared with Symbicort® Turbohaler®, resulting in further savings of £3.50 million. Conclusions: DuoResp® Spiromax® is likely to offer budgetary savings compared with Symbicort® Turbohaler®, with further cost savings potentially resulting from improved inhalation technique. PRS20 The Economic Impact of Therapy Optimization in Chronic Obstructive Pulmonary Disease in Portugal Andrade A 1, Viriato D 1, Viana R 2, Luciano T 3 Salvo, Portugal,, 2Novartis Farma-Produtos Farmacêuticos S.A., Portugal, Porto Salvo, Portugal,, 3ACES Amadora, Lisboa, Portugal . . . . 1Novartis, Porto Objectives: Chronic obstructive pulmonary disease (COPD) is an obstructive lung disease characterized by persistent airflow limitation. COPD has a significant humanistic burden, representing the 5th leading cause of death in Portugal. The prevalence of Portuguese population with more than 40 years is 14.2%. COPD symptoms are similar to other respiratory diseases, making harder its diagnosis. Furthermore, it may result in inappropriate use of medication which is associated with an economic burden. This analysis aimed to estimate the economic impact of inappropriate use of inhaled corticosteroid/long-acting beta-agonist (ICS/LABA) fixed-dosed combination in COPD patients. Methods: GOLD 2013 treatment algorithm establish that ICS therapy should only be considered in high risk population (patients classified within GOLD groups C and D), presenting FEV1 < 50% (forced expiratory volume at 1 second) and/or 2 or more exacerbations per year. Based on a Delphi panel there are 1% of patients in group C and 29% in group D. According to ECLIPSE study, only 31% of patients in group C and 37% in group D are illegible for ICS therapy. Based on these assumptions, we estimated the overtreatment of ICS/LABA in Portugal according with local COPD prevalence and number of ICS/ LABA fixed-dosed combinations prescribed. It was also calculated the economic impact associated with therapy switch of over treated patients from ICS/LABA to glycopyrronium or indacaterol, according to GOLD 2013. Results: The treatment with ICS/LABA was estimated to be approximately 1.6 times greater than expected. Considering the number of patients over treated with ICS/LABA, the therapy switch for indacaterol or glycopyrronium would result in a potential saving of 4,314,390€ or 8,694,008€ , respectively. Conclusions: Optimization of COPD therapy in compliance with GOLD guidelines would result in a better treatment for patients and potential savings, in a Portuguese National Health Service perspective. PRS21 Estimation of Increased Costs in Switching from Tiotropium to Other Lama Therapy During Maintenance Treatment of COPD in the UK Ternouth A , Schoenherr N , Edwards S C Boehringer Ingelheim, Bracknell, UK . . . . Objectives: Chronic Obstructive Pulmonary Disease (COPD) is a prevalent disease with a significant economic burden to the UK National Health Service (NHS). NICE recommends maintenance treatment including inhaled bronchodilator medications such as long-acting muscarinic antagonists (LAMAs); tiotropium is the most widely used LAMA in the UK. The objective of this model was to quantify the budget impact to the NHS of switching patients from tiotropium to another LAMA, compared to remaining on tiotropium. Methods: Two matched patient groups were considered: patients who switched from tiotropium to another LAMA and patients who remained on tiotropium. The budget impact model was conducted over a 3 month time horizon, with sensitivity analyses conducted over a 12 month time horizon. Costs included were costs of medication; costs of patient identification and new inhaler training; and primary care resource use in the 3 months following inhaler switch, priced according to PSSRU estimates. The number and duration of GP visits was estimated from the CPRD; a representative primary care patient database. Clinical efficacy endpoints were not included in the model. Results: Across the 3 month time horizon, patients who switched therapy incurred an increased incremental cost to the NHS of £77.10 per patient compared to patients who remained on tiotropium. Switching from tiotropium to other LAMA therapy also increased the NHS economic burden in the 12 month sensitivity analyses. Increased costs were driven by higher resource utilisation in terms of GP visits amongst switchers. Conclusions: Switching patients with COPD from tiotropium maintenance to another LAMA incurs an increased cost to the NHS. This is primarily due to a higher primary care resource use in the 3 months following switch. This finding has implications for prescribing practice including GP workload and patient willingness to switch, hence patients remaining on tiotropium are cost-saving for the UK NHS A592 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PRS22 Economic Burden of Hospitalized Pneumonia from a Private Health Care System Perspective in Brazil Manfrin D F 1, Ferreira C N 1, Santana C F S D 2, Paloni E D M P 2, Campi F D S 2, Gea Y 1, Rufino C S 1 1Pfizer, Inc., São Paulo, Brazil,, 2ORIZON - Companhia Brasileira de Gestão de Serviços, Sao Paulo, Brazil . . . . . . . . . . . . . . . . . . Objectives: This study aimed to evaluate the economic burden of hospitalized pneumonia and its associated costs in the Brazilian Private Health System. Methods: An administrative claims database containing over 18 million lives was used to identify hospitalized pneumonia episodes, in all ages, between Oct/2010 and Dec/2013. Pneumonia episodes were identified using ICD-10 codes of A40.3, B95.3, G00.1, J13, J15, J15.0, J15.3, J15.4, J15.8, J15.9, J18, J18.0, J18.9, J20.2, P23.3. Pneumonia costs included were taxes and exchange, medicines, materials, medicinal gases, food, medical fees, exams, procedures and personal hygiene. The costs represent a total in the evaluated period for all causes and considering the total patients identified, average cost was calculated. Results: A total of 16,227 pneumonia patients were identified with a total of 17,781 hospitalizations, representing an average of 1.1 hospitalizations per pneumonia patient in all ages. The aggregate costs for all inpatient treatment were BRL202,687,872.71 representing a total of 130,406 hospital days across the pneumonia patients. The average days for hospitalization per patient is 7.5 days. The average expenditure for the health insurance provider represents BRL12,490.78 per patient or BRL11,399.13, when evaluating per hospitalization. This translates to an average daily expenditure of BRL1,704.06 per patient or BRL1,555.13 per hospitalization. Conclusions: Costs related to pneumonia hospitalization are high and represents a large impact to the total budget spends in the private health care system. Given the significant economic burden associated with pneumonia, there is a need to better implement prevention strategies such as flu and pneumococcal vaccinations and smoking cessation. PRS23 Cost-Minimization and Budget-Impact Analysis of Fixed-Dose ICS/ LABA Combination Inhalers in the Treatment of Asthma in Slovenia Hren R 1, Trkman M 1, Stynes G 2 d.o.o., družba za promet s farmacevtskimi izdelki, Ljubljana, Slovenia, 2GlaxoSmithKline, Uxbridge, UK . . . 1GlaxoSmithKline Objectives: To compare the expected costs and outcomes associated with the novel once-daily fixed-dose combination (FDC) of inhaled corticosteroid (ICS) and long-acting beta agonist (LABA) fluticasone furoate/vilanterol (FF/VI) against those expected with currently available FDC options, fluticasone propionate/salmeterol (FP/S), budesonide/formoterol (BUD/F) and beclometasone dipropionate/formoterol (BD/F), for asthma treatment in Slovenia. Methods: We developed a Markov model based on the results of seminal study GOAL (Gaining Optimal Asthma Control) that takes into account the complexity of asthma treatment and particularly accompanying dose titration. Using available clinical evidence, we assumed that the clinical efficacy and incidence of adverse events associated with a all four FDCs were similar. We compared FF/VI with other FDCs of ICS/LABA from the Slovenian payers’ perspective at the price levels determined by international reference pricing as of April 2014. We followed a cohort of 12,300 asthma patients over the time horizon of 3 years with costs and outcoumes discounted at 3.5% per annum. Results: Model analysis showed that the expected three-year costs of FF/VI were lower than those of FP/S, BUD/F and BD/F by 13%, 40%, and 32%, respectively. The costs of treatment with FF/VI were lower than all other FDCs of ICS/LABA for patients who were previously treated with low/medium doses of ICSs (≤ 500 mg equivalent dose of BD) or high-doses of ICSs (> 500 mg and ≤ 1000 mg equivalent dose of BD). Budget impact analysis that was based on the current ICS/LABA market and projected market shares of FF/VI, FP/S, and BUD/F revealed that introduction of FF/VI could save the national Sick Fund € 863,000 over the first three years of use. Conclusions: Results of our model indicate that the introduction of the new once-daily FDC ICS/ LABA FF/VI could reduce the costs of asthma treatment in Slovenia over the first three years after introduction. PRS24 Cost-Minimization and Budget-Impact Analysis of Fixed-Dose Combination Inhalers in Treatment of Copd in Slovenia Hren R 1, Stynes G 2 1GlaxoSmithKline d.o.o., družba za promet s farmacevtskimi izdelki, Ljubljana, Slovenia, 2GlaxoSmithKline, Uxbridge, UK . . Objectives: To compare expected costs and health outcomes of the novel fixeddose combination (FDC) of inhaled corticosteroid (ICS) and long-acting beta agonist (LABA) fluticasone furoate/vilanterol (FF/VI) with currently available FDC options, fluticasone propionate/salmeterol (FP/S) and budesonide/formoterol (BUD/F), for chronic obstructive pulmonary disease (COPD) treatment in Slovenia. Methods: We developed a population-level Markov model based on the results of seminal TORCH (Towards a Revolution in COPD Health) study that accounts for disease progression of COPD and particularly the effects of changes in lung function and exacerbations. Using available clinical evidence, we assumed that all FDCs of ICS/ LABA were clinically equivalent. We compared FF/VI with two other FDCs of ICS/ LABA from the payers’ perspective at the Slovenian price levels determined by international reference pricing as of April 2014. We followed a cohort of 8,100 Slovenian COPD patients for 20 years, with costs and outcomes discounted at 3.5% per annum. Results: The life expectancy predicted by the model was 8.74 years with 2.76 years spent in moderate COPD, 2.55 years in severe COPD, and 3.43 in very severe COPD. Model analysis showed that the expected three-year survival of the cohort was 91% and that the expected corresponding costs of FF/VI (€ 1.021 per patient) were lower than those of FP/S (€ 1.535 per patient; +33% vs. FF/VI) and BUD/F (€1.260 per patient; +19% vs. FF/VI). Expected costs of maintenance, moderate exacerbations, severe exacerbations, and pneumonia were, for all treatment choices, € 617, € 65, € 1.707 and € 145 per patient, respectively. Budget impact analysis based on projected market shares of FF/VI, FP/S, and BUD/F revealed that introduction of FF/VI could save the national Sick Fund € 711,000 in the first three years. Conclusions: This model predicts that the introduction of FF/VI, a once daily FDC ICS/LABA, would reduce the overall costs of COPD treatment in Slovenia. PRS25 Benefit of Positive Airway Pressure (PAP) Therapy in Sleep Apnoea (SA) Patients With Chronic Obstructive Pulmonary Disease (COPD) in Germany: A Retrospective Comparative Cohort Analysis Based on a Statutory Health Insurance Database Doess A 1, Zucca F 2, Woehrle H 3, Brueggenjuergen B 4 1ResMed Germany Inc., Martinsried, Germany, 2HGC GesundheitsConsult, Duesseldorf, Germany, 3ResMed Science Center, Martinsried, Germany, 4Institut für Sozialmedizin, Epidemiologie und Gesundheitsökonomie, Charité - Universitätsmedizin Berlin, Berlin, Germany . . . . Objectives: It is estimated that the prevalence of moderate-to-severe SA (apnoeahypopnoea index > 15/h) is 10%. Approximately 11% of SA patients have comorbid COPD, which worsens sleep quality and desaturations. This study investigated the effects of PAP therapy on all-cause mortality and cost of illness (COI) in patients with SA and COPD in Germany. A statutory health insurance (SHI) perspective was taken. Methods: A total of > 4 million individuals covered by the SHI database were analysed (≈5% of the German SHI population). PAP therapy was initiated in 4,068 patients with SA (PAP group). Propensity score matching was used to define a control group (CG) of 4,068 SA patients matched for age, sex, risk factors/aetiology, region and medication who received usual care (no PAP). Of these, 1,300 patients in the PAP group and 1,192 patients in the CG had comorbid COPD. This subgroup of patients was followed for 3 years after initiation of PAP therapy. Results: Total COI was higher in the PAP group versus CG in the first year of follow-up (€ 8,697 vs € 6,999, p< 0.0001). However, during the second and third year the difference in COI between the PAP and CG was smaller (year 2: € 7,340 vs € 7,316, p< 0.0048; year 3: € 6,847 vs € 6,714, p< 0.001). PAP recipients had a significantly lower 3-year mortality rate compared with CG (8.2% vs 11.7%, p< 0.001; relative risk reduction 30.1%). Conclusions: SA patients with COPD treated with PAP showed significantly reduced mortality and morbidity. Total COI was higher in PAP recipients versus CG over the first 3 years of follow-up, but the difference between groups decreased over time. A follow-up period of ≥ 5 years may be required to show beneficial economic outcomes in SA patients receiving PAP therapy. PRS26 An Analysis of Us Medicare Beneficiaries: Burden of Direct Medical Costs in Patients With Idiopathic Pulmonary Fibrosis Chen S Y 1, Collard H R 2, Yeh W S 1, Li Q 3, Lee Y C 3, Wang A 3, Raghu G 4 Idec, Cambridge, MA, USA, 2University of California San Francisco, San Francisco, CA, USA, 3Evidera, Lexington, MA, USA, 4University of Washington, Seattle, WA, USA . . . . . . . . . . . 1Biogen Objectives: In the US, more than half of patients with idiopathic pulmonary fibrosis (IPF) are covered by Medicare; yet published data on the economic burden that IPF imposes to this largest US payer are limited. The objective of this study was to compare health care resource utilization (HRU) and costs between Medicare beneficiaries with IPF and matched non-IPF controls. Methods: Administrative claims from a 5% random sample of Medicare beneficiaries (aged 65+) from years 2000 to 2011 were analyzed. Incident IPF patients were identified based on ICD9-CM diagnosis codes, with at least one year enrollment before (pre-index) and after (post-index) the first diagnosis (index date). Up to 5 beneficiaries without IPF were matched to each IPF patient, based on age, gender, race, and region. Annual HRU and medical costs (excluding outpatient drug costs) during the pre-index and post-index periods were compared between IPF patients and the matched controls and univariate descriptive analyses were performed to compare the differences. Results: A total of 7,855 IPF patients were matched to 38,856 controls. During pre-index period, IPF patients had 2-3 folds higher risk of COPD, asthma and lung infections, 80% higher risk of hospitalization (28.8% vs. 15.8%), and higher total medical costs ($10,124 vs. $5,888) than matched controls (all p< 0.05). During post-index period, IPF patients had a higher risk of hospitalization (48.7% vs. 20.8%) and all other types of HRU with the total medical costs $11,955 higher than controls ($20,887 vs. $8,932) (all p< 0.05). Inpatient care accounted for 50% of total medical costs of IPF patients in both pre-index and post-index periods. Conclusions: In the US, IPF patients aged 65 or older had a greater burden of comorbidity and incurred more HRU and medical costs than matched controls. Multidisciplinary team based approach and effective therapies are needed given the high unmet needs. PRS27 Examination of the Burden of Illness of U. S. Medicare Patients Diagnosed with Chronic Obstructive Pulmonary Disease Xie L 1, Kariburyo M F 1, Du J 1, Baser O 2 1STATinMED Research, Ann Arbor, MI, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . . Objectives: To examine the economic burden and health care utilizations of patients diagnosed with chronic obstructive pulmonary disease (COPD) within the U. S. Medicare population. Methods: Patients diagnosed with COPD (International Classification of Disease 9th Revision Clinical Modification [ICD-9-CM] codes: 491, 492, 496) were identified using the U. S. national Medicare claims dataset from 01JAN2009 to 31DEC2011. The first diagnosis date was defined as the index date. A comparator group was created by identifying patients without a COPD diagnosis but of similar age, region, gender, index year and baseline Charlson Comorbidity Index scores. The index date for the comparator group was randomly chosen to reduce selection bias. One-year pre- and post-index continuous health plan enrollment was required for both groups. A 1: 1 propensity score matching was used to compare follow-up health care costs and utilizations between the COPD and comparison cohorts, adjusted for baseline demographic and clinical characteristics. Results: A total of 422,024 patients were identified for the COPD and compari- A593 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 son cohorts. After applying a 1: 1 matching, a total of 123,356 patients were matched from each cohort, with well-balanced baseline characteristics. COPD patients had higher health care utilization, including Medicare carrier (98.1% vs. 70.1%), Durable Medical Equipment (DME, 37.4% vs. 15.8%) and Home Health Agency (HHA, 17.2% vs. 4.6%) claims, outpatient visits (73.9% vs. 41.7%) and inpatient (32.5% vs. 6.8%), skilled nursing facility (SNF, 10.0% vs. 2.2%) and hospice admissions (1.2% vs. 0.6%) and prescription drug claims (53.4% vs. 49.8%), resulting in higher health care costs for Medicare carrier ($3,391 vs. $1,313), DME ($413 vs. $97), HHA ($923 vs. $228), outpatient ($10,110 vs. $3,514), inpatient ($5,983 vs. 1,045), SNF ($1,982 vs. $368), hospice ($304 vs. $143), pharmacy ($1,180 vs. $692) and total costs ($24,288 vs. $7,399) (p< 0.0001). Conclusions: COPD patients are associated with high economic burden and health care utilization. for moderate-risk RR2378 ($75) for all ages. Average reimbursement rate was RR2,000 ($63) per outpatient episode. The average cost of hospitalization was RR20,870 ($655/ n= 500), RR20,925 ($657/n= 125) for low-risk, RR20,947 ($657/n= 364) for moderaterisk, and RR17,684 ($554/n= 11) for high-risk patients; with 60%-80% of patients with CAP hospitalized and the estimated number of CAP patients of age 50 and older in Tver 3,249, the annual cost of CAP to payer is RR53,448,423 (~$1.7million); the national estimates of cost of CAP in senior adults was $124million. Employed patients comprise 12.8% and 18.3% among the inpatient and outpatient cases, respectively, and days lost from work per employed patient was 8.0 (inpatient) and 7.0 (outpatient). Conclusions: CAP in adults is a significant cause of resource use and health care cost in Russia, inpatient care constitutes the majority of cost. Cost of treatment was similar across all age and risk groups. PRS28 Estimation of the Cost of Childhood Asthma in Turkey PRS31 The Medical Costs of Chronic Obstructive Pulmonary Disease in South Korea Sekerel B 1, Malhan S 2 1Hacettepe University, Ankara, Turkey, 2Baskent University, Ankara, Turkey . . Objectives: Asthma is the most common chronic disease in childhood, reduces the quality of life of children and their families, and produces high social and health care costs. The aim of this study was to estimate the direct cost of pediatric asthma in Turkey and to examine its variability depending on asthma control level. Methods: The clinical pathway for childhood asthma was designed by and based on the data from the available Turkish literature. Unavailable data was collected by the expert’s clinical view. To calculate direct costs, the medical management of childhood asthma estimated using ‘cost-of-illness’ methodology for one year per child. All costs were calculated from the health care payer perspective. The costs were covered hospitalizations, physician visits, diagnostic tests, medications and co morbid diseases. Results: According to the recent studies, the controlled patient was 60%, partial controlled patient was 25%, and uncontrolled patient was 15% in Turkey. The cost of asthma calculation was based on weight of control. The mean annual cost per patient with controlled asthma is 542.97€ , partial controlled asthma 714.52€ and uncontrolled asthma 1047.86€. Hospitalizations, physician visits, diagnostic tests, co morbid diseases costs and medication costs estimated 4 %, 21%, 7%, 27%, 40% of total costs for controlled patient, 27%, 16%, 5%, 21%, 31% of total costs for partial controlled patient and 50%, 11%, 4%, 14%, 21% for uncontrolled patient respectively. Conclusions: The direct cost of pediatric asthma in Turkey increases depending on disease control level along with different spectrum of item distributions. To increase the utility and effectiveness of health care system, the findings of this evaluation may guide to construct future policies. Yang D W 1, Kim Y 1, Jang S 1, Hong J M 1, Yi J 1, Kim I S 1, Yang B M 1, Lee S W 2, Kim C 3, Rhee C K 4 1Seoul National University, Seoul, South Korea, 2Asan Medical Center, University of Ulsan, Seoul, South Korea, 3National Health Insurance System Ilsan Hospital, Goyang, South Korea, 4Seoul St. Mary’s Hospital, The Catholic University of Korea, Seoul, South Korea . . . . . . . . . . . . . . . . Objectives: The aim of this study is to investigate the medical costs of COPD patients in South Korea. Methods: We enrolled 300 COPD patients who had been treated and followed up for more than one year in the three hospitals from 2012 to 2013. The hospital electronic database was used to obtain medical costs and the medical records were reviewed by physicians (respiratory specialists) to assess clinical characteristics. We calculated annual maintenance costs per-patient according to disease severity, except the costs which were related to COPD exacerbation. The costs of COPD exacerbation per-case was calculated and divided into severe-exacerbation (hospitalization and emergency visit) costs and non-severe-exacerbation (outpatient visit) costs. Results: The annual maintenance costs per-patient was KRW 351,599, KRW 401,068, KRW 573,010 and KRW 999,506 for mild, moderate, severe and very-severe according to GOLD criterion, respectively. In case of the COPD exacerbation, the costs per-case for each non-severe exacerbation and severeexacerbation was KRW 163,495 and KRW 2,765,086. Conclusions: The severity of disease and exacerbation of COPD have a substantial impact on the medical costs of COPD patients. Improvement of lung function and reduction of occurrence of COPD exacerbation will be beneficial for the reduction of the health care expenditures. PRS32 The Direct Cost of Asthma in Turkey PRS29 Cost of a Pulmonary Arterial Hypertension-Related Hospitalization in Belgium Turktas H 1, Bavbek S 2, Malhan S 3 University, Ankara, Turkey, 2Ankara University, Ankara, Turkey, 3Baskent University, Ankara, Turkey Chevalier P 1, De Beule J 2, Lamotte M 1, Hunsche E 3, Régulier E 3 1IMS Health HEOR, Vilvoorde, Belgium, 2Actelion Pharmaceuticals Belgium NV, Mechelen, Belgium, 3Actelion Pharmaceuticals Ltd., Allschwil, Switzerland Objectives: Asthma is one of the mostly seen chronic illnesses in Turkey, yet there are limited studies on cost of asthma in our country. In addition they were based on a central database and weight of disease. The aim of this study was estimating the nationwide cost of managing asthma and examining its variability depending on asthma control level. Methods: The clinical pathway for asthma was determined from the literature. Part of the data was collected from the expert’s clinical view. To calculate the direct costs, the medical management of adult asthma was estimated using ‘cost-of-illness’ methodology for one year per patient. All costs were calculated according to payer perspective. The costs covered were pyhsician visits, hospitalizations, diagnostic tests, medicine and comorbid disease. Results: According to recent studies, the percentage of controlled, partially controlled and uncontrolled patients were 22%, 50%, and %28 respectively in Turkey. The cost of asthma calculation was based on weighted percentage of control. The mean annual cost per patient with controlled asthma was € 558,41, partially controlled asthma was € 594,86 and uncontrolled asthma was € 1040,63. Hospitalizations, physician visits, diagnostic tests, comorbidity and medication costs were estimated to be 4%, 11%, 11%, 28%, 46% for controlled patient, 10%, 11%, 10%, 43%, 26% for partially controlled patient and 48%, 6%, 6%, 25%, 15% for uncontrolled patient respectively. Conclusions: The cost of adult asthma in Turkey is very high and it significantly depends on disease control level. As expected, uncontrolled patient’s cost is higher than that of controlled patient. Nationwide health policies such as education of patients and physicians, selecting the right treatment and smoking cessation strategies that target the effective control of asthma will play an important role in the reduction of the economic burden of disease. . . . . . Objectives: Pulmonary arterial hypertension (PAH) is a rare disease, for which only scarce health care cost data is available in Europe. The progressive nature of the disease often requires hospitalization, the costs of which are currently unknown in Belgium, mainly due to the low number of patients affected. The objective of this research was to assess the cost and length of stay (LOS) of a PAH-related hospitalization likely related to disease worsening in Belgium. Methods: A retrospective database analysis was performed using the IMS hospital disease database from 2009 to 2011, covering 20% (2010/2011) to 34% (2009) of the hospital beds in Belgium. Data on adult patients, who were either hospitalized with a primary diagnosis of primary pulmonary hypertension (PH) (ICD-9-CM code 416.0) OR were receiving ≥ 1 medication indicated for PAH, was extracted. To ensure hospitalizations were likely related to disease worsening and not to planned procedures or routine visits, only unscheduled hospitalizations of at least 2 days were included in the analysis. Hospitalizations were excluded if primary co-diagnoses were suggestive of a non-PAH-related reason for the admission. Total hospitalization costs, including hotel, drugs, procedure costs (extrapolated to 2013), and length of stay (LOS) were analyzed with descriptive statistics. Results: 35 hospitalizations were included into the study. Patients experiencing these admissions were mainly female (71.4%), which is in line with the known female-male ratio (2-1) for this rare disease. Mean (SD) hospitalization cost was € 20,229 (9,399), including € 4,396 (9,502) drug, € 8,499 (8,999) hotel, and € 7,334 (12,386) procedure costs. Average LOS was 17.7±16.8 days. Conclusions: Long durations and high incurred costs for PAHrelated hospitalizations reveal the severe morbidity, health care, and patient burden of PAH in Belgium. . . . 1Gazi PRS33 Economic Burden in Direct Costs of Chronic Obstructive Pulmonary Disease (COPD) in Russia PRS30 The Cost Burden of Community-Acquired Pneumonia in Russia in Adults of 50 and Older: A Regional Study and National Estimates Krysanova V 1, Krysanov I 2 1I.M. Sechenov First Moscow State Medical University, Moscow, Russia, 2Postgraduate Medical Institute, Moscow National University of Food Production, Moscow, Russia Samyshkin Y 1, Roberts C S 2, Koroleva N 3, Rodionov A 4 Health, London, UK, 2Pfizer Inc., New York, NY, USA, 3Sechenov Moscow Medical University, Moscow, Russia, 4Tver State Medical Academy, Tver, Russia Objectives: COPD is one of the leading causes of morbidity and mortality worldwide and has a major burden on Russian’s health care system. It’s lead to frequent use of health care resources. The main aim of this study was to describe the direct costs for management of COPD patients with differing degrees of disease severity. Methods: The methodology for cost of illness analysis was based on a previous Russian study (Avksenteva M. V. et al., 2010). One-year costs were identified by applying cost data to medical information obtained by medical statistical records from 2007. In this study were performed 2 variants of COPD costs. In 1st variant were used epidemiological data from 2007 and medical resources costs from 2014. In 2nd variant were used extrapolated epidemiological data and medical resources costs from 2014. Due to lack of actual data in this study was modeled situation – from 2007 to 2012 the overall incidence of adult increased by 12.5%. Medical resources included hospital stays, outpatient visits and ambulance service. Results: The mean annual overall direct health care cost for 1st variant was estimated to be 54.6 billion rubles ($1.6 billion), for 2nd variant was estimated to be 61.6 billion rubles ($1.8 billion). The . . . . . 1IMS Objectives: Community-acquired pneumonia (CAP) represents a considerable burden in Russia. We assessed cost of CAP in Russia to characterize disease burden in Russian adults of age 50 and older. Methods: We conducted a retrospective chart review in a central Russian region of Tver for samples of 900 patients treated in inpatient and outpatient settings, and extrapolated data nationally. All patients were 50 years of age and represented new cases of CAP. Data were collected on demographics, comorbidities, and employment. The cost was estimated from the public payer perspective, with a productivity loss in patients below retirement age. Results: Cost of treatment was similar across age-and-risk groups in hospital and in outpatient settings. The cost of an outpatient episode was estimated far all risk groups at RR2176 ($69), ranging from RR1,939 ($61) to RR2537 (~$80) for age groups 50-64 and 75-84 respectively; the cost of episode for low-risk patients was RR1737 ($55), and . . A594 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 structure of direct costs in Russia was as follows – 77% for hospital stays, 21% for outpatient visits, 2% for ambulance service. COPD exacerbations contributed the major portion of cost and also correlated with disease severity. Conclusions: COPD associated with significant economic burden on Russian’s health care system. There is a striking direct relationship between the cost of care and severity of the disease with hospitalization leading to disease exacerbation being a major portion of cost. PRS34 The Costs of Illness of Atopic Dermatitis in South Korea Kim C M 1, Yim H W 1, Jo S J 1, Ahn S H 1, Seo S J 2, Choi W S 1 1Catholic University College of Medicine, Seoul, South Korea,, 2Chung-Ang University, Seoul, South Korea . . . . . . . . . . . . Objectives: Atopic dermatitis is a global public health concern considering its growing prevalence and mounting socioeconomic burden. However, Few studies has assessed the economic impact of atopic dermatitis in Korea. To conduct a cost analysis of atopic dermatitis and evaluate the economic impact of the disease on individual annual disease burden, quality of life, and change in medical expenses in regards to change in health related quality of life. Methods: This prospective cost analysis of atopic dermatitis by reviewing the housekeeping account books of 32 patients was conducted and evaluated the economic impact of the disease by analyzing the completed questionnaires. To handle the potential uncertainties, we compared the results with the data released by the Health Insurance Review & Assessment Board on medical costs claimed by the health care facilities. Results: In regards to the cost of illness, direct cost of atopic dermatitis per patient during the 3 month study period was 541,280 KRW and expenditure on other atopic dermatitis related products was 120,313 KRW. Extrapolated annual direct cost (including expenditures on other atopic dermatitis related product) per patient was 2,646,372 KRW. Estimated annual indirect cost was 1,507,068 KRW. Annual cost of illness of atopic dermatitis, computed by adding up direct and indirect costs, was estimated to be 4,153,440 KRW. Conclusions: The annual total social cost on a national level was estimated at 5.8000 trillion KRW. PRS35 The Cost Study of Health Services in Mongolia Damdinbazar O Mongolian national university of Medical Science, Ulaanbatar, Mongolia . Objectives: There are three main funding sources of health system In Mongolia which are state budget, health insurance and out of pocket payment. Health insurance funded health care service based on 115 DRG and total financing to health care organizations were 87.1 billion MNT in 2011. Health insurance rates 240000 MNT secondary and thirtary level of hospital by the same tariff. Aim of the study is to calculate 10 DRG costs which were spent 25 percent of the health insurance fund in 2012. Methods: We used both top down and bottom-up cost allocation method. Secondary data were used. Results: Respiratory diseases finance is 9 percent higher than the real cost. Other 9 diagnostic groups finance were less than actual cost by 8-62%. That the total funding based on the number of cases nationwide, 12,032,906,381 MNT funding was insufficient. Conclusions: Health insurance base tariff have to change based on the study while base rates should be different at secondary and thirtary level of health care organization. PRS36 A Pharmacoeconomic Care Analysis of Tuberculosis Control in Pakistan Iqbal M S 1, Iqbal M W 2, Bahari M B 3, Khalid S H 4, Iqbal M Z 1 of Clinical Pharmacy, Faculty of Pharmacy, AIMST University, Kedah, Malaysia, 2Faculty of Law, Universiti Malaya,, Kualalumpur, Malaysia, 3Faculty of Law, Universiti Malaya, Kualalumpur, Malaysia, 4Department of Pharmaceutical Technology, School of Pharmaceutical Sciences, Universiti Sains Malaysia, Pulau Pinang, Malaysia . . . . . . . . . . 1Department Objectives: To assess the direct and indirect medical costs incurred in the treatment of tuberculosis (TB) in patients attending public hospitals in Pakistan. Methods: A descriptive cross-sectional study was conducted in patients attending Accident and Emergency and TB wards of the hospitals in Pakistan by convenient-sampling technique. The direct and indirect medical costs were determined by various parameters like consultation fees, cost of medicines, travelling costs and laboratory test expenses etc. All obtained data were analyzed using descriptive and inferential statistics. Results: The mean annual direct medical cost for a TB patient was around Rs. 17317.56 (US$ 176.26) and indirect medical cost was Rs. 12918.50 (US$ 131.48). It was also observed that ccomparatively higher direct and indirect medical costs per patient (p < 0.001) were associated with large and urban hospitals. Besides, association of indirect medical costs with gender and age were the persuasive predictors of the study. Conclusions: Severity of the disease, distance to the hospital and length of stay in the hospital were proportional to the direct and indirect medical costs. In Pakistan, a significant proportion of the direct medical cost for TB treatment is subsidized for the public. culated. Methods: The costs including lab investigation charges, unit costs of treatment per bed, medication charges, food costs, transportation costs and loss of productivity were calculated per asthma episode. Data was analysed by Statistical Package for the Social Sciences (SPSS) version 18.0 using various descriptive and inferential statistical tests. Results: A median medical cost of acute exacerbation of asthma under Ministry of Health’s (MOH) perspective was USD 105.00 (RM338.47) per episode. Medication cost comprised the majority (52.38%) of the total medical costs. A median medical cost of acute exacerbation of asthma under patient’s perspective was USD 1.55 (RM4.99) per episode. Conclusions: Asthma exacerbation and length of stay in the hospital were proportional to the direct medical costs. In Malaysia, a substantial proportion of the direct medical cost of asthma treatment is heavily subsidised for the locals. PRS38 Pharmacoeconomic Evaluation and Burden of Illness of Acute Exacerbation of Copd in Patients in Malaysia Iqbal M S 1, Iqbal M Z 1, Barua A 2, Veettil S K 3, Wei L Y 3, Kit L W 3, Khan A H 4, Hussain Z 4, Iqbal M W 5 1Department of Clinical Pharmacy, Faculty of Pharmacy, AIMST University, Kedah, Malaysia, 2Division of Community Medicine, School of Medicine, IMU, Bukit Jalil, Kuala Lumpur, Malaysia, 3Department of Pharmacy Practice, School of Pharmacy, IMU, Bukit Jalil, Kuala Lumpur, Malaysia, 4Department of Clinical Pharmacy, School of Pharmaceutical Sciences, Universiti Sains Malaysia, Pulau Pinang, Malaysia, 5Faculty of Law, Universiti Malaya, Kualalumpur, Malaysia . . . . M S 1, Iqbal M Z 1, Barua A 2, Veettil S K 3, Ling T K 3, Yong N B 3, Khan A H 4, Hussain Z 4, Iqbal Iqbal M W 5 1Department of Clinical Pharmacy, Faculty of Pharmacy, AIMST University, Kedah, Malaysia, 2Division of Community Medicine, School of Medicine, IMU, Bukit Jalil, Kuala Lumpur, Malaysia, 3Department of Pharmacy Practice, School of Pharmacy, IMU, Bukit Jalil, Kuala Lumpur, Malaysia, 4Department of Clinical Pharmacy, School of Pharmaceutical Sciences, Universiti Sains Malaysia, Pulau Pinang, Malaysia, 5Faculty of Law, Universiti Malaya, Kualalumpur, Malaysia . . . . . . . . . . . . . . . . Objectives: The cost of acute exacerbations of asthma had not been well studied in literature. The aim of this study was to identify and quantify the (average) cost of moderate and severe exacerbations of asthma in patients attending tertiary-care setup in Malaysia. The related burden of exacerbations was also cal- . . . . . . . . . . Objectives: Acute exacerbation of chronic obstructive pulmonary disease (AECOPD) appears to be the main reason of hospitalization in COPD patients. Since substantial economic burden of COPD have not been previously studied in Malaysia, this study aimed at estimating and identifying different costs and related burden of illness in patients receiving treatment of AECOPD in a tertiary care hospital in Malaysia. Methods: A prospective follow-up study was performed in Department of Accident and Emergency and Respiratory Medicine of the hospital. Data were derived on the basis of per exacerbation episode. Relationship between direct medical costs and disease severity was analyzed using various descriptive and inferential statistical approaches. Results: Median actual direct medical costs and out-ofpocket costs were RM 457.68 (US$ 141.97) and RM 28.25 (US$ 8.76) per exacerbation respectively. Drug cost (41%) was the leading cost driver, followed by unit cost of treatment per bed (33.6%) and lab investigation cost (25.4%). However, food cost (44.2%) represented the largest proportion in out-of-pocket costs. More than 90% of actual direct medical costs were supported by the Government of Malaysia in the patients studied. Conclusions: Impacts of AECOPD in health care resources are worthy of attention. Cost information from pharmacoeconomic studies is important in decision making for health care professionals and policy makers in order to improve health care outcome and minimize costs. PRS39 Prospective Study on the Average Cost of Therapy for Bronchial Asthma Patients in an Indian Tertiary Care Teaching Hospital Nair S V , Abdulsalim S , Yedavalli N S , Shukla R , Mohan M K Manipal University, Manipal, India . . . . . . . . Objectives: To conduct a study to determine the average cost of therapy for bronchial asthma patients in a tertiary care center. Methods: A prospective observational study was carried out on a 100 bronchial asthma patients after ethical clearance was obtained from an Independent Ethical Review (IEC) board. The patients selected for the study were in-patients admitted to the Medicine and Pulmonary wards for bronchial asthma related complaints with and without co-morbidities. The study assessed the average cost of therapy which was obtained from patient records. Statistical analysis was performed using SPSS version 20. Results: The mean age of the study population (N=100) was 53.30±14.59. Females constituted 61% of the study population. The job profiles of the majority of study population were house wives (53%) and agriculturist (15%). The average cost of therapy among 100 patients was found to range from $1.81 to $598. The impact on the length of stay on cost of therapy per day was classified into ≤ 5, 6-10, 11-21 days and cost was found to be $9.21 ± 5.57, $12.12 ±9.65 and $15.56±10.36 respectively. Impact of co-morbidities (35%) and without co-morbidities on cost of therapy per day was found to be $13.03 ± 10.63 and $8.54 ± 6.77 respectively. Conclusions: Asthma creates a substantial financial burden on the society and results in compromise on diagnosis and treatment mainly in a developing country like India. There was a substantial increase in the cost of therapy as the duration of hospital stay increased and also in the case of patients with co-morbidities. Pharmacoeconomic analysis is needed to develop strategies to reduce the cost of therapy and thereby achieve greater medication adherence and improved quality of life in asthma patients. PRS40 Resource Use and Health Care Costs of Chronic Obstructive Pulmonary Disease in Slovakia Ondrusova M , Psenkova M , Mackovicova S Pharm-In Ltd, Bratislava, Slovak Republic . PRS37 Pharmacoeconomic Evaluation of Acute Exacerbation of Asthma in Patients in Malaysia . . . . Objectives: The objective of this cost study was to measure the resource utilisation and the direct costs associated with health care management of patients with chronic obstructive pulmonary disease (COPD) in Slovakia and to provide a basis for cost-effectiveness evaluations. Methods: The cross-sectional survey was performed to obtain the information on the management of patients with COPD and to estimate the direct costs of the disease management. The survey included 4 experts experienced in COPD treatment. The studied population were cohorts of COPD patients evaluated separately according to the stage of the disease (mild, moderate, severe and very severe). The patients were treated with standard therapy, the cost were set for one average patient per 3 months of treatment. The cost data were assessed and actualized due the 1st July 2014. All types of health care used in COPD management were evaluated (hospitalization, outpatient visits, diagnostics, laboratory tests and the management of symptoms, use of bronchodilators). Moderate A595 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 and severe exacerbations were also evaluated and the costs were set for one single event. Results: The total cost of mild COPD was € 26.22, moderate COPD € 30.26, severe COPD € 92.04 and for the very severe COPD € 267.64 for 1 patient/3 months. Expenses for bronchodilators also vary between different stages of COPD, for mild COPD it represented € 17.44, moderate COPD € 109.54, severe COPD € 219.58 and for very severe COPD € 206.15. Cost of treating exacerbations were set for one event - for moderate exacerbation € 67 and for severe exacerbation € 1060.27. Conclusions: In the management of COPD the most expensive are the costs of hospitalization, outpatient care and symptomatic treatment. The most costly is the management of the very severe COPD and severe exacerbation. This survey can be used as the source for cost inputs in pharmacoeconomic studies. PRS41 Cost-Effectiveness of Conjugate Pneumococcal Vaccination in Romania Preda A L 1, Moise M 1, Delgleize E 2, Leeuwenkamp O R 3 . . . . . . 1GlaxoSmithKline, Bucharest, Romania,, 2GlaxoSmithKline Vaccines, Wavre, Belgium,, 3Eclipse, Tervuren, Belgium Objectives: The objective was to analyze the cost-effectiveness of a national immunization program with pneumococcal conjugate vaccine (PCV): 10-valent pneumococcal non-typeable Haemophilus influenzae protein D conjugate vaccine (PHiD-CV) and 13-valent pneumococcal conjugate vaccine (PCV-13) in Romania. Methods: A published age stratified, deterministic, and static cohort model is used. This model highlights changes in cost and quality adjusted life years (QALYs) over time. Serotype specific disease incidence, age stratified disease incidence, mortality in the population, vaccine costs and resource utilization costs were obtained from a General Practitioner reports database and epidemiological sources. The model compared identical immunization programs involving PHiD-CV and PCV-13 vaccines taking the payer perspective. A cohort of 201,104 Romanian infants was followed for four years. Same net indirect protection for invasive pneumococcal disease (IPD) was assumed for both vaccines. Results: With 80% vaccine uptake and 2+1 vaccination schedule, PHiD-CV dominated PCV-13 assuming price parity. Vaccination with PHiD-CV versus PCV-13 resulted in an offset for the health care budget of £22,948 (the main driver of this difference is the decrease in Acute Otitis Media (AOM) related costs – with a total of 4,663 cases prevented) and a total of 23 QALYs gained. Sensitivity analyses revealed robustness of the model results, confirmed the dominance of PHiD-CV over PCV-13 and substantiated model outcome driven by incremental efficacy of PHiD-CV in conjunction with high incidence AOM. Conclusions: According to the model, implementation of PHiD-CV vaccination program for infants in Romania will offer substantial benefits in terms of cost savings and improved health compared to an identical vaccination program involving PCV-13. PHiD-CV’s potential to better prevent AOM translates into incremental benefits and dominance of PHiD-CV over PCV-13 given that their impact on IPD is similar. PRS42 Cost-Effectiveness of a COPD Disease Management Program in Primary Care: The Recode Cluster Randomized Trial Boland M R S 1, Kruis A 2, Tsiachristas A 1, Assendelft W 2, Gussekloo J 2, Blom C 3, Chavannes N 2, Rutten van-Molken M 1 1Erasmus University Rotterdam, Rotterdam, The Netherlands, 2Leiden University, Leiden, The Netherlands, 3Stichting Zorgdraad foundation, Oosterbeek, The Netherlands . . . . . . . . . . Objectives: Disease management programs for chronic obstructive pulmonary disease (herein, COPD-DM) are currently implemented on a broad scale in the Netherlands. However, the evidence about their cost-effectiveness is still inconclusive. We aimed to conduct a cost-effectiveness analysis of a COPD-DM program in primary care in the Netherlands, called RECODE. In RECODE, a multidisciplinary primary care team was trained in motivational interviewing to improve life style, setting-up individual care plans, early recognizing and managing of exacerbations, and implementing clinical guidelines. In addition, clinical decision making was supported by audit and feedback reports provided by an ICT program and reimbursement of physical reactivation by a physiotherapist was provided. Methods: In a two-year cluster-randomized controlled trial (1086 COPD patients, 40 clusters), the COPD-DM program was compared to usual care. As part of this trial we conducted a cost-effectiveness analysis to relate the effect of the COPD-DM on intermediate and final health outcomes to the costs from a health care and a societal perspective. Detailed self-reported health care utilization data were collected during the trial-period. Results: The 2-year intervention costs of the training for professionals, the ICT, and the audit and feedback reports were estimated to be €324 per patient. Excluding these costs, the intervention group had €584 (95% CI €86 to €1,046) higher health care costs and €645 (95% CI €28 to €1,190) higher costs from the societal perspective compared to the usual care group. Health outcomes were similar in both groups, except for 0.04 (95% CI -0.07 to -0.01) less quality-adjusted life-years in the intervention group. Conclusions: RECODE was not cost-effective during the 2-year follow-up period. This is most likely due to the fact that the interventions targeted professionals instead of patients and were suboptimally implemented, the relatively mild COPD population, and the national reforms in COPD care that affected the usual care group. PRS43 Cost Effectiveness of Bedaquiline for the Treatment of MultidrugResistant Tuberculosis Wolfson L 1, Walker A 2, Hettle R 3, Lu X 1, Kambili C 4, Murungi A 5, Knerer G 6 Pharmaceutica, Beerse, Belgium, 2Heron Evidence Development, Ltd., Luton, UK, 3PAREXEL Consulting, London, UK, 4Jansen Global Services, Raritan, NJ, USA, 5Janssen-Cilag UK, High Wycombe, UK, 6University of Southampton, Southampton, UK . . . . . . . over a 10-year horizon. A National Health Service (NHS) and personal social services perspective was considered. The effectiveness of treatment was evaluated in terms of Quality Adjusted Life Years (QALYs) and Disability Adjusted Life Years (DALYs). Data were sourced from a phase II, placebo controlled trial of bedaquiline, NHS reference costs, and the literature. Costs and effectiveness were discounted at a rate of 3.5% per annum. Probabilistic and deterministic sensitivity analysis was conducted. Results: The total discounted cost per patient on B+BR was £107,123, compared with £116,616 for BR. The total discounted QALYs per patient were 4.85 for B+BR and 3.81 for BR. The addition of bedaquiline to BR resulted in cost-savings of £9,493 and an additional 1.04 QALYs pp over a 10-year period, and is therefore considered to be the dominant (less costly and more effective) strategy over BR. B+BR remained dominant versus BR in the majority of sensitivity analyses, with a 74% probability of being dominant versus BR in the probabilistic analysis. Conclusions: In the UK, bedaquiline is likely to be cost-effective and cost-saving, compared to the current standard of care for MDR-TB under a range of scenarios. Cost-savings over a 10 year period were realized from reductions in lengths of hospital stay, which offset bedaquiline drug costs. Bedaquiline remained cost-saving in several sensitivity analyses, highlighting the certainty surrounding the results of the model. These results also indicate that the B+BR regimen can provide significant social economic benefits versus the BR only regimen. PRS44 Cost-Effectiveness Analysis of Umeclidinium Bromide Compared to Tiotropium Bromide for Symptomatic Patients with COPD in the UK Ismaila A 1, Roberts G 2, Punekar Y S 3, O’Leary M 2 1GlaxoSmithKline, Research Triangle Park, NC, USA, 2Double Helix, London, UK, 3GlaxoSmithKline, Uxbridge, UK . . . . . Objectives: To evaluate the long-term cost-effectiveness of umeclidinium bromide 62.5 mcg OD (UMEC) compared to tiotropium bromide 18 mcg OD (TIO) for the maintenance treatment of COPD from the UK National Health Service perspective Methods: We utilized a recently developed, internally and externally validated linked equations COPD Cohort disease progression model. The treatment effect, expressed as change from baseline in forced expiratory volume in one second (FEV1) at 12 and 24 weeks estimated from a Bucher method indirect treatment comparison (ITC) analysis following a systematic review. UMEC price was set at parity price of £33.5/month to TIO. Model outcomes included exacerbations, life years, quality adjusted life years (QALYs) and costs/QALY. The time horizons investigated ranged from one to 20 years (lifetime) on a sliding one-year increment. Costs, survival, and QALYs after the first year were discounted at a rate of 3.5%. Health care costs were obtained from NHS reference costs (2011-12). Sensitivity analyses were performed to evaluate the robustness of the model to variations in the underlying input parameters and assumptions. Results: The ITC estimated change from baseline in trough FEV1 of 18.06mL (95%CI: -19.11, 55.23, p= 0.341) at 12 weeks and 3.97mL (95%CI: -38.30, 46.25, p= 0.854) at 24 weeks for UMEC compared with TIO. At price parity, UMEC dominated TIO with incremental QALY of 0.0009, incremental life years of 0.0001 and cost reduction of £4.54. The sensitivity analyses suggested that variation in main parameters will not alter the behavior of the comparison between the two treatments. Conclusions: At price parity to TIO, UMEC may be considered as a cost-effective treatment alternative for maintenance bronchodilator treatment to relieve symptoms in patients with COPD in the UK. PRS45 Cost Effectiveness of Umeclidinium/Vilanterol (UMEC/VI) Combination Therapy Among Symptomatic COPD Patients Punekar Y S 1, Roberts G 2, Ismaila A 3, O’Leary M 2 1GlaxoSmithKline, Uxbridge, UK, 2Double Helix, London, UK, 3GlaxoSmithKline, Research Triangle Park, NC, USA . . . . . Objectives: UMEC/VI is a long acting muscarinic agent (LAMA) and long acting beta agonist (LABA) combination therapy. This study evaluated the cost-effectiveness of UMEC/VI compared to tiotropium (18µg) from the UK National Health Service perspective. Methods: A linked equations cohort model developed using the patient level data from ECLIPSE study and validated using patient level data from TORCH study was used. The baseline patient characteristics were derived from UMEC/VI phase IIIa clinical programme and included symptomatic COPD patients. The treatment effect expressed as change from baseline in forced expiratory volume in one second (FEV1) at 24 weeks was estimated using 3 tiotropium comparator phase IIIa trials and was assumed to last for at least 52 weeks following treatment initiation. Model outcomes included exacerbations, life years, quality adjusted life years (QALYs) and costs/QALY. The timeframe for the analysis was patient lifetime and the discount rate for costs and outcomes was 3.5%. The price of UMEC/VI was varied to estimate the points at which it would be cost effective compared with the current standard of care tiotropium. Health care costs were obtained from NHS reference costs (2011-12). Results: The random effects meta-analysis estimated treatment benefit of 92.17ml (95% CI: 61.52, 122.82; p<0.001) in FEV1for UMEC/VI compared with tiotropium. A lifetime model resulted in 0.009 fewer moderate-severe exacerbations per year on UMEC/VI. At parity price (£33.5/month), UMEC/VI dominated tiotropium with a probability of 0.81 for being cost effective at £30,000/QALY threshold. The incremental cost effectiveness ratios were £11,080 and £22,178 at 5% and 10% price premium to tiotropium, respectively. Conclusions: At an appropriate price, UMEC/VI may be considered as a cost-effective treatment alternative for symptomatic patients with COPD. 1Janssen Objectives: To evaluate the cost-effectiveness of adding bedaquiline to the intensive phase of background regimens (BR) of drugs for multidrug-resistant tuberculosis (MDR-TB) in the United Kingdom (UK). Methods: A cohort-based Markov model was developed to estimate the incremental cost-effectiveness ratio of bedaquiline plus BR (B+BR) versus BR alone (BR) in the treatment of MDR-TB in the UK, PRS46 Cost Effective Analysis of Dry Powdered Inhalers Versus Metered Dose Inhalers of Salbutamol for Asthma in Rural Secondary Care Hospital of South India Vigneshwaran E , Maddirevula M R , Dharmareddy L , Thamineni R , Kadapala P R , Golla M , Yiragamreddy P R raghavendra institute of pharmaceutical education and research, Anantapur, India . . . . . . . . . . A596 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Asthma is a chronic disorder requires continuous and long term management. Thus it makes the patient economically week and produces more burden on patient. Short acting β 2agonists in pressurized metered dose inhalers and dry powdered inhalers are the most commonly prescribed formulations in south Indian clinical settings. The present study aims to investigate and to select appropriate cost effective formulation of metered dose inhalers (MDR) or dry powdered inhalers (DPI) for salbutamol. Methods: It is a prospective comparative study conducted among subjects those who were newly diagnosed with asthma. The patients were divided into two groups based on the type of inhaler used such as MDI or DPI group. All the patients were counseled about the usage of inhalers during their treatment device allotment. Quality of life and FEV 1 were measured at the baseline visit. In addition to that data related to direct cost such as medical, laboratory and re- hospitalization costs were also measured at baseline. Follow up was done for both the groups. Similar to baseline visit quality of life, FEV1 and direct medical costs were measured during follow up. Results: The present study results shows that there is no significant difference between two groups with regards to demographic characteristics. We observed a significant difference in QOL (p < 0.05) and the mean score for two treatment devices groups was found to be 55.63 &43.72 respectively. There is no significant difference in FEV 1 (p > 0.05) and symptom free days between two treatment devices. Average cost effectiveness ratio was calculated and average cost effectiveness ratio for MDI were found to be less compared to DPI but statistically not significant. Conclusions: Overall it was found that efficacy was higher in MDI than DPI and cost was equal for both the groups to treat newly diagnosed asthma patients with salbutamol. PRS47 Cost-Effectiveness Analysis of Community-Acquired Pneumonia Treatment Zaytsev A 1, Makarevich A 2, Kondratyeva T 3 Main Military Clinical Burdenko Hospital, Moscow, Russia, 2The 301 Military Clinical Hospital, Khabarovsk, Russia, 3The 1586 Military Clinical Hospital, Podolsk, Russia . . . 1The Objectives: To evaluate clinical efficacy, safety and cost-effectiveness of treatment options for mild to moderate community-acquired pneumonia in patients with risk factors of poor efficacy (prior administration of antibiotics, comorbidities), comparing generic levofloxacin (Glevo, Glenmark Pharmaceuticals Ltd.) versus original levofloxacin (Tavanic®, Sanofi-Winthrop Ind.) and conventional treatment (b-lactam±macrolide). Methods: Patients were randomized to 3 treatment arms. Mean age was 24.3±11.5years. Treatment arm 1 included 61 patient administered GLEVO in the dose of 500 mg/day, whereas in the treatment arm 2 (n= 41) patients were treated with original levofloxacin (Tavanic) in the dose of 500mg/day, and 45 patients in the treatment arm 3 received conventional therapy. Clinical efficacy and safety were evaluated based on clinical, laboratory and radiological data analysis. Cost-effectiveness analysis included calculation of direct medical expenses and cost-effectiveness ratios (CER). Results: Clinical efficacy rate in the treatment arm 1 (Glevo) was98.4%, in the treatment arm 2 (Tavanic)–97.6%, whereas conventional therapy efficacy rate was84.4%. Adverse event incidence in the treatment arm 1 was 21.3%, in treatment arms 2 and 3 –14.6% and35.5% respectively. Treatment duration in the Glevo treatment arm was8.2±1.4days, in the treatment arm2 –8±1.2, in the conventional therapy arm–7.2±2,1 days. Time to radiological resolution of pneumonia was comparable. Mean cost of antibiotic administration cycle and the cost-effectiveness ratio in patients administered Glevo was 7.65€ (CERGLEVO = 7.8), in the treatment arm 2 – 22.4€ (CERTAV= 22.9), in the conventional therapy arm–10.8€ (CERSTAND= 12.8). Conclusions: Administration of levofloxacin for mild to moderate community-acquired pneumonia in patients with risk factors is superior in clinical efficacy compared to conventional treatment modalities. Administration of Glevo is characterized by favorable cost-effectiveness parameters. PRS48 Economic Evaluation of the Fixed Dose Combination of Indacaterol/ Glycopyrromium, as a Maintenance Bronchodilator Treatment in Adult Mexican Patients With COPD Reyes Lopez A 1, Lemus Carmona E A 2, Ruiz Miranda C I 2, Fernandez Plata M D R 3, Martínez Briseño D 3 1Center for Economic and Social Studies in Health, Mexican Children Hospital, Mexico, Mexico, 2Novartis Mexico, Mexico, Mexico, 3National Institute of Respiratory Disease, Mexico, Mexico . . . . . . . . . Objectives: To perform a cost-effectiveness analysis of Indacaterol/ Glycopyrronium (IG) against tiotropium monotherapy (TM), salmeterol/fluticasone (SF) and indacaterol/tiotropium (IT) for COPD patients from the perspective of the Mexican Public Healthcare System. Methods: A patient-simulation model structured in MS Excel, developed and validated by Asukai et al (2013), allowed us to compare IG against tiotropium monotherapy (TM), salmeterol/fluticasone (SF) and indacaterol/tiotropium (IT). The effectiveness measures analyzed were life years gained and exacerbations avoided based on efficacy data extracted from a comprehensive large phase III trial program comprising 11 studies (ILLUMINATE, SHINE, BRIGHT, ENLIGHTEN, SPARK, BLAZE, ARISE, BEACON, RADIATE, LANTERN, FLAME) with more than 10,000 patients across 52 countries. COPD drug cost, maintenance cost and exacerbation cost were estimated for six months duration of model cycle, using local prices for public health care institutions and evaluating the resources utilization data extracted from a sample of medical records. Lifetime horizon was used and a discount rate of 5%. Deterministic and probabilistic sensitivity analysis was performed. Results: Total expected costs per patient were US$30,421 and US$31,577 for the comparison IG vs TM respectively; US$28,817 and US$28,852 for IG vs SF respectively; US$28,900 and US$34,557 for IG vs IT respectively. IG delivers slightly more life years than comparators, but avoid significantly more exacerbations than the other options (1 to 4) which in turn have favorable economic impacts. Sensitivity analysis showed that base-case results are robust to variations of key model parameters. Conclusions: Indacaterol/ Glycopyrronium resulted more effective and less costly than comparators. These results showed that is possible to achieve cost-savings and a potential clinical benefit with the fixed dose of IG for Mexican patients with COPD. PRS49 Cost- Effectiveness of Real Life Asthma Pharmacotherapy Grekova D 1, Dimitrova M 2, Andreevska K 1, Petkova V 3, Madzharov V 1, Gueorguiev S 1, Petrova G 4 1Medical university of Plovdiv, Plovdiv, Bulgaria, 2Medical University of Sofia, Faculty of Pharmacy, Sofia, Bulgaria, 3Medical University, Faculty of Pharmacy, Sofia, Bulgaria, 4Medical University Sofia, Faculty of Pharmacy, Sofia, Bulgaria . . . . . . . Objectives: To analyze ambulatory prescribing practice and to assess the costeffectiveness of asthma pharmacotherapy in country settings. Methods: It is a prospective prescribing practice and cost-effectiveness analysis. During 2008-2011 were observed 238 patients in Plovdiv region and collected information about their ambulatory asthma pharmacotherapy. Prescribed medicines were systematized in INN groups of mono and fixed dose combination products. The FEO1 and percentage of patients without exacerbation were used as measure of the therapeutic results. Incremental cost-effectiveness ratio was calculated and with Tornado diagram was explored the sensitivity of the results. Results: Pharmacotherapy with fixed dose combination was performed mainly with Beclomethazone/formoterol; Budesonide/ formoterol; and Salmeterol/fluticasone. The monthly cost of pharmacotherapy is varying among 35 and 50 Euro. Incremental cost effectiveness ratio is favoring the combination Beclomethazone/ formoterol 100/6 mcg with ICER of 324 Euro for additional increase in FEO1, and 50 Euro ICER for additional patient without exacerbation, although all alternatives are cost-effective because all ICERs fall below the GDP per capita. The monotherapy was performed with Beclomethazone, Fluticasone, Budesonide, Ciclesonide, and Montelukast. Its monthly cost was among 19 and 40 Euro. Incremental cost effectiveness ratio is favoring ciclesonide that is a dominant alternative as monotherapy for both studies outcomes. Results are sensitive to the changes in therapeutic outcomes. Conclusions: The real life therapy follows the international guidelines but less fixed dose combinations were prescribed in comparison with international recommendations. Beclomethazone/ formoterol fixed dose combination and ciclesonide as monotherapy are cost-effective alternatives for the observed health care settings. PRS50 Cost-Effectiveness Analysis of High-Dose Levofloxacin Therapy of Patients With Community-Acquired Pneumonia Zaytsev A 1, Makarevich A 2 1The Main Military Clinical Burdenko Hospital, Moscow, Russia, 2The 301 Military Clinical Hospital, Khabarovsk, Russia . . Objectives: Study of clinical efficacy, tolerance and economic indicators of treatment of non-severe CAP by (5-day) course of levofloxacin 750 mg/day vs. the standard administration of levofloxacin 500 mg/day. Methods: The research included 64 patients having non-severe CAP with risk factors (administration of antibiotics during the preceding 3 months, concomitant diseases), all of them being males. The patients were randomized into 2 groups; the 1st group received high-dose therapy of levofloxacin, 750 mg/day within 5 days (Remedia, Simpex-Pharma, India). 2nd group recieved levofloxacin 500 mg/day for 7-10 days (Tavanic®, Sanofi-Winthrop Ind.). Efficacy and safety were assessed in terms of comprehensive analysis of clinical, laboratory and radiological data. For economic analysis, direct medical costs and “costs-efficacy” ratios (CER) were calculated. Results: First group included 32 patients with average age 22.7±1.8 years. Second group consisted of 32 patients with average 21.8±6 years. Clinical efficacy of high-dose levofloxacin therapy amounted to 96.9%. Average duration of antibiotic treatment was 5.2±0.9 days.. Standard regime by levofloxacin was efficient in 100% of cases. Radiological resolution was identical in both groups. Transient increase of hepatic transaminase activity was present in 3 patients (9.4%). In standard group, this adverse event was present in 4 patients (12.5%). Average treatment cost by levofloxacin 750 mg/day amounted to 13.3±6.2 euros (CERLEVO-750=13.7); cost of the standard therapy was 27.2±4.9 euros (CERLEVO-750=27.2). Conclusions : Therefore, in terms of clinical efficacy and safety, the high-dose therapy (750 mg/ day) by levofloxacin, in brief course of treatment of patients having non-severe CAP, is comparable with the standard regime of treatment (levofloxacin 500 mg/day for 7 to 10 days); and, is more efficient in economic terms. PRS51 Cost-Effectiveness of Asthma Management in a Hospital-Based Adult Asthma Clinic in Spain Perez de Llano L A 1, Villoro R 2, Hidalgo A 3, Merino M 2 Universitario Lucus Augusti, Lugo, Spain, 2Instituto Max Weber, Madrid, Spain, 3University of Castilla La Mancha, Toledo, Spain . . . . . 1Hospital Objectives: Optimal asthma control has been associated with significant reductions in mortality, morbidity, and quality of life gains for the patients. Hospital Asthma Clinics (ACs) are hospital-based units run by an experienced team composed of a pneumologist and a specialized nurse. Their aim is to provide effective treatment and optimal control to asthma patients. However, their impact on disease control and their cost-effectiveness are unknown. The objective of this study is to assess the cost-effectiveness of managing asthma patients in an AC versus traditional management. Methods: We designed a case-crossover study using the medical records of all patients submitted to one AC in Spain during 2012. We defined the case period as 365 days after the first visit to the AC, and the control period as 365 days before the index date. We calculated changes in relevant disease control indicators and estimated the Incremental Cost Effectiveness Ratio (ICER) for one additional controlled patient. Results: The percentage of controlled patients increased from 41% to 86% (n= 83, mean age was 49 ± 15.2; 66% female). Asthma control test score increased from 18.7 ± 4.6 to 22.6 ± 2.3 (p< 0.005), exacerbations decreased by 75% (p< 0.005) and FEV1 increased from 81.4% ± 17.5 to 84.4% ± 16.6 (p< 0.05). The use of ICS/LABA combinations decreased from 79.5% to 41%. On the contrary, the use of other drugs increased: anticholinergics from 3.6% to 16.9%, inhaled corticosteroids A597 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 alone from 3.6% to 45.8%, and omalizumab from 0% to 6%. Annual hospitalizations rates and emergency visits decreased by 78% and 75% respectively. The ICER was 1800€ per controlled patient per year. Conclusions: Managing asthma patients in a specialized AC is cost-effective and has significant impact on patient control, indicating better survival and quality of life for the patient according to published literature evidence. PRS52 Impact of Allergen Immunotherapy on Symptom-Free Days and Health Care Costs in Patients With Grass Pollen-Induced Allergic Rhinitis in Germany Najib M 1, Westerhout K Y 2, Verheggen B 2, Schreder C H 3 1Stallergenes, Antony, France, 2Pharmerit International, Rotterdam, The Netherlands, 3Stallergenes GmbH, Kamp-Lintfort, Germany . . . . . . Objectives: A health economic assessment was conducted to determine the relative impact of treatment with Oralair® or Grazax® on clinical effects and health care costs in patients with grass pollen-induced allergic rhinitis (AR) in Germany. Methods: The effects of three years of drug treatment on symptomfree days (SFDs) and associated costs were assessed using a health economic Markov model with a nine-year time horizon. The relative efficacy on SFDs was assessed through a network meta-analysis (i. e. indirect comparison) of 4-year, placebo-controlled, clinical trial data. Costs associated with drug treatment and other health care resources, including Statutory Health Insurance payments and patient co-payments, were calculated. The incremental costs and SFDs gained for Oralair® relative to Grazax® were generated accordingly. The uncertainty around the model outcomes was determined by means of sensitivity analyses. Results: The base case analysis over 9 years predicts a total of 206.6 discounted SFDs for Oralair®relative to 205.1 for Grazax®, thus resulting in 1.5 (95%CI: -25.0; 29.3) additional SFDs gained with Oralair®. Total discounted costs are estimated at € 1,696 and € 2,968 for Oralair® and Grazax®, respectively, with incremental costs predicted at -€ 1,272 (95%CI: -€ 1,530; -€ 999). Hence, Oralair® may be classified as the dominant strategy, as additional effects are combined with considerable cost savings. The sensitivity analyses suggest that results were mostly driven by drug-specific clinical effects on SFD, inputs for immunotherapy discontinuation, and length of the pollen season. The predicted cost savings were driven by the difference in treatment costs. Conclusions: Oralair® is cost-effective relative to Grazax®in patients with grass pollen-induced AR in Germany. Findings are confirmed by extensive sensitivity analyses. PRS53 Economic Evaluation of the Use of an Infant Formula Based on Partially Hydrolyzed Serum Protein as Compared With a Standard Whole Cow’s Milk Formula for Prevention of Atopic Dermatitis in Children Under 3 Years Old Derkach E V 1, Avxentyeva M 1, Fedyaeva V K 1, Rebrova O 2 1The Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia, 2Pirogov Russian National Research Medical University, Moscow, Russia . . . . . . Objectives: To evaluate the potential economic impact of the 100% whey-based partially hydrolyzed infant formula (PHF-W) in comparison with a cow’s milk standard formula (SF) in the prevention of atopic dermatitis (AD) in at-risk children at the age periods 0–12 and 0–36 months in Russia. Methods: The Excel model was constructed to estimate costs of artificial feeding with PHF-W vs SF and expected AD cases treatment. The model was based on the results of meta-analysis of randomized controlled trials (RCTs), literature data and the results of the expert survey. The costs of artificial feeding and AD treatment were calculated from the positions of different payers: health care system, family, and society as a whole. The incremental cost-effectiveness ratio (ICER) per averted AD case was calculated for PHF-W vs SF. Results: From health care system point of view the use of PHF-W uniquely lead to cost savings. If we consider all costs from the societal perspective PHF-W vs SF requires additional costs in the first year of baby’s life, but in leads to cost savings in a 3-year horizon. From the perspective of the at-risk child’s family artificial feeding costs will increase from 266 to 408 Euro for PHF-W vs SF. However the likelihood of AD development in a child will decrease from 15 to 8% in the first year and from 27 to 19% over three years and accordingly this will prevent AD treatment costs. Conclusions: Using PHF-W for the AD preventionin high-risk children have benefits for both the health system and for individual family. PRS54 A Cost-Effectiveness Analysis of Treatment for Mild to Moderate Obstructive Sleep Apnea-Hypopnea Syndrome (OSAHS) in France Poullié A I 1, Gauthier A 2, Cognet M 2, Clementz M 2, Späth H M 3, Perrier L 4, Scemama O 1, Rumeau Pichon C 1, Harousseau J L 1 1Haute Autorité de santé, Saint-Denis La Plaine, France, 2Amaris, London, UK, 3ISPB - Faculté de Pharmacie - Université Lyon 1, Lyon, France, 4Cancer Centre Léon Bérard, Lyon, France . . . . . . . . . . . . Objectives: In France, continuous positive airway pressure (CPAP) is recommended as first-line treatment for patients with severe OSAHS or mild-to-moderate OSAHS with high cardiovascular risk. Dental devices are recommended as second-line treatment for these patients and can be suggested as first-line treatment for mild-to-moderate OSAHS patients without high cardiovascular risk. Lifestyle advice is recommended for overweight patients. This study aims to assess the cost-effectiveness of these treatments for mild-to-moderate OSAHS patients. Methods: This study was commissioned by the French National Authority for Health (HAS) and followed their recommendations. A Markov model was developed to simulate the lifetime progression of a cohort of mild-to-moderate adult OSAHS patients. CPAP was compared with dental devices, lifestyle advice and no treatment. Daytime sleepiness, cardiovascular disease and road traffic accidents were taken into account. Clinical parameters were taken from international publications. Costs were retrieved from the French national health insurance databases: Assurance Maladie and Technical Agency of Information on Hospitals (ATIH). Costs and outcomes were discounted at 4% through 30 years and 2% thereafter. Robustness of results was assessed using sensitivity analyses. The assessed outcomes were the incremental cost per quality-adjusted life-year (QALY) gained and total life-years gained (LYG). Results: This study will inform public decision making about reimbursement of mild-to-moderate OSAHS treatments. CPAP was associated with higher costs and QALYs compared with dental devices, lifestyle advice and no treatment. Several sensitivity analyses were undertaken and it was found that the most sensitive parameters were related to sleepiness and cardiovascular inputs. Further investigation (clinical trial/observational study) of treatment effects on these parameters is needed. Conclusions: This analysis is the first to assess the cost-effectiveness of treatments in mild-to-moderate OSAHS patients in France. The technical report of this research will be available on the HAS website at the time of the congress (November 2014). PRS55 Cost-Effectiveness of Subcutaneous Immunothereapy in Allergic Rhinitis Using One or More Allergens - An Analysis Long Overdue Kiel M A 1, Röder E 2, Gerth van Wijk R 2, Rutten- Van Mölken M P M H 1 University, Rotterdam, The Netherlands, 2Erasmus Medical Center, Rotterdam, The Netherlands . . . . . . . . 1Erasmus Objectives: Allergic rhinitis – hay fever and mite hypersensitivity - is a prevalent and increasingly common condition, causing considerable morbidity and economic burden to society. A minority of patients have an indication for subcutaneous immunotherapy (SCIT). SCIT using extracts of tree pollen, grass pollen and/or house dust mite is common practice in The Netherlands, Europe and beyond. SCIT is widely reimbursed, but very few prospective cost-effectiveness studies have been done. Methods: An open-label, 2-year, multicenter, randomized controlled trial with two parallel treatment arms was performed, comparing SCIT + usual care (UC) with UC alone, using online resource use and labor productivity questionnaires, and electronic versions of EQ-5D®, SF-36® and a global subjective assessment of symptoms (GA). Primary endpoints were the costs per QALY, costs per successfully treated patient and the cost per additional symptom-free day. A Generalized Estimation Equation (GEE) model was estimated with mean two-week health care/societal costs as dependent variable, followed by a 1000-iteration bootstrap procedure. Results: A total of 183 adult patients aged 18 to 45 years with persistent moderate to severe allergic rhinitis due to one (43%) or more (57%) allergies (93 SCIT+UC, 90 UC) were included. There were no significant differences at baseline. The percentage of patients that reported to be treated successfully was 36% in SCIT and 20% in UC after two years. Other health outcomes did not differ between SCIT and Usual Care. Two-year costs of SCIT were 2946 Euro per patient. There was no difference in other costs. Cost per additional successfully treated patient were about 15,000 Euro. For the other outcomes, SCIT was dominated by UC. Conclusions: This study could not support the cost-effectiveness of SCIT. A restriction in the indication of SCIT to patients with severe persistent rhinitis, not sufficiently responsive to maximum symptomatic therapy may improve cost-effectiveness. PRS56 Impact of Allergen Immunotherapy on Quality of Life and Health Care Costs in Adults and Children With Grass Pollen-Induced Allergic Rhinitis in Germany Najib M 1, Westerhout K Y 2, Verheggen B 2, Schreder C H 3 1Stallergenes, Antony, France, 2Pharmerit International, Rotterdam, The Netherlands, 3Stallergenes GmbH, Kamp-Lintfort, Germany . . . . . . Objectives: To determine the relative impact of treatment with Oralair®, Grazax®, Alutard® or symptomatic drug treatment (SDT) on clinical effects and health care costs in subgroups of adults and children with grass pollen-induced allergic rhinitis (AR) in Germany. The cost-effectiveness of Oralair® has been demonstrated in a mixed population in previous research. Methods: The effects of three years of drug treatment on quality-adjusted life years (QALYs) and associated costs were assessed using a Markov model with a nine-year time horizon. Symptom score data were extracted, and the relative efficacy on QALYs was assessed through a network metaanalysis (i. e. indirect comparison) of placebo-controlled, clinical trial data in adults and children. Patient symptom scores were translated into the impact on quality of life by means of published sources. Costs associated with drug treatment and other health care resources, including Statutory Health Insurance payments and patient co-payments, were estimated. The incremental costs and QALYs were generated accordingly. The uncertainty around the model outcomes was determined by means of sensitivity analyses. Results: In adults, the analysis predicted more QALYs for Oralair® relative to Grazax®, 0.008 (95%CI: -0.043; 0.062) and Alutard®, 0.028 (95%CI: -0.029; 0.093), combined with incremental costs of -€ 1,272 (95%CI: -€ 1,530; -€ 999) and -€ 129 (95%CI: -€ 389; € 160) per patient, respectively. Hence, Oralair is dominant relative to Grazax® and Alutard® in adults. The incremental cost-effectiveness ratio was estimated at € 15,503 per QALY relative to SDT. Similar results were observed in children, with the exception of Alutard® (lack of data). The sensitivity analyses suggest that results were mostly driven by drug-specific clinical effects on symptom score, drug costs, inputs for immunotherapy discontinuation, and length of the pollen season. Conclusions: Oralair® is cost-effective relative to Grazax®, Alutard® and SDT in grass pollen-induced AR in Germany. Findings were confirmed again and supported by extensive sensitivity analyses. PRS57 Economic Evaluation of Omalizumab Compared With Standard Therapy in the Treatment of Severe Allergic Asthma in Adult Patients in Greece: a Cost Effectiveness Analysis Based on Clinical Trial and Real-World Data Sonathi V 1, Hatzikou M 2, Baldwin M 3, Panitti E 2, Tzortzaki E 4 Healthcare Pvt. Ltd., Hyderabad, India, 2Novartis Hellas, Metamorfosis, Greece, 3Novartis Employee by the time of the study, Horsham, UK, 4Medical School University of Crete, Heraklion, Greece . 1Novartis . . . . A598 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Severe asthma is a major cause of morbidity and mortality around the world, associated with a heavy societal burden. The aim of this study was to evaluate the economic value of omalizumab in the treatment of adult patients with severe allergic asthma in Greece, from a societal perspective, based on data collected from a clinical trial (INNOVATE) and real-world evidence (RWE) from a prospective observational study conducted in Greece. Methods: A Markov cohort model was developed in Microsoft Excel to compare the costs and outcomes of omalizumab plus standard therapy (ST, primarily comprised ICS, LABA and SABA) versus ST alone. The time horizon was that of a lifetime. Both direct and indirect costs were incorporated. Health outcomes considered were Quality Adjusted Life Years (QALYs). Costs and QALYs were discounted annually at 3.5%. Unit costs were taken from publically available sources. Productivity losses were calculated based on published data, while utility values were taken from the INNOVATE study. Deterministic and probabilistic sensitivity analyses were undertaken to test the robustness of the model results. Results: The addition of omalizumab to ST led to an incremental cost per QALY gained of € 27,888 based on INNOVATE trial and € 27,255 based on the RWE. The model appeared to be most sensitive to changes in the time horizon and the age of retirement. Results of the probabilistic sensitivity analysis showed that the probability of omalizumab being cost effective was 58% and 84%, at a willingness to pay threshold of € 30,000 and € 40,000, respectively. Conclusions: Omalizumab appears to be a cost-effective treatment option for adult patients with severe allergic asthma compared with ST in Greece, confirmed by both trial and real-world data. PRS58 Cost-Effectiveness Analysis of Indacaterol/Glycopirronium (QVA149) as a Maintenance Bronchodilator Treatment in Adult Patients With Chronic Obstructive Pulmonary Disease in Spain Granell M 1, Giovanna M 2, Paz S 3, Betoret I 1 Farmaceutica, Barcelona, Spain, 2Outcomes 10, Castellon, Spain, 3Outcomes’10, Castellon, Spain . . . . 1Novartis Objectives: To assess the cost-effectiveness (CE) of indacaterol/glycopyrronium (QVA149; 85µg/43µg) as a maintenance bronchodilator treatment of adult patients with Chronic Obstructive Pulmonary Disease (COPD) versus salmeterol/fluticasone (SFC; 50µg/500µg). Methods: A CE model of micro-simulation over a 3-, 5-, 10-year and lifetime horizon was developed from the perspective of the Spanish National Healthcare System. Patients progress through subsequent COPD stages based on their baseline characteristics and considering the natural decline of Forced Expiratory Volume in 1 second (FEV1) and exacerbation rate. In the model this is counteracted by treatment-associated FEV1 improvement from baseline and exacerbation rate reduction associated to each treatment vs. placebo, which were obtained by direct and indirect comparison of primary data from TORCH (SFC vs. placebo), SHINE (QVA149 vs. placebo) and ILLUMINATE (QVA149 vs SFC) clinical trials. The considered outcomes were life years (LY) gained and quality-adjusted life years (QALYs). Cost estimates (Euros 2014) include drugs, disease management and mild/severe exacerbation expenditures from Spanish health care cost databases and publications with a discount rate of 3% for costs and effects. Results: QVA149 has shown to be less costly and more effective than the fixed combination of SFC with respect to both LY and QALYs gained. The cost per patient treated with QVA149 over a 3-, 5-, 10-year and lifetime period was estimated to be € 108, € 182, € 305, and € 467 lower than with SFC, which resulted from avoiding exacerbation costs and decreasing maintenance cost in relation to slowing COPD progression. Therefore, QVA149 was estimated to be dominant over SFC with respect to both cost-effectiveness and cost-utility. Conclusions: Due to its higher effectiveness in improving FEV1 and reducing COPD exacerbations, QVA149 has shown to be more cost-effective than SFC. PRS59 Cost-Effectiveness Analysis of Allergen Immunotherapy in Patients With Grass Pollen-Induced Allergic Rhinitis in Spain Valero A 1, Westerhout K Y 2, van de Wetering G 2, Pérez-Alcántara F 3, Azpeitia A 4, Najib M 5 Clinic Universitari de Barcelona, Barcelona, Spain, 2Pharmerit International, Rotterdam, The Netherlands, 3Oblikue Consulting, Barcelona, Spain, 4Stallergenes Ibérica S.A, Barcelona, Spain, 5Stallergenes, Antony, France . . . . . . . 1Hospital Objectives: To determine the relative impact of treatment with Oralair®, Grazax®, Pangramin®, Pollinex Quattro®, and symptomatic drug treatment (SDT) on clinical effects and health care costs in patients with grass pollen-induced allergic rhinitis (AR) in Spain. Methods: The effects of three years of drug treatment on qualityadjusted life years (QALYs) and costs were assessed using a Markov model with a nine-year time horizon. Symptom score data were extracted, and the relative efficacy on QALYs was assessed through a network meta-analysis (i. e. indirect comparison) of 3-year, placebo-controlled, clinical trial data. Patient symptom scores were translated into the impact on quality of life by means of published sources. Costs associated with drug treatment and other health care resources were calculated. The incremental costs and QALYs gained were generated accordingly. The uncertainty around the model outcomes was determined by means of sensitivity analyses. Results: The base case analysis over 9 years estimated incremental QALYs of 0.005 (95%CI: -0.024; 0.038), 0.016 (95%CI: -0.034; 0.063), 0.059 (95%CI: 0.024; 0.107), and 0.143 (95%CI: 0.102; 0.195) when Oralair® was compared to Grazax®, Pangramin®, Pollinex Quattro® and SDT, respectively. Corresponding incremental costs were -€1,063 (95%CI: -€1,306; -€779), €109 (95%CI: €206; €428), €572 (95%CI: €321; € 864), and € 1,360 (95%CI: € 1,110; € 1,649). Hence, Oralair® was predicted as dominant relative to Grazax®, while ICERs of € 6,931/QALY, € 9,703/QALY, and € 9,517/QALY were estimated relative to Pangramin®, Pollinex Quattro®, and SDT, respectively. Apart from drug costs, the sensitivity analyses suggest that results were mostly driven by drug-specific symptom score values, duration of the pollen season, and inputs for immunotherapy discontinuation. At a willingness-to-pay threshold of € 20,000, the probability of Oralair® being the most cost-effective treatment option is 65%. Conclusions: Oralair® is cost-effective relative to Grazax®, Pangramin®, Pollinex Quattro® and SDT in grass pollen-induced AR in Spain. Findings are confirmed by extensive sensitivity analyses. PRS60 Cost-Effectiveness of Endobronchial Valve Therapy for Severe Emphysema: A Model-Based Projection Based on the Vent Study Pietzsch J B 1, Garner A M 1, Herth F J 2 Inc., Menlo Park, CA, USA, 2University of Heidelberg, Heidelberg, Germany . . . . . . 1Wing Tech Objectives: Endobronchial valve therapy (EBV) is an innovative treatment that has been shown to be safe and effective in selected subgroups of patients with severe emphysema. The objective of our study was to assess the cost-effectiveness of valve treatment in the German health care system when compared to medical management. Methods: Clinical data from a subset of the Endobronchial Valve for Emphysema Palliation Trial (VENT) provided information about clinical events, health-related quality of life, and disease staging through 12 months. This information was subsequently used as input to a previously published Markov model to project longer-term disease progression, mortality, and health resource utilization. From this combined analysis, we computed the 5-year and 10-year incremental cost-effectiveness ratio (ICER) in euros per quality-adjusted life year (QALY). Costs and effects were discounted at 3% per year. Results: EBV therapy led to clinically meaningful disease restaging at 12 months (37.8% of cohort improved staging, compared to 0% in control). Over 5 years, EBV was projected to increase survival from 66.4% to 70.7%, and to add 0.22 QALYs. Costs were estimated to increase by € 10,299, resulting in an ICER of € 46,322/QALY. Over 10 years, 0.41 QALYs were gained at additional cost of € 10,425, yielding an ICER of € 25,142/QALY. Conclusions: Our model-based analysis suggests that EBV leads to clinically meaningful changes in disease staging and progression when compared to medical management, with resulting gains in unadjusted and quality-adjusted life expectancy. Relative to the acknowledged willingness-to-pay threshold of € 50,000/QALY, our results indicate EBV is a cost-effective therapy in the German health care system. PRS61 Can Improved Treatment of Allergic Rhinitis Improve Workplace Productivity? The Role of Intranasal Formulation of Azelastine Hydrochloride and Fluticasone Propionate (Dymista) Harrow B 1, Hofmeister J 1, Gever L N 1, Karafilidis J 1, Lacey M J 2, Scheibling C M 2, Schneider J E 2 1Meda Pharmaceuticals, Inc., Somerset, NJ, USA, 2Avalon Health Economics, Morristown, NJ, USA . . . . . . . . . . . Objectives: Allergic rhinitis (AR) affects 10-20% of the US population, with treatment costs exceeding $6 billion annually and has been shown to have a substantial impact on productivity. In the U. S. AR is estimated to result in 3.5 million lost work days and 2 million lost school days annually. AZ/FP is an intranasal formulation of azelastine hydrochloride and fluticasone propionate in an advanced delivery system indicated for the relief of symptoms of seasonal AR (SAR). Patients treated with AZ/FP experience significantly greater and faster symptom relief in comparison to first-line therapy in trials, and thus have the potential to positively impact workplace productivity. We use an economic model to calculate the economic effects on workplace productivity associated with moving AZ/FP from third-tier to secondtier pricing and reimbursement. Methods: Population is SAR sufferers seeking treatment. AZ/FP is assumed to gain market share annually with second-tier pricing. Time horizon is one year and five years. Four step approach: (1) estimate total number of AR-related symptomatic days; (2) calculate total number of AR-related episodes per year multiplied by number of days per episode; (3) estimate number of these days that occur during a standard 5-day work week; & (4) Estimate proportion of AR symptomatic days resulting in absenteeism or presenteeism. Results: For a typical health plan, the estimate of expected number of absenteeism and presenteeism days per AZ/FP patient associated with moving AZ/FP from Tier 3 to Tier 2 resulted in a reduction of 4,729 AR-symptomatic days annually. Total workplace cost savings ranged from $168,838 (Year 1) to $190,937 (Year 5), with the proportional effects on absenteeism and presenteeism being roughly equal. Conclusions: AZ/ FP offers an appropriate means of adhering to AR practice guidelines and improving outcomes. This workplace productivity model shows that the added benefits to employers could be substantial. PRS62 The Potential Societal Cost Benefits of Improved Inhalation Technique With Duoresp® Spiromax® (Budesonide + Formoterol Fumarate Dihydrate) Compared With Symbicort® Turbuhaler® for the Management of Asthma and Chronic Obstructive Pulmonary Disease in Sweden Lewis A 1, Blackney M 1, Torvinen S 2, Lindqvist F 3, Safioti G 3, Grundström J 4, Polyzoi M 4, Plich A 2 1Covance Inc., London, UK, 2Teva Pharmaceuticals Europe B.V, Amsterdam, The Netherlands, 3Teva Pharmaceuticals AB, Helsingborg, Sweden, 4Parexel International, Stockholm, Sweden . . . . . . . . Objectives: DuoResp® Spiromax® (budesonide + formoterol fumarate dihydrate) is a fixed-dose combination (FDC) of inhaled corticosteroid (ICS) + long-acting beta agonist (LABA) in a novel dry powder inhaler (DPI). An economic model was developed to assess the potential societal cost benefits of improved inhalation technique with DuoResp® Spiromax® compared with Symbicort® Turbuhaler®– a DPI delivering the same FDC – in the management of adult patients with persistent asthma and chronic obstructive pulmonary disease (COPD) in Sweden. Methods: The eligible adult patient population was based on statistics from the National Board of Health and Welfare in Sweden. Societal costs (lost productivity) were based on the annual number of work-days lost for asthma and COPD patients in Sweden and the United Kingdom (UK), respectively, and the average daily cost of sick leave in Sweden. Frequency of poor inhalation technique with Symbicort® Turbuhaler® and the subsequent increased risk of unscheduled health care events were taken from a large (n=1,664) cross-sectional, Italian observational study. The estimated reduction in the proportion of patients with poor inhalation technique with DuoResp® Spiromax® A599 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 compared with Symbicort® Turbuhaler®was based on a conservative assumption. Results: An estimated 167,666 adult patients used Symbicort® Turbuhaler® annually in Sweden and were therefore eligible for treatment with DuoResp® Spiromax®, with 72,935 of these exhibiting poor inhalation technique. Based on the predicted improvement in inhalation technique with DuoResp® Spiromax® compared with Symbicort® Turbuhaler® – and assuming a hypothetical uptake of DuoResp® Spiromax®reaching 25% in years 4 and 5 – estimated societal cost savings, through the avoidance of 147,158 lost productive days, totalled SEK285.4 million (€ 31.2 million). Conclusions: DuoResp® Spiromax® has the potential to improve inhalation technique compared with Symbicort® Turbuhaler®, which would likely result in substantial societal cost savings. PRS63 Impact of Omalizumab On All-Cause and Asthma-Related Health Care Resource Utilisation in Patients With Moderate or Severe Persistent Asthma Turner S J 1, Nazareth T 2, Raimundo K 3, Zhou H 4, Ortiz B 2, Yu T C 2, Li L 5 1Novartis Pharmaceuticals, East Hanover, NJ, USA, 2Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA, 3Genentech Inc., South San Francisco, CA, USA, 4KMK Consulting Inc., Florham Park, NJ, USA, 5Career International Inc., Shanghai, China . . . . . . . . . Objectives: Increased health care resource utilisation (HCRU) is associated with inadequately controlled asthma. Here, we evaluate the impact of omalizumab on HCRU in patients with moderate or severe persistent asthma. Methods: A retrospective case-crossover study was conducted using the Truven MarketScan database. Data between 1-January-2007 to 30-September-2012 was collected for analysis. Patients included in the analysis had to have a diagnosis of moderate or severe persistent asthma, be ≥ 12 years of age during the analysis period, have had 2 years of continuous enrolment (1 year pre/post omalizumab index date), and have had exposure to omalizumab continuously for ≥ 12 months. All-cause and asthmarelated HCRU during the years, pre and post omalizumab initiation, were compared using McNemar tests and 2-sided paired Student’s t test. Data were stratified by asthma severity based on NHLBI criteria. Results: A total of 429 patients (mean age, 46.6 years; female, 59.0%; Moderate= 340 [79.3%], Severe= 89 [20.7%]) from the database were included in the analysis. The use of omalizumab was associated with 49.3% (p= 0.0003), 54.0% (p= 0.001), and 35.3% (p= 0.1466) reductions in the mean number of asthma-related ER visits and 69.2% (p= 0.0005), 65.5% (p= 0.0045), and 80.0% (p= 0.0449) reduction in the mean number of asthma-related hospitalisations among All, Moderate, and Severe asthma patients, respectively. The mean length of stay for asthma-related hospitalisations was also reduced to 71.2% (p=0.0002), 64.5% (p= 0.0016), and 90.6% (p= 0.0442) in All, Moderate, and Severe patients respectively. All-cause ER visits were reduced by 30.8% (p= 0.0014), 32.5% (p= 0.0045), and 25.0% (p= 0.1348), and hospitalisation reduced by 48.9%, 45.2%, and 64.7% (all p≤0.0155) in All, Moderate, and Severe asthma patients respectively. Cohort analysis of severe asthmatics was limited by sample size. Conclusions: In patients with moderate persistent asthma, omalizumab use was associated with significant reductions in all-cause and asthma-related HCRU. PRS64 Device Handling Errors and the Impact on Quality of Life and Health Care Resource Use in Asthmatic Patients Jha A 1, Heron L 2, Marshall J 1, Dunlop W 1 1Mundipharma International Ltd., Cambridge, UK, 2Adelphi Values, Bollington, UK . . . . Objectives: Correct device technique has a significant influence on the delivery of inhaled therapies and can thus impact the control and management of asthma. The objective of this literature review was to examine the impact of device handling errors on QOL and health care resources to understand the potential value of a novel inhaler device for health care systems and patients. Methods: A literature search of articles published 2009–2013 was undertaken using MeSH terms and key words in MEDLINE®, supplemented by a grey literature review and searching of reference lists. Article selection and relevant data extraction were based on key words relating to handling error, QOL, and health care resource use. Results: Of 575 potentially relevant publications, 22 were selected for in-depth review. Papers reported 25–73% of patients make critical handling errors that lead to no-dose or reduced-dose delivery on first use of devices. Incorrect inhaler use was four times more frequently reported in patients with uncontrolled asthma than in patients with controlled asthma. Poor asthma control also impacts resource use: poorlycontrolled patients made twice as many ER visits, and spent 2–3 times more time consulting with physicians than controlled patients (either physician visits or time speaking to physicians). Asthma control also impacts QOL: poorly-controlled patients reported health-state utility (EQ-5D) values of 0.52–0.69, compared to 0.88– 0.93 for well-controlled patients. Conclusions: Handling errors with devices can lead to poorly-controlled patients, resulting in reduced QOL and increased health care resource use. New inhaler devices represent an opportunity to reduce errors and improve asthma control, therefore improving QOL and reducing resource use. Further research is required to model the relationship between a reduction in handling error and improved asthma control status, and the subsequent impact on resource use and QOL. PRS65 Medium Term Avoided Costs: High-Dose Hypoallergenic House Dust Mite Preparation Immunotherapy VERSUS Conventional Symptomatic Treatment Hidalgo A 1, Rodríguez-Marco A 2, Arias-Muñoz M 2 1University of Castilla La Mancha, Toledo, Spain, 2MERCK S.L, Madrid, Spain . . . Objectives: Quantifying cost difference between conventional symptomatic treatment of mite allergy and subcutaneous specific immunotherapy (SCIT) with high doses of hypoallergenic dust-mite preparation. Methods: Observational, retrospective and multicenter study carried out in Spain in 2013.419 patients diagnosed with rhinitis and / or bronchial asthma for mite allergy were retrieved. Mean age 24.9 years (SD 14.4). Comparing the use of symptomatic medication (rescue and daily), diagnostic-tests, unscheduled medical care (allergist and emergency visits) and sick-leave days number associated with SCIT treatment versus no SCIT treatment. SCIT treatment vs no SCIT treatment costs ratio was performed: Used resources (Symptomatic medication, unscheduled medical care, diagnostic-tests, and 3 years SCIT treatment and sick-leave days) were quantified in € . Efficacy (decreased resource usage) of first year treatment was assumed during the remaining two years and during a three years follow-up period. Results: After a single year of SCIT all quantified resources (emergency and allergist visits, diagnostic tests, rescue medication and work absence days) diminished significantly (p< 0.05) from baseline. Reductions in resources’ cost: Hospital resources (100% in Hospitalizations; 82% in additional visits to the allergist; 79% in ER visits). In medication: (56% in rescue medication; 63% in daily medication). In diagnostic tests: (75% in spirometry testing broncho-dilation; 72% in O2 saturation measuring; 90% in FeNO measuring and 81% in chest radiographs. In leave sick days 94%. Ratio of comparative calculation described as SCIT treatment versus non SCIT treatment (or conventional symptomatic treatment) is 0.8. Conclusions: Considering 3 years of SCIT, and 3 follow up years of sustained efficacy after completing treatment, cost per patient SCIT treated is estimated at 20% below to the cost non SCIT treated patient. Direct costs are reduced by 64% and indirect costs by 94%. SCIT of hypoallergenic preparation of dust-mite allows cost savings vs conventional treatment. RESPIRATORY-RELATED DISORDERS – Patient-Reported Outcomes & Patient Preference Studies PRS66 Establishing the Relationship of Inhaler Satisfaction, Adherence, Smoking History and Allergic Rhinitis With Patient Outcomes: Real World Observations in US Adult Asthma Patients Harrow B 1, Price D 2, Pike J 3, Higgins V 3, Small M 3, Piercy J 3 Pharmaceuticals, Inc., Somerset, NJ, USA, 2University of Aberdeen, Aberdeen, UK, 3Adelphi Real World, Macclesfield, UK . . . . . . 1Meda Objectives: Improving asthma control has more recently focused on potentially modifiable clinical and behavioural characteristics including correct inhaler technique, treatment of concomitant allergic rhinitis (AR), adherence and smoking. This research aimed to establish the relationship of these factors with measures of asthma control and overall health status to add to the growing body of evidence helping to optimize asthma management interventions. Methods: Data were drawn from the USA 2013 Respiratory Disease Specific Programme, a crosssectional survey of adult asthma patients consulting for routine care. Partial Least Squares Path Modelling was used to quantify the inner model relationships between latent variables of patient-reported satisfaction of drug delivery, device functionality, device feedback (based on groupings of 12 inhaler device attributes), concomitant AR, adherence (Morisky Medication Adherence Scale), smoking history (smoking status, years smoked, number smoked per day), patient reported outcomes (Asthma Control Test, Jenkins Sleep Questionnaire, EuroQol-5D-3L) and physicianreported number of asthma exacerbations in the last 12 months. Patients not receiving inhaled maintenance therapy were excluded. Results: 243 patients met the inclusion criteria. All manifest variable loadings were positive, and a minimum Cronbach’s Alpha of 0.713 for the latent variables indicated unidimensionality of the manifest variables for each of the latent variables. Cross-loadings were also supportive of the hypothesised outer model. Better patient outcomes were significantly associated with patient satisfaction with drug delivery (p= 0.002), adherence (p= 0.049), negative smoking history (p< 0.001) and absence of concomitant AR (p= 0.005). The R2 value for outcomes was 13.3%, and the pseudo goodness of fit, which measures the overall prediction performance of the path model, was 19.5%. Conclusions: Improving patient satisfaction with inhaler drug delivery represents one potentially modifiable aspect of asthma management alongside appropriate treatment of AR, smoking cessation and improving adherence, which are likely to have a positive impact on asthma patient outcomes. PRS67 Tecepoc II Study. How to Improve the Inhalation Techniques in Patient with Copd. The Influence of Preferences Barnestein-Fonseca P 1, Vazquez-Alarcon R 1, Leiva-Fernandez F 1, Aguiar-Leiva V 1, Lobnig-Becerra M 1, Leiva-Fernandez J 2 1Distrito Sanitario Málaga (SAS), Málaga, Spain, 2Area Sanitaria Málaga Este-Axarquia, Málaga, Spain . . . . . . Objectives: to test the efficacy of two educational interventions to improve the inhalation techniques in patients with Chronic Obstructive Pulmonary Disease and the influence of patient´ preference. Methods: Design: Multicenter patients´ preference trial or comprehensive cohort design ISRCTN15106246. Patients: 465 COPD patients (to detect a difference between groups of 25%, 80% statistical power, 95% confidence level, 40% expected losses), with inhaled treatment, written consent. Non-probabilistic consecutive sampling. Allocation: Patients without strong preferences for a treatment were randomised: RCT group (block randomization), and those with strong preferences were given their choice: PPS group. Variables: Primary outcomes: Performance of correct inhalation technique. Secondary outcomes: Pick flow, Baseline dyspnea index (BDI), Functional status (forced spirometry). Interventions: Intervention-A: Written information. A leaflet with the correct inhalation technique for the main inhaler devices used in our area. Intervention-B: Intervention-A + individual training (by instructors). Follow-up: 12 month, visits: baseline, 1 month, 3rd month, 6th month, 12th month. Statiscal analysis: Mean, frequency, 95% confidence interval at baseline. Number Needed to Treat for a benefit (NNT) was calculated. Intention to treat analysis. Results: Predominance of males (91.4%), mean age 69.8 years (CI95%, 69.00-70.59); FEV1 (mean)= 55.91% (IC95%, 53.62-58.2), mixed respiratory pattern (65.9%). Severity stage: 15.7% mild, 44.1% Moderate, 40.3% Severe. Pharmacological treatment: inhaled-beta2-adrenergic (88.8%); inhaled-corticoster- A600 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 oids (76.7%); inhaled-anticholinergic (70.7%); mucolitycs (19.4%); xanthine (7.3%); oral-corticosteroids (1.3%). BDI: grade 2. Primary outcome: RCT cohorts: there was no difference between control and intervention A and there were statistically significative differences between intervention B versus control (p<0.0001), NNT=3.22 (IC95%, 2.27-5.88) and versus intervention A, NNT= 4.16 (IC95%, 2.63-10). In the PPS cohorts: there was a difference (p< 0.0001) between intervention B versus intervention A, NNT= 3.22 (IC95%, 2.32-5.55). The preferences enhanced a 6.7% the correct inhalation technique. Conclusions: The performance of a correct inhalation Technique improves with monitor training. The patients´ preferences enhance the efficacy of intervention. PRS68 Inhalation Technique Evolution After Training in Copd. The Role of the Device Barnestein-Fonseca P 1, Vazquez-Alarcon R 1, Leiva-Fernandez F 1, Aguiar-Leiva V 1, Lobnig-Becerra M 1, Leiva-Fernandez J 2 1Distrito Sanitario Málaga (SAS), Málaga, Spain, 2Area Sanitaria Málaga Este-Axarquia, Málaga, Spain . . . . . . Objectives: to test the efficacy of two educational interventions to improve the inhalation techniques per device in patients with COPD and the influence of patient´ preference. Methods: Design: Multicenter patients´ preference trial or comprehensive cohort design ISRCTN15106246. Patients: 465 COPD patients (to detect a difference between groups of 25%, 80% statistical power, 95% confidence level, 40% expected losses), with inhaled treatment, written consent. Non-probabilistic consecutive sampling. Allocation: Patients without strong preferences for a treatment are randomised: RCT group (block randomization), and those with strong preferences are given their choice: PPS group. Variables: Primary outcomes: Performance of correct inhalation technique. Independent variables: sex, age, Baseline dyspnea index (BDI), Functional status (forced spirometry). Interventions: Interv-A: Written information. A leaflet with correct inhalation technique. Interv-B: Interv-A + individual training (by instructors). Follow-up: 3 month, visits: baseline (V0), 1 month (V1), 3 month (V2). Statiscal analysis: Mean, frequency, 95% confidence interval. Intention to treat analysis. Results: Males (91.4%), mean age 69.8 years (CI95%, 69.00-70.59); FEV1 (mean)= 55.91% (IC95%, 53.62-58.2), mixed respiratory pattern (65.9%). Severity stage: 15.7% mild, 44.1% Moderate, 40.3% Severe. BDI: grade 2. Devices use: 67.3% Handihaler (Hd), 54.8% Turbuhaler (Th), 31.8% Accuhaler (Acc), 26.9% pMDI. Correct Inhalation technique: Hd: RCT-control: 11.7% V0,10% (V2); RCT-intervA: 10.9%, 17.5%; RCT-IntervB: 7.4%, 62.3% p< 0.0001. Th: RCT-control: 22% V0, 16.7% (V2); RCT-intervA: 8.7%, 24.4%; RCT-IntervB: 7.5 %, 57.5% p< 0.0001. Acc: RCT-control: 16.1% V0, 25% (V2); RCT-intervA: 17.9%, 23.1%; RCT-IntervB: 11.5%, 74.1% p< 0.0001. pMDI: RCT-control: 6.9% V0, 3.6% (V2); RCT-intervA: 12.5%, 19%; RCT-IntervB: 8.3%, 34.6% p= 0.025. There were statistically differences for all devices only in the intervention B arms (p< 0.0001). The preferences enhanced 1% for Handihaler, 12.7% for Accuhaler, 4.6% for Turbuhaler, 15.4% for pMDI the correct inhalation technique. Conclusions: The performance of a correct inhalation technique improves with monitor training for all devices. The patients´ preferences enhance the efficacy. PRS69 Identification of Dry Powder Inhaler Attributes, and their Relative Importance to Asthma and Chronic Obstructive Pulmonary Disease Patients, to Inform a Discrete Choice Experiment Hawken N A 1, Aballéa S 2, Torvinen S 3, Plich A 3 . . . . . 1Creativ-Ceutical, Luxembourg, Luxembourg, 2Creativ-Ceutical, Paris, France, 3Teva Pharmaceuticals Europe B.V, Amsterdam, The Netherlands Objectives: To identify characteristics of dry powder inhalers (DPIs) considered important by asthma and chronic obstructive pulmonary disease (COPD) patients, in order to select attributes and attribute-levels for a discrete choice experiment. Methods: Qualitative data was collected from a literature review performed to determine which inhaler attributes impact inhaler satisfaction and adherence among asthma and COPD patients using DPIs. Focus groups with asthma and COPD patients were conducted in France, with patients asked to cite and rate important features of their inhaler. Qualitative analysis of the transcripts was performed following International Society For Pharmacoeconomics and Outcomes Research (ISPOR) guidelines. Results: Results of the literature review revealed no overall consensus on the importance of different inhaler attributes in relation to inhaler satisfaction and adherence across studies. The most frequently reported attributes were: overall ease-of-use, low inspiratory flow requirements, presence of a dosing feedback mechanism, ergonomics of the inhaler mouthpiece, ease with which the device can be kept hygienic and the ease with which the medicinal dose can be prepared. Four discussion groups were held, with thirty patients participating. Overall, the degree to which the inhaler can optimise treatment convenience appeared to be the most important attribute to patients. In agreement with results of the literature review, patients also rated the following inhaler attributes as being important: size of the dose counter, ability to keep the mouthpiece hygienic, ergonomics of the inhaler mouthpiece, presence of a dosing feedback mechanism, ease with which a dose can be prepared, low inspiratory flow requirements. Conclusions: Results of this study provide an insight inhaler attributes most valued by asthma and COPD patients. Patients described their ideal inhaler to be small, with an ergonomic mouthpiece and an easy to use dose preparation mechanism, and providing enough medicine for at least a month of treatment. PRS70 Symptom Burden and Health Related Quality of Life in Patients With Idiopathic Pulmonary Fibrosis in Clinical Practice: Insights-Ipf Registry Pittrow D 1, Klotsche J 2, Kreuter M 3, Hoeper M M 4, Wirtz H 5, Koschel D 6, Claussen M 7, Andreas S 8, Grohé C 9, Geier S 10, Koppe U 10, Behr J 11 1Technical University Carl Gustav Carus, Dresden, Germany, 2Deutsches RheumaForschungsinstitut, Berlin, Germany, 3Thoraxklinik am Universitätsklinikum Heidelberg, . . . . . . . . . . . . . Member of the DZL, Heidelberg, Germany, 4Hanover Medical School, Member of DZL, Hanover, Germany, 5Universitätsklinikum Leipzig AöR, Leipzig, Germany, 6Fachkrankenhaus Coswig, Coswig, Germany, 7LungenClinic Grosshansdorf, akademisches Lehrkrankenhaus Universität Schleswig-Holstein; Mitglied des Deutschen Zentrums für Lungenforschung, Grosshansdorf, Germany, 8Lungenfachklinik Immenhausen, pneumologische Lehrklinik Universität Göttingen, Immenhausen, Germany, 9Evangelische Lungenklink, Berlin-Buch, Germany, 10Boehringer Ingelheim, Ingelheim, Germany, 11Ludwig Maximilian University, and Asklepios Clinics Gauting, Member of the DZL, Munich, Germany Objectives: We aimed to assess the symptom burden and health related quality of life in patients with idiopathic lung fibrosis. Methods: Patients have been consecutively enrolled in an ongoing prospective non-interventional registry in Germany, investigating clinical characteristics, clinical management practices and quality of life. IPF diagnoses were in agreement with the international IPF guideline published in 2011. Clinical parameters and treatment practice were recorded by the physician. Patients filled out the EQ-5D-5L, St George’s Respiratory Questionnaire (SGRQ), WHO-5 and the UCSD shortness of breath (SOB) scale. The time trade off (TTO) score was calculated for the EQ-5D. Results: To date (04 June 2014), 421 patients with IPF have been enrolled in the registry (mean age 68.6±9.5; 77% male). The mean six-minute walk distance was 271±200, mean % of predicted forced vital capacity was 72±20 and the mean % predicted DLCO was 35±16. Patients were treated with oral steroids (22.1%, as monotherapy in 7.1%); N-acetylcysteine (34.8%), pirfenidone (47.2%), and long-term O2 therapy (34.4%). The physician rated the disease in 35.6% as stable, in 31.1% as slowly progressing and in 11.9% as rapidly progressing. One in four patients described their current state of health as at least good, and every fifth as poor. The mean EQ-5D TTO score was 0.8±0.2. 45% of the patients showed depressive symptoms based on the WHO-5. The mean SGRQ sum score was 47.7±20.1 describing difficulties with breathing in the previous 3 months. Higher EQ-5D TTO scores were significantly associated with a lower number of comorbid diseases (r= -0.31), higher 6-minute walk distance (r=0.20), higher FVC % pred (r=0.27), less depression (r=0.66) and lower SGRQ scores (r= -0.72). Conclusions: The IPF patients in this large registry had a more severe disease, a higher symptom burden and more compromised quality of life compared to recent randomised controlled trials. PRS71 Translation and Linguistic Validation of Two COPD Symptom Diaries (Nicsi And Emsci) for Use in 14 Countries Eremenco S 1, Albuquerque P 2, Arnold B J 3, Trundell D 4, Hareendran A 5 1Evidera, Inc., Bethesda, MD, USA, 2Almirall, S.A., Barcelona, Spain, 3FACITtrans, Elmhurst, IL, USA, 4Evidera, Inc., London, UK, 5Evidera, London, UK . . . . . . Objectives: The Nighttime Symptoms of COPD Instrument (NiSCI) and Early Morning Symptoms of COPD Instrument (EMSCI) were developed to support treatment benefit endpoints in global clinical trials. Translations that were conceptually equivalent to the English source version and easily understood by the target country populations were required. The purpose of this study was to translate and assess conceptual equivalence of the NiSCI and EMSCI for use in 14 countries: Austria, Bulgaria, Canada, Czech Republic, France, Germany, Hungary, Italy, Lithuania, The Netherlands, Poland, South Africa, Spain, and United Kingdom. Methods: The NiSCI and EMSCI were translated following ISPOR guidelines for linguistic validation of PRO measures (Wild et al., 2005) using the universal approach discussed in the second Task Force Report (Wild et al., 2009). The universal English, Spanish and French versions were previously translated (Eremenco et al., 2012). For the remaining languages, two forward translations by native translators, reconciliation of the forwards, one back-translation by an English-speaker fluent in the target language, and final reconciliation by a native speaking language coordinator were conducted for both measures. Harmonization was performed to ensure conceptual equivalence across languages. Interviews were conducted with five native-speaking COPD patients for each language/country combination. Interview data were analyzed to assess linguistic and cultural validity in each language and confirm conceptual equivalence. Results: Mean age of the sample (N= 80) was 60 years (range 41-83) and 54% were male. The translations were well understood and considered relevant, with patients raising only minor issues during interviews. Changes were made to the universal French (chest congestion), Hungarian (wheezing, chest congestion), Italian (chest congestion, moderately), and Lithuanian (instructions, wheezing, shortness of breath, experienced) following the patient interviews. Conclusions: All translated versions of the NiSCI and EMSCI in this study were found to be conceptually equivalent and acceptable for use in the 14 countries evaluated. PRS72 Testing E-PRO Device Usability During the Translation Process: A Case Study of the Exact in 7 Countries Eremenco S , Murray L Evidera, Inc., Bethesda, MD, USA . . Objectives: Usability testing of electronic Patient-Reported Outcomes (ePRO) instruments is typically conducted during instrument development, in the language/country of origin. It has been suggested that usability testing also be performed during the translation process. It is unclear whether this additional step is necessary. In this study, usability testing was conducted as part of the linguistic validation process in Simplified Chinese (China), German (Germany), French (France), Russian (Russia), and Spanish (Universal, tested in Chile, Spain, and US) for the Exacerbations of Chronic obstructive pulmonary disease Tool (EXACT), an e-PRO developed and tested in English (US). Methods: The translation process followed ISPOR guidelines (Wild et al., 2005). Cognitive interviews were conducted with 2-3 native-speaking COPD respondents per language/country combination in 2008. Subjects completed the EXACT in paper-pen screenshot format and were interviewed for translation validation. Subsequently they were instructed to use a PDA (Tungsten E2; CRF, Inc.) to complete the first 5 EXACT items and were interviewed regarding device usability. Interviewers rated subjects’ ability to use the device. Results: Subjects (N=20) were 45-84 years, 60% male, and 60% with secondary education or less. Most (n=18) had not used a PDA previously; all (n=20) reported A601 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 it was easy to use, the screens were easy to read, and that they could sign on and move through the questions with no difficulty. Subjects from the Spanish-speaking countries had the lowest reported difficulty turning the device on, with higher levels of difficulty observed by the interviewers in China, who reported “somewhat” for two of the three respondents. Interviewers observed that the majority (n= 15) could “easily” or “very easily” use the device. Conclusions: Given the consistent ease-of-use findings in these diverse, device naïve subjects across 7 countries and the emphasis on subject training in clinical trials, it was determined that usability testing with future translations was unnecessary. PRS73 A Comparison of The Reliability and Validity of the Four-Item and Six-Item NISCI Symptom Summary Scores Mocarski M 1, Trundell D 2, Zaiser E 2, Garcia Gil E 3, Lamarca R 3, Hareendran A 2 1Forest Research Institute, Jersey City, NJ, USA, 2Evidera, Inc., London, UK, 3Almirall S.A., Barcelona, Spain . . . . . . Objectives: The Nighttime Symptoms of COPD Instrument (NiSCI), developed through qualitative research with patients, includes six symptom items: coughing, wheezing, shortness of breath, difficulty bringing up phlegm, chest congestion, and tightness in the chest. A symptom severity score is computed based on these items. In situations where patient burden is a major consideration, a smaller set of items may be preferable. Clinicians identified coughing, wheezing, shortness of breath, and difficulty bringing up phlegm as the most relevant for COPD patients. Exploratory psychometric analyses were conducted for the symptom summary score based on these four items compared with all six items. Methods: Psychometric properties of the four-item versus six-item symptom summary scores were examined using phase 3 clinical trial data from a random split-half sample. Symptom summary scores were tested for: internal consistency using Cronbach’s Alpha; test-retest reliability using Intraclass Correlation Coefficients (ICC) and Concordance Correlation Coefficients (CCC); convergent validity using Spearman Rank Order Correlation Coefficients; and known-groups validity using ANOVA and Scheffe’s test for pair-wise comparisons. Results: Patients (n=832) were aged 40–93 years (mean 63.78 ±-9.07 [SD]) and 51% were male. Both scores were internally consistent and valid. Cronbach’s alpha was slightly higher for the six-item symptom summary (0.85) versus the four-item symptom summary (0.78). ICC and CCC scores were 0.85 for the six-item and 0.84 for the four-item symptom summary scores. Likewise, construct validity and known-groups validity were similar for both scores (p> 0.05). Conclusions: The NiSCI symptom summary score based on four items has similar psychometric properties to the six-item symptom summary score. Both scores have measurement properties suitable for use in clinical trials. Further work will test the psychometric properties of the instrument administered with four items. PRS74 Health-Related Quality of Life (HRQOL) in Patients With Idiopathic Pulmonary Fibrosis Yount S 1, Beaumont J 1, Kaiser K 1, Wortman K 1, Chen S Y 2, Van Brunt D 2, Cella D 1 1Northwestern University, Chicago, IL, USA, 2Biogen Idec, Cambridge, MA, USA . . . . . . . . Objectives: Idiopathic pulmonary fibrosis (IPF) is a progressive disease characterized by declining lung function, leading to debilitating limitations on activity which may negatively impact HRQOL. However, HRQOL data in this population are limited. The primary objective was to evaluate HRQOL in IPF using measures from the Patient Reported Outcomes Measurement Information System (PROMIS). The secondary objective was to examine the association between key symptoms and HRQOL. Methods: Individuals with IPF were recruited via patient advocacy organizations to complete an online survey consisting of PROMIS-29 health profile, PROMIS-Dyspnea, dyspnea measured by Modified Medical Research Council Dyspnea Scale (MMRC), self-reported cough, and cough subscale of the ATAQ (A Tool to Assess Quality of life)-IPF. PROMIS-29 scores have mean= 50, SD= 10 in the US general population; PROMIS-Dyspnea scores are referenced to a chronic obstructive pulmonary disease (COPD) sample. Results: The 275 survey participants showed worse mean PROMIS-29 scores than the general population in all measured domains (mean value: anxiety=63.9; depression=61.9; fatigue=60.1; pain=62.6; sleep disturbance= 55.8; physical function= 36.2; social role= 42.2). Dyspnea severity was associated with worse mean PROMIS-29 scores (all p< 0.05). PROMIS-Dyspnea (mean= 58.7) and Functional Limitations Due to Dyspnea (mean= 58.4) scores were worse than the COPD reference population. Cough severity was associated with worse HRQOL measured by ATAQ-IPF. Reliability of PROMIS-29 scores exceeded 0.65 and were moderately correlated with measures of similar constructs. A limitation of the study is that data drawn from a sample from advocacy organizations might not be generalizable to the entire IPF population. Conclusions: Patients with IPF report substantial deficits in HRQOL, particularly with respect to physical function, anxiety, pain, depression and fatigue. Patients suffering from dyspnea and cough had poorer HRQOL. These deficits should be monitored in clinical practice and evaluated in investigational trials aiming to improve the HRQOL of IPF patients. PRS75 Asthma and Copd In Spain: Quality of Life and Health RESOURCES CONSUMPTION Collar JM Mundipharma Pharmaceuticals, Madrid, Spain Objectives: To analyze the impact that asthma and chronic obstructive pulmonary disease (COPD) have on patients Health Related Quality of Life (HRQOL) and on health resources consumed by the Spanish Health System. Methods: Data obtained from the last Spanish National Health Survey (SNSHS) 2011-12, on adult population (> 15y), identifying patients with a diagnostic of asthma or COPD. Descriptive statistic analysis was carried out, specially focused on self-reported HRQOL (EQ-5D-5L) and health resources utilization. The EQ-5D results were translated into QALYs, with the social tariffs validated in Spain. Other demographic factors, potentially related with the mentioned respiratory diseases prevalence, as age, sex, smoking habits were analyzed. Four groups were compared: 1) asthma patients, 2) COPD patients, 3) patients with other chronic conditions, and 4) global Spanish population. Results: Data from 21,007 adults was recorded. Prevalence of asthma was slightly above COPD (5.4% vs 4.7%), and 47.2% were suffering from chronic diseases. Asthma patients were younger than those with COPD (51y vs 61y, as an average, respectively). Concerning EQ-5D results, the dimensions mainly affected either in COPD or asthma were, respectively, pain/discomfort (53.7% / 38.7%), mobility (42% / 27.4%), usual activities (34.4% / 21.8%), and depression/anxiety (32.4% / 25.4%). Based on the EQ-5D questionnaire, the QALYs calculation were 0.75 for COPD and 0.90 for asthma patients, the latter results were similar to the other chronic conditions and to the global population. Health resources in COPD patients were significantly higher than in the other 3 groups analyzed, considering office visits (GPs/Specialists), hospitalization and emergency unit visits during the last year. Conclusions: Patients diagnosed with asthma have a similar HRQOL compared to the rest of the population, but COPD patients have it worse. COPD is also associated with a higher health resources consumption, which implies a relevant impact on the Spanish National Health System. PRS76 Health Related Quality of Life Among Young Smokers Muragundi P M 1, Dharmagadda S 2, Ligade V S 3, Udupa N 3, Naik A N 4 College of Pharmaceutical Sciences, Manipal, India, 2Manipal College of Pharmceutical Sciences, Manipal, India, India, 3Manipal University, Manipal, India, 4Manipal College of Pharmaceutical Sciences, Manipal University, Manipal, India . . . . . . . . 1Manipal Objectives: As young students are more prone to smoking and associated dangers it affects their health related quality of life (HRQoL). This study was done to assess the HRQoL and other associated factors affecting HRQoL among smokers and non-smokers who are students. Methods: In this exploratory study a pilot tested questionnaire was used to collect information regarding young students, demographics, BMI, food habits, alcohol consumption, family history of smoking as well as family history of diseases. For health related quality of life measurement both descriptive and visual analogue score (VAS) of EQ 5D 5L questionnaire ware used. The data collected was analyzed using SPSS 16.0.0. The test of significance was done by using Chi Square test for checking the associated habits and Mann Whitney U test is done to check the significance of association between HRQoL and smoking and other associated habits. Results: Total 126 students were included in the study. The age was 22.9±2.17 (Mean±SD) and BMI was 22.03±4.27 (Mean±SD) and all of them were residing in an University town. 63 % of the students were non-vegetarians and 44.8% were consuming alcohol at least once in a month. The mean EQ 5D visual analogue score of the studied population was found to be 83.3±11.5 (Mean±SD). The Chi square test showed the association between smoking with family history of smoking, alcohol consumption and soft drinks consumption (p< 0.001, 0.001 and 0.05 respectively). Mann Whitney U test showed as significant difference in the VAS scores among smokers and non-smokers (p< 0.05). Conclusions: Health Related Quality of Life is severely compromised by smoking, hence there is an urgent need to create awareness among young students. PRS77 Health Related Quality of Life And Health Care Utilization in Primary Care Patients With Moderate/Persistent Severity Asthma Garcia Ruiz A 1, Garcia-Agua Soler N 1, Quintano J 2 1University of Malaga, Malaga, Spain, 2Respiratory group of SEMERGEN, Córdoba, Spain . . . Objectives: The aim of this study is to evaluate the health related quality of life (HRQoL) and health care utilization in primary care (PC) patients with asthma treated with beclomethasone/formoterol. Methods: This study analyzed the HRQoL and health care utilization from a cohort of 65 patients over 6 months. Inclusion criteria were patients aged 18 years and older with moderate/severe persistent asthma (GINA criteria) treated with beclomethasone/formoterol at least 1-3 months before the inclusion in the study. Sociodemographic variables such as age, sex, duration of disease, severity of asthma, concomitant pathology were evaluated. The evaluation of the HRQoL were measured with the Asthma Quality of Life Questionnaire (AQLQ), and two generic questionnaires, EuroQol-5D and SF-36. The evaluation of the health care utilization included visits to PC, visits of care nursing and visits to accident and emergency (A&E) department and admissions. Statistical analysis: average ± standard deviation (SD); frequency and proportions. Inferential statistics in terms of average HRQoL and health care utilization were calculated using T-Student, Chi-square and ANOVA. Results: Average patients were female (60%), aged 49 years old (SD 2.16) with disease duration of 92 months (SD 18.34). The average health care utilization was: 3.43 (SD 0.35) visits to PC and 1.42 (SD 0.29) visits to nursing; analytical: 0.63 (SD 0.09); chest x-ray: 0.38 (± 0.08); ECG 0.32 (SD 0.08). The average of exacerbations without hospital admission was 1.09 (SD 0.19) and the A&E department visits of PC related with asthma was 0.43 (SD 0.11). Statistically significant differences (p < 0.05) and clinically significant between the beginning and end of the study on all forms of quality of life measured in these patients were found measured with AQLQ, EuroQol-5D and SF36. Conclusions: Beclomethasone/formoterol improved HRQOL in patients with asthma representing a good cost/utility relationship. PRS78 Health-Related Quality of Life Among Tuberculosis Patients in Pakistan: A Cross Sectional Study Using WHOQOL-BREF Iqbal M S 1, Iqbal M W 2, Bahari M B 1, Iqbal M Z 1 of Clinical Pharmacy, Faculty of Pharmacy, AIMST University, Kedah, Malaysia, 2Faculty of Law, Universiti Malaya, Kualalumpur, Malaysia . . . . . . . . 1Department Objectives: The aim of this study was to assess impairment in health-related quality of life (HRQoL) of pulmonary tuberculosis (PTB) patients in Pakistan. This study also evaluated utility of various socio-demographic and clinical factors that A602 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 are directly and indirectly linked with HRQoL of TB patients. Methods: A prospective cross-sectional study was conducted at a tertiary care hospital in a province of Pakistan, the Punjab. Data was collected by using WHOQOL-BREF questionnaire (Urdu version, pretested for reliability and validity) by means of face-to-face interviews and where possible by gender-based focus group sessions. In addition, facilitators also used few open-ended questions in order to get patients’ demographic and socioeconomic data. Participants were also asked to share their personal experiences of being diagnosed and treated with TB and what impact it had on their life style. All obtained data were analyzed using descriptive and inferential statistics. Results: The overall Cronbach’s alpha coefficient of the revalidated WHOQOL-BREF questionnaire was 0.785. The confirmatory factor analysis also provided an acceptable fit to a four-factor model in the studied sample. The scores for negative feelings, blue mood, depression, living place, personal relationships and sex life were significantly different in the psychological health and social relations domains. Age, gender and physical exercise were also significantly associated with the HRQoL of the patients. Conclusions: The WHOQOL-BREF was reliable and valid in the assessment of the HRQoL of TB patients in Pakistan. Despite the ability to cure TB, there was a significant impact on HRQoL of the TB patients. Till today, much attention is spent on curative and preventative mechanisms whereas the impact of TB on HRQoL is often neglected. patients recruited through an agency, Twitter and Asthma UK’s website. Participants had experienced a NCS and a SS <3 years previously. Medication history and Asthma Control Test (ACT) scores were collected. Interviews were audio-recorded, and thematically analysed. Results: The sample was 69% Caucasian, with six males and seven females, and a mean age= 47. Most were working full time or retired, and had many years since asthma diagnosis (mean= 22 years). NCS were related to reliever (n= 7), preventer (n= 4) or both (n= 2) medications with the most common NCS from one brand to another (n= 4). Participants reported negative feelings, difficulty using the medication and ineffective symptom control. Most participants used the NCS medication < 6 months (range < 24 hours–2 years). The SS for most was a return to their original pre-NCS medication (n= 11). Although most found getting their SS straightforward, this required at least two HCP visits for half the sample, and took up to two months to obtain. Most expected symptoms to improve following the SS but nine participants still had sub-optimal ACT scores. Lasting impacts included damaged relationships with HCPs, strong views about NCS and concern about NCS reoccurrence. Conclusions: Failure to inform and involve patients in medication changes can have lasting impacts. It is important to note that NCS can incur costs associated with rejection of medicines and additional consultations, thus negating any attempt at cost saving. PRS79 Systematic Literature Review Assessing Data on the Burden of Allergic Rhinitis from a Cost and Quality of Life Perspective RESPIRATORY-RELATED DISORDERS – Health Care Use & Policy Studies Hahn-Pedersen J 1, Boxall N 2, Maier W 2, Linneberg A 3, Serup-Hansen N 1 1ALK, Hørsholm, Denmark, 2Mapi, London, UK, 3Glostrup University Hospital, Glostrup, Denmark . . . . . Objectives: To assess published data on perennial allergic rhinitis/house dust mite allergic (PAR) patients and seasonal allergic rhinitis/grass pollen allergic (SAR) patients in order to establish the burden of allergic rhinitis (AR) and allergic asthma from both a quality of life (QoL) and cost perspective. Methods: A systematic literature review was conducted using Medline and Embase in eight pre-specified countries for the time period of January 2000 to January 2014. Search terms were related to QoL and/or cost. A total of 2963 abstracts and titles were identified. 50 abstracts met predefined criteria and provided data for calculations and collation. Results: Based on the RQLQ, the overall QoL of PAR patients was significantly worse than that of SAR patients (2.73 ± 0.12 cf. 2.04 ± 0.18, p≤ 0.001). In general, practical problems (mean: 3.80 ± SE: 0.08) and activities (mean: 3.70 ± SE: 0.11) were the most affected domains. Measured by the physical component score from the SF-36 questionnaire, the overall QoL of PAR patients was worse (p= 0.002) than that of SAR patients (49.06 ± 1.26 cf. 64.19 ± 7.89). For both PAR and SAR patients, domains measured by the SF-36 showed that vitality (mean: 59.95 ± SE: 3.24) was the most affected domain. Based on RQLQ, allergy immunotherapy improves eye symptoms, nasal symptoms, activities and practical problems the most in grass allergic patients. Direct comparison of total direct and indirect costs was complicated by differences in individual costs assessed among studies, however, the primary burden of costs is indirect and caused by high absenteeism and presenteeism. Conclusions: The QoL and economic burden of AR was substantial in the countries included in this review. However, limitations regarding the number of studies per country, heterogeneity between studies, and the lack of presented data have restricted the conclusions that could be drawn. PRS80 Prospective Study on Quality of Life (QOL) of Bronchial Asthma Patients in a Tertiary Care Teaching Hospital Yedavalli N 1, Shukla R 2, Nair S V 2, A S 2, Mohan M K 2 College of Pharmaceutical Sciences, manipal, India, 2Manipal University, Manipal, India . . . . . . . . 1Manipal Objectives: To determine the Quality of Life (QoL) of bronchial asthma patients in a tertiary care teaching hospital at baseline and at follow up. Methods: The study was conducted for a period of 6 months among 100 bronchial asthma patients, with and without co-morbidities, admitted to the pulmonary and medicine wards in the university hospital after obtaining the ethical clearance. The quality of life was assessed using Asthma Quality Of life Questionnaire (AQLQ). Patients were followed up after 4weeks from the date of discharge. Statistical analysis was performed using SPSS version 20. Results: The mean age of the study population was 53.30±14.59 having 61% of the patients as females. Data was analyzed by Wilcoxon signed rank test. The total score of Qol at follow up (5.94 ±0.76) showed an improvement with a P value < 0.0001 (wilcoxon signed rank t test) when compared with baseline (3.92 ± 1.04). A significant improvement in symptoms was seen at follow up (6.05 ± 0.82) from baseline (3.93 ± 1.14). The activity limitation improved with a P value < 0.0001 (Wilcoxon signed rank t test) from baseline (4.00 ± 1.08) to follow up (5.92 ± 0.83). At follow up (5.97 ± 0.82) the emotional function showed an enhancement on comparing with baseline (3.92 ± 1.14). The environmental stimuli showed no significant changes at baseline (3.66 ± 1.29) and follow up (3.84 ± 1.34). Conclusions: The results of the study suggests that clinical pharmacists have a major role in improving patient knowledge and thereby significantly improve the quality of life of the patient. PRS81 Impact of Non-Consented Switch and Subsequent Switch in Asthma Medication: Qualitative Study of Patient Perspective in the UK Ballinger R 1, Friedemann C 1, Golics C J 1, Lloyd A 2, Doyle S 3 1ICON PLC, Oxford, UK, 2ICON Plc, Oxford, UK, 3GlaxoSmithKline, Uxbridge, UK . . . . . . Objectives: Patient satisfaction with asthma treatment is associated with better adherence and achievement of treatment goals. While medication changes are sometimes instigated for clinical or financial reasons, changes without patients’ knowledge or involvement (‘non-consented switch’ [NCS]) can impact symptom control and patient-health care provider (HCP) relationships. This study explored experiences of patients who had a NCS and subsequently requested and received a second switch (SS). Methods: Interviews were conducted with 13 UK adult asthma PRS82 Evaluation of Selected Ken-Drgs in Greek Public Hospitals: The Degree to Which they Reflect Actual Expenditure and Average Length of Stay Siskou O 1, Galanis P 2, Kaitelidou D 2, Kalogeropoulou M 1, Kouli E 2, Thireos E 3, Vafeiadis J 4, Theodorou M 5, Prezerakos P 6, Lemonidou C 2, Liaropoulos L 2 for Health Services Management and Evaluation, National and Kapodistrian University of Athens, Athens, Greece, 2National and Kapodistrian University of Athens, Athens, Greece, 3Athens Medical Society, Athens, Greece, 4National Organization for Health Care Services Provision -EOPYY, Marousi, Greece, 5Open University of Cyprus, Latsia, Nicosia, Cyprus, 6University of Peloponnese, Sparta, Greece . . . . . . . . . . . 1Center Objectives: To evaluate the degree to which the officially pre-determined cost and average length of stay (ALoS) per selected KEN-DRG reflect actual resource use and to investigate other factors (e. g. size, type and location of hospital) correlated with charges and ALoS. Methods: Data were from the Health Regions data base and included discharges, ALoS, and charges in NHS Hospitals for the selected KEN-DRGs. Predetermined ALoS and cost per KEN-DRG were derived from Gazette 946/27/03/2012. Continuous variables did not follow normal distribution, so non parametric methods (Spearman’s correlation coefficient, Kruskal-Wallis test and Mann-Whitney test) were used. A two sided p-value ≤ 0.05 was considered statistically significant. Results: Higher length of stay compared with the officially pre-determined was found for Chronic Obstructive Pulmonary Disease (COPD) with complications (10.8 vs 9 days), Bronchitis & Asthma with complications (5.8 vs 4 days), Bronchitis & Asthma without complications (2.7 vs 2 days) and Breast Cancer with complications (7.9 vs 5 days). Consequently, officially pre-determined cost for Breast Cancer with complications (€ 965), Bronchitis & Asthma with complications (€ 792) and COPD with complications (€ 1,446) represents less than 90% of total hospital charges (total actual average charge: € 1,186; € 929; € 1,634 respectively). However, for the rest of the DRGs, under study, there were none or limited additional hospital charges. For all KEN-DRGs, size and population density was positively correlated with total charges (rs= 0.12, p= 0.018; median for Athens = 863 & semi-urban=752 p=0.09 respectively). Moreover, for Bronchitis & Asthma with and without complications, population density was found to be positively correlated with ALoS (median for urban areas= 5.1 and semi-urban= 4.7; p= 0.09 and median for urban areas= 3 and semi-urban= 2.5; p= 0.03 respectively). Conclusions: The introduction of KEN-DRGs was a useful first step to modernize the hospital reimbursement system. However, further revisions are required in order for KEN-DRGs to become more useful. PRS83 Patterns of Asthma Treatment Utilization in Newly Diagnosed Elderly Patients are Inconsistent With Asthma Management Guidelines Ali A K Eli Lilly and Company, Indianapolis, IN, USA . . Objectives: Asthma is under-diagnosed and under-treated in elderly, and utilization of asthma medications in this population is not well characterized. This study describes utilization patterns of asthma medications in newly diagnosed ≥ 50 patients. Methods: Medicare enrollees are followed for 12 months after first asthma diagnosis between Q1 2004- Q3 2012 (ICD-9-CM code: 493*). Treatment pattern analysis was applied to initiators of asthma medications. Patients with > 1 prescription of second-line therapy following exposure to first-line therapy are considered therapy switchers or augmenters. Patients with history of chronic obstructive pulmonary disease are excluded (ICD-9-CM codes: 490, 491*, 492*, 494*, 495* & 496). Results: Among 126,176 elderlies with asthma, 24,021 initiated asthma therapy within 12 months of asthma diagnosis (70% women, median age 70 years). 319 (1.3%) continued therapy with one drug class (first-line therapy); 13,940 (58%) discontinued first-line therapy; 3,469 (14.4%) switched to another drug class (second-line therapy); and 6,293 (26.2%) added second-line therapy (augmented). About 34% of patients initiated therapy with inhaled short-acting beta-agonists (SABA); majority of them either discontinued SABA or added a controller medication. 19% of elderly patients started with oral corticosteroids after asthma diagnosis, 77% discontinued them and 12% switched to another asthma controller. Among controller medications, inhaled corticosteroids (ICS) /long-acting beta-agonists (LABA) combination therapy, ICS monotherapy, and leukotriene antagonists respectively accounted for 19%, 12%, and 11% of treatment initiators. The majority of these anti- A603 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 inflammatory formulations were discontinued during 12 months after asthma diagnosis. Conclusions: Asthma treatment utilization patterns reflect poor asthma control among newly diagnosed elderly patients, and initiation of anti-inflammatory treatment after asthma diagnosis appears to be inconsistent with asthma management guidelines. PRS84 Doctors’ Failure in Observance of the COPD Management Guidelines: Case of the CZECH Republic Tichopad A 1, Koblizek V 2, Kolek V 3, Pecen L 1 s.r.o., Prague, Czech Republic, 2Charles University Faculty of Medicine and University Hospital, Hradec Králové, Czech Republic, 3Olomouc University Hospital, Olomouc, Czech Republic . . . . 1CEEOR Objectives: The primary objective of this study was to examine the accuracy of the GOLD 2011 strategy implementation among the Czech respiratory specialists, esp. with respect to the patients’ classification. The secondary objective was to explore what effect a misclassification has on inadequate use of inhaled corticosteroids (ICS). Methods: Multicentre cross-sectional study was conducted among COPD specialists, consisting of general questionnaire and patient-specific forms. A subjective classification into the GOLD 2011 groups as practiced by the health care professionals was examined and then compared with the objective classification achieved by rigorous software-computed classification. Adequacy of the ICS prescription was evaluated with regard to the subjective classification. Results: GOLD 2011 were claimed to be the leading guidelines for 143 out of 144 specialists involved, often accompanied by CPPS guidelines (83.3%) and the ACP/ACCP/ATS/ERS standards (50.7%). Based on 1355 patient forms, a discrepancy between the subjective and objective classification was found in 32.8% of cases. The most common reason for incorrect classification was erroneous symptoms assessment resulting in either under-estimation in 23.9% of cases and over-estimation in 8.9% of the examined patients’ records. Specialists seeing more than 120 patients per month were most likely to misclassify their condition, i. e. in 36.7% of all seen patients. In general, whilst examining the subjectively-driven ICS prescription, it was found that 19.5% of patients received ICS incorrectly, while in 12.2% of cases the ICS was erroneously omitted. Furthermore, with consideration to the objectively computed classification, it was discovered that 15.4% received ICS unnecessarily, whereas in 15.8% of cases the ICS was not prescribed though, in fact, it would be adequate seeing the patient’s condition. Women failed in correct prescription more frequently than men, predominantly by overprescribing ICS. Conclusions: Despite high awareness of the GOLD 2011 guidelines, its implementation is insufficient. Czech specialists tend to either under-classify or overuse the ICS. PRS85 Social Media Meets Population Health: A Sentiment And Demographic Analysis of Tobacco and E-Cigarette Use Across The “Twittersphere” Clark E M 1, Jones C 2, Gaalema D 1, White T J 1, Redner R 1, Everett R 1, Dodds P S 1, Couch M 2, Danforth C 1 1University of Vermont, Burlington, VT, USA, 2University of Vermont - College of Medicine, Burlington, VT, USA . . . . . . . . . . . . Objectives: Twitter, a popular social media outlet, has become a useful tool for the study of social behavior through user interactions called tweets. The location time, and message content of tweets provide invaluable social and demographic information for an applied comparison of social behaviors across the world. Our goal is to determine the density and sentiment surrounding tobacco and e-cigarette tweets and link prevalence of word choices to tobacco and e-cigarette use at various localities. Methods: All tweets with geo-spatial coordinates are salvaged from the twitter-feed, representing approximately 1% of the entire twitter-sphere. Pattern matching by tobacco and e-cigarette related keywords yield approximately 20,000 affiliated tweets per month from North America. The emotionally charged words that contribute to the positivity of various subsets of regional tweets are quantitatively measured using hedonometrics. We examined the density of these behavioral tweet indicators by region and tested the relationship between tweeted smoking sentiments and time-space-type coordinates over a 4-month span. Results: For states with ≥ 600 tobacco related tweets (N= 30), we find a strong positive correlation (Pearson’s r= 0.54, p< 0.01) between the relative tweet density per state and the average positivity of tobacco related tweets. However, state-to-state sentiment comparisons suggest the attitude toward tobacco use can vary. We also explore the relationship between the ratio of tobacco tweets per state-to-state smoking rate estimates. Our results illustrate significant variation in smoking sentiments by state and at varying regional scopes. Conclusions: It is anticipated that realtime analysis of nicotine and tobacco products using tweets will allow for more targeted forms of health policy planning and intervention. Regional density of nicotine and tobacco use related tweets yield insight to the prevalence of tobacco usage per capita. Sentiment analysis across the twitter-sphere can help illuminate hazardous health behavioral trends, which may lead to better targeting of health behavior interventions. PRS86 Sustainable Policy: Higher Medication Use & Adherence During Reimbursement of Pharmacologic Smoking Cessation Treatments Van Boven J F , Vemer P University of Groningen, Groningen, The Netherlands . . analysis was performed on real-world observational data from the years 2010-2013 in The Netherlands. Data on use and adherence was collected, in patients who were dispensed bupropion or varenicline in community pharmacies for the first time. Adherence was defined a using minimal 80% of the in guidelines recommended duration and intensity of use. Results: The study cohort consisted of 4,412 users of pSCT. The number of prescriptions was stable at 0.5 prescriptions per 1,000 inhabitants (dispensing prevalence, dp) during 2010. The prevalence was on average 0.8 dp, with peaks in the the 1st and 4th quarters of 2011. In 2012, the prevalence was stable at 0.4 dp. In 2013 was on average 0.5 dp, with a small peak in the 1st quarter. Adherence was 18% in 2010 and 2012 (non-reimbursement period), and 21% in 2011 and 2013 (reimbursement period). Conclusions: Not only the likelihood of starting smoking cessation, but also the extent of adherence to pharmacologic smoking cessation is higher during reimbursement. Increasing the awareness of health care providers on adherence issues is warranted. PRS87 Monte-Carlo Simulation to Estimate the Health Care Costs Avoided With Fluticasone Furoate/Vilanterol Due to Exacerbation Rate Reduction in Spanish COPD Patients Mayoralas S 1, Huerta A 2, Parrondo J 2, Rubio-Terrés C 3, Rubio-Rodríguez D 3 1Hospital Ramón y Cajal, Madrid, Spain, 2GlaxoSmithKline, Madrid, Spain, 3Health Value, Madrid, Spain . . . . . Objectives: Exacerbations are considered one of the main drivers of costs of Chronic Obstructive Pulmonary Disease (COPD). In moderate to severe COPD patients with a history of exacerbations, the addition of an inhaled corticosteroid (ICS) to a long acting β 2 agonist (LABA) has been associated with a decreased rate of exacerbations versus treatment with LABA alone. This study aims to estimate the health care costs that the addition of the ICS Fluticasone Furoate (FF) to the LABA vilanterol (VI) could avoid versus LABA monotherapy in Spanish patients due to the reduction of the rate of exacerbations. Methods: The number of moderate to severe COPD patients > 40 years old with a history of exacerbation potentially treated with FF+VI was estimated from Spanish prevalence data. 1-year Monte-Carlo simulations (one simulation per patient) were developed to simulate the number of moderate and severe exacerbations and the health care costs avoided with FF+VI versus VI from the National Health System (NHS) perspective. Monte-Carlo simulation was chosen as it allows simulating the effect of changes in different parameters obtained from clinical studies to describe real-life distributions. Parameters used in the simulations were the yearly rate of moderate-severe exacerbations with FF+VI and VI obtained from pooled-analysis of two head-to-head clinical trials (NCT01009463 and NCT01017952) and the costs of moderate and severe exacerbations obtained from an observational study in real-life Spanish setting. Results: 18,098 patients were included. FF+VI could avoid 7,424 moderate and severe exacerbations (95% confidence interval 7,411; 7,438) vs VI. The reduction in the number of exacerbations could lead to average health care costs avoided to the NHS of 3,278, 382€ (3,066, 703€ ; 3,336, 392€ ) in a year. Conclusions: Treatment with FF+VI could decrease the economic burden associated with COPD reducing the health care costs for the Spanish NHS due to the decreased rate of exacerbations compared with LABA (VI) monotherapy. PRS88 Determination of Availability of Antimicrobial Preparations for Treatment of Community-Acquired Pneumonia in Ukraine Iakovlieva L , Vasilieva A , Rybka A , Yurchenko O National University of Pharmacy, Kharkiv, Ukraine . . . . Objectives: Data of british specialists show, that 5-11 out of 1000 adults have community-acquired pneumonia (CAP) every year, which is 5-12% of all cases of lower respiratory tract infections. The rate of CAP took over 3% in the structure of total respiratory organs disease rate in Ukraine within years 2007-2011. Methods: The data of use of antimicrobial preparations (AP): cephalosporins (ceftriaxone), penicillins (amoxicillin and enzyme inhibitor), macrolides (azithromycin) and fluoroquinolones (ciprofloxacin, levofloxacin) for CAP treatment in Kharkiv hospital were used. Analysis of affordability as of index of payment capacity (Ca. s.) of trade names (TNs) provided in pharmaceutical market of Ukraine on indicated INN has been carried out to determine the availability of such preparations for wider population of Ukraine. Results: Results of the AP affordability analysis made in five groups of INN, showed, that the ratio of therapy of high, middle and low availability for each AP group is different, but in all pharmacological groups except for ceftriaxone group, highly available therapy prevails. Percentage of highly available preparations in the studied groups of INN is: azitromycin (92.2%) > ciprofloxacin (74.3%) > amoxicillin and enzyme inhibitor (67.5%) > ceftriaxone (49.02%) > levofloxacin (46.94 %). As of preparations of middle availability, costing 5% - 15% of average monthly salary, ceftriaxone preparations are most (49.02%), azitromycin preparations are least (6.49%). There are no low availability preparations in ciprofloxacin fluoroquinolone group. Cephalosporins and macrolides group preparations have one AP of low availability. Pharmacotherapy with application of preparations-analogous to levofloxacin, is costly, as 32.7% of preparations of this group are preparations of low availability (16 AP). Conclusions: Antibacterial preparations needed for treatment of CAP are present in the pharmaceutical market of Ukraine in wide range of preparations and cost. It makes them available to various social groups. . Background: The discussion on the reimbursement of Smoking Cessation Treatment (SCT) has known many stages in The Netherlands. From January 2011, SCTs were reimbursed, until January 2012 when the reimbursement of nicotine replacement therapies (NRTs) and pharmacotherapeutic SCT (pSCT) was discontinued. As of 2013, NRTs and pSCTs were again reimbursed for a maximum of one attempt per calendar year, provided they are accompanied by behavioural counselling. Objectives: To assess the impact of changes in reimbursement policy of pSCT on use and adherence. Methods: A retrospective dispensing database PRS89 How Much the Appropriate Tobacco Price Would Be?: A Discrete Choice Experiment of General Public in Japan Igarashi A 1, Goto R 2 of Tokyo, Graduate School of Pharmaceutical Sciences, Tokyo, Japan,, 2Kyoto University, Kyoto, Japan . . 1University Objectives: Though tobacco price increases are effective in reducing tobacco consumption and prevalence of smoking, tobacco-tax in Japan is still lower than those in other developed countries. General public, particularly non-smokers, may think that A604 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 tobacco tax is important as a source of revenue. The objectives of this research is to quantitatively assess about how general public value several aims of tobacco policy and the amount of tax revenue and to estimate “ideal” tobacco price where the price elasticity of tobacco consumption is given. Methods: We conducted a web-based survey from 9-15, Aug 2012 including a discrete choice experiment (DCE) to elicit general public’s preference for targeted smoking rate, tax revenue, tobacco policy for the younger generation and the usage of tax. Then, we numerically forecast general publics’ support rate of a tobacco policy as a function of tobacco price where price elasticity is given. Results: 1,077 completed questionnaires and were eligible for analysis. They statistically prefer the tobacco policy which results in lower prevalence of smokers, higher tax revenue, targeted for young people and tax revenue is used for general purposes. Given the price elasticity of 0.3, 0.4 and 0.5, highest tobacco price, which acquire the highest support from general public, was around JPY871 (EUR6.2), JPY555 (EUR4.0) and JPY454 (EUR3.2), respectively. Conclusions: The present price of cigarette per one pack is around JPY460 (EUR 3.3). When smokers are highly responsive to price elevation and are successfully quit after the elevation, present price of cigarette is suitable when we consider public’s preference for tax revenue as well as normal policy measures. Rooms for further price elevations depends on the price elasticity of cigarette consumption. Sensory Systems Disorders – Clinical Outcomes Studies PSS1 Efficacy of Treatments for Macular Oedema Secondary to Branch Retinal Vein Occlusion: A Network Meta-Analysis Regnier S , Bezlyak V Novartis Pharma AG, Basel, Switzerland . . Objectives: To indirectly compare the efficacy of approved treatments for macular oedema secondary to branch retinal vein occlusion (BRVO). Methods: Randomized, controlled trials (RCTs) evaluating the efficacy at month 6 (or month 12) of at least two of ranibizumab 0.5mg pro re nata (after 6 monthly doses), aflibercept 2.0mg monthly, dexamethasone implant (retreatment interval ≥ 6 months), laser photocoagulation or sham were identified from a published systematic review, congress abstracts, clinicaltrials. gov and Novartis data on file. Outcomes of interest were the mean change in best-corrected visual acuity (BCVA, measured as Early Treatment Diabetic Retinopathy Study letters), and the percentage of patients gaining ≥ 15 letters. For two studies, the standard deviation around mean BCVA gain was estimated. A Bayesian network meta-analysis with random treatment effects was used to compare mean BCVA gains and odds ratios (ORs) for gaining ≥ 15 letters. Results: The analysis included seven RCTs. Patient-level data for three ranibizumab trials were re-analysed to match key inclusion criteria (BCVA and duration of disease at baseline) from aflibercept studies. Statistically significant (p< 0.05) BCVA letter gains were found for ranibizumab monotherapy (+11.6), ranibizumab plus laser combination therapy (+10.1) and aflibercept (+10.1) versus laser, and for ranibizumab monotherapy versus dexamethasone (+8.0). ORs versus laser for gaining ≥ 15 letters were 3.24 (95% credible interval, 1.03–12.56; p< 0.05) and 3.07 (0.63–14.8) for ranibizumab monotherapy and aflibercept, respectively. Mean BCVA letter gains were numerically higher with ranibizumab monotherapy compared with aflibercept (+1.5 letters [-5.3–8.8]; OR for gaining ≥ 15 letters, 1.06 [0.16–8.94]). The probability of being the most efficacious treatment based on letters gained was 57% for ranibizumab, 15% for ranibizumab plus laser and 28% for aflibercept. Conclusions: The analysis confirms the superiority of ranibizumab over dexamethasone implant and laser therapy, and shows that ranibizumab has the highest likelihood of being the most efficacious treatment for BRVO. PSS2 Comparative Effectiveness of An Acellular Synthetic Matrix As An Adjunct to Standard Care In The Treatment of Venous And Mixed Leg Ulcers: Modeling of Clinical Data And Routine Data Augustin M 1, Gutknecht M 1, Anastasiadou Z 1, Heyer K 1, Harding K 2, Shannon R J 3, Heinrichs E L 4 1University Medical Center Hamburg-Eppendorf, Hamburg, Germany, 2Cardiff University, Cardiff, UK, 3Global Health Economic Projects LLC, Clifton Park, NY, USA, 4Daresbury Innovation Centre, Daresbury, Cheshire, UK . . . . . . . . . Objectives: Recently, advanced treatments such as topical growth factors, biological dressings, and tissue engineered products have been developed in order to promote healing as well as decrease the time to healing of chronic leg ulcers. Nevertheless, if standard care fails, there is no widely accepted, easy to use secondline treatment primarily because most of these advanced treatments are expensive and need further evidence of effectiveness in everyday clinical practice. The purpose of our study was to assess the effectiveness of a new acellular synthetic matrix (ASM) as an adjunct to standard care, using healing time as endpoint, in venous and mixed leg ulcers in German everyday clinical practice. Methods: Prospective data on venous and mixed leg ulcers from a safety and effectiveness study on an acellular synthetic matrix was compared retrospectively to matched data from a German Registry of Chronic Wounds (DRCW) database containing German patients with equivalent leg ulcer disease, patient age and gender, baseline wound area, wound duration, same outcome measures, and comparable follow-up periods. The outcome of interest was healing time over 12-weeks of treatment. Results: Analysis using Kaplan Meier survival curves showed a mean of 73.1 days (95% CI: 66.4 - 79.9 days) to healing for acellular synthetic matrix treated ulcers, in comparison to 92.2 days (95% CI: 87.1 – 97.2 days) for equivalent ulcers treated in everyday clinical practice in Germany (Log rank test χ 2 11.46, p= 0.001). Conclusions: Results from this study predict a highly significant effect of a new acellular synthetic matrix as an adjunct to standard care in the repair of venous and mixed leg ulcers. The predictive model may be useful in determining expected effectiveness of the acellular synthetic matrix in the treatment of venous and mixed leg ulcer patients treated in everyday clinical practice in Germany. PSS3 Cost-Effectiveness of Omalizumab in Chronic Idiopathic Urticaria Refractory to H1-Antihistamines in Turkey Tatar M 1, Sezen S 2, Senturk A 3, Balp M M 4, Saylan M 2, Keskinaslan A 2 University, Ankara, Turkey, 2Novartis Pharma, Istanbul, Turkey, 3Polar Polar Health Economics & Policy, Ankara, Turkey, 4Novartis Pharma AG, Basel, Switzerland . . . . . . . 1Hacettepe Objectives: Omalizumab is a humanised anti-IgE recombinant monoclonal antibody approved in Turkey for treatment of chronic idiopathic (or spontaneous) urticaria (CIU) in patients resistant to standard treatment. We aimed to assess the cost-effectiveness of omalizumab compared with the standard care in the treatment of CIU patients in Turkey. Methods: A Markov model of 10 years horizon was developed to estimate costs and outcomes associated with omalizumab 300mg/every 4 weeks and standard of care for 24 weeks (SoC= licensed dose H1 antihistamines) using data from two phase III studies (ASTERIA I&II includes patients with CIU refractory to licensed dose of H1-antihistamines). The study was undertaken from a Turkish health care payer perspective. At the end of this period, omalizumab was stopped and background medication continued. Patients who have responded were retreated by the same strategy upon relapse (UAS7 ≥ 16 or ≥ 28). We used Urticaria Activity Score 7 (UAS7) to measure the outcome of the treatment and we used symptom free days (UAS7 ≤ 6) as the outcome measure in this cost effectiveness analysis. Resource utilization data were obtained via a Delphi panel. Unit costs were taken from the Social Security Institution’s official list. Costs and outcomes were evaluated over a lifetime and discounted at 3%. Results were presented as incremental cost/symptom free days. Results: The incremental symptom free days with omalizumab versus SoC was 27,965 days and the incremental cost was 1,610,349 Turkish Liras for 100 patients over 10 years. The incremental cost per symptom free day was 57.59 Turkish Liras. One-way sensitivity analyses confirmed the robustness of the model results. Conclusions: CIU affects the quality of life of patients and is associated with direct and indirect costs to payers and society. We believe that omalizumab 300mg Q4wks is a cost-effective treatment option for patients with refractory CIU in Turkish setting. PSS4 Outcomes of Patients With Neovascular Age-Related Macular Degeneration (Namd) In Greece Under Ranibizumab Yfantopoulos J 1, Rouvas A 2, Chatzaras A 1, Xanthopoulou P 2, Theodosiadis P 2 1National and Kapodistrian University of Athens, Athens, Greece, 2Attikon University Hospital of Athens, Chaidari, Greece . . . . . Objectives: To evaluate the efficacy of treatment with ranibizumab in a sample of Greek patients with nAMD. Methods: In this observational, non-interventional, retrospective study, 194 cases with nAMD were enrolled from the Ophthalmology Department of “Attikon” hospital. Patients treated with ranibizumab for at least one year in one eye under routine clinical practice, and had at least one follow-up visit, were considered eligible. Patients participating in another study and/or were receiving pharmaceutical product containing VEGF inhibitors were excluded. Primary endpoints were the mean change in Best Corrected Visual Acuity (BCVA), and in Central retinal thickness (CRT), assessed by Optical Coherence Tomography. Results: The average BCVA score per eye at baseline was 0.284 in decimal points (Snellen equivalent≈ 20/84; ETDRS letters≈ 57.7), which improved at final visit (≈14 months after baseline) reaching 0.336 (Snellen equivalent≈ 20/60; ETDRS letters≈ 61.3), i.e. 6,33% or approximately 3.7 ETDRS letters improvement (p< 0.001, r= 0.317) with 4.9 ranibizumab injections overall (4.3 injections/year). CRT similarly improved, as it decreased by 28.4%, from 290.31 to 207.77 μ m (p< 0.001, r= 0.683). An extra ranibizumab injection per eye per year, increased the odds of an at least 0.1 increase in BCVA from baseline to last visit by 20.8% (95%CI 1.068-1.366), and an at least 0.2 increase by 15.7% (95%CI 1.013-1.322). Conclusions: Ranibizumab provides significant benefit in patients with nAMD under routine clinical practice by improving BCVA and reducing CRT, benefit which may be more important in a more frequent dosing regimen. PSS5 Results of A Decubitus Prevention and Wound Care Project Dozsa C 1, Borcsek B 2 of Miskolc, Miskolc, Hungary, 2Med-Econ Ltd., Veroce, Hungary . . 1University Objectives: Background: The project focused on the preventive use, effectiveness and the better knowledge about the potential effects and mechanisms of dressings. Aims: To improve the effectiveness of wound care, to prevent effectively the appearance of pressure ulcers during hospitalization. Methods: During 2 months of the project 80 patients of 3 active and 2 chronic care departments were involved, treatment of 84 pressure ulcers of 46 patients occurred. After the health check the attachment of dressings and the wound management were carried out according to the protocols and were documented on special datasheets. Subsequently, the checking and processing of questionnaires – continuing the monitoring of patient pathways – were done. Results: Pressure ulcers did not develop in case of 79% of the patients after prevention, effectiveness of targeted prevention reached 95.5%. The incidence of decubitus fell by 1.6-1.7 % in the project (formation rate was 58.890.9 % lower in average). In the background of successful prevention min. 3-day use was observed, but even use of 1-1 preventive dressing was effective. The average length of wound care was14 days, frequency was 8.2 times/patient. The healed or recovering wounds were more frequently treated, but rather for more days, however, their treatment cost less than the average and those wounds which were not healing. Conclusions: Wound care starts with prevention. The quality of wound care should not decrease after the patient’s discharge, the home follow-up would be essential, nevertheless its “best practices” has not spread widely in Hungary yet. In the short term Decubitus Team doctors should be involved in patients’ discharge planning, however in the long term creation of a so-called Wound Care Centre could offer a sustainable solution. Hereinafter controlled trials of higher number of cases could prove the cost-effectiveness of intelligent dressings in wider aspects of the health care system. A605 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PSS6 Clinical Effectiveness of Fumaric Acid Esters (Fumaderm) in Psoriasis: A Systematic Review of Literature PSS9 The Epidemiology of Medical Treatment for Glaucoma and Ocular Hypertension in Germany Ceglowska U , Wlodarczyk A , Slomka M Agency for Health Technology Assessment In Poland (AHTAPol), Warsaw, Poland Ernst F Santen GmbH, Munich, Germany Objectives: To review the literature evaluating the clinical efficacy of Fumaderm (FAE) in patients suffering from psoriasis and psoriatic arthritis. Methods: Systematic literature search in electronic databases (MEDLINE, Cochrane and Embase) was performed (from inception to March 14,2014). Results were presented by narrative synthesis including a quality assessment using Jadad scale. Results: Three randomized controlled trials were included (Fallah Arani 2011, Altmeyer 1994, Peeters 1992). Quality of studies ranged from 3 to 4 points. Observation period was 16 weeks for all studies. One open-label study comparing FAE with methotrexate (MTX) in moderate to severe chronic plaque psoriasis was identified. The primary efficacy endpoint was the difference in mean change from baseline in Psoriasis Area and Severity Index (PASI). After 12 weeks of treatment the mean PASI (± SD) decreased to 10.5 ± 6.7 compared with starting value of 18.1 ± 7.0 in patients treated with FAE and to 6.7 ± 4.5 compared to starting value of 14.5 ± 3.0 in patients treated with MTX. The absolute difference between groups (FAE vs. MTX) in the mean values was 1.4 (95%CI: 2.0; 4.7) and wasn’t statistically significant (p = 0.417). In the remaining two double-blind studies the efficacy and safety of Fumaderm was assessed against the placebo. One study shows that in psoriasis vulgaris population the mean PASI in FAE group decreased from 21.53 at baseline to 10.77, whereas in placebo group mean PASI has not changed (p< 0.0001) after 16 weeks of treatment. Moreover, the results of study in psoriatic arthritis population revealed that FAE significantly (p< 0.04) decrease joint pain compared to placebo. Conclusions: The results revealed that FAE and MTX have similar clinical efficacy in the treatment of patients with moderate to severe psoriasis. Moreover, FAE is more effective than placebo in the treatment of psoriasis vulgaris and psoriatic arthritis. Objectives: The purpose of this study is to review the epidemiology and the treatment paradigm of German glaucoma population and to assess the frequency of switches from monotherapy to second- or third line combination therapy in primary open angle glaucoma (POAG) patients. Glaucoma is the second leading cause of blindness globally. The newly published EGS guidelines state that if the first choice monotherapy is well tolerated and has effective intra ocular pressure (IOP) lowering but has not succeeded in reaching the target pressure, the addition of a second drug should be considered. Methods: German patient databases were searched in the following areas: glaucoma prevalence and incidence studies in German population as well as treatment paradigm in glaucoma patients. Results: Our analysis underlines the high number of glaucoma patients in Germany and the relevance of this disease for the German health care system. The study also demonstrates that a considerable number of glaucoma patients do not reach IOP target under monotherapy and have to be treated with a combination therapy. In general, treatment with a combination of agents of different classes is associated with superior IOP lowering efficacy. With an increasing number of available fixed dose combination products, more options become available for combination therapy of glaucoma patients. Conclusions: Glaucoma care needs to be given high priority in public health programs. Especially treatment options for glaucoma patients in need for a combination therapy should be in the focus of health care system decision makers as well as further research in glaucoma clinical trials and clinical care. . . . PSS7 Knowledge and Perception of Medical and Pharmacy Students Toward The Usage of Sunblock Hadi H , Elkalmi R , Awadh A , Jamshed S , Al-Shami A International Islamic University Malaysia, Kuantan, Malaysia . . . . . Objectives: To evaluate the knowledge and perception of medical and pharmacy students toward the usage of sunblock as skin protection against ultraviolet (UV). Methods: This cross-sectional study was conducted among the undergraduate final year medical and pharmacy students at the International Islamic University Malaysia (IIUM). Validated questionnaire were used to collect the data. The questionnaires were distributed to 134 students from medicine and 100 pharmacy students. Descriptive and inferential statistics are used whenever appropriate. Results: Overall, 161 participants out of a total of 234 completed the questionnaire with 101 medical students (75.4%) and 60 pharmacy students (60.0%). Majority of the respondents were female 64 (63.4%) and 37 (36.6%) were male. The median of knowledge scores of the final year medical students was significantly lower than the final year pharmacy students (p< 0.01). There is no significant differnce between the knowledge of the female and male students (Mann Whitney U Test value = 0.27, p< 0.01). This study reported that 24 (39.3 %) of pharmacy students were influenced by the media to use sunblock whereas 35 (34.7%) of medical students had the highest influence from friends to use sunblock. However, this study showed there was no significant difference in the perception of pharmacy and medical studnets p= 0.020. Conclusions: In conclusion, the knowledge of pharmacy students is significantly higher than the knowledge of medical students had on the usage of sunblock. Both medical and pharmacy students have the same level of perception towards the usage of sunblock. PSS8 Pharmacoepidemiology of Cellular/Tissue Derived Products for the Treatment of Venous Leg Ulcers in Outpatient Care Settings Gilligan A M , Waycaster C Smith & Nephew Biotherapeutics, Fort Worth, TX, USA . . . Objectives: Venous leg ulcers (VLUs) are a debilitating condition for patients with venous insufficiency. Compression therapy is the standard care for treatment of VLU; however the success rate is approximately 50% at 6 months. Clinical trials with cellular/tissue derived products (CTPs) have shown promising efficacy for the treatment of VLU. The objective was to identify patient and clinical characteristics in the VLU population and examine patterns of CTP utilization. Methods: Retrospective, de-identified electronic medical records from 2007-2013 were extracted from the Intellicure Limited Data Set (I-LDS). The I-LDS extracts records from 96 hospital-based outpatient wound centers. Patient, wound and encounter level characteristics were examined. CTPs of interest included extracellular matrix (ECM), human skin equivalent (HSE), and living skin equivalent (LSE). Results: A total of 9,091 patients, 25,734 wounds, and 222,666 encounters for VLU were identified. The majority of patients was male (50.5%), Caucasian (74.1%), and reported Medicare as their primary insurance (53.4%). The average age was 68.9 (SD= 14.6) and the average number of physician visits was 17.7 (SD= 22.5). The mean wound surface area was 20.1cm2 (SD= 83.4). The overall average wound duration was 5.8 months (SD= 26.7). Of the 25,734 wounds, approximately 7.1% received ECM (3.4%), HSE (3.5%), or LSE (0.2%). The average number of applications for ECM was 2.7 (SD= 2.8), 0.6 (SD= 1.3) for HSE, and 3.1 (SD= 3.3) for LSE. Wounds treated with CTPs were, on average, several months older: 12.5, 13.7, and 11.1 months for ECM, HSE, and LSE, respectively. Overall average wound treatment time was 2.9 months (SD= 4.8). However, treatment time was substantially longer with CTP utilization with an average time of 9.6, 9.7, and 7.5 months for ECM, HSE, and LSE, respectively. Conclusions: CTP utilization was relatively low within outpatient wound centers. Results from this analysis indicate that health care providers are using CTPs on older, more difficult-to-heal VLUs. . Sensory Systems Disorders – Cost Studies PSS10 A Us Hospital Economic Impact Model for Oritavancin In Absssi Patients With Risk of Mrsa Infections Wu C 1, Fortier K J 1, LaPensee K 2, Fan W 2, Mitchell M 2, Cyr P L 1, Jensen I S 1 1ICON plc, Cambridge, MA, USA, 2The Medicines Company, Parsippany, NJ, USA . . . . . . . . . . Objectives: It is estimated that acute bacterial skin and skin structure infections (ABSSSI) account for about 10% of hospital admissions in the US. Analyses of hospital claims indicate 74% of ABSSSI admissions involve empiric treatment with methicillin-resistant Staphylococcus aureus (MRSA) active antibiotics. Hospitalization costs could be reduced if moderate-severe ABSSSI patients were treated to a greater extent in the observational unit followed by discharge to outpatient parenteral antibiotic therapy (OPAT). Oritavancin is a novel single-dose regimen, intravenous lipoglycopeptide antibiotic for ABSSSIs caused by gram-positive bacteria, including MRSA. The aim of our analysis was to quantify the economic value of using oritavancin for ABSSSI patients at risk of MRSA from a US hospital perspective. Methods: A decision analytic model based on current clinical practice was developed to estimate the economic value of decreased hospital resources by using oritavancin. Utilization of antibiotics was informed by analysis of the Premier hospital database. Demographic and clinical data were derived from a targeted literature review. ER, observation, laboratory, administration costs were from Medicare National Limitation amounts. Drug costs were 2014 wholesale acquisition costs. To estimate the economic impact of reducing resources using oritavancin we set its price to $0. Results: For a hypothetical US hospital treating 1,000 ABSSSI patients eligible for IV MRSA antibiotics/year, the administration of oritavancin in 25.75% of patients facilitates shifting patients to the OPAT setting (441 to 561 patients) with a total annual economic impact of $2,752K. Inpatient and outpatient costs were reduced by $2,543K and $209K, respectively. Inpatient cost savings were derived from a reduction in hospitalizations and lower administration burden drove decreased OPAT costs. Conclusions: Using oritavancin in moderate-severe ABSSSI patients, including those at risk of MRSA, is estimated to deliver an estimated cost reduction of $2,752/patient by shifting patient care to the OPAT setting, and decreasing resource utilization. PSS12 Economic Impact of Visual Impairment: A Pilot Study in Singapore Wang X 1, Luo N 2, Lamoureux E 3, Ang M 4 1National University of Singapore, Singapore, 2Saw Swee Hock School of Public Health, National University of Singapore, Singapore, 3University of Melbourne, Melbourne, Australia,, 4Singapore National Eye Center, Singapore . . . . Objectives: To examine the economic impact and independent determinants of visual impairment (VI) in Singapore. Methods: 100 patients with VI associated with the most common eye diseases were recruited from the Singapore National Eye Centre. VI was classified as mild and moderate/severe VI based on the presenting visual acuity (VA) in the better-seeing eye. Medical costs (MC) and loss of productivity (LP) in the patients and their families were estimated based on the center’s billing data and self-reported data, respectively. LP was calculated for working patients based on absenteeism due to VI. Linear regression models were used to assess the association between costs and VI, generic (EQ-5D), and vision-specific quality of life (VF-14). Results: The median (range) age of participants was 73.0 years old (47.0-92.0). The proportion of male was 48.0% and the median (range) presenting VA was 0.54 (0.30-2.00). The yearly median (range) MC and LP were S$1.53K (S$0.13S$83.59K) and S$0 (S$0-S$7.62K) per person, respectively. The yearly MC for those with mild VI (S$3.50K) was significantly lower than those with moderate/severe VI (S$5.21K) (P< 0.0001). The yearly MC in participants reporting full health or better vision function were lower than those not in full health or better vision function (e. g. MC for participants in full health= S$2.96K; MC for participants not in full health= S$5.19K; P< 0.0001). After adjusting for socio-demographic characteristics, the association between MC and VI, EQ-5D and VF-14 remained the same. LP had a similar association with EQ-5D and VF-14 with or without adjustment; however, LP A606 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 decreased with VI severity (the adjusted difference in LP between participants with mild and moderate/severe VI= S$0.22K, P= 0.3846). Conclusions: Visual impairment poses substantial economic burden on individuals and families in Singapore. The economic burden of VI is mainly medical costs. Self-reported health problems seem to be better predictors of costs of VI than clinical measures. PSS13 Costs of Burn Care: A Systematic Review Hop M J 1, Polinder S 2, Middelkoop E 3, van Baar M E 1 of Dutch Burn Centres, Rotterdam, The Netherlands, 2Erasmus MC, Rotterdam, The Netherlands, 3Association of Dutch Burn Centres, Beverwijk, The Netherlands . . . . . . 1Association Objectives: Burn care is traditionally considered expensive care. However, detailed information about the costs of burn care is scarce, despite the increased need for this information and the enhanced focus on health care cost control. In this study, economic literature on burn care was systematically reviewed to examine the problem of burn-related costs. Methods: Cost or economic evaluation studies on burn care that had been published in international peer-reviewed journals from 1950-2012 were identified. The methodology of these articles was critically appraised by two reviewers, and cost results were extracted. Results: A total of 156 studies met the inclusion criteria. Nearly all of the studies were cost studies (n= 153) with a health care perspective (n=139) from high income countries (n=127). Hospital charges were often used as a proxy for costs (n= 44). Three studies were cost-effectiveness analyses. The mean total health care cost per burn patient in high-income countries was $88,218 (range $704- $717,306, median $44,024). Conclusions: A wide variety of methodological approaches and cost prices was found. We recommend that cost studies and economic evaluations employ a standard approach to improve the quality and harmonization of economic evaluation studies, optimize comparability and improve insight into burn care costs and efficiency. PSS14 Costs of Psoriasis in Europe. A Systematic Review of the Literature Obradors M 1, Figueras M 1, Paz S 2, Comellas M 3, Lizán L 3 Farmaceutica S. A., Barcelona, Spain, 2Outcomes’10, Castellon, Spain, 3Outcomes 10, Castellon, Spain . . . . . 1Novartis Objectives: To appraise the literature referred to direct and indirect costs of psoriasis in Europe. Methods: A systematic review of the literature was performed. Electronic databases [MedLine/PubMed, Scopus, Cochrane Library, ISI Wok, MEDES, IBECS, CSIC] and Google Scholar were searched to identify publications referred to direct and indirect costs of psoriasis in Europe. Bibliographic references were hand searched. European studies published in English or Spanish until October 2013 were selected. Economic evaluations of specific drugs and of preventive or diagnostic interventions were excluded. Costs were updated to € , 2013. Results: 12 studies were reviewed (4 German, 2 Swedish, 2 Italian, 1 Dutch, 1 French, 1 English and 1 Spanish). From the social perspective, the mean annual cost of psoriasis in Europe was heterogeneous and varied between € 1,340.25 (Spain) and € 8,253.74 (Italy), mainly due to differences of study sample characteristics, methodology used and national health system features. The highest proportion (68%-82.5%) of total costs was attributable to direct costs (hospitalizations, medications and laboratory tests). The annual total cost of severe psoriasis was, at least, 2.5 fold superior to the costs of mild disease, mostly because of a more frequent use of hospital resources and loss of productivity. Out-of-pocket expenditure ranged from € 480.67 to € 797.00, being mostly explained by the use of dermatological products (OTC) (29%-59.7%) and alternative therapies (balnerotherapies) (24%-49%). The initiation of biologic therapies implied reductions of up to 76% on the length of hospital stay and up to 98% hospital admissions, contributing to a decline of inpatient cost [€ 2,357.30 (SD: 722.29) vs. € 564.19 (SD: 257.62); p= 0.005; UK]. Conclusions: Costs results vary across European countries. Severe psoriasis is a costly disease. The use of biologic agents may contribute to a more efficient management of severe psoriasis due to a more steady control of symptoms that improve clinical outcomes and decrease the needs for hospital care and inpatient cost. PSS15 Economic Burden of Eye Disease in Diabetic Patients: Literature Review Riera M 1, Granell M 2, Sarycheva A 2, Roura M 2 Consultant, Molina de Segura, Spain, 2Novartis Farmaceutica, Barcelona, Spain . . . . 1Independent Objectives: To provide a literature overview on the costs of diabetic complications focused on the economic burden of Diabetic Macular Edema (DME) and associated vision loss and blindness. Methods: Literature searches in PubMed (including Medline) and Embase were conducted to identify economic data related to diabetic retinopathy, macular edema, blindness and low vision. Posterior filters included year of publication (from 2000 to May 2014), country (Canada, USA or Europe) and English language. Relevant full text articles were reviewed and major findings summarized. Results: A total of 468 citations from 5 different searches were retrieved and 16 studies fulfilled the criteria to be summarized in this review. Microvascular diabetic complications were found to be frequent and led to a 70% increase in direct costs compared to patients with no complications. When microvascular and macrovascular complications were given together, management costs increased by up to 250%. Eye disease was associated with a 2,5 ratio of extra cost compared to diabetic patients without complications, with DME incurring in twice to 3 times higher costs related to Diabetic Retinopathy. Direct costs of DME accounted for 65-80% of total costs, with 73% of those being associated to hospital expenditure. Data on the economic burden of visual impairment and blindness were limited: vision loss had an annual per capita financial cost of $19.370 in Canada (2007); including the value of lost wellbeing, the annual cost of vision loss was $33.704 per patient. Conclusions: DME appears to be a costly complication of diabetes. Progression to visual impairment and blindness can lead to high economic and social burden. PSS16 Cost-Analysis of Chronic Idiopathic Urticaria Disease Tatar M 1, Senturk A 2, Ergin G 2, Tuna E 2, Mat C 3, Bulbul Baskan E 4, Artuz F 5, Gurer M A 6, Atakan N 7, Inaloz S 8, Sezen S 9, Keskinaslan A 9, Saylan M 9 1Hacettepe University, Ankara, Turkey, 2Polar Polar Health Economics & Policy, Ankara, Turkey, 3Cerrahpasa Medical Faculty, Istanbul, Turkey, 4Uludag Medical Faculty, Bursa, Turkey, 5Ankara Numune research and Training Hosptial, Ankara, Turkey, 6Gazi University Medical Faculty, Ankara, Turkey, 7Hacettepe University Faculty of Medicine, Ankara, Turkey, 8Gaziantep University Medical Faculty, Gaziantep, Turkey, 9Novartis Pharma, Istanbul, Turkey . . . . . . . . . . . . . . Objectives: Chronic idiopathic (or spontaneous) urticaria (CIU) is estimated to occur in 0.3% of the general population in Turkey. Although the burden of the disease is substantial for the payer, there has been no cost-analysis performed in our country. The purposes of this study are to determine the resources and costs related to the diagnosis and treatment of patients with CIU, and to estimate the annual economic burden to the Social Security Institution (SGK). Methods: Delphi technique was applied to determine the type and the amount of resources used in different stages of CIU defined as mild or moderate to severe based on physicians clinical assessment. The Delphi method solicits the opinion of an expert panel through a carefully designed questionnaire which in this case included questions on: epidemiology, diagnosis, treatment of symptoms and angioedema, adverse events, follow-up visits, hospital and emergency service admissions. The responses were analysed and discussed in a face to face meeting followed by consensus building steps. Unit of resources used for different CIU severity levels were determined from expert opinion. Unit costs of resources used in outpatient clinics were obtained from SGK’s Reimbursement Guideline-List of Procedure Fees Per Service. Results: Hospitalizations, emergency admissions, outpatient visits and treatments, are the key cost drivers in the management of CIU The annual cost per patient is calculated to be 725.36 Turkish Liras (TL) for mild CIU, 1.322,61 TL for moderate CIU and 2,478.75 TL for severe CIU. The total annual cost of CIU to SGK is estimated to be 262 million TL in 2014. Conclusions: This is the first study that aims to estimate the resource utilisation and cost burden of CIU in Turkey by using the Delphi technique. Cost effective treatment of CIU is an unmet need given the heavy burden to SGK identified by the Delphi Panel. PSS17 The Direct and Indirect Costs of Wet Age-Related Macular Degeneration (Wamd) and Diabetic Macular Edema (Dme) in Greece Geitona M 1, Karagiannis D 2, Pantelopoulou G 3, Hatzikou M 3, Kousoulakou H 1 of Peloponnese, Corinth, Greece, 2Eye Hospital of Athens, Athens, Greece, 3Novartis Hellas, Metamorfosis, Greece . . . . . 1University Objectives: Treatment delays in patients with wAMD and DME are very important since they are associated with disease progression. This is the first study in Greece to highlight patient’s access delays’ to treatment and to estimate the burden of diseases in Greece. Methods: An expert panel with 11 ophthalmologists was convened. The experts came from six out of seven Regional Health Authorities, covering geographically the largest part of the country. A 13-page questionnaire was developed and validated. The Delphi technique was used to collect data on resource use and disease management, patient access delays, and indirect costs. Unit costs were retrieved from NHS sources and the analysis was conducted from the societal perspective, including direct costs incurred by patients, their families and the health care system, and indirect costs, of productivity losses. Results: Total cost of managing wAMD during the first year of treatment was estimated at € 20,660,32.8% of which consisted of direct costs (drug acquisition and administration, follow-up, lab and imaging tests), and 67.2% of which included non-medical and indirect costs (caregivers). Direct medical cost per DME patient for the first year of treatment was estimated at €6,066, while the non-medical and indirect costs were estimated at € 12,271, resulting in a total cost of € 18,337 per patient per year. 91% of the experts agreed that there are significant delays in patient access to specific treatments (average 20 days); all agreed (100%) that these delays lead to faster disease progression, and a strong majority (>80%) supported that they lead to deterioration of patients’ QoL. Conclusions: There are significant costs associated with the management of wAMD and DME in Greece, 70% of which is incurred by patients and their families. Administrative procedures leading to patient access delays should be minimized in order to avoid disease progression and associated costs. PSS18 Economic Burden of Burn Injuries in the Netherlands Hop J M 1, Wijnen B F M 2, Nieuwenhuis M 3, Dokter J 1, Middelkoop E 4, Polinder S 5, van Baar ME1 of Dutch Burn Centres, Rotterdam, The Netherlands, 2Maastricht University, Maastricht, The Netherlands, 3Association of Dutch Burn Centres, Groningen, The Netherlands, 4Association of Dutch Burn Centres, Beverwijk, The Netherlands, 5Erasmus MC, Rotterdam, The Netherlands . . . . . . . . . . . 1Association Objectives: Burn care has rapidly improved in the past decades. However, health care innovations can be expensive, demanding careful choices on their implementation. Obtaining knowledge on the extent of the costs of burn injuries is an essential first step for economic evaluations within burn care. The objective of this study was to determine the economic burden of patients with burns admitted to a burn center and to identify important cost categories until three months post-burn. Methods: A prospective cohort study was conducted in the burn center of Maasstad Hospital Rotterdam, the Netherlands, including all patients with acute burn related injuries from August 2012 until July 2013. Total costs were calculated from a societal perspective, until three months post injury. Subgroup analyses were performed to examine whether the mean total costs per patient differed by age, etiology or percentage total body surface area (TBSA) burned. Results: In our population, with a mean burn size of 8%, mean total costs were € 24,246 per patient varying from € 11,498 to € 71,756. Most important cost categories were burn center days, surgical interventions and work absence. Flame burns were significantly more costly than other types of burns, adult patients were significantly more costly than children and adolescents A607 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 and a higher percentage TBSA burned also corresponded to significantly higher costs. Conclusions: Mean total costs of burn care in the first three months post injury were estimated at € 24,246 and depended on age, etiology and TBSA. Mean total costs in our population probably apply for other high-income countries as well, although we should realize that patients with burn injuries are diverse and represent a broad range of total costs. To reduce costs of burn care, future intervention studies should focus on reducing length of stay and enabling an early return to work. PSS19 The Cost of Blindness in the Republic of Ireland 2010-2020 Green D 1, O Neill C 2, Ducorroy G 1, Skelly A 1, Keegan D 3, Kenny D 4, Naughton A 3, Keegan D 3 1Novartis Ireland, Dublin 4, Ireland, 2National University of Ireland, Galway, Galway, Ireland, 3Mater Misericordiae Ubiversity Hospital, Dublin, Ireland, 4National Council for the Blind Ireland, Dublin, Ireland . . . . . . . . Objectives: Aims The aim of this study is to estimate the prevalence of blindness in the Republic of Ireland and estimate the financial and economic cost of blindness between 2010 and 2020. Methods: The prevalence of blindness was based on the National Council for the Blind of Ireland blind register and adjusted for under registration found in previous literature. The financial cost of blindness was based on the sum of total direct and indirect health care costs (direct health care costs due to blindness, cost of depression due to blindness, cost of injurious falls due to blindness) and non-health care costs (productivity losses, informal care and deadweight welfare loss). Where possible, methods adopted reflect those used elsewhere in the literature. Results: The total financial cost of blindness in the ROI is estimated to have been € 276. million in 2010. This is projected to increase by 32.6% to € 367 million in 2020 if current trends in disease burden continue. The total economic cost of blindness in the ROI is estimated to be € 809 million in 2010 and is predicted to increase to over € 1.1 billion in 2020 based on current trends. Conclusions: A significant proportion of blindness can be avoided through the implementation of existing technologies by the health service. However a significant portion (98.04%) of the burden of illness falls beyond the health service (primarily to the Department of Social Protection and the Department of Finance) and may serve to reducethe priority of policies aimed at avoiding blindness that might otherwise be received. PSS20 Cost-of-Illness Study of Senile Cataract in the Czech Republic Kruntoradova I 1, Kruntorádová K 2, Rogalewicz V 2, Barták M 3 1Charles University, Kladno, Czech Republic, 2Czech Technical University in Prague, Kladno, Czech Republic, 3University of J.E.Purkyn? in Ústí nad Labem, Ústí nad Labem, Czech Republic . . . . Cost-of-illness study of senile cataract in the Czech RepublicIlona Kruntorádová, Klára Kruntorádová, Vladimír Rogalewicz, Miroslav Barták. Objectives: In the Czech Republic, 7% of patients with senile cataract are indicated for surgery each year, which means ten thousand of interventions annually. The aim of this study is to quantify direct costs in relation to the implanted intraocular lens type from both the public health insurance perspective and the patient’s perspective. The senile cataract surgery with monofocal IOL is covered by public health insurance. Methods: Direct costs were calculated through a panel of experts that provided expert opinions, taking into account consumption of paid medical services provided to patients. The panel consisted of 6 experts (4 physicians from eye clinics, 2 physicians from outpatient departments). Patient´s costs were obtained through a patient questionnaire asking the direct (out-of-pocket) expenses associated with the cataract. The study covered a cohort of 200 patients with senile cataract in non-productive age (65+). Data were collected from 4 clinics performing cataract surgery. Results: The structure of direct costs associated with senile cataract, its surgery, and treatment of possible complications has been specified. Direct costs of senile cataract surgery from the perspective of public health insurance are CZK24900 (»EUR907) per person. Patient´s direct costs are lower for monofocal lenses than for multifocal ones. Patients’ costs for an IOL and the related treatment reach up to CZK35000 (»EUR1275), out of pocket expenses after treatment average out at CZK4450 (»EUR162). Conclusions: Due to absence of laws governing medical care premium services and the age of patients, monofocal IOLs are implanted predominantly. As a consequence, up to CZK 30 million (»EUR 1 1 million) are saved on the public health insurance budget annually. Patients´ expenses for implanting multifocal lenses are offset by savings due to unrealized spending for eyeglasses and related medical care in the course of time. PSS21 Costs of Dental Outpatient Care – Resource Use Differentials Across Clinical Dentistry Branches Rancic J 1, Rancic N 2, Majstorovic N 3, Biocanin V 4, Milosavljevic M 4, Jakovljevic M 4 Dentistry Clinic Dr D. Lopicic, Belgrade, Serbia and Montenegro, 2Military Medical Academy University of Defence Belgrade, Belgrade, Serbia and Montenegro, 3The Faculty of Dentistry University of Belgrade, Belgrade, Serbia and Montenegro, 4The Faculty of Medical Sciences University of Kragujevac, Kragujevac, Serbia and Montenegro . . . . . . 1Specialist Objectives: Dental medical care bears particular financial burden for Eastern European transitional economies due to its lack of insurance coverage in most countries of the region and almost complete out-of-pocket payments by citizens. This study estimates real costs of these services in the field and describe resource use patterns and differentials across clinical dentistry branches, ICD-10 diagnostic groups, and across particular medical services. Methods: In a case-series design prospective cost-comparison study has been conducted from the perspective of the patient. Sample size was 752 complete episodes of treatment, selected randomly in 2012/2013 throughout an array of several specialist state-owned (university associated) and private-owned dental clinics in upper-middle income Serbia. All direct medical costs of dental care were taken into account. Costs were expressed in national currency Republican Serbian Dinar (RSD). Results: Mean total direct medical costs of dental care were 5,018.96±17,109.61RSD per single dentist visit while total costs incurred by this population sample were 3,774,256RSD. Highest unit utilization of services belongs to conservative dentistry (31.9%), oral surgery (19.5%) and radiology (17.4%), while highest value based turnover belongs to implantology 90,765.33±43,012.31RSD, orthodontics 77,361.82±73,123.92RSD and prosthetics 60873.89±26788.87RSD. Most frequently treated diagnosis was tooth decay (33.8% unit services provided), pulpitis (11.2%) and impacted teeth (8.5%), while most expensive to treat were anomalies of tooth position (70,998.33±73122.73RSD), abnormalities of size and form of teeth (55,662.50±77,304.45RSD) and loss of teeth due to accident, extraction or local periodontal disease (36,835.35±37,128.28RSD). Conclusions: Although range of dental medical costs currently falls behind EU average, Serbia’s emerging economy is likely to expand in the long run while market demand for dental services will grow. Due to threatened financial sustainability of current health insurance patterns in Eastern Europe, getting acquainted with true size and structure of dental care costs might essentially support informed decision making in future. PSS22 Cost-Effectiveness Analysis of Ustekinumab Compared With Etanercept for the Treatment of Moderate to Severe Psoriasis in Costa Rica Obando C A , Desanvicente-Celis Z , Herrera J A , Moreira M , De Castro J Janssen, Panama, Panama . . . . . . . Objectives: To assess the cost-effectiveness of Ustekinumab (UST) compared with Etanercept (ETN) in Costa Rica, in patients with Moderate to Severe Psoriasis. Methods: A cohort simulation Markov Model was developed based on response rates for UST and ETN [Psoriasis Area Severity Index (PASI) ]. The time frame was 10 years. The perspective was that of the Public System of Health of Costa Rica. The health outcome of interest was Quality Adjusted Life Years (QALYs). Efficacy data was taken from the ACCEPT clinical trial; this phase III clinical trial directly compares UST and ETN, what strongly supports the efficacy data that is used in the model. Utilities for health states were taken from published studies. The base year was 2013. All costs are presented in Costa Rican currency (Colones – CRC). Costs and outcomes were discounted at 3.5%. Probabilistic sensitivity analysis (PSA) was conducted to assess uncertainty around the parameters. Results: UST resulted in 3.85 QALYs and ETN in 3.58, per patient, respectively. Mean total costs per patient were: CRC 8.441.031 for UST and CRC 5.401.222 for ETN. UST resulted both more costly and more effective than ETN. The Incremental Cost Effectiveness Ratio comparing UST and ETN was 11.142.470 per QALY Gained. According to the classification of the World Health Organization (WHO), the acceptable threshold for QALY Gained for Costa Rica is CRC 14.140.792 (3 times the Gross Domestic Product per capita). The results of the probabilistic sensitivity analysis showed that, at the threshold suggested by the WHO, the probability of UST of being cost effective, compared with ETN is around 70%. Conclusions: UST can be considered cost effective when compared to ETN, according to the threshold suggested by the WHO, in patients with moderate to severe Psoriasis, from the perspective of the Public System of Health of Costa Rica. PSS23 Cost-Effectiveness of Ranibizumab on Patients With Diffuse Diabetic Macular Edema Within the Public Mexican Health Care System Ruiz Miranda C I , Ubiarco Lopez V Novartis Mexico, Mexico, Mexico . . . Objectives: To perform a cost-effectiveness analysis of Ranibizumab plus laser photocoagulation vs monotherapy with laser photocoagulation in patients with diffuse Diabetic Macular Edema (DME). Methods: A Markov model was designed to analyze laser Photocoagulation vs laser Photocoagulation plus Ranibizumab. Transition probabilities were obtained from RESTORE. The base patient was a diabetic with 53 years presenting DME, according to Mexican context. General mortality rates were elicited locally from CONAPO*. A panel Delphi was performed to get use of resources locally. The time horizon was 5, 7, and 10 (lifetime) years according to life expectancy from ENSANUT** 2013, discount rate 5%. The outcome was life years without visual impairment. The cost values were from Guidelines for the Exchange of Services in the Health Sector. Probabilistic sensitivity analysis (PSA) was performed using Monte Carlo technique. Results: Incremental Cost Effectiveness Ratios of the combination versus the monotherapy were: $5,019.57, $2,375.62 and $622.67 USD per life year without visual impairment in a time horizon of 5, 7 years and lifetime. Cost effectiveness curve showed Ranibizumab be a cost effective option at 98.7% vs monotherapy before reaching the GDP per capita. Conclusions: Ranibizumab demonstrated to be a more cost-effectiveness alternative than monotherapy with laser photocoagulation. These results show the possibility of achieving potential clinical benefit with Ranibizumab in patients suffering from loss of vision. *CONAPO National Council of Population **ENSANUT National Health and Nutrition Survey PSS24 Modeled Outcomes and Overall Costs of the 13-Valent Pneumococcal Conjugate Vaccine in the Tunisian National Vaccination Program Zigmond J 1, Pecen L 1, Tichopad A 1, Roberts C S 2, Jomaa I 3 1CEEOR s. r. o., Prague, Czech Republic, 2Pfizer Inc., New York, NY, USA, 3Pfizer Pharmaceuticals Tunisia, Tunis, Tunisia . . . . . . Objectives: Like other North African countries, Tunisia has a substantial burden of pneumococcal disease, with high resistance to antibiotics. The Tunisian population remains largely unprotected in the absence of a national immunization program (NIP). Methods: A decision-analytic model was developed to evaluate the potential outcomes and costs of the PCV13-based NIP compared to no vaccination. The model estimates bacteremia and meningitis (jointly IPD), all-cause community acquired pneumonia (CAP), and all-cause otitis media (OM). The demographics and disease A608 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 characteristics were obtained from WHO estimates or local sources, adjusted to local conditions. PCV13 direct and indirect effectiveness was extrapolated from PCV7 trials and surveillance records, adjusted to local serotype distribution. Cost of vaccine was USD 16.34. A discount rate for cost and life-years was 3%. The payer and societal perspectives were considered. Results: The budget impact in a single year with PCV13-based NIP in place would amount to USD 1.82 million, or USD 7.93 million without indirect vaccine protection considered. From this investment, 141 971 illnesses (1071 IPDs, 12477 CAPs and 128423 OMs) and 347 deaths could be avoided annually. Without indirect vaccine protection, 58 524 illnesses (601 IPD, 4721 CAP, 53202 OM) and 184 deaths could be avoided. The cost-effectiveness analysis produced ICER of USD 340/LYG or USD 367/QALY from the payer’s perspective. From the societal perspective, the NIP is dominant. Not considering indirect protection, the ICER would be USD 140/LYG or USD 152/QALY from a societal perspective and USD 1157/LYG or USD 1254/QALY from a payer perspective. Conclusions: PCV13based NIP delivers benefits and cost savings that greatly offset the investment into vaccine. WHO strongly encourages investment in interventions that deliver an additional year of life in full quality for less than one GDP per capita (USD 4237); hence, a PCV13-based NIP with the above ICER presents an attractive option. PSS25 Ranibizumab for the Treatment of Visual Impairment Due to Myopic Choroidal Neovascularization: Cost-Effectiveness Versus Aflibercept Leteneux C 1, Haig J 2, Xue W 3, Bhattacharyya S 4 1Novartis Pharma AG, Basel, Switzerland, 2Optum, Burlington, ON, Canada, 3Optum, Uxbridge, UK, 4Novartis Healthcare Pvt. Ltd., Hyderabad, India . . . . Objectives: Ranibizumab has demonstrated efficacy in patients with myopic choroidal neovascularization (mCNV) and is the first anti-VEGF licensed in this indication. Aflibercept is being evaluated for use in mCNV. An existing model demonstrating the cost-effectiveness of ranibizumab versus verteporfin photodynamic therapy was adapted to provide an initial evaluation of ranibizumab versus aflibercept. Methods: A Markov model in mCNV with a lifetime horizon and visual acuity health states was adapted to evaluate the cost-effectiveness of ranibizumab and aflibercept from a UK health care perspective. Baseline characteristics, injection frequency and ranibizumab efficacy were based on the disease activity treatment arm from the RADIANCE study (n= 116, Caucasian, Indian and East Asian patients). Data for aflibercept were derived from initial results for the aflibercept treatment arm from the MYRROR study (n= 90, East Asian patients only). Relative efficacy was assessed by indirect comparison. An evaluation using the East Asian subgroup of the ranibizumab disease activity treatment arm in RADIANCE (n= 35) was also conducted. Results: Ranibizumab dominated aflibercept in both evaluations. Based on the disease activity arm from RADIANCE, ranibizumab was associated with a lower lifetime cost (incremental cost -£1770) and higher lifetime qualityadjusted life-years (QALYs) (incremental gain 0.02) than aflibercept. Results were similar for the evaluation based on the East Asian subgroup. Ranibizumab was associated with a lower lifetime cost (incremental cost -£2856) and higher lifetime QALYs (incremental gain 0.06) than aflibercept. These results were driven by the greater number of injections, higher treatment and recurrence costs, and smaller proportion of patients gaining ≥ 20 letters visual acuity for aflibercept compared with ranibizumab. Conclusions: This initial analysis suggests that ranibizumab is less costly and is associated with a gain in QALYs relative to aflibercept based on the disease activity arm and the East Asian subgroup from RADIANCE, as well as initial data from MYRROR. PSS26 Cost-Effectiveness of Aflibercept in the Treatment of Macular Oedema Secondary to Central Retinal Vein Occlusion In Sweden Eriksson M 1, Castelo-Branco A 2, Nilsson J 2 1Bayer AB, Solna, Sweden, 2OptumInsight, Stockholm, Sweden . . . Objectives: Central retinal vein occlusion (CRVO) is caused by a blood clot in the central retinal vein, which slows or stops blood from leaving the retina. As a result, blood and fluids can accumulate, causing retinal injury and vision loss. Thus, a major complication in eyes with CRVO is macular oedema (ME) and is the primary factor for poor visual acuity and visual fields in non-ischemic CRVO. A global costeffectiveness model was developed and adopted to estimate effects and associated costs, in Sweden, for treatment of ME secondary to CRVO with aflibercept compared to ranibizumab. Methods: A Markov model was developed, including health states that reflect the clinical treatment and disease progression/regression of the ME. The simulated patient population consisted of adults treated for ME secondary to CRVO with an average starting-age of 64 years. Patients were treated and monitored for two years and followed for 15 years in the base case. Treatment regimens were taken from clinical trials with aflibercept (GALILEO & COPERNICUS) and ranibizumab (CRUISE & HORIZON), with 8.2 vs. 8.8 injections the first year and 2.9 vs. 3.5 injections the second year, respectively. Results: Aflibercept can be regarded as a cost-effective, i. e. dominating, treatment-alternative compared to ranibizumab as aflibercept is both less costly (total incremental cost of more than -35,000 SEK) and more effective (total incremental QALYs of 0.061) than ranibizumab. Due to the more treatments, ranibizumab had higher drug (incremental cost: -8,537 SEK) and administration (incremental cost: -5,793 SEK) costs compared to aflibercept. Probabilistic sensitivity analysis showed that aflibercept was dominating over ranibizumab in 70% of the simulations. Conclusions: Aflibercept is more cost-effective than ranibizumab for the treatment of ME secondary to CRVO in Sweden. PSS27 Cost-Effectiveness of Laser Doppler Imaging in Burn Care in The Netherlands; A Randomised Controlled Trial Hop M J 1, Stekelenburg C 2, Hiddingh J 3, Kuipers H 3, Goei H 1, Middelkoop E 4, Nieuwenhuis M 3, Polinder S 5, van Baar M E 1 . . . . . . . . . . . 1Association of Dutch Burn Centres, Rotterdam, The Netherlands, 2Association of Dutch Burn Centres, Beverwijk, The Netherlands, 3Association of Dutch Burn Centres, Groningen, The Netherlands, 4VU University Medical Centre, Amsterdam, The Netherlands, 5Erasmus MC, Rotterdam, The Netherlands Objectives: In patients with burns an early accurate diagnosis of burn depth is essential to determine optimal treatment. The combination of Laser Doppler imaging (LDI) and clinical assessment leads to an accurate estimate of burn depth. However, the actual effects of the introduction of LDI on therapeutic decisions, clinical outcomes and costs are unknown. The aim of our study was to analyse the effectiveness and cost-effectiveness of LDI in burn care. The effects of LDI on decision-making, clinical outcomes, costs, and cost-effectiveness were assessed. Methods: A randomised controlled trial was conducted in all three Dutch burn centres, including subsequent patients with burns of indeterminate depth. In the standard care (SC) group, burn depth and treatment choices were based on clinical assessment only, in the other group (LDI) clinical assessment and LDI results were combined. Primary outcome was the effect of the introduction of LDI on wound healing time. The economic evaluation was performed from a societal perspective with a bottom up approach, following the micro-costing method. Results: Mean time to wound healing from randomisation was 14.3 days in the LDI group and 15.5 days in the SC group (p= 0.258). In the subgroup of clinical patients requiring surgery earlier decision for surgery and a shorter wound healing time were observed in the LDI group (16.0 versus 19.9 days, p= 0.029). Mean total costs per patient were € 18 549 versus € 18 896 (p= 0.837). Conclusions: LDI proved to provide guidance for therapeutic decisions with a significantly shorter wound healing time in the subgroup of clinical patients requiring surgery. When time to surgery can be reduced by 2.4 days, similar to the time to decision for surgery in our study, cost savings of € 794 per scanned patient can be achieved. PSS28 Cost-Effectiveness Analysis of Ingenolo Mebutato VersUS miquimod in the Treatment of Actinic Keratoses in the Perspective of the Italian Health System Di Matteo S 1, Colombo G L 2, Pellacani G 3, Bruno G M 1 e Valutazioni economiche, Milan, Italy, 2University of Pavia, Milan, Italy, 3Università degli Studi di Modena e Reggio Emilia, Modena, Italy . . . . . . 1S. A. V. E. Studi Analisi Objectives: Actinic Keratosis (AK) is the most common neoplastic lesion of the skin, its prevalence in Italy is 1.4% in the adult population, over the age of 45 years. The objective of this study is to evaluate through the development of a decision-tree model, the impact in terms of cost-effectiveness of treatment of patients with actinic keratosis (on the face), of ingenolo mebutato gel vs. imiquimod cream. Methods: The effectiveness was expressed in terms of utility; the ratio of cost effectiveness was expressed in terms of cost per Quality Adjusted Life Years (QALYs). The time horizon of the simulation was 12 months. For ingenolo mebutato was considered the price to the public starting from the ex-factory price currently lower in Europe (Spain price), while for imiquimod has been adopted the reference price, because of the drug generication. It was also considered the adherence rate of patients to the two treatment alternatives, due to the different duration of treatment (2-3 days Vs. 4-8 weeks) and adverse events, which in the case of imiquimod may persist for all the therapy lenght. Results: Based on these assumptions, ingenolo mebutato therapy is found to be less expensive and more effective, and so dominant, compared to imiquimod. The cost-effectiveness analysis has been tested with univariate sensitivity analysis, which confirmed the validity of the base case. Conclusions: Based on these statement, it seems clear that ingenolo mebutato, due to its way of administration combined with its expected cost, represents a rational investment for the treatment of AK in the landscape of our national health system. PSS29 Cost-Effectiveness of 13-Valent Versus 10-Valent Pneumococcal Conjugate Vaccine Use in Croatia National Vaccination Program Tichopad A 1, Pecen L 1, Roberts C S 2, Uglesic L 3, Tesovic G 4, Rogier K 5 s. r. o., Prague, Czech Republic, 2Pfizer Inc., New York, NY, USA, 3Pfizer Croatia d. o. o., Zagreb, Croatia, 4University of Zagreb Medical School, Zagreb, Croatia, 5Pfizer bv, Capelle a/d IJssel, The Netherlands . . . . . . . 1CEEOR Objectives: The national immunization program (NIP) on a voluntary basis started in 2010 in Croatia, including the 10-valent PCV10 and the 13-valent PCV13. We compare the cost-effectiveness of PCV10 and PCV13 use in the NIP. Methods: A Markov model was developed to examine cost-effectiveness of PCV13 versus PCV10 from the payer’s perspective in 10 years. The simulated diseases were invasive pneumococcal disease (bacteremia and meningitis), all-cause community acquired pneumonia (CAP), and all-cause acute otitis media (AOM). Direct effectiveness was extrapolated from PCV7 clinical trials, adjusted by local serotype. Indirect effect (IE) was extrapolated from the US surveillance data following universal PCV7 use. Vaccine prices per dose for PCV10 and PCV13 were € 45.16 and € 47.71, respectively. The epidemiology inputs were based on national sources or adopted from neighboring Slovenia. Costs were obtained from local reimbursement lists and the DRG system. The IE for PCV10 was separately taken at 0%, 50% and 100% level. Results: Compared to PCV10 with presumed no IE, PCV13 could avoid additional 985 IPD cases, 15583 cases of inpatient and 26481 cases of outpatient CAP, and 53555 AOM cases, whereas for modeled 50% IE of PCV10 only 679,10568,17 641 and 35026 cases would be avoided, and for modeled 100% IE of PCV10 372,5552,8798,16498 cases would be avoided, respectively. There would be 2778 or 1958 or, 1137 deaths avoided, respectively. PCV13 compared to PCV10 with assumed no IE leads to € 3.060 million more spent on vaccination and € 28.585 million saved, giving thus overall saving € 25.524 million in 10 years. Conclusions: The cost-effectiveness analysis showed PCV10 to be dominated by PCV13 by its overall lower costs and higher number of QALY as well as LYG gained, regardless of the IE level. The results were most sensitive to the cost and incidence of hospitalized pneumonia. A609 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PSS30 Cost-Effectiveness of Ranibizumab Versus Photodynamic Therapy for the Treatment of Neovascluar Age-Related Macular Degeneration Based in China Cost Setting Feng S 1, Ying X 1, Qi F 1, Chang J 2 1Fudan University, China, Shanghai, China, 2Beijing Novartis Pharma Co., Ltd, Beijing, China . . . . Objectives: To compare ranibizumab with photodynamic therapy (PDT) for the treatment of predominantly classic chorodial neovascularization associated with age-related macular degeneration (AMD) by means of cost-utility analysis in China. Methods: This study compared cost-effectiveness between Ranibizumab group (each delivering 0.5 mg of ranibizuamb monthly) and PDT group for the first year (short-term) using decision tree model, from third-payer perspective. During a 10-year time horizon (long-term), a Markov model was constructed to extrapolate effects of treatment beyond clinical trials from a societal perspective (discounted at 5%). Visual acuity data from the Anti-Vascular Endothelial Growth Factor Antibody for the Treatment of Predominantly Classic Choroidal Neovascularization in AMD (ANCHOR) trial were applied. Direct-vision-related medical costs were based on Clinical Pathway of AMD in China and experts consultation. We performed a literature review and meta-analysis to estimate the costs related to comorbid states, including fall and depression resulting from low vision after suffer from AMD, which were rarely inclusive in most previous studies. Informal care costs and utilities data were derived from published studies. All costs were adjusted according to 2012 price index in China. Probabilistic sensitivity analyses (PSA) were performed to test the robustness in structural assumptions and parameter inputs. Cost-effective acceptability curves were performed after a 1000-sample Monte Carlo simulation. Results: In short-term model, the incremental cost-effectiveness ratio (ICER) was $81,029/ QALY for ranibizumab compared with PDT. The ICER was reduced to $13,206/QALY for a 10-year time frame, when follow-up costs were included. From a societal perspective, PSA revealed a 97.75% probability of ranibizuamb being more cost-effective than PDT at a threshold of $18,278 per QALY gained (3 times Chinese GDP per capita in 2012) in China. Conclusions: Ranibizumab can be a long-term cost-effective option for the treatment of AMD compared with PDT from a societal perspective. PSS31 Cost-Utility Analysis of Recommended Ranibizumab Regimen for AgeRelated Macular Degeneration in China Ying X 1, Feng S 1, Wu W 1, Chang J 2 1Fudan University, China, Shanghai, China, 2Beijing Novartis Pharma Co., Ltd, Beijing, China . . . . Objectives: To conduct the cost-utility analysis (CUA) of treatment strategy with ranibizumab recommended by the Chinese Clinical Pathway of age-related macular degeneration (AMD) in China. Methods: Visual acuity data for the as-needed dosing regimen with ranibizumab (RBZ-PRN, administrated every month for three doses, additional reinjections were determined by physicians’ need) were derived from pivotal trials, based on ranibizumab dose of 0.5 mg. Decision tree model and Markov model were developed to estimate RBZ-PRN compared with best supportive care (BSC) for short and long term. The first year decision tree model from a thirdpayer perspective was performed to estimate CUA in short-term. Then a 10-year Markov model from a societal perspective were constructed (discounted at 5%) in long-term CUA estimation. Resource utilization was obtained from official recommendations in China. The costs of low-vision related disease were extrapolated from meta-analysis. Informal care costs and utility values were estimated by published studies. The uncertainty was identified in a one-way sensitivity analysis and probabilistic sensitivity analysis (PSA). Cost-effective acceptability curves were obtained by a Monte Carlo approach with 1000 repetitions. Results: For RBZ-PRN compared with BSC, the incremental cost-effectiveness ratios (ICER) ranged from $90,546/QALY for the 1-year time frame to $9,787/QALY for the 10-year time horizon. As indicated in the PSA, RBZ-PRN was the optimal strategy in 100% of cases below the willingness to pay threshold of $18,278 per QALY gained (3 times Chinese GDP per capita in 2012) for the 10-year model. Conclusions: When administered as needed, ranibizumab is cost-effective compared with BSC from a societal perspective. In China’s current clinical practice, this can be a useful tip for decision-makers who should consider cost-effective issues for the treatment of wet-AMD. PSS32 Non-Proliferative Diabetic Retinopathy: is It Cost-Effective to Treat Early? Mistry H 1, Auguste P 1, Lois N 2, Waugh N 1 of Warwick, Coventry, UK, 2Queen’s University Belfast, Belfast, UK . . . . 1University Objectives: Diabetic retinopathy (DR) is a leading cause of sight loss in people of working age. There have been recent developments in laser photocoagulation techniques, along with anti-vascular endothelial growth factor drugs for the treatment of severe non-proliferative and proliferative DR. The aim of the study was to evaluate the cost-effectiveness of panretinal photocoagulation administered at the severe non-proliferative diabetic retinopathy (NPDR) stage (early treatment), compared with waiting until high-risk proliferative (HR-PDR) characteristics (deferred treatment) developed. Methods: A Markov model with a 30-year time horizon was developed, with clinical pathway options for patients presenting with moderate NPDR through to irreversible-severe vision loss and blindness (and to death). Once patients entered a post-treatment health state they can progress to more severe health states, regress back to earlier stages of the disease, persist where they are, or die. NHS and personal social services perspective was adopted. Transition probabilities were based mainly on data derived from the Early Treatment Diabetic Retinopathy Study. Health state utilities, costs and complications were based on information from the literature, supplemented by expert opinion. Costs and outcomes were discounted at 3.5%. Both deterministic and probabilistic sensitivity analyses were conducted. Results: Administering panretinal photocoagulation at the severe NPDR stage was more effective and less costly than waiting until HR-PDR developed. Sensitivity analyses gave similar results, with early treatment continuing to dominate deferred treatment. The probabilistic sensitivity analysis suggests that at willingness-to-pay threshold of £20-£30,000 per quality-adjusted life year, the probability of early treatment being cost-effective is 60%. Conclusions: Panretinal photocoagulation administered at the severe NPDR stage is likely to be cost-effective. However, given the limitations of the evidence on current treatments, these results to be interpreted with caution. A trial of early versus deferred laser therapy is needed to provide better data based on modern treatments. PSS33 Cost-Utility Analysis (Cua) of First-Line Actinic Keratosis (Ak) Treatments in Finland Soini E 1, Hallinen T 1, Sokka A L 2, Saarinen K 3 1ESiOR Oy, Kuopio, Finland, 2LEO Pharma Oy, Vantaa, Finland, 3PlusTerveys Oy, Nastola, Finland . . . . . Objectives: CUA of cryosurgery, topical treatments (diclofenac 3% 12 weeks, imiquimod 3.75% 6 weeks or 5% 4/8 weeks, ingenol mebutate gel (IMG) 0.015%/ head 3 days or 0.05%/body 2 days), and methyl aminolevulinate + photodynamic therapy (MAL+PDT) in the treatment of 25cm2 AK-plague affecting any body part. Methods: A sequential probabilistic decision-tree with 2-year time-horizon was used to assess the cost-utility (incremental cost-effectiveness ratio, ICER) of AK-treatments, and to determine the cost-effectiveness acceptability frontier (CEAF) and expected value of perfect information per patient (EVPI) from health care payer perspective. In the model, the first-line AK-treatment resulted in complete clearance (CC) with or without adverse events (AE), non-CC or AK-recurrence. Non-CC AK was retreated with PDT and AK-recurrence was retreated with the previous treatment. Incident AK-patients (year 2009, n= 3409, organ transplant patients excluded) were identified from the Finnish hospital discharge register to assess AK-related 2-year secondary health care costs for patients initiating different treatment regimens. Other costs included general practitioner, AE-management, and outpatient drugs (5/2014 without VAT; other costs in 2013 value). Quality-adjusted life-years (QALY) were based on EQ-5D. Results were discounted with 3% annually. Results: The mean per patient 2-year QALYs (costs) were 1.519 (€727) for IMG 0.015%, 1.518 (€ 887) for MAL+PDT, 1.516 (€ 802) for IMG 0.05%, 1.514 (€ 995) for diclofenac, 1.512 (€ 815) for imiquimod 3.75%, 1.511 (€ 707) for imiquimod 5%, and 1.507 (€ 1010) for cryosurgery. IMG 0.015% had € 2806/QALY gained ICER against imiquimod 5%, and dominated other AK-treatments. IMG 0.05% dominated diclofenac, imiquimod 3.75% and cryosurgery, and had € 21,550/QALY gained ICER against imiquimod 5%. MAL+PDT had € 32,848/QALY gained ICER against IMG 0.05%. Based on the CEAF, IMG 0.015% was the optimal treatment when willingness-to-pay/QALY gained exceeded € 2806. The EVPI was € 25/€ 81/€ 211 with the willingness-to-pay of € 0/€ 15,000/€ 30,000 per QALY gained. Conclusions: IMG 0.015% was the most cost-effective first-line AK-treatment. PSS34 Cost-Effectiveness of Ranibizumab VersUS Aflibercept in Treatment of Treatment of Visual Impairment Due to Diabetic Macular Oedema (Dmo) Haig J 1, Regnier S A 2, Malcom W 3, Xue W 4 . . . . . 1Optum, Burlington, ON, Canada, 2Novartis AG, Basel, Switzerland, 3Novartis UK LTD, Frimley/ Camberley, Surrey, UK, 4Optum, Uxbridge, UK Objectives: To estimate the cost-effectiveness of ranibizumab 0.5mg pro re nata (PRN) compared with aflibercept 2mg bi-monthly in the treatment of visual impairment (VI) due to diabetic macular oedema (DMO) taking a UK health care perspective. Methods: A Markov model previously reviewed by the National Institute for Health and Care Excellence (NICE) was used to simulate the long-term outcomes and costs (at 2012 price level) of treating DMO. The health states were defined by increments of 10 letters in best-corrected visual acuity (BCVA) with a 3-month cycle length. Patients could gain (or lose), at most, 2 health states between two cycles. A lifetime time horizon was implemented. Future costs and health outcomes were discounted at 3.5% per annum. Baseline characteristics, ranibizumab effectiveness and adverse events were estimated with data from the RESTORE trial (36 months). A published network meta-analysis was used to assess the relative effectiveness of ranibizumab to aflibercept. Aflibercept injection frequency was calculated with VIVID/VISTA phase III trials. Different utilities were used if the treated eye was the better or the worse-seeing eye. Results: Ranibizumab monotherapy leads to an incremental gain of 0.05 quality-adjusted life-years (QALY) (0.04 from the better-seeing eye and 0.01 from the worse-seeing eye) with a cost savings of ₤5,841 relative to aflibercept. Therefore, ranibizumab provides greater health gains with lower overall costs than aflibercept. Probabilistic sensitivity analysis shows that ranibizumab has a 58% probability of being dominant and 79% probability of being cost effective compared with aflibercept at a willingnessto-pay threshold of £20 000/QALY. Conclusions: Ranibizumab is dominant over aflibercept in the treatment of VI due to DMO. PSS35 The Future Health Economic Potential of Next Generation Artificial Vision Devices for Treating Blindness in Germany: an Early Cost-Utility Assessment Schwander B AHEAD GmbH, Loerrach, Germany . Objectives: The next generation of artificial vision devices (AVDs), which is currently tested in clinical trials, has the potential to improve the vision of blind patients with retinitis pigmentosa (RP) in a manner that they will be categorized as visual impaired but no longer as blind. This unprecedented vision improvement will result in a mentionable quality of life gain which poses the question at which costs the next generation AVDs are to be regarded as cost-effective. Methods: In order to answer this research question a Markov model, with the health states blind, visual impaired and death, was developed to simulate and to compare the costs and effects of next generation AVDs versus best supportive care (BSC) over a lifetime horizon. Health care costs and health utilities for the Markov health A610 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 states were determined on the basis of published literature. For next generation AVDs, which are currently tested in clinical trials, various possible effect and pricing scenarios have been simulated. Results: Applying the base case settings resulted in incremental costs of € 107,925, in 2.03 incremental quality-adjusted life years (QALYs) and in a cost-effectiveness ratio of € 53,165 per QALY gained. Probabilistic and deterministic sensitivity analyses as well as scenario analyses for the effect size and the AVD costs were performed in order to investigate the robustness of results. In these analyses a strong variation of the cost-effectiveness results was obtained ranging from € 23,512 (best case) to € 176,958 (worst case) per QALY gained. Conclusions: The innovative nature, the high unmet medical need and the expected unprecedented efficacy of next generation AVDs will highly likely lead to the case that even relatively high incremental cost effectiveness ratios, that have been obtained when simulating various effect and pricing scenarios, will be regarded as acceptable from a German health care payer perspective. PSS36 Implications for Time Savings Using New Intraoperative Measuring Technology Tavardkiladze G 1, Bakhshinyan V 1, Deger M 2, Irwin C 2, Rose S 2 Research Centre for Audiology and Hearing Rehabilitation, Moscow, Russia, 2Cochlear AG, Basel, Switzerland . . . . . 1National Objectives: Intraoperative threshold measurement is a part of the cochlear implantation procedure and in the current setting conducted by the clinicians with a standard set-up. The newly released CR220 Intraoperative Remote Assistant is a handheld device and can also be used by someone already in the operating theatre (OT). The aim of this study was to compare measurement time with the new CR220 and standard set-up and to investigate from the clinic’s perspective any cost-savings created as a result of time-savings with the new device. Methods: Stages of the measurement process are identified and the time is measured for each stage during 113 patients’ implantation procedure. A literature review was conducted to identify the reimbursement level of this process in order to translate any time-savings to cost savings. Results: When the clinician travels to the OT, the mean time spent per procedure with CR220 is 8.4% less than the computer set-up (163.7 minutes vs 149.9 minutes). If the measurement is conducted by someone already in the OR, the measurement time is reduced by 95.5% with the CR220 (163.7 minutes vs 7.3 min). Literature review revealed that the fee for measurement as $18.99-22.57 per 15 minutes in the US setting and in most of the other settings this procedure is not reimbursed separately but covered under cochlear implantation. Conclusions: The analysis showed that considerable time is spent for the clinician to travel to OT and waiting in the OT. This “unproductive” time is not only wasteful, but also means the clinician is not available in the clinic seeing patients where their expert skills are of most value. Moreover the clinic is either underpaid or is not paid at all for this expertise and time demanding process. The new CR220 gives clinics the opportunity to allocate their limited resources efficiently. Sensory Systems Disorders – Patient-Reported Outcomes & Patient Preference Studies PSS37 Drug Survival Rates And Cost of Biological Agents for the Treatment of Moderate to Severe Psoriasis in the Balearic Islands (Spain) Ventayol P Hospital Son Espases, Palma de Mallorca, Spain . Objectives: There are few studies combining dose regimen in routine clinical practice, drug survival rates and costs of biological agents for the treatment of naïve patients with moderate-to-severe psoriasis in the clinical practice. To assess the dose regimen in routine clinical practice, drug survival rate (persistence rate) and efficiency (cost per persistence) for etanercept (ETN), adalimumab (ADA) and ustekinumab (UST) in a real practice clinical setting. Methods: A retrospective study on psoriasis patients aged ≥ 18 years, naïve to a biological agent and a minimum of 6 months of treatment was performed in 5 public health system hospitals in the Balearic Islands (Spain) for the period from January 1st 2010 to December 31st 2013. The recorded variables were: sex, weight, age, indication (psoriasis or psoriatic arthritis), discontinuation reason and pharmacy dispensation records. Costs were based on the average wholesale price, estimating annual cost according to the first treatment received. Persistence rates were reckoned taking into account the current total days of therapy comparing posology with pharmacy supplied dose, and were estimated using the method of Kaplan-Meier. Results: During the study period a cohort of 112 psoriatic patients (57% men) were evaluated: 37 patients with ADA (81 kg, 51 years, 27; mean weight, mean age, and prevalence of psoriatic arthritis respectively), 34 with ETN (82 kg, 52 years, 25%) and 41 with UST (76 kg, 43 years, 19%). The persistence rate at 2 years was, 48%, 62% and 81% and the cost per persistence at 2 years was 52.961 € , 40.160 € , and 30.657 € (for ADA, ETN and UST respectively). Conclusions: UST showed better overall drug survival compared to ETN and ADA. UST has been the most efficient alternative for the treatment of naïve patients and has shown the least budget-impact per persistent-patient at 2 years analysis. PSS38 Medication Adherence and Discontinuation Predicted by Disease Duration in Glaucoma Patients: Findings From A Cross-Sectional Study in Korea Park K H 1, Cha J H 2 1Seoul National University College of Medicine, Seoul, South Korea, 2Pfizer Pharmaceuticals Korea Ltd., Seoul, South Korea . . . . Objectives: Although several studies reported patients with chronic disease were found to have lower medication adherence and higher discontinuation rates as disease duration increased, it is still not evident in glaucoma patients. With this perspective, this study was designed to assess the association of disease duration with medication adherence and discontinuation in glaucoma patients in Korea. Methods: It was a cross-sectional, multi-centered and observational study where glaucoma outpatients with less than two years of drug use were recruited at 15 eye clinics from March to November 2013. All patients completed a self-administered questionnaire asking about their daily use of glaucoma medications to estimate adherence and discontinuation. Medication adherence and discontinuation were defined as patients administering the drug for ≥ 80% of prescribed days and if patients stopped taking medication for 7 consecutive days respectively. Results: A total of 1,050 glaucoma patients were enrolled in the study. Of the total, 14.4% showed to be non-adherent to their glaucoma therapy and 7.5% had the experience of medication discontinuation. All patients were categorized into 3 groups according to disease duration: group A ≤ 1 year (n= 600,57.1%), B > 1 year and ≤ 2 years (n= 415,39.5%), and C > 2 years (n= 35,3. 3%). The patients of group A with the disease duration ≤ 1 year were likely to be non-adherent to glaucoma therapy compared to those with longer disease duration. (A: 84.9% vs. B: 86% vs. C: 100%, p=. 045) Highest discontinuation rate was found in group B with the disease duration bewteen 1 and 2 years. (A: 6.7% vs. B: 8.9% vs. C: 5.7%, P= . 380) Conclusions: The study results highlight more attention should be paid to the patients who newly started glaucoma therapy because in the patients with less than 2 years of disease duration the adherence was low and the discontinuation rate was high. PSS39 Health State Utilities for Pressure Ulcers – A Comparison of Condition-Specific and Generic Measures and Time-Trade-Off (Tto) Meads D M 1, Czoski-Murray C 1, Rutherford C 2, Dealey C 3, McGinnis E 4, Stubbs N 5, Wilson L 1, Nixon J 1, Hulme C T 1, McCabe C 6 1University of Leeds, Leeds, UK, 2University of Sydney, Sydney, Australia, 3Birmingham Hospitals NHS Trust and University of Birmingham, Birmingham, UK, 4Leeds Teaching Hospitals NHS Trust, Leeds, UK, 5Leeds Community Healthcare NHS Trust, Leeds, UK, 6University of Alberta, Edmonton, AB, Canada . . . . . . . . . . . . Objectives: To compare a newly developed condition-specific utility index (CSUI), the Pressure Ulcer Quality of Life Utility Index (PUQoL-UI) with generic and directly elicited TTO values. Methods: The PUQoL-UI was completed by a group of patients (n= 100) in England with pressure ulcers (PUs) along with the EQ-5D and own health TTO. The discriminatory power of the utility measures was assessed across PU grade and health and PU severity ratings. Multivariate regression was conducted to explore determinants of utility values. Results: The mean sample age was 77.2 years (range 22.7-101.7), 49% were female and 50% wheelchair users. Mean (SDs) utility for superficial PUs (grades 1-2) were 0.72 (0.17), 0.70 (0.35) and 0.24 (0.16) and for severe PUs (grades 3-4) 0.67 (0.17), 0.65 (0.35) and 0.15 (0.38) for the PUQoL-UI, TTO and EQ-5D, respectively. Mean (SDs) utility by self-reported PU severity was: [Mild] 0.78 (0.16), 0.66 (0.35), 0.29 (0.36); [Moderate] 0.72 (0.17), 0.63 (0.38), 0.25 (0.34); [Severe] 0.58 (0.17), 0.70 (0.33), 0.04 (0.40) for the PUQoL-UI, TTO and EQ-5D, respectively. Regression analyses indicated both EQ-5D and PUQoL-UI values were explained by perceived severity and general health ratings but not demographics or PU grade. Duration and body part affected were additional significant explanatory factors of the EQ-5D while wheelchair use approached significance. Conclusions: Values were much lower for the EQ-5D than the other assessments which may be partly explained by the range in EQ-5D and partly due to background mobility issues being captured. The PUQoL-UI appears to have good discriminatory power and is recommended for use in trials of PU interventions. The utilities presented here will be useful for decision-analytic models that incorporate PU impact. Probabilistic sensitivity analyses including the PUQoL-UI will likely generate lower levels of uncertainty than the EQ-5D due to the smaller SDs for health states. PSS40 Estimating Utility Data for Patient Symptom Severity in Chronic Spontaneous Urticaria Hawe E 1, Stull D E 1, McBride D 1, Balp M M 2 1RTI Health Solutions, Manchester, UK, 2Novartis Pharma AG, Basel, Switzerland . . . . . . Objectives: To obtain utility estimates suitable for use in economic models for chronic spontaneous (idiopathic) urticaria (CSU). Methods: Patient-level data from three randomised clinical trials: ASTERIA I, ASTERIA II, and GLACIAL were analysed. Health states were derived from Urticaria Activity Score (UAS7), a patient-completed diary of signs and symptoms which calculates an average daily score over 7 days. Higher score means more severe symptoms. UAS7 scores for the health states were: Urticaria-free: 0; Well-controlled urticaria: 1-6; Mild urticaria: 7-15; Moderate urticaria: 16-27; Severe urticaria: 28-42. Mean EQ-5D utilities were calculated for each health state. Individual trial analyses showed inconsistent utilities across the UAS7 health states due to small subsample sizes. A mixed model was used to predict EQ-5D according to UAS7 health states in a pooled dataset containing all treatment arms and time-points from the three trials. The predictor variable was UAS7 health state and the dependent variable was EQ-5D utility. Fixed/random effects for trial and patient were included and the following covariates: UAS7 health state at baseline (Moderate or Severe), presence of angioedema at baseline and during follow-up, duration of CSU, number of previous CSU medications, and gender of the patient. A parsimonious model was selected using the approach of backwards elimination; UAS7 health state was forced into the model. The validity of pooling trials was considered through visual comparisons and interaction terms. Results: There was a consistent improvement in EQ-5D utilities as severity of urticaria improved. Mean utilities at Week 12 ranged from 0.712 in patients with severe urticaria to 0.897 in patients who were urticaria-free. Sensitivity analysis confirmed the robustness of results. Conclusions: The results suggest that EQ-5D utility score increased with decreasing severity of urticaria. EQ-5D utility scores allow the comparison of HRQoL across diseases by calculating QALYs in economic models. A611 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PSS41 Health Related Quality of Life in Patients With Actinic Keratosis - Results From Patients Treated in Dermatology Specialist Care In Denmark Ragnarson Tennvall G 1, Norlin J M 2, Malmberg I 3, Erlendsson A 4, Hædersdal M 4 Swedish Institute for Health Economics, Lund, Sweden, 2LEO Pharma A/S, Ballerup, Denmark, 3LEO Pharma AB, Malmö, Sweden, 4Bispebjerg Hospital, Copenhagen, Denmark . . . . . . 1IHE, The Objectives: Actinic keratosis (AK) is a common skin condition associated with cumulative sun exposure that may progress to non-melanoma skin cancer. The disease can potentially influence Health Related Quality of Life (HRQoL), but studies of HRQoL in patients with AK are limited. The objective was to analyze HRQoL in patients with AK using generic and disease-specific HRQoL instruments and to analyze the relationship between instruments. Methods: AK patients who visited dermatological clinics in Denmark were included in an observational, crosssectional, study in a multi-center setting. Dermatologists assessed AK severity and patients completed: Actinic Keratosis Quality of Life Questionnaire (AKQoL), Dermatology Life Quality Index (DLQI), EQ-5D (5L), and EuroQoL Visual Analogue Scale (EQ-VAS). Results: A total of 312 patients from 10 clinics were included in the analyses. In general, patients with AK reported impaired HRQoL. The mean values (possible range) were: AKQoL 6.7 (0-27), DLQI 2 (0-30), EQ-5D-5L 0.88 (0-1), and EQ-VAS 79 (0-100). HRQoL was least affected in patients with mild actinic disease, whereas patients with severe actinic damage suffered from further impaired HRQoL (mean AKQoL 10.1 and DLQI 4.6). The correlation between DLQI and AKQoL was moderate (0.52), whereas the correlations between DLQI and EQ-5D (-0.36) and between AKQoL and EQ-5D (-0.10) were weak. Conclusions: All patients with AK had impaired HRQoL. Patients with severe actinic damage were considerably more affected than those with mild disease. Correlations between instruments demonstrate that they are complementary as they measure different aspects of HRQoL and are used for different purposes. EQ-5D is essential for economic evaluations, the DLQI is responsive to changes in relation to treatment and AKQoL captures important aspects related to sun damaged skin. PSS42 Categorical Health States In Chronic Spontaneous Urticaria (Csu) Based On The Weekly Urticaria Activity Score (Uas7): Are They Distinct, Discriminative, And Reproducible? Stull D E 1, McBride D 1, Gimenez-Arnau A 2, Grattan C 3, Khalil S 4, Balp M M 4 Health Solutions, Manchester, UK, 2Hospital del Mar and Universitat Autònoma, Barcelona, Spain, 3Norfolk and Norwich University Hospital, Norfolk, UK, 4Novartis Pharma AG, Basel, Switzerland . . . . . . . . 1RTI Objectives: Specific ranges of scores reflecting patient severity or changes in severity have not been established for average daily urticaria activity summed over 7 days (UAS7; range= 0-42), a common measure for assessing CSU disease activity. This study evaluates whether five non-overlapping health states derived from the continuous UAS7 score can discriminate between patients with different severities of urticaria and are reproducible across multiple studies. Methods: Data come from three randomised, double-blind, placebo-controlled Phase III clinical trials evaluating the effect of omalizumab on symptoms of patients with refractory CSU. Five CSU health states were defined: Urticaria-Free (UAS7= 0); WellControlled Urticaria (UAS7= 1-6); Mild Urticaria (UAS7= 7-15); Moderate Urticaria (UAS7= 16-27); Severe Urticaria (UAS7= 28-42). Comparison variables included the Dermatology Life Quality Index (DLQI), a 10-item dermatologic QoL instrument (range= 0-30; higher scores= greater QoL impairment); patient diary questions asking about sleep and activity interference; presence of angioedema; and number of diphenhydramine 25mg pills taken in previous 24 hours. Analyses established whether different UAS7 health states showed different values on comparison variables. Analyses were replicated across the trials at baseline and weeks 12,24, and 40 (ASTERIA I and GLACIAL) and baseline and weeks 12 and 28 (ASTERIA II). Results: Mean values for comparison variables were lowest (zero or very close to zero) for patients who were Urticaria-Free and highest for those with Severe Urticaria. For Well-Controlled and Mild Urticaria comparison variable values increased. Larger increases in values occurred for Moderate and Severe Urticaria. Changes in categorical health state severity were highly related to categorical changes in DLQI (p< 0.001 for all trials and time points). Conclusions: Categorical UAS7 health states show meaningful differences in mean values on comparison variables and are highly related to established levels of effect on dermatological QoL. Categorical UAS7 health states could be informative about subgroups for economic models and useful for clinical practice. PSS43 The Burden Of Primary Hyperhidrosis On The Patient: Eq-5d-5l Utilities, Willingness To Pay And Daily Time Spent In Managing The Condition Kamudoni P 1, Salek M S 1, Mueller B 2 1Cardiff University, Cardiff, UK, 2Riemser Pharma GmbH, Greifswald - Insel Riems, Germany . . . . Objectives: The objective of this study was to estimate the burden associated with primary hyperhidrosis by assessing patient’s health utilities, willingness to pay (WTP) for a complete cure and daily time spent in managing the condition. Methods: The data used in this study were collected under a longitudinal multi-stage research undertaken to develop and validate a new HRQoL instrument from patients with hyperhidrosis recruited through online social networking communities (Hyperhidrosis support group UK and International hyperhidrosis society) from January to August 2013. Only the baseline assessment is used in this analysis. Disease severity was measured using the Hyperhidrosis Disease Severity Scale (2 = for tolerable sweating, 3 = …barely tolerable sweating, 4 = intolerable sweating). The EuroQoL 5D-5L was used for assessing health utility index. Results: EQ-5D health utility index was lower in patients with more severe hyperhidrosis [mean utility value = . 85± 0.13 for HDSS = 2,0. 8±. 15 for HDSS = 3, and 0.69±. 2 for HDSS = 4, chi-square = 25.86, df = 2, p < 0.001]. Further, the health utility index was. 64 ±. 22 for WTP £0,0. 81±0.16 for £ 1 to 49,. 81±15 for £ 50 to 99,. 76±. 16 for £100 to 199,. 79± for £ 200 to 299,. 71±. 19 for £300 or more. Patients spent a mean of 50±134 minutes (HDSS = 2), 65±119 minutes (HDSS = 3) and 161±293 minutes (HDSS = 4) for daily management of hyperhidrosis. WTP showed the lowest correlation to disease severity. Conclusions: The current study underscores the multidimensionality of the burden of hyperhidrosis, with all aspects showing greater impairment with greater disease severity. Health utility and daily time spent in managing the condition offered significant discrimination of patients. PSS44 Subjective Expectations Regarding Life Expectancy And HealthRelated Quality Of Life In Moderate To Severe Psoriasis Patients Rencz F 1, Gulacsi L 1, Remenyik É 2, Szegedi A 2, Holló P 3, Kárpáti S 3, Péntek M 1, Brodszky V 1 University of Budapest, Budapest, Hungary, 2University of Debrecen, Debrecen, Hungary, 3Semmelweis University, Budapest, Hungary . . . . . . . . 1Corvinus Objectives: To assess psoriasis patients’ subjective expectations regarding their future health-related quality of life (HRQOL) and life-expectancy, and to explore variables associated with under- or overestimating behaviour. Methods: A crosssectional questionnaire survey of adult moderate to severe psoriasis patients was carried out. Patients were asked to indicate the age they expect themselves to live. HRQOL expectations were measured by the EQ-5D descriptive system for 6 months ahead and for future ages of 60,70,80 and 90, respectively. Current health state was evaluated with EQ-5D and visual analogue scale (EQ VAS), Dermatology Life Quality Index (DLQI) and Psoriasis Area and Severity Index (PASI). Results: Overall 167 patients (71% males) were included in the analysis with mean age of 50.38±12.35 years, mean EQ-5D, EQ VAS, DLQI and PASI scores were 0.71±0.30,65.3±21.08,5. 89±7.10 and 7.82±10.13, respectively. Currently 56% of the patients were on biological therapy. Patients expected 0.1±0.23 mean improvement in EQ-5D scores within 6 months (p< 0.001); inverse or palmoplantar psoriasis, and using only topical treatment or initiation of the first biological at the time of the survey were likely associated with higher expectations. Males overestimated their life-expectancy by 2.94±11.86 years whereas females underestimated by 5.23±9.34 years (p< 0.001) compared to the gender- and age-matched statistical life-expectancy. Expected mean EQ-5D scores for ages from 60 to 90 were: 0.56±0.48,0. 38±0.50,0. 15±0.55, and -0.17±0.54 (p< 0.001), respectively that are lower than the general population norms in Hungary. Both for 6 months ahead and older ages, expected EQ-5D correlated moderately with current EQ-5D and EQ VAS and only weakly with DLQI and PASI (p< 0.05). Conclusions: Patients expected considerable improvement in their HRQOL for the near future and large-scale deterioration for older ages. Exploring unrealistic expectations might help to prevent dissatisfaction with treatment benefits and to improve compliance. PSS45 The Decision Making Process In Receiving Bone Conduction Implants (Bci) For Single Sided Deafness Kosaner M , Urban M VIBRANT MED-EL Hearing Technology GmbH, Innsbruck, Austria . . Objectives: The main objective of this study was to evaluate the process in which patients with single sided deafness proceed to receive bone conduction implants. Factors contributing to decisions for or against implantation were also compiled. Methods: Using a comprehensive search strategy, several online databases were searched to identify studies published since 2002. Research involving adults and children with single sided deafness (SSD); and reporting on patient preference for receiving BCIs were included. Screening of titles, and data extraction and quality assessment of full papers were undertaken by one reviewer with any uncertainties resolved by consultation with a second reviewer. Results: 16 studies were identified covering a total of 914 individuals diagnosed with SSD. All patients who trialled a CROS device preferred to receive a BCI. Acceptance of new generation CROS devices is suggested to be better but still low. Following a BCI Headband trial 19% to 77% of patients across studies (mean 51%) proceeded to receive a BCI. When reported, the most common reason for rejecting implantation was insufficient benefit with the Headband in speech in noise or insufficient/no relief from tinnitus. Studies assessing factors in decision making found that age, gender, aetiology, duration of hearing loss or the presence of contralateral hearing loss did not differ between individuals who decide for or against implantation. One study so far suggests the role of transcranial attenuation at 2 kHz and tinnitus loudness to play a role in decision making. Conclusions: When given the option to trial traditional treatments and BCI simulators/Headbands many patients with SSD reject BCIs. This research highlights the importance of providing such trials before implantation. It is still unknown which aspects play a role in decision making and identifying better candidates. PSS46 The Burden Of Chronic Urticaria In Europe: A Systematic Literature Review Betoret I 1, Lambert C 1, Paravisini A 1, Tribaldos M 2, Paz S 3, Lizán L 2 Farmaceutica, Barcelona, Spain, 2Outcomes 10, Castellon, Spain, 3Outcomes’10, Castellon, Spain . . . . . . 1Novartis Objectives: To synthesize and analyze the available information on the burden of chronic urticaria (CU) [Patients’ Reported Outcomes (PROs): Health related quality of life (HRQoL), adherence, satisfaction, preferences, use of medical resources and costs] in Europe. Methods: A systematic review on PROs and costs of CU was performed. International (Pub Med, WOK, Scopus, Cochrane Library) and national (CSIC-IME, IBECS, MEDES) databases were consulted. Original articles, narrative/systematic reviews of studies developed in Europe, until December 2013 were retrieved. Editorials, letters/commentaries, and efficacy or economic evaluations of specific drugs were excluded. Costs were updated to € , 2013. Results: 9 studies assessed HRQoL (3, Germany; 1, France, Greece, Italy, Spain, UK, Germany/France, respectively) and 1 satisfaction with treatments (Germany/France). No studies on adherence or preferences for treatments were identified. The CU-Q2oL instrument, (0-100, higher value, worse HRQoL), was the most frequently used (n= 4). Scores ranged from 18.4 A612 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 (Greece) to 42.8 (Germany) revealing an acceptable perception of HRQoL. Sleep, itching/embarrassment and mental health were the HRQoL dimensions most impaired. Patients taking prescription drugs were more satisfied than those taking over the counter (p< 0.01). Severely ill patients were willing to change therapies if new, more effective alternatives became available (p< 0.05). Only 1 study assessed the costs of CU in Europe while another one described the use of medical resources. CU total cost in France was € 2,139.48 per patient/year. Patients lost 2.2 working days/month, being productivity losses 92% of total costs. CU patients were mostly cared for a dermatologist according to findings in Germany. A mean of 11.7 (SD: 11.5) visits/ month to the dermatology clinic were reported. Conclusions: PROs and costs in CU are infrequently addressed in the literature. Findings show patients reduced HRQoL and their willingness for more effective therapies. Frequent medical visits and loss of productivity make CU a burdensome disease in European countries. PSS47 A Review Of Patient Reported Outcomes (Pros) In Psoriasis According To The Food And Drug Administration (Fda) Pro Guidance Criteria Cheng R Covance Market Access Services, Inc., San Diego, CA, USA . Objectives: Psoriasis is a chronic inflammatory skin condition that affects an estimated 4.5 to 7.5 million people in the United States. Numerous clinical trials have been conducted in psoriasis, many of which have included general dermatologic or psoriasis-specific PROs as study outcome measures. Commonly used PROs in these trials include the Dermatology Life Quality Index (DLQI), Psoriasis Disability Index (PDI), Psoriasis Life Stress Inventory (PLSI), Psoriasis Quality of Life Questionnaire (PQOL-12), and Skindex. This study assessed how well the development and validation process of psoriasis-specific PROs align with the FDA PRO guidance. Methods: We reviewed the development and validation studies of the five PROs to assess if they align with the FDA PRO guidance criteria with regard to: (1) content validity, including patient input in concept elicitation, item generation, and cognitive debriefing; and (2) psychometric testing, including construct validity, internal-consistency and test-retest reliability, and responsiveness. Results: The PDI was developed without patient input in terms of concept elicitation, item generation, or cognitive debriefing. Variability and oftentimes inadequately documented evidence of patient input for concept elicitation, item generation, and cognitive debriefing were observed for the other four PROs. Test-retest reliability was not evaluated in the PLSI. Evidence of responsiveness was not available for three of the PROs. Conclusions: Based on this review, additional qualitative and quantitative research is needed to eliminate the identified gaps before these PROs could fulfill the FDA guidance for inclusion as a measure for a PRO label claim. 40.4±17.5 years with 76% males, respectively. Amongst the Hungarian patients 18% used none or only topical therapy in the last 12 months, 31% systemic non-biological treatment and 52% biologicals whereas in Iran 48% of the patients applied only topicals and 39% treated with non-biological systematic therapy. Mean EQ-5D, DLQI and PASI of the Hungarian and the Iranian sample were 0.69±0.3, 6.29±7.3, 8.01±10, and 0.62±0.4, 10±6.5, 13±8.3, respectively. In both countries psoriasis of the neck and/or décolletage was associated with the greatest HRQOL reduction using either instrument (p< 0.05). Regarding clinical types, in Hungary the palmoplantar involvement while in Iran nail psoriasis patients reported the worst general HRQOL (mean EQ-5D scores: 0.36±0.3 and 0.47±0.4). Correlation between EQ-5D and DLQI was found very similar across the two countries (rs= -0.43- and -0.44, p< 0.001), but EQ-5D showed significant correlation with PASI only in Hungary (rs= -0.43, p< 0.001). Strong positive correlation was identified between DLQI and PASI in both countries but only in those patients who received systemic therapy: Iran (rs= 0.72, p< 0.001) and Hungary (systemic non-biological: rs= 0.65, p< 0.001, biological: rs= 0.76, p< 0.001). Conclusions: Our results suggest that disease severity, treatments, and country-specific differences might lead to variations in the relationship between the outcome measures used in psoriasis. PSS50 How Can the Quality of Life in Hand Eczema Questionnaire (qolheq) be Interpreted? A Banding Study Ofenloch R F 1, Weisshaar E 1, Dumke A -K 2, Molin S 3, Diepgen T L 1, Apfelbacher C 4 Hospital Heidelberg, Heidelberg, Germany, 2University Hospital Jena, Jena, Germany, 3University Hospital Munich, Munich, Germany, 4UR - University of Regensburg, Regensburg, Germany . . . . . . . . . 1University PSS48 Current Management and Barriers to Treatment for Wet AgeRelated Macular Degeneration (Wamd): Perspectives From Patients and Caregivers Objectives: The Quality of Life in Hand Eczema Questionnaire (QOLHEQ) is a diseasespecific instrument to assess health-related Quality of Life (HRQOL) in hand eczema (HE) patients. The QOLHEQ assesses four domains of HRQOL: (a) symptoms, (b) emotions, (c) functioning and (d) treatment/prevention. The QOLHEQ total-score ranges from 0-127 points. The aim of this study was to assign bands of the QOLHQ-score to an anchor question (AQ) on HE severity to aid the interpretation of the QOLHEQscore. Methods: Data was drawn from the German validation study of the QOLHEQ. We calculated median, mean and mode of the AQ for 30 categories of the QOLHEQscore to devise separate sets of bands. Weighted kappa was calculated in order to identify the set of bands with the best agreement between QOLHEQ and AQ. Results: Overall n=316 HE patients were included in the study. Their mean age was 46.7 years (SD=12.9) and 54.1% of the sample were female. With a weighted kappa of 0.54 the best agreement was found for the following band: QOLHEQ<9 = minimal impairment; QOLHEQ 9-25 = slight impairment; QOLHEQ 25-58 = moderate impairment; QOLHEQ 59-79 = severe impairment; QOLHEQ>79 = very severe impairment. Conclusions: This is the first study which uses an anchor-based approach in order to devise a banding for the QOLHEQ-score. This banding represents a standardized means of interpreting the QOLHEQ total score. Further studies are needed to explore which banding may be adequate for different language version of the QOLHEQ. Varano M 1, Eter N 2, Winyard S 3, Wittrup-Jensen K U 4, Heraghty J5 1Fondazione G. B. Bietti-IRCCS, Rome, Italy, 2University of Muenster, Muenster, Germany, 3Royal National Institute of Blind People, London, UK, 4Bayer Pharma AG, Berlin, Germany, 5Macular Disease Foundation Australia, Sydney, Australia PSS51 Factors Conditioning Health Related Quality Of Life In Patients With Psoriasis In Europe: A Systematic Review Of The Literature . . . . . Objectives: The aim of this global survey was to evaluate the management of wet age-related macular degeneration (wAMD) from a patient/caregiver perspective. Methods: Patients with a wAMD diagnosis and current or prior use of intravitreal injections and caregivers from nine countries (Australia, Brazil, Canada, France, Germany, Italy, Japan, Spain, and the UK) completed a questionnaire. Results: 910 patients and 890 caregivers were surveyed. 55% of patients had wAMD in 1 eye and 64% had been receiving intravitreal injections for > 1 year. Many caregivers were a child/grandchild of the patient (47%), or partner (23%); only 7% were a professional caregiver. Most (73%) patients visited a health care professional (HCP) within 1 month of first noticing a change in vision; 41% of patients who delayed visiting a HCP thought the symptoms would resolve. Following diagnosis, 54% of patients began treatment immediately, and a further 37% scheduled an appointment within 1–3 weeks. 52% of patients reported a temporary improvement or stabilization in vision as a result of current treatment, and 22% reported a return to pre-diagnosis vision or that their vision was still improving. Most patients and caregivers reported a number of obstacles in managing wAMD, including the treatment itself (35% and 39%, respectively), treatment costs (28% and 29%), and finding the right treatment options (27% and 31%). Additionally, 27% of caregivers found the patient’s treatment extremely inconvenient with 57% of employed (non-professional) caregivers having to take time off work or miss personal obligations. 16% of patients missed a HCP appointment because their caregiver was unable to take them (26%), fear of injections (21%), illness (19%), forgetfulness (15%), inconvenience (10%), costs (6%), and discomfort of injections (3%). Conclusions: Many patients and caregivers identified a number of obstacles in managing wAMD. These included the treatment itself (patients and caregivers) and loss of productivity (non-professional caregivers). PSS49 A Comaprative Cross-Sectional Study On Health-Related Quality Of Life In Psoriasis From Hungary And Iran Moradi M , Rencz F Corvinus University of Budapest, Budapest, Hungary . . Objectives: To compare health-related quality of life (HRQOL) of Hungarian and Iranian psoriasis patients and to analyze possible differences in the relationship between EuroQol 5 dimensions (EQ-5D), Dermatology Life Quality Index (DLQI) and Psoriasis Area and Severity Index (PASI). Methods: Same cross-sectional questionnaire survey was carried out in two countries. Altogether 200 adult psoriasis patients enrolled from two Hungarian university clinics and 62 from an Iranian clinic. Besides HRQOL assessment, data on demographics, applied treatments, affected body sites and clinical types were collected. Results: Mean age of the Hungarian and Iranian patients were 51.2 ±12.9 years with 69% males, and Obradors M 1, Figueras M 1, Paz S 2, Comellas M 3, Lizán L 3 Farmaceutica S. A., Barcelona, Spain, 2Outcomes’10, Castellon, Spain, 3Outcomes 10, Castellon, Spain . . . . . 1Novartis Objectives: To identify conditioning factors of Health Related Quality of Life (HRQoL) of patients with psoriasis as reported in the literature in the last 5 years in Europe. Methods: Electronic databases [PubMed, ISI-WOK, Cochrane Library, MEDES, CSIC-IME, IBECS] and grey literature [Google Scholar], were searched to identify studies written in English or Spanish on HRQoL in patients with psoriasis, published in Europe between January 1, 2009 and December 31, 2013. Bibliographic references were hand searched. Editorials, letters, commentaries, opinion papers and studies related to specific treatments were excluded. Results: 27 studies accomplished the inclusion criteria. Most of the publications (66.7%, n=18) were cross-sectional studies; 25.9% (n= 7) had a prospective design while 7.4% (n= 2) were retrospective. 12 studies analyzed the relation between HRQOL and demographic characteristics, determining in 75% and 50% of them, there was a relation between sex (women) and age (younger patients) with poorer HRQoL. 3 publications demonstrated the HRQoL impairment associated with visibility of skin lesions. 4 studies appraised the relationship between disease activity and HRQoL determining a more compromised HRQOL in those patients with active psoriasis. Disease Severity was the most frequent assessed factor, being studied in 13 publications, describing in 92% of them that patients with higher PASI had worse HRQoL. 2 articles indicated that disease symptoms as skin discomfort and pruritus were elements that negatively influenced the HRQoL of these patients. Emotional disturbances were identified as predictors of poorer HRQOL in 5 studies. 3 publications determined that the use of biologic therapy contributed to improve HRQoL. Conclusions: HRQoL has been broadly addressed in patients with psoriasis in Europe. Several disease- and patient-related factors contributed to its deterioration. Therapeutic measures with proved effectiveness in controlling disease symptoms and reducing PASI should be considered in patients with severe disease who have a relation with poorer HRQoL. Sensory Systems Disorders – Health Care Use & Policy Studies PSS52 Cost-Effectiveness and Value of Information Analyses of Nutritional Support in Preventing Pressure Ulcers in High Risk Hospitalised Patients Tuffaha H W 1, Shelley R 2, Chaboyer W 2, Gordon L G 1, Scuffham P A 1 University, Meadowbrook, Australia, 2Griffith University, Gold Coast, Australia . 1Griffith . . . . . . . A613 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To assess the cost-effectiveness of nutritional support (high protein supplementary diet) versus standard care (regular hospital diet) in preventing pressure ulcers in hospitalised patients at high risk of pressure ulcers and malnutrition. Further, to evaluate the need and value of additional research using value of information analysis. Methods: Analyses were undertaken from the perspective of the State health department in Queensland, Australia, using a Markov decision model. Evidence for the relative risk (RR) was estimated from a meta-analysis of randomised controlled trials; other parameters were systematically identified from the literature to populate the model. The incremental net monetary benefit (INB) was calculated and a probabilistic sensitivity analysis using Monte Carlo simulation was conducted. The expected value of perfect information (EVPI), expected value of perfect parameter information (EVPPI), expected value of sample information (EVSI), expected total cost of additional research, expected net benefit of sampling (ENBS), and the return on investment (ROI) were calculated for an estimated population of 125,000 over ten years. Results: At a willingness-to-pay of AU$ 50,000 per quality-adjusted life year, the INB was AU$ 530, with a probability of 84% for nutritional support to be the preferred intervention. The population EVPI was AU$ 4.75 million, the highest EVPPI was for RR at AU$ 2.25 million. For a future randomised study investigating the RR of the two interventions, the ENBS would be maximised at AU$ 380,000 with 1,200 patients in each arm; from an EVSI of AU$1.6 million and total study cost of AU$ 1.2 million. The expected ROI would be 32%. Conclusions: Nutritional support is cost-effective in preventing pressure ulcers in high risk hospitalised patients; however, there is uncertainty surrounding the decision and the value of this uncertainty is high. A future clinical trial to resolve this uncertainty is worthwhile. PSS53 The Impact of the German Drg-System on Policy Decision Making in Ent Fichtenbauer M , Kosaner M , Urban M Vibrant Med-El Hearing Technology Gmbh, Innsbruck, Austria . . . Objectives: The German diagnosis-related groups (G-DRG) have been introduced as a reimbursement system for in-patient care. The aim of this study was to report on clinic-based data available from the G-DRG browser regarding the impact of middle ear implantation as a basis for future health care decision making in ear, nose and throat medicine (ENT). Methods: Using a clinical algorithm, the unevaluated D23Z database from 2012 was analyzed to determine epidemiological characteristics of patients with different types of hearing losses and receiving middle ear implants (MEI). All ICD-10 Codes used for the diagnosis of conductive (CHL), mixed (MHL) or sensorineural hearing loss (SNHL) were extracted. The number of MEI cases reported in the D23Z was analyzed regarding Patient Clinical Complexity Level (PCCL), age, sex and duration of stay. The incidence of MEI candidates to the total number of ICD-10 codes and per indication were calculated. Results: According to the D23Z, approximately 30`000 individuals admitted to ENT clinics were diagnosed with a hearing loss. About 38% suffered from CHL, 8% suffered from MHL and 48% suffered from SNHL. The proportion of patients who received a MEI per indication were respectively 0.8% for SNHL, 8% for MHL and 1% for CHL. Overall, about 466 (1%) patients received a MEI. 49.8% of them were men and 50.2% were women. Children aged as young as 1-2 years could also receive treatment. All patients had normal-term stay in hospitals with no comorbidities or complications (PCCL < 4 in 99.2% of cases) The incidence of MEI to the ENT patient base and the German population were respectively 15.5/1,000 and 0.6/100,000. Conclusions: The overall number of middle ear implantations compared to standard treatments is quite low. Analyzing data on MEI available from highly-developed DRG systems can be used by other countries for policy decision making in ENT. PSS54 Understanding Trends In Ophthalmologist Patient Selection And Care Based On Patterns Of Billing Culp J L 1, Connelly N 1, Kristopik J 1, Doyle J J 2 Consulting, Durham, NC, USA, 2Quintiles, Hawthorne, NY, USA . . . . . . 1Quintiles The recent U. S. release of Medicare Part B billing data provides unprecedented insight into a health care system that reimbursed $77 billion in 2012. Of the 100 physicians receiving the greatest payments, nearly half were ophthalmologists. Objectives: Examine ophthalmologist trends in patient selection and care, and implications on system value delivery, based on Medicare and other data. Methods: We examined aggregate and line-item data, comparing the 100 highest billing ophthalmologists to each other, and to the remaining 16,971, identifying patterns in services billed, beneficiaries served, service location, and dollars billed, allowed and paid. We also compared volume and cost of a key officebased code (J2778: Ranibizumab injection) to a key facility-based code (67042: vitrectomy for macular hole), and incorporated epidemiology data to understand provider incentives and drivers of care. Results: Ophthalmologists represented 2% of all practitioners billing Part B, but received 7% of reimbursement; the Top 100 billing ophthalmologists accounted for 0.6% of ophthalmologists billing, but received 8.8% of payments; disparities between Top 100 billers and non-Top 100 include: variety of patients treated, percent of charges reimbursed, and practice location; high volume J2778 performers were much more likely to be Top 100 billers than were high performers of 67042; and based on prevalence among those 65 and older, a significantly greater percentage of beneficiaries with wet AMD received treatment than did beneficiaries with macular hole. Conclusions: A large portion of payments were received by relatively few ophthalmologists, whose practices demonstrate disparities in variety and volume of procedures over non-top billers; incentives—some financial—may be factors that contribute to greater treatment of wet AMD than of macular hole; and a focus on conducting high-reimbursement services may drive disproportionate system costs. Greater transparency may drive changes in practitioner behavior, and reduce unnecessary health care system costs. PSS55 Fabrication Of Voriconazole Solid Lipid Nanoparticles For Effective Ocular Delivery Kumar R , Sinha V R UIPS, Panjab University, Chandigarh, India . . . Objectives: Preparation of Voriconazole (VCZ) solid lipid nanoparticles (SLNs) for effective ocular delivery for the treatment of fungal keratitis. Methods: SLNs were prepared by solvent emulsification technique using Compritol (lipid), Pluronic F-68 (surfactant) and sodium taurocholate (co-surfactant). Characterization of SLNs was performed by size measurement, in-vitro release, ex-vivo corneal permeation studies and in-vitro antifungal activity. Results: Particle sizes were found in the range of 150-300 depending upon lipid/Smix ratio with good zeta potential. Entrapment efficiency of SLNs was found between 40-60% with sustained in vitro drug release (> 70% in 12h). The ex-vivo corneal permeation studies exhibited good ocular permeation of VCZ from SLNs. Ex-vivo study also supports good ocular permeation of VCZ from SLNs when compared with drug suspension. Further, in-vitro antifungal activity exhibited the potential of VCZ SLNs. Conclusions: The sustained release property with good corneal permeation of VCZ from SLNs encourages its application for in-vivo studies and hence could be proposed as an effective carrier for ophthalmic administration. PSS56 The Analysis of Dental Care in Ukraine at the Regional Level Zaliskyy O , Shcherba P , Zalis’ka O Danylo Halytsky Lviv National Medical University, Lviv, Ukraine . . . Objectives: The structure of the overall incidence diseases the mouth and teeth morbidity occupy the third place in Ukraine. Thus 99% of the patients served in outpatient clinics institutions. The structure of diseases requiring hospitalization (approximately 1% of patients), the top spot is occupied odontogenic inflammatory diseases and injuries of the maxillofacial face. Methods: In Ukraine, the availability of dentists is 4.0 per 10 000 population and 4.5 per 10 000 children. We have analyzed the statistical data of the Lviv Regional Department of Health. We found that 98% of the population have dental problems. Prevalence of dental caries, temporary occlusion in 6-years children reached 87.9%, a 12-year-olds - 72.3%. Prevalence of chronic catarrhal gingivitis among adolescents aged 12-15 years ranged 70-98% and teeth abnormalities in children 7-18 years more than 80%. In Ukraine everyone requires a prosthetics after 50 years. Results: Main social burden have the dental institutions of the state and municipal property. By 2013 there were about 1.5 million causes to the dental care. We determined that in Lviv region on the basis of licenses to practice medicine in dentistry are 248 dentists, who working in cities and towns, and only about 10% - in rural areas. From 196th private dental surgeries 89 are situated in the regional center. Danylo Halytsky Lviv National Medical University opened and acting university Dental Center from 2012, where dentists had treated about 400 000 people, including more than 172 000 children each year. Such preventive examinations make it possible to carry out monitoring as indicators of dental public health field, to identify the most important risk factors for dental diseases. Conclusions: The management and pharmacoeconomic studies of dental care, identifying optimal funding for state and municipal health care institutions for cost-effectiveness use of state funds. PSS57 Dental Care Use And Associated Factors Among People With Rheumatoid Arthritis: A Nationwide, Population-Based, Propensity Score-Matched Follow-Up Study Chen Y T Yang-Ming University, Taiwan, Taiwan . . Objectives: Patients with RA suffer from a higher risk of periodontal attachment loss and increased oral inflammation. There were few studies to access the utilization of dental care among RA in the Taiwan. The purpose is identify realize analyze and discuss the dental use of diabetic patients, and the association between the risk of rheumatoid arthritis (RA) and a history of periodontitis. Methods: retrospective cohort studybased on the nationwide, population-based, NHIB used administrative data, case group consisted of 5,506 (age ≥ 18 years) patients with rheumatoid arthritis (RA group) as the study group and 22,024 patients without RA attending the Outpatient wing of Department of General Medicine formed the control group (NRA group). Matched for Age, gender and RUB, Both groups were matched on 1:4. Results: More advanced forms of periodontitis were found in RA patients compared with controls. The results showed that RA patients (66.9% of RA) had 5-years utilization rate of dental care than non-RA patients (13.9% of non-RA). However, people have RA or not, the characteristics of dental use were similar. Only has the gender aspects to differ from, when the male suffered from RA, the utilization of dental care were not different with the female. Conclusions: we propose that the consulting rheumatologists inform the patients that they have a higher risk of periodontal. this study demonstrates an association between periodontitis and incident RA. And the study is limited to lack of BMI, smoking, alcohol status. PSS58 Macular Oedema Due to Retinal Vein Occlusion Methods for the Identification of Treatment Guidelines and Areas of Unmet Clinical Needs by Means of Systematic Review Patel A S Abacus International, Bicester, UK . . Objectives: Retinal vein occlusion (RVO) causes macular oedema (MO), which can lead to vision loss. The present study sought to identify treatment guidelines internationally by conducting a systematic review and extensive hand searches. The aims were to i) develop a systematic methodology for the identification of such guidelines ii) review the guidelines and treatment pathways identified in order to propose optimal positioning for an hypothetical intervention for the treatment of MO in RVO, and iii) to identify areas of unmet clinical needs. Methods: A614 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Systematic searches in the electronic databases MEDLINE, EMBASE and The Cochrane library were conducted and 53 online databases (including HTA agency websites, international ministries of health, and clinical trials. gov) were hand searched for clinical guidelines in the treatment of MO caused by RVO. Results: Fifteen documents on treatment pathways or guidance used internationally were identified from the hand searches. No papers or abstracts were found from the electronic database searches. There were considerable between-jurisdiction differences in the guidance for the management of MO caused by RVO. These differences were consolidated to produce two amalgamated treatment pathways. In total, eight treatment positions for interventions in the treatment of RVO subtypes were identified. For one of the identified positions – treatment of ischaemic branch RVO – no licensed treatment currently exists. Conclusions: The described systematic methodology for the construction of treatment pathways may be used by manufacturers in early drug development decisions to identify unmet clinical needs, understand which treatment positioning may provide the most value, and identify future treatment comparators in the same indication. Guidelines to inform such commercial strategies may not be identifiable from electronic database searches alone with extensive hand searches being a necessity. Between jurisdiction guideline nuances also need to be taken into account when considering the target market for an intervention in development. PSS59 Ophthalmology: Therapy Trends in Europe Based on Clinical Trial Registry Data Gupta A , Mukherjee B , Mahal S S , Mathews A Novartis Healthcare Pvt Ltd, Hyderabad, India . . . . . Objectives: Ophthalmology pharmaceutical market is growing worldwide due to rising aging population, new delivery technologies and changing lifestyle. However, challenges like patent expiry of major brands and lack of awareness still persists. Therefore, it is important to be aware of the upcoming treatment options, changing patients’ needs and requirement of cost effective therapies. This analysis provides an overview of the recent trends and future scenario in Ophthalmology market. Methods: Pharmaceutical companies sponsored clinical trials initiated from January 2011 to April 2014 in Glaucoma, Age-related Macular Degeneration (AMD), Diabetic Retinopathy (DR) / Diabetic Macular Edema (DME), Dry eye syndrome (DES) and Retinal vein occlusion (RVO) have been considered. Only Phase I - III trials listed on public registries have been considered. Results: The data showed that > 30% of the trials are being conducted on Glaucoma in USA, Europe, Asia and Australia. This is followed by AMD (24%), DR/DME (21%), DES (17%) and RVO (7%). Also, > 50% ophthalmological trials are being conducted in USA; followed by Europe (~25%) and Asia (~20%). In Europe, 71 trials have been conducted on 48 molecules, of which 69% are chemical entities, 19% are biologicals and > 10% are entities like RNAi (oligonucleotide, aptamers), DARPin. Eye drops (46%) and intravitreal injections (37%) are the key topical and parenteral formulations, respectively. 10% of the trials have been conducted on oral formulations. In Europe, EU5 countries comprise of 43% of the trials and Germany has maximum 37 trials. Novartis has conducted trials in maximum 30 countries, followed by Santen (19), Pfizer (15) and Allergan (13) in Europe. Conclusions: Based on the analysis, currently, Glaucoma, AMD and DR/DME are the major focus of the companies in ophthalmology. Though, biologicals and RNAi are being tested routinely, chemical entities are foremost modalities. Similarly, eye drops remain as preferred method of delivery with respect to other newer delivery techniques. Research Poster Presentations - Session V Disease-Specific Studies CANCER – Clinical Outcomes Studies PCN1 Treatment Patterns and Health Outcomes Among Patients with Radioiodine-Refractory Differentiated Thyroid Cancer in the United States and Western Europe Gianoukakis A G 1, Flores N M 2, Pelletier C L 3, DiBonaventura M 2, Forsythe A 3, Wolfe G 2, Rege J 3, Taylor M H 4 1University of California, Los Angeles, Torrance, CA, USA, 2Kantar Health, New York, NY, USA, 3Eisai Inc, Woodcliff Lake, NJ, USA, 4Oregon Health & Science University, Division of Hematology & Medical Oncology, Portland, OR, USA . . . . . . . . . . . . Objectives: Most patients with differentiated thyroid cancer (DTC) have an excellent prognosis after receiving standard treatment, consisting of surgery and often adjuvant radioactive iodine (RAI). However, a subgroup of patients prove to have progressive DTC which is refractory to RAI (RRDTC). Treatment options for RRDTC are limited. This study investigated the treatment patterns and health care resource utilization of patients with RRDTC. Methods: Data were collected by performing a retrospective chart review study in the US and 5EU (France, Germany, Italy, Spain, UK) with physicians recruited from an online panel. Physicians provided clinical information on 1 to 4 of their RRDTC patients in an online survey. Demographics, disease history, treatment information, and health care resource were included and reported descriptively. Health care resource use was compared across treatment classes using general linear models. Results: 231 physicians participated and provided a total of 700 patient charts (44.1% of charts were from the US and 11-12% from each 5EU country). 45.0% of patients were male with a mean age at diagnosis of 55.1 years [SD= 12.4]. 52.0% of patients were treated with systemic treatment (e.g., 16.9% tyrosine kinase inhibitors [TKIs] only; 13.3% chemotherapy only). The remaining 48.0% were either in a watch and wait (“WW”) period (20.1%) or were managed with non-systemic palliative therapies (27.9%; eg, external beam radiation). Overall, patients averaged 15.87 days hospitalized per year (due to disease related complications or side effects). Although not statisti- cally significant (p> .05), a trend toward more days hospitalized from disease-associated complications was observed for patients managed with WW (Mean= 9.21, respectively) and non-systemic treatment (Mean= 8.27) than patients treated with chemotherapy (Mean= 7.25) or TKIs (Mean= 8.22). Conclusions: Among patients diagnosed with RRDTC, watch and wait and non-systemic treatment options remain common. A large direct cost burden may be observed given the frequent and long hospital stays. PCN2 Approving Drugs Based on Early Stage Data - How Phase II Trial Data Correlates with Phase III Outcomes. Case Study: NSCLC Macaulay R , Tan H HERON Commercialization, London, UK . . Objectives: There is increasing pressure on regulators from patients, physicians and industry for earlier access to pharmaceuticals for serious diseases. In reaction, in March 2014 the European Medicines Agency (EMA) announced it was piloting adaptive licensing, and the Medicines and Health care products Regulatory Agency (MHRA) unveiled their Early Access to Medicines Scheme. Nevertheless, there are questions over how, and if, Phase II trial benefits can be predictive of clinical advantages in Phase III studies, which this research aims to address. Methods: Phase III data of any Non-Small Cell Lung Cancer (NSCLC) oncologic appraised by the EMA, or that had failed Phase III clinical trials, since 2002 was extracted along with its corresponding Phase II data. Statistical tests were conducted using Pearson’s coefficient correlation. Results: 12 oncologics were identified with both Phase II and III readouts, 6 of which met their Phase III trial primary endpoint. Overall Response Rates (ORRs) reported in Phase II trials varied from 0%-61% (mean 24%). 4/4 (100%) drugs with Phase II ORRs > 30% met their primary endpoint vs. only 2/8 (25%) with ORRs ≤ 30%. Phase II ORRs were strongly correlated with Phase III Progression-Free Survival (PFS) (r2= 0.864, p< 0.0005) and Overall Survival (OS) outcomes (r2= 0.858, p< 0.001). Nevertheless, 5/6 drugs that failed their Phase III primary endpoints had comparative Phase II data indicating benefits versus these same comparators, most notably onartuzumab, whose Phase III trial was terminated early due to lack of efficacy, despite demonstrating significant OS benefits of 8.8 months in Phase II. Conclusions: In NSCLC, Phase II ORRs can be strongly predictive of the magnitude of PFS and OS readouts in Phase III trials. However, comparative advantages in Phase II trials seem to be poorly predictive of OS benefits in Phase III studies, raising questions over the appropriateness of approving drugs on early stage comparative data. PCN4 Cervical Human Papilloma Virus (HPV) DNA Primary Screening Test Results of the Experience of a Regional Laboratory in Central Italy Passamonti B U 1, Bulletti S 1, Gustinucci D 1, Martinelli N 1, D’Amico M R 1, Spita N 1, Malaspina M 1, Carlani A 1, DI Dato E 1, Galeazzi P 1, Tintori B 1, D’angelo V 1, Calvi C 2 1Centro Unico di Screening Regione Umbria, PERUGIA, Italy, 2ROCHE DIAGNOSTICS, MONZA, Italy . . . . . . . . . . . . . . . Objectives: To investigate feasibility and effectiveness of a cervical screening program with DNA tests as preliminary assay versus usual cytology protocols in Umbria Region. Methods: A large cohort of 35-64 aged women afferent to the unique regional laboratory was considered. The usual algorithm with cervical cytology as primary test was followed in January 2008–June 2010, whereas in August 2010–October 2011 high-risk human papillomavirus (HR-HPV) DNA test was used as primary screening. The cohorts were compared in terms of acceptance rate of invitation, cytological results, molecular results including HPV genotype, detection rate of histological lesions. Results: A total of 31,228 women were invited: 21,249 were suggested to undergo classical cervical cytology screening, 9,979 HR-HPV DNA test as primary screening. A similar rate of adhesion (56.6% vs. 56.5%) was observed. Age-related differences were evidenced, with younger women (35-49) more prone to accept the invitation to HR-HPV DNA testing rather than usual cytology screening (61.6% vs. 55.5%; p< 0.0001); analogously, uninvited younger women spontaneously requesting cervical screening were more prone to specifically request molecular than classical cytological testing (24.8% vs. 10.8%; p< 0.0001). Among the 6,272 HR-HPV DNA testing women, 396 (6.4%) were positive, and, among them, 141 (36%) featured an altered cytology. All patients with altered cytology were suggested to undergo colposcopy and 106 out of 141 (75.1%) answered to the invitation. Among them, 89 (84%) featured abnormal histology with 48 (45.3%) CIN1 and 41 (38.7%) CIN2. If comparing the CIN2 detection rate within the two studied periods, it was almost doubled using the HR-HPV DNA than pap test as primary assay (0.64% vs. 0.37%; p= 0.005). Finally, the implementation of the DNA test screening program did not increase total costs. Conclusions: although with some limits, the introduction of HR-HPV DNA primary testing resulted feasible and effective, significantly increasing detection of severe lesions. PCN5 Comparative Effectiveness of Treatments for Relapsed or Refractory Mantle Cell Lymphoma (R/R MCL), Using Matching Adjusted Indirect Comparison Tongbram V 1, Sengupta N 2, Gaudig M 3, Sidhu M 1, Exuzides A 4, Colby C 4, Sanden S V 5, McGovern A 1 1ICON Plc, Morristown, NJ, USA, 2Janssen Pharmaceuticals, Inc, Raritan, NJ, USA, 3Janssen Pharmaceuticals, Inc, Neuss, Germany, 4ICON plc, San Francisco, CA, USA, 5Janssen Pharmaceuticals, Inc., Beerse, Belgium . . . . . . . . . Objectives: Prognosis for relapsed or refractory (R/R) MCL patients with existing treatments is poor; most patients progress within ~4 months. Ibrutinib, an oral once daily Bruton’s tyrosine kinase inhibitor showed durable single agent activity with good response rate in 111 R/R MCL patients and a median progression free survival (PFS) of 13.9 months. Ibrutinib received breakthrough designation and United States Food and Drugs Administration approval for use in MCL patients who received at least one prior therapy (R/R MCL). This indirect analysis aims to compare the efficacy A615 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 of ibrutinib to available treatments for R/R MCL patients. Methods: A systematic literature review was conducted to identify clinical trials containing treatments of R/R MCL. Matching adjusted indirect comparison (MAIC), described by Signorovitch et al 2012, was utilized to obtain indirect relative treatment effect for ibrutinib compared to other treatments. Using individual patient level data (IPD), baseline characteristics of the ibrutinib trial patients were matched with the patients in the published studies to obtain overall response (ORR) and complete response (CR) rates based on balanced population between the ibrutinib and published studies. Kaplan Meir curves for overall survival and PFS of comparators were plotted alongside those of the matched ibrutinib patients. Results: Nineteen studies evaluating various treatments were identified. Five trials evaluating bortezomib, BR (bendamustine, rituximab), FCM (fludarabine, cyclophosphamide, mitoxantrone), FCM-R (fludarabine, cyclophosphamide, mitoxantrone, rituximab), and rituximab-hyper-CVAD were considered for matching. Complete matching of the IPD was possible for the bortezomib, FCM and FCM-R studies. Ibrutinib showed statistically significant better odds of achieving ORR compared to bortezomib (OR 3.62; 95% CI 1.18-11.14) and FCM (OR 3.22; 95% CI 1.01-10.26). Conclusions: The indirect analysis suggests a potential for improved ORR compared to a few relevant treatments in patients with R/R MCL. Phase III comparative confirmatory data with ibrutinib are anticipated in late 2014. PCN6 Overall Survival in Patients with HER2+ Early Stage Breast Cancer Patients Treated with Trastuzumab in the US Department of Defense Practice Setting Gallagher C M 1, More K 2, Masaquel A S 3, Kamath T 3, Guerin A 4, Ionescu-Ittu R 4, GauthierLoiselle M 4, Nitulescu R 4, Sicignano N 5, Barnett B 3, Wu EQ6 1Walter Reed National Military Medical Center, Bethesda, MD, USA, 2Naval Medical Center Portsmouth, Portsmouth, VA, USA, 3Genentech, South San Francisco, CA, USA, 4Analysis Group, Inc., Montréal, QC, Canada, 5Health ResearchTx, Trevose, VA, USA, 6Analysis Group, Inc., Boston, MA, USA . . . . . . . . . . . . Objectives: The NSABP/NCCTG trial (Romond et al. NEJM 2005; 353: 1673-1684) established the efficacy of trastuzumab in the adjuvant treatment of HER2+ early stage breast cancer (HER2+BC). Yet, little is known about the patterns of use and outcomes of adjuvant trastuzumab in clinical practice. The study aimed to estimate the overall survival (OS) and relapse-free survival (RFS) of HER2+BC patients treated with adjuvant trastuzumab in the US Department of Defense (DOD) practice setting. Methods: Adult women initiating adjuvant trastuzumab within 1 year of BC surgery were identified in the DOD health claims database (01/200312/2012). An algorithm based on secondary neoplasm ICD9 codes and treatment gaps and initiations was used to identify relapses. OS and RFS unadjusted rates at 3 and 4 years after the initiation of the adjuvant trastuzumab treatment were estimated from Kaplan-Meier plots. Results: The study sample included 3,188 women (median age 63 years), followed for a median of 3.3 years after the initiation of trastuzumab and treated continuously with trastuzumab for a median of 12 months. Of these 3,188 women, 13.8% received neo-adjuvant therapy prior to the surgery, 17.7% relapsed, and 7.9% died during the follow-up. The OS rates at 3 and 4 years were 93.2% (95% CI 92.1%-94.2%) and 90.0% (88.6%-91.2%), respectively. The corresponding RFS rates were 78.8% (77.1%-80.3%) and 75.8% (74.0%-77.5%), respectively. Conclusions: The findings suggest that most HER2+BC patients in the DOD practice setting received per-label trastuzumab treatment (for 52 weeks) and had OS rates that are similar to the OS rates that were previously observed in the NSABP/NCCTG clinical trial (90.0% vs. 93% at four years). The lower RFS rates observed in this study versus the NSABP/NCCTG trial (75.8% vs. 85.7% at 4 years), may be partially explained by differences in the characteristics of the patients, including age. PCN7 The Relative Efficacy of Treatments in First-Line Management of Newly Diagnosed Chronic Myeloid Leukaemia: Systematic Literature Review and Indirect Comparison Kroes M 1, Zagorska A 2, Osei-Assibey G 1, Paine A 3 1Abacus International, Bicester, Oxfordshire, UK, 2Bristol-Myers Squibb, Rueil-Malmaison, France, 3Zedediah Consulting, Wokingham, Berkshire, UK . . . . Objectives: To assess the relative efficacy of first-line treatments in chronic myeloid leukaemia (CML), an updated systematic literature review (SLR) and indirect comparison (IC) were conducted with follow-up period up to 48 months. Methods: We updated a SLR initially conducted in 2011. Medical databases were interrogated systematically in January 2014 to identify trials comparing first-line treatments for CML. Using a fixed-effect Bayesian model implemented in WinBUGS, ICs were made to calculate relative efficacy (cumulative complete cytogenetic response (CCyR) and major molecular response (MMR)) for dasatinib, nilotinib and imatinib. Results: Nineteen randomised controlled trials (RCTs) were included in the SLR, 10 were eligible for inclusion in the IC. Compared with imatinib 400mg by 12 months, odds of cumulative CCyR were significantly greater for dasatinib 100mg [odds ratio (OR) 2.25,95% credible interval (CrI) 1.55-3.15], nilotinib 600mg [OR 2.23 (95% CrI 1.503.21)] and 800mg [OR 1.94 (95% CrI 1.31-2.78)]. By 24 months compared with imatinib 400mg, the odds remained significantly higher with nilotinib 600mg [OR 2.03 (95% CrI 1.28-3.10)] and 800mg [OR 1.70 (95% CrI 1.08-2.55)] and higher, but not significant with dasatinib 100mg [OR 1.41 (95% CrI 0.85-2.22)]. By 12,24,36 and 48 months respectively, compared with imatinib 400mg, the odds of a MMR were: dasatinib 100mg [OR 2.22 (95% CrI 1.52-3.15) / 2.09 (1.45-2.93) / 1.76 (1.22-2.48) / 1.90 (1.27-2.74)] nilotinib 600mg [OR 2.86 (95% CrI 1.95-4.08) / 3.24 (2.26-4.54) / 2.45 (1.70-3.43) / 2.54 (1.75-3.59)] and 800mg [OR 2.76 (95% CrI 1.88-3.94) / 2.60 (1.82-3.62) / 2.12 (1.48-2.95)/ 2.18 (1.51-3.06)]. For both outcomes at all time points, there was no significant difference between dasatinib and nilotinib. Conclusions: Analysis including all available RCTs suggests that second-generation tyrosine kinase inhibitors dasatinib and nilotinib are more efficacious than imatinib 400mg and should be treatments of choice in newly diagnosed CML. PCN8 Treatments for EGFR Mutation-Positive (M+) NSCLC Patients – A Network Meta-Analysis (NMA) by Mutation Type Popat S 1, Yang J C H 2, Lungershausen J 3, Griebsch I 3, Marten A 4, Wu Y L 5 1Royal Marsden Hospital, London, UK, 2National Taiwan University, Taipei, Taiwan, 3Boehringer Ingelheim GmbH, Ingelheim am Rhein, Germany, 4Boehringer Ingelheim Pharma GmbH & Co. KG, Ingelheim, Germany, 5Guangdong Lung Cancer Institute, Guangzhou, China . . . . . . . . . Objectives: Lung cancer is one of the most common causes of cancer-related deaths world-wide. Afatinib is an irreversible ErbB family blocker showing superior efficacy in EGFRm+ NSCLC as 1st-line treatment compared to standard-of-care chemotherapy. To date, no head-to-head trial results exist to compare afatinib, gefitinib or erlotinib. Previous NMAs have compared the three treatments on an ITT level, but not by mutation type. This analysis attempts to fill this gap. Methods: Based on the EGFRm+ study network requested by NICE a Bayesian approach NMA was conducted to estimate relative treatment effects of afatinib versus erlotinib and gefitinib for progression free survival (PFS) and overall survival (OS) per mutation type (common mutations: Del19 and L858R). Results: 9 studies were included, 8 reported PFS and 5 OS by mutation type, respectively. Results from fixed effects models are reported. Afatinib significantly improved PFS in common mutations versus gefitinib (HR 0.43: CrI 0.24; 0.75) and erlotinib (HR 0.60; CrI 0.39; 0.91). Results also favored afatinib for both EGFR mutation subgroups Del19 and L858R, but did not reach statistical significance. Afatinib showed a high probability (> 80%) of being the best treatment both for common mutations and per mutation type. For OS, a trend favoring afatinib was shown in particular for Del19. The probability of afatinib being best for Del19 was > 70%. For L858R no difference in OS was detected between the TKIs. Conclusions: In line with findings from previous NMAs, this analysis by mutation type confirms both for PFS and OS a consistent trend towards superiority of afatinib versus reversible TKIs. Afatinib appears to be the best treatment option for patients with common mutations, in particular Del19 mutations. A direct trial-based comparison of the efficacy of these agents is warranted to clarify their relative benefits. PCN9 Health Care Costs in Patients Treated with Ipilimumab for Advanced Melanoma Results of a Retrospective Chart Review Tarhini A 1, Rao A S 2, Corman S 3, Botteman M 4, Ji X 4, Mehta S 3, Margolin K 5 1University of Pittsburgh Cancer Institute, Pittsburgh, PA, USA, 2Bristol-Myers Squibb, Plainsboro, NJ, USA, 3Pharmerit International, Bethesda, MD, USA, 4Pharmerit US Bethesda, Bethesda, MD, USA, 5Seattle Cancer Care Alliance, Seattle, WA, USA . . . . . . . . Objectives: This analysis described health care costs over time—excluding ipilimumab drug costs—stratified by survival duration and baseline ECOG status, in patients receiving ipilimumab for advanced melanoma in the US community setting. Methods: We analyzed data from a retrospective chart review of patients with unresectable stage III/IV melanoma treated with ipilimumab as first-line monotherapy between 04/2011 and 09/2012. Hospitalizations, emergency department visits, subsequent chemotherapy, radiation, surgeries, nursing home, and hospice visits costs were estimated using published sources and tariffs. Total costs, excluding ipilimumab drug costs, were calculated for 3 periods: treatment regimen (between first and last ipilimumab doses); post-regimen; and pre-death (within 90 days of death). Monthly costs were compared for the total population and stratified by baseline ECOG status (0 vs. ≥ 1, when available) and survival (< 1 year vs. ≥ 1 year) using Wilcoxon rank sum tests. Results: Data were abstracted from 273 patient charts at 34 sites. Excluding ipilimumab drug costs, total monthly costs during the treatment regimen, post-regimen, and pre-death periods were $690, $2151, and $5123, respectively. Total monthly costs across all study periods were higher for patients with ECOG ≥ 1 (n= 135) vs. ECOG= 0 (n= 104) (p= 0.0294), particularly in the pre-death period ($5987 vs. $3460, respectively; p= 0.0143). A similar pattern was observed for patients surviving < 1 year (n= 109) vs. ≥ 1 year (n= 122) (p< 0.0001), with a difference of $9524 vs. $2955 (p< 0.0001) during the pre-death period (42 patients still alive after < 1 year follow-up were excluded from this analysis). Key cost drivers were hospitalizations (32.4% of total costs), followed by non-ipilimumab chemotherapy (23.1%), hospice care (19.1%), and nursing home stays (12.5%). Conclusions: In this population, monthly costs were significantly lower during the treatment regimen period than in subsequent periods. Survival ≥ 1 year and baseline ECOG= 0 were associated with significantly lower total monthly costs, particularly in the pre-death period. PCN10 Systematic Review of Relapsed or Refractory Mantle Cell Lymphoma (MCL) Clinical Trials: Implications for Decision Modeling Sorensen S 1, Dorman E 1, Xu Y 2, Sallum R 2, Pan F 1, Szatkowski A 2, Gaudig M 3, Sengupta N 4 Pharmaceuticals, Inc, Neuss, Germany, 4Janssen Pharmaceuticals, Inc, Raritan, NJ, USA . . . . . . . . 1Evidera, Bethesda, MD, USA, 2Evidera, Lexington, MA, USA, 3Janssen Objectives: No standard of care exists for patients with relapsed or refractory mantle cell lymphoma (MCL) and treatment options are limited. This study sought to synthesize the clinical evidence of current treatments for relapsed or refractory MCL, its limitations, and discuss the implications for decision making. Methods: A systematic literature review was conducted in MEDLINE of phase II, III, or IV clinical trials published in English between January 1, 2001 and May 2, 2013. Supplemental searches included EHA, AACR, ASCO, and ASH 2011–2013 conference proceedings. Results: Results of the review indicate a paucity of evidence relevant for decision making. Of 808 records reviewed, 17 trials in the R/R MCL population were identified, only three of which were randomized controlled trials; all others were single-arm trials. For most treatments, only one trial was available. These factors made it infeasible to conduct a meta-analysis or indirect comparison. Furthermore, there was a large amount of heterogeneity in the patient populations, trial designs, and reported outcomes, making it difficult to compare outcomes across trials. Finally, of the 17 trials identified, five reported progression-free survival (PFS) Kaplan–Meier (K-M) graphs and only three reported overall survival (OS) A616 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 K-M graphs. Conclusions: Due to very limited availability of trials with robust endpoints and long-term follow-up, alternative options for establishing comparative efficacy must be used for decision making in relapsed or refractory MCL. These alternatives include implementing comparisons of single-arm trial data without adjustment (i.e., via naïve comparison) or methods such as match-adjusted indirect comparison (MAIC) to derive comparative estimates. MAIC is a relatively novel method and may be difficult to implement given the heterogeneity in trial designs and patient-level characteristics in MCL trials. The scarcity of K-M data to inform PFS and OS of certain comparators further limits the comparisons that can be made through modeling. PCN11 The Efficacy of Current Treatment Options for Metastatic Cervical Cancer Palin H J 1, McCormick A L 1, Sabaté E 2 Clarity, Macclesfield, UK, 2F. Hoffmann - La Roche, Basel, Switzerland . . . . . 1Complete Objectives: The prognosis of patients with metastatic cervical cancer (CC) remains poor, and treatment options are limited, with no single agent or combination of agents recognised as standard of care; cisplatin/paclitaxel is the therapy most cited by guidelines. This study aimed to assess the efficacy of reported treatment options for patients with metastatic CC. Methods: Searches of PubMed were conducted, with no date restrictions, to identify published randomised controlled Phase II/III clinical trials (RCTs) of chemotherapies recommended by treatment guidelines, and radiotherapy and/or surgery, that reported overall survival (OS) in patients with metastatic (systemic recurrent, persistent or de novo-metastatic) CC. Treatment guidelines and the Cochrane Library were also explored to identify additional citations. Results: Of 65 articles identified, 10 articles published between 1987 and 2014 proceeded to data extraction. Evidence supporting the use of chemotherapy was limited to cisplatin-monotherapy or platinum-based combination therapy. Overall the OS benefit of these agents ranged from 0.9 to 2.9 months and 0.79 to 1.32 for hazard ratio (HR). The latest innovation, bevacizumab plus chemotherapy, demonstrated the greatest significant gain in OS versus chemotherapy (OS gain 3.7 months; HR 0.71; p= 0.004). The study did not identify any RCTs that supported the use of surgery and/or radiotherapy in this setting; the evidence was limited to seven retrospective hospital based studies. Conclusions: This study highlighted an unmet need for additional treatment options for metastatic CC. Use of cisplatin-monotherapy or platinum-based combination therapy has provided limited survival benefits for many decades. The novel combination of bevacizumab plus chemotherapy has demonstrated an increase in survival in these patients. However, since there is no RCT evidence supporting the use of surgery and/or radiotherapy, a health technology appraisal of these alternative interventions is not currently feasible. Additional clinical research is urgently needed to assess the comparative clinical value of these therapies. PCN12 Comparison of Mean Overall Survival (OS) and Radiographic Progression Free Survival (RPFS) Based on Matching Adjusted Indirect Comparison of Abiraterone Acetate and Enzalutamide for the Treatment of Castration-Resistant Prostate Cancer in Chemotherapy Naïve Patients Dearden L 1, Majer I 2, Heeg B 2, Liwing J 3, Sandstrom K 3, Diels J 4 1Janssen, High Wycombe, UK, 2Pharmerit International, Rotterdam, The Netherlands, 3JanssenCilag AB, Sollentuna, Sweden, 4Janssen Research & Development, Beerse, Belgium . . . . . . Objectives: Abiraterone acetate plus predniso(lo)ne (AA) and enzalutamide (E) are novel therapies for the treatment of metastatic castration-resistant prostate cancer in chemotherapy naïve patients. Pivotal trials have been conducted evaluating the efficacy of the drugs using different comparators. In the COU-AA-302 trial, patients were randomised between AA and active comparator predniso(lo) ne whereas in the PREVAIL trial, E was compared against placebo. For health economic purposes, the mean overall survival (OS) and radiographic progressionfree survival (rPFS) of both novel agents need to be compared in the absence of head-to-head trial data. Methods: Due to the difference in the comparator arms, only survival data from AA and E were used for the comparison. Observed individual level survival data with baseline patient characteristics were available for AA. Individual survival data were simulated for E to replicate the rPFS and OS curves published for the pivotal trial. rPFS and OS were modeled and extrapolated by fitting parametric survival functions. The Weibull, exponential, and lognormal models were evaluated based on statistical and clinical considerations, i.e. assessing the model fit and the implied hazard profiles, respectively. To control for differences in baseline patient characteristics (PSA, ECOG, Gleason score, BPI, LDH, metastasis, age, race) rPFS and OS estimates for AA were adjusted using a matching algorithm. Results: The Weibull models were selected for extrapolation of both OS and rPFS. The mean rPFS was estimated to be 23.9 (95% CI: 21.5-26.3) and 19.5 (95% CI: 16.0-23.9) months for AA and E respectively. Mean OS was estimated to be 38.7 (95% CI: 36.4-40.7) and 34.6 (95% CI: 31.8-37.8) months respectively. Conclusions: Based on currently available data and the presented modeling approach, these findings suggest that AA is associated with longer mean rPFS and OS than E. PCN13 Clinical Effectiveness of Robotic Image-Guided Stereotactic Radiosurgery (CyberKnife) in Selected Primary and Secondary Soft Tissue Neoplasms: A Systematic Review Wlodarczyk A , Ceglowska U , Slomka M , Luchowska K Agency for Health Technology Assessment In Poland (AHTAPol), Warsaw, Poland . . . . Objectives: Effectiveness of radiosurgery for head and spinal neoplasms is established. The aim of this study was to systematically review the clinical literature of CyberKnife for people with selected primary and secondary soft tissue lesions. Methods: A systematic search was conducted for best available clini- cal data for lung metastasis, primary and secondary liver cancer, locally advanced pancreatic cancer population treated with CyberKnife radiation. Searching using Medline, EMBASE, Cochrane Library took place in September 2013. Results: Only one relevant comparative clinical study (matched-pair analysis) met the inclusion criteria, assessing effectiveness and safety of stereotactic radiosurgery and radiofrequency ablation for colorectal liver metastasis. For other neoplasms single-arm studies were found. Compared to RFA CyberKnife for liver metastasis was significantly better in median local disease free survival, which was 34.4 months vs. 6.0 months, (p< 0.001). 1 and 2-year local control rates also favored CK (85.0% vs. 65.0% and 80.0% vs. 61.0%, respectively) but the difference wasn’t significant. However, trend for better OS was found with RFA (34.4 vs. 52.3 months). For lung metastasis, treatment with CK resulted in 24.0-62.0% complete or partial response, 38.0-76.0% patients stabilized. In primary liver tumors OR (CR + PR) was observed in 63.0-86.0% patients, 0.0-29.0% stabilized, median PFS reached 10.0-15.8 months. Inconsistent results were seen in locally advanced pancreatic cancer population. In one study 92.0% responded or stabilized but in other only 1 patient of 77 had PR. Median OS was 6.4-10.3 months. All studies reported mostly mild adverse events after CK. Serious AE were rare. Conclusions: There is limited quality evidence on the effectiveness and safety of robotic image-guided stereotactic radiosurgery in patients with soft tissue neoplasms. Available studies are highly heterogenic in methods, patients characteristics and outcomes but suggest that CyberKnife may be beneficial in local tumor control. There is a need of well-designed comparative studies. PCN14 Analysis of Treatment Options for Relapsed or Refractory Chronic Lymphocytic Leukemia (CLL) Sallum R 1, Dorman E 2, Xu Y 1, Tran-Kerr K 1, O’Donnell M 1, Sorensen S 2, Szatkowski A 1, Sengupta N 3, Gaudig M 4 1Evidera, Lexington, MA, USA, 2Evidera, Bethesda, MD, USA, 3Janssen Pharmaceuticals, Inc, Raritan, NJ, USA, 4Janssen Pharmaceuticals, Inc, Neuss, Germany . . . . . . . . . Objectives: For patients with relapsed or refractory (R/R) chronic lymphocytic leukemia (CLL), treatment outcomes are poor and treatment options are limited. Ibrutinib is an oral, once-a-day, first-in-class covalent inhibitor of Bruton’s tyrosine kinase approved by the Food and Drug Administration (FDA) for R/R CLL. In a recent phase III trial (PCYC-1112), ibrutinib was associated with improved progression-free survival (PFS, hazard ratio [HR] = 0.215) and overall survival (OS, HR= 0.387) versus ofatumumab. The aim of this study is to provide a summary and analysis of results observed with current therapies in high-risk patients with R/R CLL. Methods: A systematic literature review and targeted literature search of clinical trials and international treatment guidelines in PubMed/MEDLINE (January 1,2001–April 28,2013) and ASCO/ASH/EHA conference proceedings (2011–2013) were conducted to identify and evaluate current treatment options for R/R CLL, including alemtuzumab, rituximab, bendamustine, chlorambucil, and ofatumumab. Results: Study results highlight poor outcomes with existing treatment options and continuously high unmet need in patients. Sixteen trials were identified; the majorities were single-arm with small sample sizes, making comparative effectiveness difficult to establish. Time-to-treatment failure was 5.8 months with alemtuzumab, while median PFS was 5.5 months with rituximab, 5.5–5.7 months with ofatumumab, 8 months with chlorambucil-ritiximab, and 15.2 months in previously-treated patients and 6.8 months in previously-treated patients with del (17p) with bendamustine-rituximab. Ofatumumab has demonstrated activity in patients with difficult-to-treat, high-risk CLL and is the only recognized and approved treatment by health authorities globally in this treatment setting and recommended in treatment guidelines. Conclusions: The lack of standard of care creates challenges for defining comparators in clinical trials and health technology assessments. In R/R CLL with high-risk features, ofatumumab is an appropriate comparator. Interim results from the phase III RESONATE trial showed that ibrutinib achieved significantly improved efficacy versus ofatumumab, even in high-risk disease patients. PCN15 An Indirect Treatment Comparison of Cabozantinib Verse Vandetanib in Progressive Medullary Thyroid Cancer (MTC) Rinciog C 1, Myrén K J 2, Aldén M 2, Diamantopoulos A 1, LeReun C 3 Limited, Borehamwood, UK, 2Sobi, Solna, Sweden, 3Independent Biostatistician, Carrigaline county, Ireland . . . . . . 1Symmetron Objectives: MTC is a rare form of thyroid cancer with prevalence of less than 7 per 100,000. A majority of MTC patients have RET mutations, and RET M918T mutations are associated with especially poor prognosis. In 2012, EMA approved the first tyrosine kinase inhibitor (TKI) CAPRELSA® (vandetanib, VDB) for the treatment of MTC. In March 2014, the EMA approved another TKI -COMETRIQ® (cabozantinib, CBZ) for the treatment of adult patients with progressive, unresectable locally advanced or metastatic MTC, with orphan drug status. The objective of this study was to assess the relative efficacy in PFS and OS of CBZ vs VDB. Methods: Since there are no clinical trials directly comparing the two treatments, an adjusted indirect comparison (Bucher et al. method) was used. Evidence on PFS for the two treatments was collected from the pivotal clinical trials in MTC. The analysis considered all patients and a subgroup of RET M918T mutation positive (RET+) patients. Our analysis focused on PFS due to lack of evidence for the VDB OS in the RET M918T mutation subgroup. In the all patients analysis three different scenarios were explored: a logrank model to ensure comparability with the VDB data; a Cox model stratified on age at randomization and prior TKI status; and a Cox model without stratifications. Results: In the subgroup analysis (logrank model) PFS was estimated to increase by 65% with CBZ comparing to VDB (HR 0.35; 95% CI 0.14-0.87). In the all-patients analysis the estimates were less conclusive: logrank model (HR 0.72; 0.40-1.28), Cox model with stratifications (HR 0.61; 0.35-1.04), Cox model without stratifications (HR 0.66; 0.39-1.13). Conclusions: The results showed a positive trend in favour of CBZ in PFS. Given the limited evidence a direct head-to-head comparison is necessary to validate the study findings. A617 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PCN16 A Systematic Literature Review to Identify Trials in First-Line Ras Wild-Type (WT) Metastatic Colorectal Cancer (MCRC) Patients Jarrett J 1, Weijers L 2, Hnoosh A 3, Harty G 3, von Hohnhorst P 4 - HEOR & Strategic Market Access, Houten, The Netherlands, 3Merck Serono Ltd., UK, UK, 4Merck Serono, Darmstadt, Germany . . . . . 1MAPI, London, UK, 2Mapi Objectives: The aim was at assembling published evidence on treatments for RAS wild-type (wt) (KRAS and NRAS exon 2,3,4) metastatic colorectal cancer (mCRC) to evaluate the efficacy and safety of cetuximab in combination with FOLFOX or FOLFIRI chemotherapy as first line treatment of these patients. The outcomes of interest were progression free survival (PFS), overall survival (OS), overall response (ORR), and adverse event data. Methods: A systematic literature search was performed on the 24th March 2014 in Medline, Embase, and The Cochrane Library to identify all randomized controlled trials (RCTs) as well as single arm trials concerning the efficacy and safety of first line interventions of interest in patients with RAS wt mCRC. Abstract and article selection was performed by two independent researchers, with a third person resolving disagreements, according to predefined standards, which were based on criteria for patient, intervention, comparator, outcomes and study design (PICOS). Results: 596 citations were identified after removing duplicates. 520 citations were excluded for reasons including: patient population (231), study design (174), intervention (74), outcomes (23) and duplicates (18). A total of 76 abstracts were included and the full-text version of these publications were retrieved and screened resulting in excluding 69 further publications due to the study design (20), patient population (43), outcomes (5), and one (1) duplicate, not being in line with the review selection criteria. Ten citations on five different RCTs were identified that met the inclusion criteria. Conclusions: This systematic review provided the latest studies on RAS wild-type (wt) metastatic colorectal cancer (mCRC) treatments. Despite the small number of studies available for a relatively new biomarker-specific mCRC, anti- EGFR treatments such as cetuximab in combination with FOLFOX and FOLFIRI have demonstrated a significant benefit in progression free survival (PFS), overall survival (OS) and overall response rate (ORR) versus chemotherapy alone in mCRC RAS wt patients. PCN20 Characteristics of Patients with Pleural Mesothelioma in the Russian Federation Gorbounova V A 1, Bychkov M B 1, Vladimirova L Y 2, Levchenko E V 3, Karpenko T D 1, Mamontov O Y 3, Burmistrov V 4, Novick D 5, Rajan N 6, Sholokhova E 4, Solovyenko Y 4, Tjulandin S A 1 1N.N. Blokhin Russian Cancer Research Center, Moscow, Russia, 2Rostov Scientific Research Oncological Institute, Rostov-on-Don, Russia, 3N.N. Petrov Research Oncology Institute, SaintPetersburg, Russia, 4Eli Lilly Vostok S.A., Moscow, Russia, 5Eli Lilly Holdings Limited, Windlesham, UK, 6Eli Lilly Australia Pty Ltd, West Ryde, NSW, Australia . . . . . . . . . . . . . . . . . . . Objectives: Malignant pleural mesothelioma (MPM) is a rare tumour with a poor survival and prognosis. In Russia, the epidemiology of this disease has not been well studied. We conducted an epidemiological study to understand the characteristics of patients with MPM in Russia. Methods: Retrospective study of the characteristics of patients with MPM. Patients treated for 1st and 2ndline therapy were included in ten centers of the Russian Federation. The data were generated from hospital records, electronic databases, and other sources of information between June and December in 2013. Results: One hundred and twelve patients were enrolled. The average age was 60.0±10.87 years and 42% were women. Harmful work environment was indicated in 21/112 (18.8%) of the patients. The exposure to asbestos was reported in 10/112 (8.9%) patients, and contact with erionite in 8/112 (7.1%) patients. Fifty percent (56/112) of the patients were smokers or are current smokers; smoking period was 25.3±13.25 years. The disease stage at diagnosis was I-II in 31 (27.6%), III-IV in 81 (72.4 %) patients. The ECOG performance status was 0 in 16 (14.3%), 1 in 46 (41.1%), 2 in 30 (26.8%), 3 in 15 (13.4%) and 4 in 1 (0.9%) patients, in 4 patients - unknown. 50/112 (44.6%) patients had immunohistochemical verification of the diagnosis. Antineoplastic drugs were used to treat 85/112 (75.9%) patients. The main drugs used were cisplatin (n= 75), gemcitabine (32), doxorubicin (34), pemetrexed (29) and carboplatin (21). Conclusions: In Russia, the majority of MPM patients were diagnosed at advanced stages of disease. In 80% of the cases, a harmful work environment was not identified. Given the low prevalence and the frequent use of off-label medicines, it can be considered an orphan disease. Limitations: Only limited number of centres was included. PCN21 First-Line Therapy for Patients With Multiple Myeloma: Direct and Indirect Comparison of Treatment Regimens on the Existing Market Kuhr K 1, Wirth D 2, Srivastava K 3, Lehmacher W 1, Hellmich M 1 1University of Cologne, Cologne, Germany, 2Janssen-Cilag, Neuss, Germany, 3Heron India, Chandigarh, India . . . . . Objectives: Motivated by the discussion whether the german AMNOG is applied to currently marketed drugs we compared first-line therapies for patients with multiple myeloma (MM). Methods: A systematic literature search for randomized controlled trials (RCTs) was conducted and VMP (bortezomib (Velcade), melphalan and prednisone), MPT (melphalan, prednisone and thalidomide) and MP (melphalan and prednisone) were identified as therapies of interest. We extracted information on overall survival (OS), progression-free survival (PFS), response criteria (CR, VGPR, PR), and grade 3-4 AEs (any, hematological, non-hematological, DVT, PNP). Random-effects meta-analysis was used for direct and the Bucher method for adjusted indirect treatment comparison. Results: Seven RCTs with a total of 2,367 patients were included in our analyses, one RCT (n= 682) comparing VMP vs. MP and six RCTs (n= 1,685) comparing MPT vs. MP. Direct head-to-head comparison of VMP vs. MPT was lacking. For MPT vs. MP, data were extracted from a recently published meta-analysis of individual patient data if available. VMP was superior to MP regarding OS. Both VMP and MPT were superior to MP regarding PFS and response criteria, but had a higher risk of developing AEs. The indirect comparison of VMP vs. MPT via MP showed a statistically not significant advantage for VMP regarding survival outcomes. Significant benefits were observed for CR and development of any grade 3-4 AEs favouring VMP. Conclusions: Analysis of both aggregated and individual patient data essentially lead to the same conclusions, i.e. VMP and MPT seem more effective than MP, VMP seems ahead of MPT regarding response criteria and adverse events. We found significant between-trials heterogeneity, however no consistent relationship of effect and study-level covariates (e.g. maintenance dosing) was apparent. Thus, we relied on the random effects approach to metaanalysis to cope with the unexplained trial-to-trial variability. Our results may best be confirmed by a head-to-head trial of VMP vs. MPT. PCN22 What is the Clinical Effectiveness and Cost- Effectiveness of Erythropoietin-Stimulating Agents for the Treatment of Patients with Cancer-Treatment Induced Anaemia? Insights from Cumulative Meta-Analyses (CMA) and Lessons for Cost-Effectiveness Analyses Huxley N , Haasova M , Crathorne L , Hyde C University of Exeter, Exeter, UK . . . . Objectives: A health technology assessment (HTA) informing the recent NICE guidance regarding the use of erythropoiesis-stimulating agents (ESA) in cancertreatment induced anaemia (CIA) identified uncertainty around the overall survival hazard ratio (OSHR). We investigated how the understanding of OS in CIA patients treated with ESAs has shaped over time and the effects of accumulating OS evidence on cost-effectiveness. In addition, the effects of narrowing inclusion criteria, by comparing the HTA results to a recent Cochrane review, were investigated. Methods: CMA was applied to both HTA review and Cochrane review OS data to identify patterns in results; study results were accumulated by the year of publication. Annual OSHR results from the CMA were applied to an economic model developed in the HTA to calculate the cost-effectiveness of ESAs. Results: Precision of the OSHR estimate appeared to improve with additional evidence, but the true location of the estimate remained uncertain and the best estimate varied over time. Using the HTA CMA, results from 2001 and 2002 suggested survival benefits to using ESAs (0.77, 95% CI 0.60–0.98 and 0.78, 95% CI 0.65–0.93 respectively), with ESAs being cost-effective at a willingness to pay threshold of £30,000 per QALY for all values of the OSHR 95% CI. HTA CMA for all other years and all Cochrane CMA results did not suggest any significant effects of ESAs on OS. Cost-effectiveness results were therefore uncertain. Conclusions: Current evidence suggests we cannot reject the possibility of no difference in OS between patients receiving or not receiving ESAs, regardless of study inclusion criteria. However, there is also insufficient evidence to support such conclusions, particularly as earlier results from narrower inclusion criteria suggested some survival benefits. This analysis highlights the additional uncertainty of the current evidence base on cost-effectiveness analyses, which cannot be captured in standard sensitivity analyses. PCN23 Bortezomib Re-treatment in Patients with Multiple Myeloma (MM). A Real World Medical Practice Experience from a Swedish National Registry Thilakarathne P 1, Diels J 2, van Sanden S 3, Liwing J 4, Chirita O 5, Van Agthoven M 6, Nahi H 7 1Janssen Pharmaceutica N.V., Beerse, Belgium, 2Janssen Research & Development, Beerse, Belgium, 3EMEA HEMAR Analytics, Janssen EMEA, Beerse, Belgium, 4Janssen-Cilag AB, Sollentuna, Sweden, 5Janssen, High Wycombe, UK, 6Janssen-Cilag BV, Tilburg, The Netherlands, 7Karolinska University Hospital, Solna, Sweden . . . . . . . Objectives: The question of sequencing multiple myeloma (MM) treatments is a key one, as is the retreatment in patients where the same treatment was effective in earlier lines. We investigated bortezomib retreatment in a Swedish MM national registry. Methods: Patients diagnosed with MM since January 2000 until June 2011 from 7 university clinics, 5 regional centers and 3 local hospitals in Sweden were included. Time to response and overall survival (OS) were analyzed using stratified Kaplan-Meier analysis. Results: Of the 541 patients treated with bortezomib (out of a total population of n= 1638), 93 were retreated with bortezomib. Median followup from start of retreatment was 10.2 mos. Median age was 63.5 (range 38–83), 57.3% were male, 34.2%/15.9%/20.7% had stage I/II/III disease (ISS); median number of prior therapies at initial bortezomib and retreatment was 1 and 3. 26.8%/32.9% of pts initiated retreatment as 3rd/4th line therapy. 37.8%/22.0% initiated bortezomib retreatment in combination with dexamethasone, 7.3% in monotherapy, compared to 40.3%, 21.8% and 8.5%, at initial bortezomib. ≥PR/VGPR-rates at re-treatment were 59.1%/12.7%, compared to 82.0%/39.9% at initial bortezomib. Median time to ≥ PR/ VGPR was 2.4/1.3 months at retreatment versus 1.9/2.1 at initial bortezomib. The ≥ PR rate at retreatment was numerically longer in patients with < = 2 (75.1%) vs > = 3 (20%) therapies prior to retreatment. Median PFS/OS from start of re-treatment was 5.5 [95%CI: 3.7, 10.0]/17.4 [10.3, 26] months, compared to 8.1 [6.9, 9.7]/25.9 months [21.0, 31.6] for initial bortezomib. Number of prior therapies at retreatment did not affect PFS; however, OS was longer in pts with fewer prior therapies (p= 0.0112). There was a trend towards longer PFS (p= 0.087) in retreated patients who achieved > = PR compared to non-responders. Conclusions: These data suggest that, in everyday medical practice, bortezomib retreatment is effective in relapsed/refractory MM, with more than half of pts who responded to initial bortezomib achieving ≥ PR at retreatment. PCN24 Treatment Sequencing Survival Model for Patients with Multiple Myeloma Ineligible for Stem Cell Transplantation (SCT) Heeg B 1, Van Agthoven M 2, van Beurden-Tan C 3, Liwing J 4, Mellqvist U H 5, Plesner T 6, Logman F 7, Aschan J 4, Einsele H 8, Treur M 1, Barendse M 4, Richardson P G 9, Palumbo A 10, Nahi H 11, Sonneveld P3 1Pharmerit International, Rotterdam, The Netherlands, 2Janssen-Cilag BV, Tilburg, The Netherlands, 3Erasmus University Medical Center, Rotterdam, The Netherlands, 4Janssen-Cilag . . . . . . . . . . . . . . . . A618 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 AB, Sollentuna, Sweden, 5Department of Hematology, Gothenburg, Sweden, 6Vejle Hospital, Vejle, Denmark, 7Eastbridge Pharma BV, Utrecht, The Netherlands, 8Medizinische Klinik II, Würzburg, Germany, 9Dana-Farber Cancer Institute, Boston, MA, USA, 10University of Torino, Torino, Italy, 11Karolinska University Hospital, Solna, Sweden Objectives: Treatment of multiple myeloma (MM) consists of a consecutive application of various treatment modalities. The optimal sequence of different treatment combinations is unclear. There is a need for the establishment of an analytic framework comparing overall survival (OS) of different treatment sequences in MM. This analysis contains the first attempt to create such a framework. Methods: We performed a systematic literature review (SLR) on treatments compared in randomized controlled trials in newly diagnosed and relapsed MM patients. The following treatment/combinations were considered; melphalan/prednisone (MP), dexamethasone (D), thalidomide (T), bortezomib (V) and lenalidomide (R). Response, time to next treatment (TTNT) and OS data were extracted. Response rates were derived and pooled for each treatment and treatment line using ordered logistic network meta-analysis. The pooled estimates were combined in a Markov model framework with treatment switch and mortality transition probabilities to predict OS. Seventeen sequences were tested, one starting with MPR and four for each starting with MP, MPT, MPV and MPVT. Results: Estimations on the basis of the Markov model showed that median OS varied between 40-43,42-45,43,44-47 and 46-49 months for sequences starting with MP, MPT, MPR, MPV and MPVT, respectively. Conclusions: This study shows a first attempt to build an analytic framework comparing OS of treatment sequences in MM. The absolute results should be interpreted with caution, as MM is a heterogeneous disease and the model does not yet capture all subtleties related to sequential treatment in clinical practice. However, face validity of these first results seem promising and warrant further development of this analytic framework. PCN25 Burden of Disease in Stage IV Non-Small Cell Lung Cancer: Comparing ALK Positive and ALK Negative Mutation Status in Egypt, Russia and the United Arab Emirates Fisher M 1, De Renteria J 2, Kuehl M 3, Suponcic S 4, Dharamshi K 1 Health, Munich, Germany, 3Kantar Health, Epsom, UK, 4Kantar Health, New York, NY, USA . . . . . 1FIECON, London, UK, 2Kantar Objectives: The effect of having a positive anaplastic lymphoma receptor tyrosine kinase (ALK) mutation on the burden of disease in Stage IV non-small cell lung cancer (NSCLC) is unknown. This study compared the burden of disease in ALK positive Stage IV NSCLC with ALK negative Stage IV NSCLC. Methods: Burden of disease by ALK mutation status was evaluated in Egypt, Russia, and the United Arab Emirates (UAE) using disability adjusted life years (DALYs). A burden of disease model was developed in line with World Health Organisation (WHO) guidelines. Epidemiological data were collected by a survey disseminated during interviews in at least three centres per country. Disability weights were obtained from a Dutch study. Discounting, age-weighting and life expectancies were based on the WHO reference case. Sensitivity analyses were conducted to test the level of uncertainty in the model results. Results: Total DALYs across the countries were lower in ALK positive disease compared to ALK negative disease (833 vs. 2611). On a per country basis, total DALYs were higher in ALK positive disease compared to ALK negative disease for Egypt (22 vs. 11) and UAE (229 vs. 219) but lower for Russia (582 vs. 2381). DALYs per incident case of the disease were consistently higher across the countries in ALK positive disease compared to ALK negative disease (Egypt: 11.24 vs. 10.85; Russia: 9.38 vs. 4.97; UAE: 15.29 vs. 7.82). Results were sensitive to changes in discounting, age-weighting, deaths, and incidences. Conclusions: The study would suggest that the greatest burden of disease lies in the population with ALK negative Stage IV NSCLC compared to ALK positive Stage IV NSCLC, although this finding varied between countries. However, it was consistently found that a single patient with Stage IV NSCLC will have a greater burden of disease if their ALK mutation test is positive. PCN26 Real-World Data on the Epidemiology and Treatment of HR +, HER2Advanced Breast Cancer in Postmenopausal Patients in Different Regions of Russia for Forming Markov Models of Managment of Patients Krasnova L , Vorobiev P , Holownia M Russian Society for Pharmacoeconomics and Outcomes Research, Moscow, Russia . . . Objectives: In Russia, breast cancer (BC) took first place in the structure of cancer incidence and mortality in the female population. Modern therapeutic approaches advanced breast cancer (ABC) can achieve clinically significant regression of symptoms, prolong life and improve its quality. Aim of this study was to conduct clinical and epidemiological analysis of reference postmenopausal patients with hormone(HR+), HER2-ABC in Russia. Methods: Survey of experts from different regions of Russia. The questionnaire included the following questions: the proportion of patients with different types of breast cancer; drugs and modes of their appointment as the first, second and third-line treatment of patients postmenopausal HR+, HER2-ABC; hospitalization rate per year; laboratory and instrumental methods of research, consulting, adverse events with the use of drugs in the treatment of patients and their drug regimes correction. Results: From 13 regions of Russia 7 experts refused to provide information on the questionnaire, which may indicate a reluctance to disclose information on the epidemiology and the tactics of the disease. Data were obtained from the regions - Moscow, Saint Petersburg, Republic of Khakassia, Omsk region, Primorsky Krai, Krasnoyarsk Krai. Share of the growth of new breast cancer patients is 10,5-29,6% (median 13.4%) per year. Number of patients with ABC from patients with breast cancer is 9,2-53% (median-30%). On average, 70% of patients with ABC are in menopause. Number of patients with HR+, ER2-ABC at the end of 2013 amounted to Moscow-1772, St. Petersburg-492, Republic of Khakassia-188, Omsk region-344, Primorsky Krai-358, Krasnoyarsk Krai-3471. The average frequency of hospitalizations in these patients is 6 times a year. Proportion of these patients receiving 3-line therapy differed significantly in Moscow and St. Petersburg, as compared with other regions. Conclusions: Obtained clinical and epidemiological data made it possible to fill in the model of management of patients with breast cancer in particular regions of Russia. PCN27 Projecting Health Gains: The Population Impact of New Treatments to Metastatic Breast Cancer Lucas R 1, Andrade S 2, Barros H 1 1Medical School, University of Porto, Porto, Portugal, 2Roche Farmaceutica Quimica, Amadora, Portugal . . . Objectives: Breast cancer is a leading cause of oncological death among Portuguese women but a decline in age-standardized mortality rates is well documented and due to an increasing frequency of early diagnosis (more than 85% of the women aged 45-69 years, underwent a screening mammography according to the most recent national health survey) and access to more efficient treatments, including adjuvant and neoadjuvant chemotherapy. However, certain tumor characteristics, such as Human epidermal growth factor receptor 2 (HER2) -positive breast cancer, are associated with poor prognosis for patients with both early and metastatic breast cancer. Methods: Previous studies showed that trastuzumab was expected to prevent metastases in about 700 women during 10 years if used in early breast cancer, in Portugal. T-DM1 significantly prolonged progression-free (3 m) and overall survival (6 m) compared to lapatinib plus capecitabine in patients with HER2positive advanced breast cancer previously treated with trastuzumab and a taxane. Also, the combination of pertuzumab, trastuzumab, docetaxel, as compared with placebo, trastuzumab, docetaxel, when used as first-line treatment for HER2-positive metastatic breast cancer, significantly prolonged progression-free survival. These trials data and population epidemiological information is used to fit predictive models adjusting for available confounders. Results: Although high quality data on patients with HER2-positive metastatic disease regarding age and other potential health differences is lacking, particularly in Portugal, we modeled population measures of frequency and applied data describing the clinical epidemiology of metastatic breast cancer to anticipate the potential impact of these treatment options on the quality and the quantity of life of affected individuals, taking into consideration the period 2014-24. Additionally, by contrasting the available alternatives, we will present an estimation of the averted expenses regarding treatment options for the same period. Conclusions: The study provides information on health gains and evidence for decision making. PCN28 Long-Term Trends in Breast Cancer Epidemiology in the Slovak Republic Ondrusova M , Psenkova M Pharm-In Ltd, Bratislava, Slovak Republic . . Objectives: The latest available published oncology national data from the Slovak Republic (SR) are for year 2008. The objective of this analysis was to evaluate longterm national trends of breast cancer (BC) incidence, mortality, clinical stages and prevalence to obtain more actual data for the purpose of cost-of-illness studies and budget impact analysis in the SR. Methods: Time-trends of national data on incidence, mortality, clinical stages and prevalence of BC in 1978-2008 were analyzed by using joint-point regression (Version 4.0.4. - May 2013). The data were predicted to the current year 2014. The trends are presented with a corresponding 95% Confidence Intervals (CI) and p-value with null hypothesis being constant with time. Results: In 2008, BC age-standardized incidence in the SR represented 59.8/100,000 females (n= 2.639 cases, mean age of women= 61.9 y.), in 1978-2008 the incidence was increasing continuously with annual percentage change (APC) of 2.2% (95% CI= 2.1-2.3, p< 0.0001), estimation for 2014 is 63.8/100,000 (n= 3.081). Mortality in 2008 represented 15.2/100,000 (n= 751) with four join-points and decreasing trend from 2001 by APC -3.6% (95% CI= -7.2-0.1, p< 0.04), estimation for 2014 is 14.0/100,000 (n=820). National prevalence in 1978-2008 represents 22.118 cases, for 2014 estimate is 29.012. According to the national data, in 2014 we estimate 32.6% of patients diagnosed in the 1st clinical stage, 36.2% in the 2nd, 16.4% in the 3rd, 7.0% in the 4th and 7.8% in the undefined clinical stage. The number of cases diagnosed in the 1st clinical stage is increasing during the time. Despite the observed increase in the values of BC incidence, the decrease in mortality has manifested itself in recent years. Conclusions: Actual data on incidence, mortality, prevalence and clinical stages of BC in the SR can be used as the source for setting the size of population in cost-of-illness studies and budget impact analysis. PCN29 The Importance of Long-Term Surveillance of Stage IB Melanomas: Unexpectedly Low Survival Subsequent to Recurrence Leeneman B 1, Franken M G 1, Blommestein H M 1, van Gils C W M 2, van der Meijde E 3, Wouters M W J M 4, Plaisier P W 5, Kruit W H J 6, van Ruth S 7, ten Tije A J 8, Hendriks M P 9, Coupe V M H 3, Uyl- de Groot CA1 1Erasmus University, Rotterdam, The Netherlands, 2GlaxoSmithKline, Zeist, The Netherlands, 3VU University Medical Center, Amsterdam, The Netherlands, 4Netherlands Cancer Institute–Antoni van Leeuwenhoek, Amsterdam, The Netherlands, 5Albert Schweitzer Hospital, Dordrecht, The Netherlands, 6Erasmus Medical Center, Rotterdam, The Netherlands, 7Tergooi Hospital, Hilversum, The Netherlands, 8Amphia Hospital, Breda, The Netherlands, 9Medical Center Alkmaar, Alkmaar, The Netherlands . . . . . . . . . . . . . . . . . . . . . . . . . . Objectives: Stage IB cutaneous melanomas are characterized by a low risk for recurrence and a high survival. However, empirical evidence on recurrence patterns and survival subsequent to recurrence is limited. We investigated overall survival (OS) and recurrence patterns, and assessed whether the time-to-recurrence was associated with prognostic factors. Methods: The Dutch Comprehensive Cancer Centers registry was used to identify patients diagnosed with stage IB melanoma between January 2003 and December 2011 in six Dutch hospitals including two academic and four general hospitals (n= 971). Data were retrospectively collected A619 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 using electronic hospital records. Survival was assessed using the Kaplan-Meier estimator. Recurrence patterns were investigated by type of first recurrence and time-to-recurrence. A multivariate cox regression was used to analyze whether time-to-recurrence was associated with gender, age and tumor thickness. Emigrated patients (n= 10) and patients with an unknown recurrence status (n= 144) were excluded. Results: Of all 817 patients, 111 patients (13.6%) experienced disease progression (median follow-up: 5.5 years). Patients who developed a recurrence had a lower survival compared to patients who did not developed a recurrence (median OS: 9.4 years versus median has not yet been reached; 5-year survival rate: 69.9% versus 96.6%; p< 0.001). The most frequent type of first recurrence was lymphatic (36.9%), followed by distant (22.5%), local (21.6%) and intralymphatic (9.9%), respectively. The median time-to-recurrence has not yet been reached; however, in case of a recurrence, the median time-to-recurrence was 2.5 years (minimum: 0.01 years; maximum: 9.8 years). The time-to-recurrence was not statistically significantly associated with gender (HR= 0.81; p= 0.29), age (HR= 1.01; p= 0.38) and tumor thickness (HR= 1.03; p= 0.76). Conclusions: Long-term surveillance of stage IB melanomas is of utmost importance, because survival subsequent to recurrence is much lower than expected. The risk of developing a recurrence was substantial; however, the time-to-recurrence was not associated with gender, age and tumor thickness. PCN30 Epidemiology of Patients with Metastatic Castrate Resistant Prostate Cancer in Europe and Australia Marteau F 1, Gimonet G 1, Gabriel S 2, Dinet J 1, Flinois A 3, LE Cleac’h J Y 3 Pharma, Boulogne-Billancourt, France, 2IPSEN Pharma, Boulogne Billancourt, France, 3Kantar Health, Paris, France . . . . . . . 1IPSEN Objectives: The objective of this study was to evaluate both the incidence of metastatic Castrate Resistant Prostate Cancer (mCRPC) and the number of mCRPC patients who receive specific mCRPC treatments (mCRPCTT): chemotherapy and second generation Hormone Therapies (ADT manipulations were not included). Methods: This study was conducted in 8 European countries and Australia The incidence of mCRPC patients was assessed using several sources: national cancer registries, a literature review and an ad-hoc chart review where 292 oncologists, 76 onco-radiotherapists and 357 urologists reported information about 4171 prostate cancer patients. Of these, 2401 had metastatic castrate resistant disease. Patient characteristics and treatments received were assessed and reported separately by country. Results: Across all 9 countries, 76 200 new patients were diagnosed mCRPC over the past year. Of these patients, 35% (26 400 patients) went to supportive care without receiving any mCRPC TT while 65% (49 800 patients) received a 1L mCRPCTT. Prior to receiving any 1L mCRPCTT, 43% of patients had ADT manipulations during a short transitional period (median duration = 1 month). Of the 49 800 patients who received a 1L mCRPCTT, 59% (29 250) went to a 2L mCRPCTT, 15% deceased during or just after the 1L TT and 26% went on to receive supportive care only. Of the 29 250 patients who received a 2L mCRPCTT, 46% (13 500) went to a 3L mCRPCTT. Conclusions: Our methodology enabled us to assess incidence figures and the volume of mCRPC patients who receive specific mCRPCTT: over one-third of mCRPC patients did not receive any mCRPCTT. Among the 65% who received a 1L TT, 59% receive a 2L mCRPCTT. PCN31 Comparison of Epidemiology and Drug Treatment in HER2 Negative Metastatic Breast Cancer (MBC) in EU5 Nersesyan K 1, Robinson D 1, Pomerantz D 2 Health, St Louis, MO, USA, 2Kantar Health, New York, NY, USA . . . 1Kantar Objectives: Explore differences/similarities in epidemiology and drug treatment of metastatic breast cancer (MBC) in EU5. Methods: All data was derived from the Kantar Health CancerMPact database, sources for which include country specific cancer registries, published scientific studies and proprietary physician surveys comprising 85 doctors seeing 9,255 patients per month. Age and gender specific incidence rates, annual stage specific progression rates and annual stage specific survival rates are used to calculate total number of surviving patients at a specific stage up to 10 years after diagnosis. Results: Prevalence of BC ranged between 41-73 per 100K population across EU-5. Among BC patients, prevalence of MBC was similar: 8% (UK, Italy) to 10% (Germany, France, Spain). Overall 62% of MBC patients were diagnosed with HER2-negative disease (56% Germany-70% France). Among these patients 35-40% had active disease and were treated with chemotherapy. Patients with triple negative disease had had fewer lines of treatment than did not triple negative. Patients who are HER2-negative generally receive between two and three lines of chemo therapy on average. Second line chemotherapy regimens varied. Capecitabine was the most common therapy (mono and combination) in all countries ranging from 36% (UK) to 40% (Germany). 2nd and 3rdmost common therapies were vinorelbine (23%-26%) and paclitaxel (20%-23%) in Germany, Italy and Spain vs. docetaxel (18%-34%) and paclitaxel (11%-19%) in UK and France. In third line, the most commonly used agents were capecitabine (16%-44%) and vinorelbine (18%-26%). Eribulin was used in second line (3%-6%) and third line (11%19%) in all countries except for Spain. Conclusions: Capecitabine is the most utilized chemotherapeutic agent in the second and third lines chemotherapy in Western Europe for HER2-negative patients. A variety of other regimens, primarily monotherapies, may also be used in later lines, including vinorelbine, gemcitabine, eribulin, and docetaxel. PCN32 Association of Diabetes and Cancer Diagnosis in Primary Care Practices in France Grandfils N , Ricarte C , Solomiac A IMS Health, PARIS LA DEFENSE, France . . . Objectives: Several studies suggest that diabetes carries an increased risk for a number of different cancer types. The aim of this study was to investigate the incidence of 14 different cancer types in the diabetic and non-diabetic popula- tion. Methods: IMS Disease Analyzer™ (DA) database was used, focusing on patients with or without diabetes in general practice in France. The analysis was performed retrospectively from 2001 to 2013. The Hazard Ratios (HR; Cox regression) for the risk of cancer in diabetic versus non-diabetic patients were adjusted for demographic and clinical variables. IMS Oncology Analyzer™ (OA) database was also used to describe the cancer patient profile in 2013. Results: Overall 3.1% of patients in DA are diabetic. 75,104 patients were diagnosed with type-2 Diabetes Mellitus (T2DM). This population was matched to a control group in terms of age (60±15 years) and gender (male: 52%). The overall risk of cancer was lower in T2DM patients than in patients without diabetes (HR: 0.81; CI: 0.76 to 0.85). Those patients were less likely to develop breast, prostate, lung cancer or non-hodgkin lymphoma. Nevertheless, their risk was significantly higher regarding liver cancer (HR: 1.99; CI: 1.46 to 2.72), pancreatic cancer (HR: 2.13; CI: 1.51 to 3.00) and endometrial cancer (HR: 1.25; CI: 1.01 to 1.55). No significant increase in risk was observed in colorectal, stomach, kidney, thyroid, urinary bladder, gall bladder and oesophageal cancer. OA showed a higher proportion of diabetic patients among the cancer population (14.6%, n= 10,621). Consistently, this rate was even higher in patients diagnosed with pancreatic (28.7%, n= 362), liver (27.0%, n= 256) and endometrial cancer (21.7%, n=161). Conclusions: This retrospective analysis showed that T2DM may increase the risk of certain cancer types but seems to prevent from some others. Further research is required to evaluate the factors involving the diabetes-cancer correlations, such as the anti-diabetic drugs. PCN33 Patient Count Projections for Advanced Melanoma by Line of Therapy and Other Clinical Characteristics in EU Countries: Results from The UK, Germany, France, Italy and Spain (EU-5) Gueron B 1, Kish J K 2, O’Day K 2, Martel M J 2, Manley Daumont M 1 1Bristol-Myers Squibb, Rueil Malmasion, France, 2Xcenda, LLC, Palm Harbor, FL, USA . . . . . . . Objectives: To forecast the number of advanced melanoma (AM) patients (Stage III unresectable and Stage IV) newly initiating treatment over 5 years (2014-2018) by line of therapy and clinical/tumor characteristics including BRAF/PD-L1 mutations status and rate of brain metastases. Methods: A patient count model was developed to forecast the AM population using historical rate data (1991-2012) and other population parameters including incidence rate annual percent change, stage at diagnosis, rates of disease progression and survival obtained through a comprehensive literature review and hand-search of cancer registry websites. Analysis of a cross-sectional sample (Ipsos Global Oncology Monitor) of 1,297 patients in the EU-5 was used to address any clinical data gaps. The model was validated by comparing projected 5-year prevalence rates to GLOBOCAN 2012 estimates. Results: The model-projected number (rounded to nearest 100) of incident melanoma cases for 2014 was: Germany= 23,100; UK= 18,900; France= 12,400; Italy= 12,000; Spain= 5,800. Of incident cases, 11.3%-13.0% were treatment eligible AM. Incidence rates increases of 1.7-7.8% per year were applied based on historical trends. Analysis of IPSOS data and review of the literature showed BRAF and PD-L1 prevalence rates of 45.4%-56.2% and 15.9%-16.7%, in AM patients, respectively. Literature-derived, brain metastasis prevalence ranged from 15.9-36.0% in Stage IV patients. Considering case progression, resection and adjuvant treatment rates, the forecasted number of AM patients eligible for 1stand 2nd line treatment in 2018 is, respectively: Germany= 3,700 and 1,700; UK= 3,100 and 1,400 France= 1,900 and 500; Italy= 1,800 and 1,000; Spain= 1,100 and 400, representing approximately 10.8-12.0% of incident cases. Conclusions: While melanoma incidence is projected to increase over the next 5 years the majority of incident cases will be diagnosed in earlier disease stages. Under these assumptions, the largest proportion of the incident melanoma population that is AM patients initiating treatment is expected to be 12% in 2018, a slight decline from 13% in 2014. PCN34 A Validated Prediction Model and Nomogram for Risk of Recurrence in Early Breast Cancer Patients Witteveen A 1, Vliegen I M H 2, Siesling S 3, IJzerman M J 4 1University of Twente, MIRA Institute for Biomedical Technology & Technical Medicine, Enschede, The Netherlands, 2University of Twente, Centre for Healthcare Operations Improvement and Research (CHOIR), Enschede, The Netherlands, 3Comprehensive Cancer Centre the Netherlands (IKNL), Utrecht, The Netherlands, 4MIRA Institute for Biomedical Technology & Technical Medicine and University of Twente, Enschede, The Netherlands . . . . . . . Objectives: The objective of this study is to develop and validate a conditional logistic regression model for the prediction of locoregional recurrence (LRR) of breast cancer. To make a translation to clinical practice a web based nomogram was made. Methods: Women first diagnosed with early breast cancer (without distant metastasis or ingrowth in the chest wall or skin) between 2003-2006 were selected from the Netherlands Cancer Registry (n= 39,929). Risk factors for LRRs within five year of the primary treatment were determined using logistic regression. Risks were determined per year, conditional on not being diagnosed with recurrence in the previous year. The presence of interaction and collinearity in the nomogram was assessed, as well as the discrimination by means of the area under the ROC curve and calibration by the Hosmer-Lemeshow goodness-of-fit test in deciles. Data on primary tumours diagnosed between 2007-2008 from a selection of Dutch hospitals was used for external validation of the performance of the nomogram (n= 13,792). Results: The final model included the variables grade, size, multifocality, and nodal involvement of the primary tumour, type of surgery, and whether patients were treated with radio-, chemo- or hormone therapy. The modelling group showed an area under the ROC curve of 0.82,0. 74,0. 67,0. 70 and 0.60 respectively per subsequent year after primary treatment. The calibration was sufficient. All effects in the validation group were in the same direction, and the estimates in the validation group did not differ significantly from the modelling group. Conclusions: This validated nomogram can be used as an instrument to aid clinical decision-making and to identify patients with a high risk of breast cancer recurrence who might benefit from a more intensive follow-up after breast cancer. A620 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PCN35 Survival After Locoregional Recurrence or Second Primary Breast Cancer: Impact of the Disease-Free Interval Witteveen A 1, Kwast A B G 2, Sonke G 3, IJzerman M J 4, Siesling S 2 1University of Twente, MIRA Institute for Biomedical Technology & Technical Medicine, Enschede, The Netherlands, 2Comprehensive Cancer Centre the Netherlands (IKNL), Utrecht, The Netherlands, 3Netherlands Cancer Institute (NKI), Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands, 4MIRA Institute for Biomedical Technology & Technical Medicine and University of Twente, Enschede, The Netherlands . . . . . . . . Objectives: The association between the disease-free interval (DFI) and survival after a locoregional recurrence (LRR) or second primary (SP) breast cancer remains uncertain. The objective of this study is to clarify this association to obtain more information on expected prognosis. Methods: Women first diagnosed with early breast cancer between 2003-2006 were selected from the Netherlands Cancer Registry. LRRs and SP tumours within five years of first diagnosis were examined. The five-year period was subsequently divided into three equal intervals. Prognostic significance of the DFI on survival after a LRR or SP tumour was determined using Kaplan-Meier estimates and multivariable Cox regression analysis. Follow-up was complete until January 1,2013. Results: A total of 36,255 women was included in the analysis. LRRs or SP tumours were diagnosed in 1,646 (4.5%) patients: 55% developed a LRR and 45% SP breast cancer. Longer DFI was strongly and independently related to an improved survival after a LRR (long versus short: HR 0.63, 95% CI 0.460.86; medium versus short HR 0.80, 95% CI 0.64-1.00; P for trend 0.01). Other factors related to improved survival after LRR were younger age (< 70 years) and surgical removal of the recurrence. No significant association was found between DFI and survival after SP tumours. Conclusions: This is the first study to explore the association between the DFI and survival after recurrence in a nationwide populationbased cancer registry. The DFI before a LRR is an independent prognostic factor for survival, with a longer DFI predicting better prognosis. PCN36 Long Term Survival of Patients with Various Lung Cancer Histology in Seer Between 2004-2011 Schmaus K1, Benedict A2 Francisco, CA, USA, 2Evidera, Budapest, Hungary 1Evidera, San Objectives: Overall survival (OS) data from clinical trials in oncology are often incomplete, thus modelling over the lifetime horizon requires long term extrapolation and it is a critical input to cost-effectiveness studies. Data from the Surveillance, Epidemiology, and End Results (SEER) program may provide good validation on the long term OS. The objective was to examine the parametric functions that best fit data in lung cancer (LC) of various histologies in SEER. Methods: SEER data (2004-2011) were analyzed for patients diagnosed with stage IV small cell, large cell, squamous cell carcinoma and adenocarcinoma of the lung with complete followup. Mean age was 68.03 (sd 11.67) and 55.5% were males, with varying baseline age and gender distribution by histology. Treatment status could not be established. Parametric models for OS were fitted using exponential, Gompertz, loglogistic, lognormal, and Weibull distributions. Models were fitted with and without covariates. Fits were inspected and compared graphically using survival and quantile-quantile plots, and statistically using the Akaike Information Criterion (AIC). Modelled mean life expectancy results were compared to the restricted mean life expectancy of the Kaplan-Meier estimator. Results: The lognormal distribution was found to have the best fit within the SEER population, both with and without covariates indicating that a small proportion of patients survive for a long time despite the poor general prognosis of any type of LC. Loglogistic and gamma distributions were 2nd and 3rd best, followed by Weibull, Gompertz and exponential, for all histologies. The last three fitted the data poorly, and underestimated mean life expectancy. Conclusions: Only small proportions of LC patients are alive at 5-8 years, nevertheless the mean OS estimates are impacted by the choice of survival function. The lognormal distribution fit best across all histologies indicating a higher propotion of patients alive than estimated with Weibull models. PCN37 Impact of Hospital Volume on Breast Cancer Outcome: A Population Based Study in the Netherlands Siesling S 1, Tjan-Heijnen V 2, de Roos M 3, Snel Y 4, van Dalen T 5, Wouters M 6, Struikmans H 7, van der Hoeven K 7, Maduro J 8, Visser O 9 1Comprehensive Cancer Centre the Netherlands (IKNL), Utrecht, The Netherlands, 2Maastricht University Medical Center, Maastricht, The Netherlands, 3Rivierenland Hospital, Tiel, The Netherlands, 4Association of General Hospitals, Utrecht, The Netherlands, 5Diakonessen Hospital, Utrecht, The Netherlands, 6Netherlands Cancer Institute, Amsterdam, The Netherlands, 7Leiden University Medical Centre, Leiden, The Netherlands, 8University Medical Centre Groningen, Groningen, The Netherlands, 9Comprehensive Cancer Centre the Netherlands, Utrecht, The Netherlands . . . . . . . . . . Objectives: For low-volume tumours, high surgical hospital volume is associated with better survival. For high volume tumours like breast cancer this association is unclear. The aim of this study is to determine to what extent the yearly surgical hospital breast cancer volume is associated with overall survival. Methods: All patients, diagnosed with primary invasive non-metastatic breast cancer in the period 2001-2005, were selected from the Netherlands Cancer Registry. Hospitals were grouped by their annual volume of surgery for invasive breast cancer. Cox proportional hazard models were used including patient and tumour characteristics as covariates. Follow-up was completed until the 1thof February 2013. Primary endpoint was 10-year overall survival rate. Results: In total 58,982 patients with invasive non-metastatic breast cancer were diagnosed during the period 2001-2005. Hospitals were grouped by their (mean) annually surgical volume: < 75 (n= 19), 75-99 (n=30), 100-149 (n= 29), 150-199 (n= 9) and ≥200 (n=14). The 10-year observed survival rates were 77%, 81%, 80%, 82% and 82%, respectively. After case-mix adjustment patients in low volume hospitals had a HR of 1.08 (< 75 vs ≥ 200; 95%CI 1.02-1.14). Age at diagnosis (continuous, HR 1.05, 95%CI 1.05-1.05), socioeconomic status (lowest vs highest; HR 1.12, 95%CI 1.07-1.16), grade (high vs low, HR 1.72, 95%CI 1.63-1.82), tumour size (2-5 cm vs 1-2 cm; HR 1.46, 95%CI 1.40-1.51), and a higher number of positive lymph nodes (1-3 vs 0; HR 1.40, 95%CI 1.34-1.46 and >10 vs 0; HR 3.19, 95%CI 3.00-3.39) influenced death, all to a larger extent than surgical volume did. Conclusions: In the Netherlands, surgical hospital volume influences 10-year overall survival only marginally, and far less than patient and tumour characteristics. No difference in survival was revealed for invasive non-metastatic breast cancer patients in hospitals with 75-99 operations per year compared with hospitals with over 200 operations per year. PCN38 Simulation Model of Ibrutinib in Treatment of Relapsed or Refractory Mantle Cell Lymphoma (MCL) Peng S 1, Sorensen S 1, Pan F 1, Dorman E 1, Sun S 2, Van Sanden S 3, Sengupta N 4, Gaudig M 5 Companies of Johnson and Johnson, Raritan, NJ, USA, 3University Hospital Leuven, Leuven, Belgium, 4Janssen Pharmaceuticals, Inc, Raritan, NJ, USA, 5Janssen Pharmaceuticals, Inc, Neuss, Germany . . . . . . . . 1Evidera, Bethesda, MD, USA, 2Janssen, Pharmaceutical Objectives: For patients with relapsed or refractory (R/R) mantle cell lymphoma (MCL), prognosis is poor, with a median survival of one to two years, and treatment options are very limited. In a recent phase II trial (PCYCY-1104), Ibrutinib (Imbruvica™), a first-in-class oral once a day covalent Bruton’s tyrosine kinase inhibitor, was associated with a median progression-free survival (PFS) of 13.9 months. After a median follow-up of 15.3 months, 63% of patients were alive. The aim of the current study was to evaluate the projected life years (LYs) and quality-adjusted LYs (QALYs) associated with ibrutinib and other treatments for R/R MCL. Methods: Patients with R/R MCL were simulated to receive treatment in a health state, survival partition model. Patients received ibrutinib, bendamustine and rituximab (BR), fludarabine, mitoxantrone, and cyclophosphamide (FMC), temsirolimus, or other comparators until death or until progression of disease, at which point they were modeled to receive a subsequent line of treatment or best supportive care. Clinical inputs for ibrutinib were informed by PCYC-1104 trial data; OS was extrapolated to estimate survival outcomes. Clinical inputs for comparators were informed by published sources identified through a systematic literature review. Utility values were informed by published studies. Outcomes were discounted by 3.5%. Results: Treatment with ibrutinib resulted in better health outcomes, incrementally increasing overall LYs by 0.92, 0.86, and 0.92 and PFS LYs by 0.87, 0.87, and 0.87 compared to BR, FMC, and temsirolimus, respectively. Ibrutinib was associated with 0.71, 0.70, and 0.72 overall incremental QALYs compared to BR, FMC, and temsirolimus, respectively. Conclusions: Compared with other therapies, Ibrutinib yielded an average incremental benefit of 0.90 LYs for R/R MCL patients, largely driven by the significant incremental improvement in duration of PFS. Currently a phase III trial is ongoing, the data from which will be used to validate the model. PCN39 The Benefit of Her-2 Targeted Therapies on Overall Survival of Patients With Metastatic Breast Cancer – A Systematic Review Alves C 1, Mendes D 1, Andrade S 2, Batel Marques F 3 1AIBILI, Coimbra, Portugal, 2Roche Farmaceutica Quimica, Amadora, Portugal, 3University of Coimbra, Coimbra, Portugal . . . . Objectives: This study was aimed at evaluating the overall survival (OS) gains associated with HER-2 directed therapies in patients with metastatic breast cancer. Methods: A bibliographic search was conducted in the MEDLINE (PubMed) and in the Cochrane Central Register of Controlled Trials databases, from their inception through March, 2014. Only phase III clinical trials (RCTs) including HER2-positive metastatic breast cancer patients have been included in this review, irrespective of the treatment administered (i.e., chemotherapy and/or hormone therapy, chemotherapy and/or hormone therapy plus HER2-targeted therapy). OS was defined as time from randomisation until the occurrence of death from any cause. Studies have been grouped according to the line of treatment, i.e. first-line or second-line or beyond. Results: Seventeen RCTs were eligible for inclusion, of which 12 assessed therapies targeting metastatic breast cancer HER2+ in the first line setting. OS improved from 20.3 months in the first RCT (standard chemotherapy; Slamon et al, 2001) evaluating HER-2 targeting therapies to 48 months (estimated median OS) in the study of Swain and colleagues (2013), with triple combination of pertuzumab, trastuzumab and docetaxel. Four studies evaluated OS of HER-2 targeting therapies in second-line setting of metastatic breast cancer. The OS in second-line setting improved from 15.3 months (capecitabine; Cameron et al, 2008) to 30.7 months (trastuzumab emtancine; Verma et al, 2012). In the third-line setting, the association of lapatinib + trastuzumab has demonstrated to improve OS in 4.5 months compared with lapatinib alone (14 months vs 9.5 months; Blackwell et al, 2012). Conclusions: The HER-2 directed therapies had an undeniable benefic impact in the overall survival of patients with HER-2 positive metastatic breast cancer. Triple combination of docetaxel, pertuzumab and trastuzumab is associated with a survival extent of more than 3.5 years, compared with a life expectancy of 1,5 years achieved 13 years ago. PCN40 Simulation Model of Ibrutinib for Chronic Lymphocytic Leukemia (CLL) With Prior Treatment Pan F 1, Peng S 1, Sorensen S 1, Dorman E 1, Sun S 2, Gaudig M 3, Sengupta N 4 Companies of Johnson and Johnson, Raritan, NJ, USA, 3Janssen Pharmaceuticals, Inc, Neuss, Germany, 4Janssen Pharmaceuticals, Inc, Raritan, NJ, USA . . . . . . . 1Evidera, Bethesda, MD, USA, 2Janssen, Pharmaceutical Objectives: Treatment options for chronic lymphocytic leukemia (CLL) who received prior therapy are limited; no standard of care exists. In a recent phase III clinical trial (PCYC-1112), ibrutinib, an oral, once-a-day, first-in-class covalent Bruton’s tyrosine kinase inhibitor, was associated with improved progression-free survival (PFS, HR= 0.215) and overall survival (OS, HR= 0.387) compared with ofa- A621 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 tumumab. The current study aimed to evaluate the projected life years (LYs) and quality-adjusted LYs (QALYs) associated with ibrutinib, ofatumumab, and other therapies for treatment of CLL with prior therapy. Methods: A health state model simulated treatment of a cohort of CLL patients who had received prior therapy. Patients were simulated to receive either ibrutinib or ofatumumab until death or disease progression, at which point they received subsequent treatment or best supportive care. Clinical inputs for ibrutinib and ofatumumab were informed by PCYC-1112 trial data (N=391). Long-term follow-up data from PCYC-1102 and PCYC1103 trials (combined N=101) was used in sensitivity analysis. Long-term OS and PFS were extrapolated from clinical trials to estimate survival outcomes. Utility were informed by published studies. Evaluation of ibrutinib versus other existing agent and emerging agents including idelalisib and ABT-199 was included in a sensitivity analysis. Long-term health outcomes were discounted by 3.5%. Results: Treatment with ibrutinib resulted in better health outcomes, incrementally increasing LYs by 0.63 and progression-free LY by 0.87 over a 5-year time horizon compared to ofatumumab, which lead to 0.47 incremental QALYs. In a 10 year time horizon analysis, ibrutinib increased LYs by 0.79. Ibrutinib was also associated with increased LYs and QALYs compared to other existing and emerging treatments. The model results are most sensitive to the approaches used to extrapolate OS. Conclusions: Ibrutinib was demonstrated to yield better health outcomes for CLL patients with prior therapy compared to ofatumumab, largely driven by significant improvements in PFS and OS. CANCER – Cost Studies PCN41 Budget Impact Analysis of Aflibercept in yhe Treatment of Metastatic Colorectal Cancer (MCRC) in Poland Kaczor M P 1, Pawlik D 1, Wójcik R 1, Glasek M 2, Pieczonka A 2, Kraska A 2, Zelazowski K 2 s. c., Kraków, Poland, 2Sanofi, Warsaw, Poland . . . . . . . . PCN43 Forecasting Outpatient Pharmaceutical Expenditure for Cancer Treatment in Germany Schoch G G 1, Blank S 1, Tamminga M 1, Steimle T 1, Stargardt T 2 1Techniker Krankenkasse, Hamburg, Germany, 2University of Hamburg, Hamburg, Germany . . . . . . Objectives: To allow budgeting of pharmaceutical expenditure for cancer drugs in Germany, we forecasted future outpatient pharmaceutical expenditure for cancer treatment from the perspective of the statutory health insurance (SHI) for 2016. Methods: Based on data of the Techniker Krankenkasse (TK), a large German sickness fund with more than 8.2 million insured, we forecasted pharmaceutical expenditure for 12 cancer indications in 2016 (according to ICD-10: C16, C18-21, C22, C26.9/C49.9, C34, C43, C50, C56, C61, C73, C90, C91.1). To extrapolate results to whole SHI, we adjusted for differences in demographics of insured between TK and SHI using publicly available data, i.e. KM6 statistics. We also incorporated trends in membership to SHI. To assess the impact of new drugs, we obtained expert opinion by IMS Health on (a) the timing of drug launches in the German cancer market, (b) the expected prices of new drugs and (c) the extent to that new drugs will replace existing pharmaceuticals. For calculations, we assumed that newly launched drugs will reach on average a diffusion of 20% of their market potential until 2016. Results: According to our model, SHI outpatient pharmaceutical expenditure for these 12 cancer indications was million € 2,780 in 2012, i.e., 9.5% of total outpatient pharmaceutical expenditure. In 2016, we expect annual pharmaceutical expenditure for these indications to increase by 17.2% to million € 3,258. Of the 26 new drugs identified to be launched until 2016,10 will at least partly replace existing pharmaceutical treatments. Thus, million € 526 of our budget estimate will be due to new drugs, €2,650 million will be due to pharmaceuticals that were already launched in 2012 while € 82 million will be due to demographic change. Conclusions: The expected increase in costs for cancer drugs are a financial challenge for German SHI. Whether benefit of new drugs and expected costs can be considered fair value needs to be investigated elsewhere. 1Aestimo Objectives: To estimate the budget impact resulting from the introduction of aflibercept for the treatment of metastatic colorectal cancer (mCRC) within drug program in Poland. Methods: Analysis was performed in 3-year time horizon (2014-2016) from the public payer (NHF) perspective. Target population is defined as adult patients with mCRC that is resistant to or has progressed after an oxaliplatincontaining regimen (including patients who experienced distant relapse within 6 months of completion of oxaliplatin-based adjuvant therapy). Eligible patient population was estimated by compilation of following data: epidemiological studies, local market study, IMS data, survey among Polish oncologists. Market shares of different regimens (aflibercept 4 mg/kg +FOLFIRI, bevacizumab 10 mg/kg +FOLFOX-4, FOLFIRI) were projected based on the NHF data and experts’ opinion. Following cost categories were included: drug acquisition and administration (anti-VEGF, chemotherapy), diagnostics, monitoring and adverse events (grade 3-4). Results: With the introduction of aflibercept, estimated annual number of patients starting aflibercept treatment will be 90, 209 and 224 in year 2014, 2015 and 2016, respectively. Total annual expenditures in year 2014, 2015 and 2016 were calculated to be 39.3, 40.3 and 41.2 million PLN in scenario without aflibercept, compared with 37.4, 34.9 and 35.0 million PLN, respectively, with the introduction of aflibercept. In case of aflibercept reimbursement, the NHF would save 1.9 million PLN in year 2014, 5.3 million PLN in year 2015 and 6.1 million PLN in year 2016. Conclusions: The introduction of aflibercept would result in savings for the NHF in Poland, mainly as a consequence of reduced pharmacological costs compared to bevacizumab. PCN42 Estimating the Economic Impact of Sorafenib in Treatment of Locally Recurrent or Metastatic, Progressive, Differentiated Thyroid Carcinoma (DTC) That is Refractory to Radioactive Iodine (RAI) Treatment Sussman M 1, Munsell M 1, Valderrama A 2, Seal B S 2, Wen L 2 1Boston Health Economics, Inc., Waltham, MA, USA, 2Bayer HealthCare Pharmaceuticals Inc., Whippany, NJ, USA . . . . . . Objectives: Sorafenib, a multikinase inhibitor, received Food and Drug Administration (FDA) -approval in 2013 for treatment of patients with locally recurrent or metastatic, progressive, radioactive iodine-refractory (RAI-r) differentiated thyroid carcinoma (DTC). A budget impact model (BIM) was developed from a United States (US) payer perspective to estimate the costs of adding sorafenib to the set of available treatments in a hypothetical health plan in the RAI-r DTC population. Methods: An Excel-based BIM evaluated costs of RAI-r DTC with other FDA-approved and compendia-recommended treatments using baseline and projected market shares. Clinical inputs included the prevalence of RAI-r, average monthly dosage, and average duration of sorafenib and other FDAapproved and compendia-recommended treatments. Economic inputs for each treatment included the wholesale acquisition cost (WAC) per dose and hospital administration costs per month. A net per-month cost to the payer for sorafenib was $6,872. Laboratory testing costs were derived from product-specific package inserts and the Centers for Medicare & Medicaid Services (CMS) Physician Fee Schedule. Sorafenib market share was assumed to increase from 35% at baseline to 54% at 1 year, with shift from other treatments coming mostly (12%) from clinical trial/no treatment. The duration of sorafenib treatment was 11 months based the DECISION trial. Results: An estimated 25 patients with RAI-r DTC were eligible for treatment with sorafenib. Costs increased 25% ($282,467) or $0.02 per member per month (PMPM) from baseline to 1 year post baseline. Sensitivity analyses, varying default inputs for duration of treatment (±2 months) and estimated market share for sorafenib (±10%), showed greatest sensitivity to sorafenib market share (incremental total costs: $180,812–$384,122). Conclusions: Our findings indicate that adding sorafenib to a hypothetical health plan’s formulary has a manageable budget impact of $282,467, or a PMPM increase of $0.02, given the small RAI-r DTC population. PCN44 Budget Impact Analysis of Everolimus for the Treatment of Hormone Receptor Positive, Human Epidermal Growth Factor Receptor-2 Negative (HER2-) Advanced Breast Cancer in Kazakhstan Lewis L 1, Taylor M 1, Suriya Y 2, Kuanysh N 2, Kaldygul S 2, Ramil A 2 1York Health Economics Consortium, York, UK, 2Kazakh Research Institute of Oncology and Radiology, Almaty, Kazakhstan . . . . . . Objectives: To determine the budget impact of everolimus (in combination with letrozole/anastrozole) as a second-line treatment for ER+ HER2-negative advanced and metastatic breast cancer in postmenopausal women in Kazakhstan. Methods: A cumulative cohort model was developed to estimate the five-year costs associated with introducing everolimus to the Kazakh health care system, with two scenarios: “with everolimus” and “without everolimus”. Treatment-specific PFS and OS data were extrapolated from trial data using a Weibull function. It was assumed that data from the BOLERO-2 trial (everolimus+exemestane vs exemestane alone) were representative of everolimus+letrozole/anastrozole and letrozole/anastrozole used in the model. Per-patient drug, health state, adverse event costs were calculated. The per-patient costs were multiplied by the number of patients expected to receive each treatment according to predicted market share, which was split between everolimus+letrozole/anastrozole, letrozole/anastrozole alone, chemotherapy and tamoxifen. Results: The within-trial data from BOLERO-2 reported 17 month OS of 74.7% and 67.6% for everolimus+exemestane and exemestane alone, respectively. The utilities reported in BOLERO-2 (data available up to week 78) were 0.67 and 0.70 for everolimus+exemestane and exemestane alone, respectively. The five year results demonstrate that the introduction of everolimus leads to a 12% increase in drug costs, a 2% reduction in pre-progression health state costs, a 1% increase in post-progression health state costs and a 2% reduction in adverse event costs. The net result is a 2% increase in total costs, from T16.97 billion to T17.389 billion over a period of five years. Conclusions: The analysis estimated that, if everolimus were to be introduced to the Kazakh health care market for the treatment of ER+ HER2advanced breast cancer, there would be a small impact upon overall health care expenditure. An increase in drug acquisitions costs was largely offset by a reduction in other health care costs due to improved disease management. PCN45 Budget Impact Analysis of CYP2C19 Genotyping to Target Voriconazole Prophylaxis During Induction-Consolidation Therapy in Acute Myeloid Leukemia (AML) in The United States Mason N T , Bell G C , McLeod H L H. Lee Moffitt Cancer Center, Tampa, FL, USA . . . . . . Objectives: To assess the impact of genotyping acute myeloid leukemia (AML) patients for CYP2C19*17 gene variant status prior to induction-consolidation therapy from the perspective of a United States (U. S.) payer. Methods: Developed to aid U. S. payers regarding the budgetary impact of DNA genotyping, this model examines the predicted economic outcomes of a hypothetical cohort of 100 neutropenic AML patients under two alternatives: (1) standard voriconazole prophylaxis and (2) genotyping patients for targeted prophylaxis. Published allelic frequencies estimate 27% of the general population may have at least one *17 allele. The presence of the CYP2C19*17 allele results in more rapid metabolism and clearance of voriconazole, which can lead to underdosing and ineffective prophylaxis on the standard regimen. The incidence of invasive fungal infection is 15% without effective prophylaxis and is reduced to 6.6% upon adequate prophylaxis. Targeted prophylaxis based on genotyping prescribes an alternative drug or higher voriconazole dose in patients with the *17 allele. Further model parameters were taken from published literature and 2014 CMS Laboratory Fee Schedule. Results: The average total cost of care for AML patients receiving standard versus targeted voriconazole prophylaxis was $46,795 and $46,385 per patient, respectively. In addition to the $410 saved per patient, the number of invasive fungal infections was reduced from 6.6 to 4.3 A622 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 by targeting prophylaxis based on patient genotype. The deterministic sensitivity analysis showed that the savings is most dependent on the incidence of invasive fungal infection, cost of treating an invasive fungal infection, and frequency of *17 in the population. Conclusions: Genotyping AML patients for CYP2C19*17 prior to induction-consolidation is expected to be cost-neutral or potentially cost-saving by reducing the incidence of invasive fungal infections compared to standard prophylaxis. These results may mitigate potential budgetary concerns, thereby reducing barriers to a test that can be clinically beneficial to AML patients. PCN46 Budget Impact Analysis of the Introduction of New Therapeutic Agents for the Treatment of Metastatic Castration Resistant Prostate Cancer (MCRPC) Patients After Docetaxel Failure in the Brazilian Private Health System Asano E , Vitale V Janssen-Cilag Farmaceutica, São Paulo, Brazil . identified at IMSS and 631 at ISSSTE, 2,570 and 536 of them with NSCLC. A 4.2% ALK+ rate was assumed. Direct medical costs of standard treatment for LC were obtained from a published source. Cost for crizotinib was given by the manufacturer. All costs are expressed in 2014 USD ($1USD= $13MXN). Two scenarios are presented: 1) world without crizotinib, where all patients with LC are treated with standard treatment; 2) world with crizotinib, where patients with ALK+ advanced NSCLC are treated with crizotinib and all other LC patients are treated with standard treatment. Results: 81 and 17 ALK+ advanced NSCLC patients were identified in IMSS and ISSTE, respectively. Total costs in a world “without” and “with” crizotinib using a one-year time horizon is $50.3 and $52.9 million, respectively, for IMSS. For ISSSTE, total costs were $10.5 versus $11.1 million. The combined incremental budget impact across both public health care institutions is 5.2%. Conclusions: Crizotinib, the only drug approved for the treatment of ALK+ advanced NSCLC patients has a minimal incremental budget impact on the overall expenditure within the two main Mexican public health care institutions. . Objectives: In the latest years several drugs demonstrated to increase survival in mCRPC patients post chemotherapy failure. However, issues remain related to the treatment sequencing of these drugs. The aim of this study is to estimate the budget impact of the introduction of enzalutamide in the Brazilian Private Health System. Methods: A deterministic state transition budget impact model (BIM) was developed to estimate treatment costs of mCRPC patients after chemotherapy failure over a 3-year time horizon. Budget impact was estimated comparing a baseline scenario including mandatory coverage drugs (abiraterone, cabazitaxel) with an alternative scenario including all treatment options (abiraterone, cabazitaxel and enzalutamide). Target population, dosing, duration of therapy and sequencing was based on scientific literature. Pharmaceutical direct costs were based on factory price, assuming parity price of enzalutamide and abiraterone. Univariate Deterministic sensitivity analysis was conducted to determine the impact of parameters on results. Results: The BIM estimates that a total of 5,789 patients will be treated in the next three years, with annual cost estimates in the baseline scenario of R$83,944,041 in year 1, reaching R$198,507,065 in year 3 of the simulation. The introduction of enzalutamide would incur a total increase in costs of R$16,649,325 after 3 years. In deterministic sensitivity analysis, enzalutamide price, proportion of patients receiving additional treatment line and duration of therapy were the most important variables that impacted results, with the alternative scenario remaining more costly than the baseline scenario in all simulations, incurring additional costs ranging from R$6,634,955 to R$38,818,233. Conclusions: Considering current available evidence regarding treatment sequencing, the introduction of enzalutamide is expected to increase costs to the Brazilian Private Health System. PCN47 Budget Impact Model of Ceplene® As Maintenance Therapy in Adult Patients with Acute Myeloid Leukemia in First Remission Kaskens L 1, Gehenio D 2, Munzel U 2, Darba J 3 Health Economics & Outcomes Research S.L., Barcelona, Spain, 2Meda Pharma GmbH & Co KG, Bad Homburg vor der Höhe, Germany, 3Universitat de Barcelona, Barcelona, Spain . . . . 1BCN Objectives: To assess the economic impact of Ceplene® with low-dose Interleukin-2 (IL-2) for the treatment of adult patients with Acute Myeloid Leukemia (AML) in first complete remission (CR-1) which previously received intensive chemotherapy in Spain. Methods: A budget impact model was developed using the perspective of the Spanish National Health System with a 4-year time horizon. Ceplene®/IL-2 was compared with no treatment and an unrelated allogenic hematopoietic stem cell transplant (allo-HSCT). For both treatment options and no treatment, health care costs (EUR 2013) including medical visits, hospitalisations, laboratory and diagnostic tests, prophylactic measures, treatment of complications and infections were considered. Average treatment costs per patient up to a maximum of three years were estimated for both treatment options and no treatment depending on the probability of overall survival without or with a relapse, and death without relapse, as well as the duration until relapse or death without relapse during this period. Total annual health care costs were estimated based on the annual per patient cost, the target population, and the market shares associated with each option, before and after the introduction of Ceplene®/IL-2. Results: Patients eligible for Ceplene®/ IL-2 were estimated at 1,502 in 2013 with a small increase up to 1,509 in 2016. The overall budget impact with the introduction of Ceplene®/IL-2, is estimated to decrease with € 674,149 and € 728,945 in 2014 and 2016, and an increase of € 202,322 in 2015. Overall budget impact savings over the period 2014-2016 are estimated at € 1,130,894. Conclusions: The introduction of Ceplene®/IL-2 as maintenance therapy supposes savings in the budget impact for the treatment of AML patients in CR-1 in Spain. Ceplene®/IL-2 is expected to fulfil a direct medical need for patients not eligible or having an unfavourable profile for an unrelated allo-HSCT receiving no treatment, and those who received an unrelated allo-HSCT with unfavourable prognostics. PCN48 Budget Impact Analysis of the Use of Crizotinib for Non-Small Cell Lung Cancer and ALK+ Mutation in the Two Main Public Health Care Institutions in Mexico Guirant-Corpi L , Muciño-Ortega E Pfizer S.A. de C.V., Ciudad de México, Mexico . . Objectives: Standard treatment for lung cancer (LC) in Mexico is chemotherapy. Crizotinib is the only therapy approved for patients with ALK+ advanced non-small cell lung cancer (NSCLC), a low prevalence condition. This analysis aims to estimate the economic impact of using crizotinib for patients with ALK+ advanced NSCLC in the Mexican setting from public health care institution perspective. Methods: A budget impact analysis with a one-year time horizon was developed to compare expected costs that the Mexican Social Security Institute (IMSS) and the Safety and Social Services for State Workers Institute (ISSSTE) public health care institutions would incur if they were to include crizotinib on their formularies. Using epidemiology data from published sources, a total of 3,023 potential patients with LC were PCN49 Budget Impact Analysis of Everolimus Plus Exemestane Versus Gemcitabine Plus Paclitaxel and Capecitabine Plus Docetaxel in Metastatic Breast Cancer Patients in Egypt Elsisi G 1, Saeed E 2, Elmahdawy M 3 of Health, Faculty of Pharmacy Helwan University, Cairo, Egypt, 2Central Administration for Pharmaceutical Affairs, Cairo, Egypt, 3Ministry of Health, Cairo, Egypt . . . 1Ministry Objectives: To estimate the budget impact of everolimus-exemestane versus the most commonly used regimens in the Egyptian practice; gemcitabine-paclitaxel and capecitabine-docetaxel for a health care plan that introduces everolimus for post-menopausal hormone receptor positive, human epidermal growth factor receptor-2 negative metastatic breast cancer (HR+, HER2-MBC) patients over three years. Methods: Drug and medical budget impacts (2013 EGP) were estimated over the first three years of the three drug regimens use from the health insurance perspective. Epidemiology data were used to estimate target population size. The treatment data for MBC patients were obtained from published and nonpublished sources. The model considered 2 scenarios—without (pre) and with (post) everolimus-exemestane. Monthly medical costs were calculated for the pre- and post-progression phase. Results were considered on a per member per month (PMPM) basis to examine the relative impact on the plan. Deterministic sensitivity analyses were conducted. Results: In a real-world 6,055,902 targeted patients, 288,261 of them were found to be candidates for everolimus-exemestane regimen. For patients taking gemcitabine-paclitaxel and capecitabine-docetaxel regimens, the estimated incremental cost PMPM was LE3.00 and LE2.94 respectively for each after three years. The estimated incremental cost PMPM for the gemcitabine-paclitaxel population was LE0.62, LE2.60 and LE5.77 for year 1, 2 and 3 respectively while for the capecitabine-docetaxel population was LE0.59, LE2.54 and LE5.70 for year 1, 2 and 3 respectively. The capecitabine-docetaxel results were most sensitive to the cost of everolimus while gemcitabine-paclitaxel results were most sensitive to the number of eligible patients. Conclusions: Increased acquisition costs of everolimus-exemestane for HR+, HER2-MBC treatment are expected to be obviously offset by both the reduced number of progressed patients and the relatively small medical costs due to avoided adverse events of each of gemcitabine-paclitaxel and capecitabine-docetaxel regimens. The expected budget impact of covering everolimus for this group of patients was relatively small. PCN50 Buget Impact Analysis of Rituximab for Chronic Lymphocytic Leukemic: The Case of Brazilian Public Health Koury C D N 1, Nunes A A 2, Nita M 1 -Fundação de Ensino e Pesquisas Econômicas, Brasilia, Brazil, 2University of São Paulo USP, Ribeirão Preto, Brazil . . . . . . 1FIPE Background: Chronic Lymphocytic Leukemia (CLL) is a malignant disease incurable of the lymphoid system, that affects predominantly elderly, especially in Western countries. Your treatment when necessary is based on the administration of chemotherapy, with association of fludarabine plus cyclophosphamide (FC), the most widely used schema. Recently the addition of rituximabe, a monoclonal antibody has been associated with this scheme, known as FCR. Objectives: To elaborate a budget impact analysis (BIA) of rituximab for chronic lymphocytic leukemic for help the decision making. Methods: A BIA of association of fludarabine plus cyclophosphamide in SUS compared to rituximab with this scheme was performed. The analysis’ time horizon was 5 years, using a CLL prevalence of 4.4% and 25% of CLL refractory between them (1.634), considering an annual growth rate of 0.8143% and a market share of 25% and 75% according the classification of diagnosis and stage of Rai & Keating. The mean total rituximab dose considered was 375/mg/ m2, with an average personal weight and size of 70kg and 1,70m, which means 681,75mg per cicle. All cost purchase prices and remission rate of rituximab (22%) and stand chemiterapy (9%) were obtained at one year trial in the onco-hematology high complexity Clinical Hospital of the Faculty of Medicine of Ribeirão Preto HCFMRP / USP hospital measured in real 2012. Results: The budget impact of FC per year would be 38.7m reais ($17.5m) in the 1st year, considering 25% of target population, reaching approximately 135.3m reais ($60.9m) in 75% of patients. For RFC, the budget impact would be 97 million reais ($43.6m) in the 1st year, reaching 340 million reais ($153m) in 75% of patients. Conclusions: Treatment costs still impressive, considering that rituximab ’ values reach 2.5 times the clicle unit values of standart chemioperapy, fact that did not happen in other countries where they are already covered. PCN51 Real-World Costs of Laboratory Tests for Non-Small Cell Lung Cancer van der Linden N 1, Smit E F 2, Uyl-De Groot C A 1 for Medical Technology Assessment (iMTA), Erasmus University, Rotterdam, The Netherlands, 2VU University Medical Centre, Amsterdam, The Netherlands . 1Institute . . . . A623 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Cancer patients undergo a wide range of laboratory procedures, from simple blood tests to complex molecular diagnostics. In cost-effectiveness analyses, costs of laboratory testing are often ignored or estimated inappropriately. We present real-world costs of laboratory procedures for non-small cell lung cancer (NSCLC) patients, per category of laboratory testing. Methods: In a Dutch academic hospital, all laboratory tests performed for NSCLC patients between 2009 and 2011, were recorded and categorized in clinical chemistry; pathology; microbiology; serology, hematology, transfusion; pharmacology; and other or unknown. Number of tests per type were multiplied with unit costs per test obtained from The Dutch Healthcare Authority. Results: 1,015 patients were included, with a total of 171,632 laboratory procedures. 392 different types of tests were performed. Mean cost for laboratory testing is EUR1,175 (95%CI 1,066-1,283) per patient. For cost allocation and modeling purposes, cost per month with laboratory testing (EUR265,95%CI 247-282) and cost per day with laboratory testing (EUR96, 95%CI 91-100) are presented. Costs are mainly driven by (molecular) pathology (26%), other (25%, mainly order processing fees) and clinical chemistry (24%, due to high test volumes). Conclusions: Costs of laboratory procedures for NSCLC patients are substantial. Relatively simple blood tests contribute significantly to these costs due to high test volumes. Main cost driver however is molecular testing by the pathologist, for the use of targeted therapies. In pharmacoeconomic evaluations, taking laboratory costs into account significantly impacts results, especially when testing practices differ between treatment alternatives. recombinant human granulocyte colony stimulating factor (pegfilgrastim) to reduce the risk of FN, the incidence rate of FN was 39.5% in the control group, while it was 3.04% in the pegfilgrastim group. Pegteograstim (Neulageg®) was shown to be non-inferior compared to pegfilgrastimin the phase three clinical trial, so the result was directly applied. The number of patients to whom the result applies was estimated as 1,440 patients, which is 8% of the annual average number of breast cancer patients in South Korea. The hospitalization costs for FN was estimated by an average costs of 18 patients admitted to a Catholic University hospital in 2013, which was 15,396,014 (±18,847,475) KRW. Costs for isolated ward, 3rd generation antibiotics and G-CSF were included on the hospitalization costs for FN. Weighted average costs given by Health Insurance Review & Assessment Service were used for micro-costing. Results: When assuming an average incidence rate of 20% for FN, FN will occur in 1.72% (25 patients), which is a 8.6% decrease with pegfilgrastim as well as with pegteograstim. Therefore, when estimating at the sum of the costs of pegfilgrastim, pegteograstim and the average cost of hospitalization, the total cost is 5,572,986,587 KRW for pegfilgrastim while it is 3,499,386,587 KRW for pegteograstim. On the other hand, when pegfilgrastim or pegteograstim are not used, the incidence rate of FN is 20% (288 patients), and the average cost of hospitalization after FN occurs is 4,434,052,032 KRW. Conclusions: When pegteograstim is reimbursed to reduce the incidence of FN during chemotherapies for breast cancer with moderate risk of FN, about 1 billion KRW saving is expected from a payer perspective. PCN52 Positron Emission Tomography/Computed Tomography Imaging for Non-Small Cell Lung Cancer: A Budget Impact Analysis PCN55 Costs of Pneumonia in Patients With Cancer Diagnosis from the Private Health System Perspective in Brazil Wichmann R M 1, Freitas M G 2, Silva M T 3 1Brazilian Ministry of Health, Brasilia, Brazil, 2Brazilian Ministry of Health, Brasília, Brazil, 3Universidade Federal de Manaus, Manaus, Brazil Manfrin D F 1, Ferreira C N 1, Santana C F S D 2, Paloni E D M P 2, Campi F D S 2, Gea Y 1, Rufino C S 1 1Pfizer, Inc., São Paulo, Brazil, 2ORIZON - Companhia Brasileira de Gestão de Serviços, Sao Paulo, Brazil . . . . . . . . Objectives: To estimate the budgetary impact of the introduction of PET-CT for staging non-small-cell lung carcinoma (NSCLC) in Brazilian Public Healthcare System (SUS). Methods: For estimation of the budgetary impact the work considered patients diagnosed with NSCLC that would be submitted to the PET-CT for tumor staging. It was considered the time horizon of 2013, 2014 and 2015. The number of procedures in those years was calculated from an estimate of the years 20082012, obtained from SUS. It was assumed that by 2013 the demand for PET grows due to its incorporation and, from then on, there would be a tendency to annual decrease of 7.5%, the same as in previous years. Cost estimates were obtained from recent cost-effectiveness literature. Results: The average cost of the PET-CT procedure calculated in 2012 is $1,229.92, the value used for 2013 approximately. The 2014 and 2015 values were adjusted for inflation at rate of 5% per year, resulting $1,291.40 and $1,355.97, respectively. The total budget impact for each year was calculated by multiplying the number of procedures to be performed by its base value minus the savings achieved. The values of $2,072,300.84, $2,030,645.11 and $1,988,842.77, for the years 2013, 2014 and 2015, were found respectively. Conclusions: The introduction of PET-CT in the staging of NSCLC affects the budget of the Ministry of Health in 21.8%, 19.6% and 17.5% in the years 2013, 2014 and 2015, respectively. PCN53 Budget Impact and Incremental Survival Benefit of Eribulin Mesylate as a Treatment for Metastatic Breast Cancer in Brazil Majethia U 1, Tremblay G 2, Borges L 3, Jones T 1, Forsythe A 2, Pomerantz D 4, Clark O A C 5 Lake, NJ, USA, 2Eisai Inc, Woodcliff Lake, NJ, USA, 3Evidências, Campinas, Brazil, 4Kantar Health, Princeton, NJ, USA, 5Evidencias, Campinas, Brazil . . . . . . . . Objectives: The objective of this study was to estimate the incremental Budget Impact (BI) and survival benefit of utilizing eribulin for treatment of Metastatic Breast Cancer (MBC) in patients with 2-5 prior chemotherapy regimens including anthracycline and taxane. Methods: Epidemiology was derived from 2013 CancerMPact report and National databases (FOSP and INCA). Treatment of Physician’s Choice (TPC) arm included capecitabine, gemcitabine, vinorelbine, docetaxel and paclitaxel. TPC market shares, efficacy and Adverse Events (AE) data were taken from Phase III clinical trial. Total costs comprised of drug costs, administration costs, direct medical and AE costs. A micro-costing analysis of resource utilization for AE treatments and disease management pre and post progressions was performed. Local Brazil tariffs for each costs unit were applied to an Excel-based model to compare total costs and survival rates with and without eribulin for MBC patients across a 5-year horizon from private payers perspective (assumed to cover 25% of Brazil population). Results: Applying an MBC prevalence rate, proportion of patients with active disease and treated with 3rdline chemotherapies the model estimates up to 801 patients treated with eribulin out of 3864 eligible patients over 5 years. Assuming eribulin market share of 2%, 5%, 9%, 14% and 20% in years 1, 2, 3, 4 and 5, the BI is R$294K, R$741K, R$1345K, R$2111K and R$3043K (net increase of 0.12% - 1.21%). The main cost offsets include the displacement of more widely-used TPC therapies. Eribulin MBC treatment in Brazil is estimated to yield an incremental 245 progression free patient years and 408 life years in population covered by private insurance. Conclusions: Given the limited number of effective treatment options available to patients receiving third line chemotherapy, eribulin represents a much needed therapy option for this population. With additional survival benefits eribulin represents an effective innovative approach to MBC management. PCN54 Budget Impact on the Use of Pegfilgrastim to Reduce the Febrile Neutropenia During Chemotherapy for Breast Cancer with Moderate Risk Compared to a Standard Therapy Choi W S , Lee Y J , Ju S Y , Heo S B , Kim C M Catholic University College of Medicine, Seoul, South Korea . . . . . . . . . . . . . . . . . . . . . . Objectives: Cancer patients are susceptible to infections, including pneumonia, due to immunosuppressive therapies associated with cancer treatment. This study aimed to evaluate the budget impact of pneumonia in patients with previous diagnosed cancer in the Brazilian Private Health System. Methods: Orizon database (N= 18 million lives) was used to identify patients with any type of cancer followed by a pneumonia hospitalization between October 2010 and December 2013. Pneumonia was identified using code of A40.3, B95.3, G00.1, J13, J15, J15.0, J15.3, J15.4, J15.8, J15.9, J18, J18.0, J18.9, J20.2, P23.3. Inpatient and related outpatient costs were included. Results: A total of 68,717 patients with a pneumonia hospitalization were identified. Of those, 2,769 were diagnosed with cancer (WCa) before the pneumonia hospitalization for a total of 3,605 hospitalizations. This translated to a mean of 1.30 hospitalization per WCa patient. The group without cancer diagnosis (WoCa), 65,948 pneumonia patients had a total of 81,583 hospitalizations for a mean of 1.24 hospitalizations. The average costs per patient are BRL2,863.08 for the WoCa group and BRL9,288.07 for the WCa group. When considered the costs per hospitalizations the values are BRL2,314.60 and BRL7,134.16 respectively. Conclusions: Although the number of pneumonia hospitalization per patient was slightly higher in WCa compared with WoCA patients, the cost per patient and cost per hospitalization was at least 3 times higher in the WCa compared with WoCA patients. This suggests pneumonia has a substantial financial impact in patients with cancer who are in the Brazilian Private Health System. . 1Eisai, Woodcliff . . . . Objectives: To conduct a budget impact analysis on the use of Pegteograstim (Neulageg®) to reduce the risk of febrile neutropenia (FN) during chemotherapies for breast cancer with moderate risk compared to a standard treatment with antibiotics and G-CSF after occurrence of FN. Methods: The efficacy of pegylated PCN56 Economic Impact of a Genomic Companion Diagnostic Test for Breast Cancer Patients in French Private Hospitals Vataire A L 1, Aballéa S 1, Katz G 2 1Creativ-Ceutical, Paris, France, 2ESSEC Business School, Paris-Singapore, Cergy, France . . . . Objectives: Several multigene prognostic and predictive tests have recently been launched. The 21-gene assay (OncotypeDX®), a validated gene expression profiling test that predicts the likelihood of adjuvant chemotherapy benefit in patients with early stage breast cancer, was found to be cost-effective and recommended in several guidelines. Its use in clinical practice in France is limited because of the absence of reimbursement. This study aims to determine if the utilisation of the 21-gene assay in private hospitals would provide good value for money from a collective perspective in France and whether hospitals can afford using the test under the current payment system. Methods: A multicenter retrospective study was conducted to estimate the cost of adjuvant chemotherapy from societal and national insurance perspectives. The resulting estimate was used as an input of a Markov model to assess the cost-effectiveness of the 21-gene assay from the French collective perspective and the economic impact of the test on the revenue in private hospital organizations. Results: The cost of adjuvant chemotherapy in private hospitals was estimated at € 8,218 per patient from the national insurance perspective (€ 10,305 from the societal perspective). The 21-gene assay was found cost-effective compared to standard practice and cost-saving with inclusion of productivity costs. The absence of reimbursement involves a deficit for private hospitals of € 3,200 per patient tested. Conclusions: Providing the 21-gene assay in French private hospitals would be cost-effective in the French collective perspective. In the absence of reimbursement from primary payers, some private hospitals may cover the costs of companion diagnostics to improve their attractiveness, but the test will be underused, thus depriving patients from a technology that could improve their quality of life and using resources that could be freed up for other patients. PCN58 Cost-Effectiveness and Budget-Impact Analysis of Braf Inhibitors in Patients With Metastatic Malignant Melanoma (MMM) in Slovenia Hren R GlaxoSmithKline d.o.o., družba za promet s farmacevtskimi izdelki, Ljubljana, Slovenia . Objectives: To analyze cost-effectiveness and assses budget impact of novel BRAF inhibitors – vemurafenib and dabrafenib – in patients with MMM. Methods: In the absence of head to head data we derived a decision model from indirect comparison A624 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 of progression-free survival (PFS) curves for dabrafenib and vemurafenib, from their respective clinical trials with dacarbazine as a common comparator. The model was applied to the conditions of locally-specific population data and treatment costs, including adverse events, of patients with MMM. In the model, individuals moved from progression-free state to post-progression state or death and were followed for 45 weeks. We compared vemurafenib and dabrafenib from the payers’ perspective at the price level determined by international reference pricing as of December 2013 in terms of costs and progression-free life years (PFLYs). Based on the cost-effectiveness model, we carried out budget impact analysis for a scenario with vemurafenib only and a scenario with vemurafenib and dabrafenib using their projected market shares. Results: Our model has shown that more than 99% of total treatment costs of MMM patients in Slovenia were due to drug costs of BRAF inhibitors in spite of the fact that treatment of associated adverse events were 5 times higher in vemurafenib than in dabrafenib, primarily due to higher incidence of cutaneous squamous cell carcinoma and keratocanthoma. Treatment with dabrafenib vs. vemurafenib allowed to save € 13,009 and gain 0.022 PFLY at a discount rate of 3.5%; the sensitivity analysis showed robustness of findings and retained dominance of dabrafenib over vemurafenib. Budget impact analysis for a 3-year period revealed that introduction of dabrafenib would save the national Sick Fund € 402,000 (€ 5,318,000 in the scenario without dabrafenib vs. € 4,916,000 in the scenario with dabrafenib). Conclusions: Study results suggest that introduction of dabrafenib in Slovenia could reduce costs and improve outcomes in MMM. PCN59 Budget Impact Analysis of a Return-To-Work Intervention for Cancer Patients Shows Hospitals Bear the Costs, for Society to Enjoy the Benefits Mewes J C 1, Steuten L M G 1, de Boer A 2, Frings-Dresen M 2, IJzerman M J 3, van Harten W 4 of Twente, Enschede, The Netherlands, 2Academic Medical Centre, Amsterdam, The Netherlands, 3University of Twente and MIRA institute for Biomedical Technology & Technical Medicine, Enschede, The Netherlands, 4Netherlands Cancer Institute, Amsterdam, The Netherlands . . . . . . . . . . 1University Objectives: Multidisciplinary return-to-work (RTW) interventions effectively support cancer survivors to resume work and potentially increase quality of life, but are not or only partly reimbursed by health insurers. To ensure optimal support for cancer patients in resuming work, it is essential that hospitals can offer RTW in a financially viable way. We analysed the budget impact of a RTW intervention (counselling by occupational physicians + physical exercise) and explored how financing of a return-to-work intervention can be arranged. Methods: The budget impact analysis compared costs of RTW support for all patients able and willing to resume work versus no standardised support, for a large cancer centre serving a population of 1 million inhabitants. Costs and financial benefits relevant from a societal perspective were considered, including intervention costs, productivity losses, and patients´ costs. We identified which stakeholders, including hospitals, employers, health insurances, social security, and patients accrue what costs; and which enjoy the financial benefits under different financing arrangements. Results: RTW costs are ≈ € 2,000 per patient. For a large cancer centre, the annual budget impact is €817k in 2014, rising to €14.7m in 2017. Ccosts for patients with a multidisciplinary rehabilitation need are typically covered by health insurance, leaving € 735k to be financed by the cancer centre. Small improvements in return-to-work and quality of life led to substantial reductions in productivity loss and future health care costs. These savings outweigh the costs of the intervention, rendering RTW cost-saving from a societal perspective. Conclusions: From a societal perspective return-to-work is expected to be cost-saving. Hospitals bear the largest share of the cost, while most financial benefits fall upon other stakeholders. Re-distributing costs and financial benefits among stakeholders would result in feasible financing of the intervention. PCN60 Cost of Treating Acute Promyelocytic Leukemia (APL) Patients in Italy Kruse M M 1, Wildner R 1, Barnes G 2, Martin M 3, Mueller U 4, Lo Coco F 5, Leyman S 6, Pathak A 2 1Optum, Waltham, MA, USA, 2Teva Pharmaceutical, 19355, PA, USA, 3OptumInsight, Uxbridge, Middlesex, UK, 4Teva Pharmaceutical, Ehrenkirchen, Germany, 5University Tor Vergata, Rome, PA, USA, 6TEVA Pharmaceuticals Europe, Amsterdam, The Netherlands . . . . . . . . . Objectives: To estimate the total costs of arsenic trioxide and all-trans retinoic acid (ATO+ATRA) versus ATRA and idarubicin (AIDA) regimens in Italy when used in 1st-line APL treatment. ATO+ATRA is approved in 2st-line treatment, but commonly used globally in 1st-line. Methods: A Markov model was developed with three health states: non-progressive disease, progressive disease and death. Each month, patients could move from non-progressive to progressive disease or die from either state. After progression, patients discontinued treatment and switched to the other regimen. Treatment regimens, efficacy and adverse events were derived from published sources and expert opinion, while each arm’s unit costs (induction and consolidation for both; maintenance for AIDA only) were collected from standard Italian sources. Per-patient costs were reported, and extensive one-way sensitivity analyses were conducted. Results: Expected 2-year pharmacy costs for ATO+ATRA were € 46,600 versus € 5,300 for AIDA. However, direct medical (DM) costs (e.g., monitoring, hospitalizations, etc.) for ATO+ATRA were € 11,300 versus €28,500 for AIDA. The higher costs stemmed from AIDA consolidation taking place in a hospital setting and 2-year maintenance monitoring costs. The treatment failure likelihood for patients on AIDA was 14% versus 3% on ATO+ATRA. As these patients switched to 2nd-line treatment, relapse costs for AIDA were € 1,500 compared to € 400 for ATO+ATRA. AIDA patients incurred higher costs from adverse events (AEs) than ATO+ATRA (€ 600 vs. € 300, respectively). Results were most sensitive to consolidation cost changes: ATO+ATRA pharmacy costs and AIDA DM costs. Conclusions: The results suggest potential cost savings due to DM, progression and AE costs for ATO+ATRA, although AIDA pharmacy costs are lower. Additionally, ATO+ATRA patients have lower clinical AE risks and shorter treatment duration (8 months versus 28 months for AIDA), which could lead to patients’ improved quality of life and cost savings. Further research is needed to determine the cost-effectiveness of ATO+ATRA. PCN61 Economic Impact of Centralized Histological Reviews in Patients with Sarcoma, Gist, and Desmoid Tumors Perrier L 1, Kembou Nzale S 1, Rascle P 1, Bui B 2, Morelle M 1, Ranchère Vince D 1, Terrier P 3, Neuville A 2, Decouvelaere A V 1, Le Cesne A 3, Gomez F 1, de la Fouchardière C 1, Meeus P 1, Trédan O 1, Pérol M 1, Fayette J 1, Neidhardt E M 1, Biron P 1, Boyle H J 1, Marec Bérard P 4, Farsi F 5, Ducimetière F 1, Blay J Y 1, Ray Coquard I 1, Coindre J M 2 1Cancer Centre Léon Bérard, Lyon, France, 2Institut Bergonié, Bordeaux, France, 3Insitut Gustave Roussy, Villejuif, France, 4IHOP, Lyon, France, 5Regional oncology network Réseau Espace Santé Cancer, Lyon, France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Objectives: The aim of this study was to compare the costs of disease management based on revised diagnoses after centralized histological reviews for sarcoma, GIST, and desmoid tumors with the costs based on diagnoses before reviews. Methods: A decision tree was constructed. For both options, the initial pathway was the World Health Organization (WHO) Classification of soft tissue and bone tumors. Diagnoses were considered concordant only when the final diagnosis was categorized in the same manner as the initial finding, as defined by the WHO classification. The decision tree was evaluated over a time horizon of 12 months. Disease management and the probabilities were based on a cohort of patients who had a histological review performed within the RRePS (Réseau de Référence en Pathologie des Sarcomes) network in 2010. The characteristics of the patient and disease, as well as any relevant guidelines, were used. All of the disease managements were defined by the authors of this study. The costs were considered from the French National Health Insurance (NHI) perspective and the costs of the histological review were extracted from the literature. The costs were assessed for each pathway and expressed in Euros 2013. The expected costs were calculated. One-way and probabilistic sensitivity analyses were performed. Results: A total of 2,425 patients underwent a histological review. Of these, 341 patients were found to have a discordant diagnosis. Ten patients were excluded due to missing data. The costs reached € 8,420 (histological review included) when disease management was based on revised diagnoses and € 8,610 when not. Conclusions: In addition to the positive impact of centralized histological reviews on the quality of diagnosis for sarcoma, GIST, and desmoid tumors, our model demonstrated that histological reviews lower the cost of disease management for the French NHI. PCN62 A Cost-Analysis of Complex Radiotherapy in Patients with Head and Neck Cancer Results from the Art-Orl Study Perrier L 1, Morelle M 1, Pommier P 2, Boisselier P 3, Lartigau E 4, Gallocher O 5, Alfonsi M 6, Bardet E 7, Rives M 8, Calugaru V 9, Chajon E 10, Noel G 11, Mecellem H 12, Pérol D 1, Dussart S 1, Giraud P 13 1Cancer Centre Léon Bérard, Lyon, France, 2Leon Berard Cancer Centre, Lyon, France, 3Institut Régional de Cancérologie de Montpellier, Montpellier, France, 4Centre Oscar Lambret, Lille, France, 5Groupe OncoRad Garonne, Toulouse, France, 6Institut Sainte Catherine, Avignon, France, 7Centre René Gauducheau, Saint Herblain, France, 8Institut Claudius Regaud, Toulouse, France, 9Institut Curie, Paris, France, 10Centre Eugène Marquis, Rennes, France, 11Centre Paul Strauss, Strasbourg, France, 12Institut de Cancérologie de Lorraine, Vandœuvre-lès-Nancy, France, 13Hôpital Européen Georges Pompidou, Paris, France . . . . . . . . . . . . . . . . Objectives: A cost analysis investigating TomoTherapy ® (Accuray), Elekta Volumetric-modulated Arc Therapy (VMAT®) and Varian RapidArc®was conducted in patients with head and neck cancer. Methods: The cost-analysis, funded by the National Institute of Cancer (INCa), was performed prospectively based on a multicenter study. Cost calculations were strictly based on a micro costing approach according to the hospitals’ point of view. Only resources which are likely to vary between the strategies being compared were considered. Data on consumption of resources were collected from the treatment planning until the end of the last irradiation session. Productivity losses of radiotherapy involved personnel related to organisational constraints or absenteeism, costs of administrative personnel, costs of logistics and general management were not taken into account. All costs were given in 2013 euros. Numbers of irradiation sessions were compared using Kruskal-Wallis test. Uncertainty was captured by one-way and probabilistic sensitivity analyses using a non-parametric bootstrap method. Results: 174 patients were enrolled in 16 French centers from February 2010 to February 2012. 173 economic questionnaires were exploitable. The mean numbers of sessions were 34.33 (SD: 2.90) for TomoTherapy® (n= 73) and 34.53 (SD: 2.57) for Varian RapidArc® (n= 92, p= 0.603). Eight patients were treated with Elekta Volumetricmodulated Arc Therapy (VMAT®). For irradiation (all sessions included), the over cost of TomoTherapy® (n= 73) reached € 1,109 per patient compared to Varian RapidArc® (n= 92). Sensitivity analyses showed that the annual operating time of the accelerators played a major role in irradiation costs. Conclusions: This is to our knowledge the first study highlighting costs incurred by different IntensityModulated ArcTherapy (IMAT) modalities in this setting. Costs of TomoTherapy® appeared more expensive than RapidArc®. The study should be now completed by a cost-effectiveness analysis in order to shed further light on which modality to focus on. PCN63 Costed Treatment Pathways of Diffuse Large B Cell Lymphoma in a UK Population-Based Cohort: A Patient Level Simulation Model Wang H I 1, Smith A 1, Roman E 1, Crouch S 1, Jack A 2, Patmore R 3 1University of York, York, UK, 2St James’s University Hospital, Leeds, UK, 3Castle Hill Hospital, Hull, UK . . . . . . . Objectives: Diffuse large B-cell lymphoma (DLBCL) is the most common type of non-Hodgkin lymphoma and treatment is usually given with curative intent. Using restricted datasets derived from clinical trials, previous studies examining the cost of treating this cancer have generally focussed on first-line therapy alone; meaning A625 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 that their findings can neither be extrapolated to the general patient population nor to other points along the pathway. Based on empirical data from a representative population-based patient cohort, the objective of this study was to develop a simulation model that could predict costs at an individual level and estimate the real medical costs of treating DLBCL. Methods: All patients newly diagnosed with DLBCL in the UK’s population-based Haematological Malignancy Research Network (www.hmrn.org) in 2007 were followed until 2013 (n= 271). The mapped treatment pathways, alongside cost information derived from the National Tariff 2013/14, were incorporated into a patient level simulation model in order to reflect the heterogeneities of patient characteristics and treatment options. Results: The expected total medical costs were £23,184 for those treated curatively with chemotherapy and £2,125 for those with a palliative approach over a period of five years. The predicted medical cost of the first line was £15,243, and £16,325 and £7,145 for the second line and the third line treatments respectively. The predicted costs captured 94% of the actual costs and the proportion of patients simulated in each health state was identical to the empirical data, supporting the validity of the model. Conclusions: This is the first cost modelling study to use empirical data to provide ‘real world’ evidence to estimate medical costs of entire DLBCL treatment pathways. Future application of the model developed here could be used to evaluate new technologies/treatments and support health care decision makers, especially in the era of personalised medicine. PCN64 Cost-Consequence Analysis of an Aprepitant Regimen Versus a Standard Antiemetic Regimen for the Prevention of Highly Emetogenic Chemotherapy-Induced Nausea and Vomiting in Italy Rosti G 1, Ravasio R 2 9 Treviso (Italy), Treviso, Italy, 2Health Publishing & Services, Milano, Italy . . 1ULSS Objectives: Prevention of chemotherapy-induced nausea and vomiting (CINV) remains an important goal for patients receiving cytotoxic treatment. The objective of this study was to assess, from the Italian National Health Service payer perspective, the costs of an antiemetic regimen using aprepitant, a selective neurokinin-1 receptor antagonist, for patients receiving highly emetogenic chemotherapy. Methods: A decision-analytic model was developed to compare an aprepitant regimen (aprepitant, ondansetron, and dexamethasone) with a standard antiemetic regimen (ondansetron and dexamethasone) for expected costs after highly emetogenic chemotherapy. The model was populated with clinical results from patients with high-dose cisplatin treatment in a randomized trial of CINV preventative therapy. Only direct medical costs (pharmacological treatment and emesis) – Euro 2014 – were considered. Sensitivity and Threshold analyses on key clinical and economic parameters were performed. Results: Aprepitant regimen showed the lower expected treatment cost (aprepitant regimen: € 109.88; standard antiemetic regimen € 115.57). The drug acquisition cost of the aprepitant regimen (aprepitant regimen: € 86.16; standard antiemetic regimen € 79.37) was offset by reduced health care resource utilization costs for CINV (aprepitant regimen: € 23.72; standard antiemetic regimen: € 36.20). Sensitivity and Threshold analysis confirmed the base case results. Conclusions: The results of this cost-consequence analysis suggest that, from the Italian National Health Service payer perspective, aprepitant regimen is a cost-saving strategy compared with antiemetic regimen without aprepitant in the prevention of highly emetogenic chemotherapy-induced nausea and vomiting. PCN65 Cost of Non Small Cell Lung Cancer by Vietnamese and European Treatment Standards in Vietnam Dinh H T 1, Nguyen T T T 2 of Medicine and Pharmacy in HCMC, HCMC, Vietnam, 2University of Medicine and Pharmacy in HCMC, Ho Chi Minh City, Vietnam . . . . . 1University Objectives: Evaluate and compare the cost of non-small cell lung cancer (NSCLC) treatment by different standards of treatment in Vietnam. Methods: A tree-decision model has been developed to estimate the cost of NSCLC. The analysis has been conducted based on the perspective of health insurance companies, therefore only medical direct costs, including cost for drugs and medical services has been evaluated. The list of medical services and drugs was derived from NCCN Clinical Practice Guidelines in Oncology (NCCN therapy) and Lung Cancer Therapy of Vietnam National Cancer Hospital (Vietnam therapy). The transition rates in the model were derived from clinical researches and consultation with experts. The cost of drugs and medical services has been averaged from the relevant medical services of some major hospitals in Vietnam. Results: The total treatment costs increase following the severity of stage whether treatment was implemented by Vietnamese or European standards. Moreover, in all stages of disease, the total treatment costs of NSCLC by Vietnamese guideline are less than that by NCCN standard (74,425,114 vs 79,729,000 in phase I; 139,031,940 vs 210,585,139 in phase II, 172,333,617 vs 339,542,672 VND in phase III, and 160,690,121 VND vs 266,197,825 VND in phase IV, respectively). This can be explained by the lack of some expensive drugs and medical services in Vietnam due to the socio-economic conditions of Vietnam. In the structure of costs, cost for drugs take the dominant part with approximately 76.61% by NCCN standard and 74.92% by Vietnamese standard, which is almost 3 times higher than that of medical services. Conclusions: The high cost of NCSLC should be considered to conduct the relevant health care policies, especially with high-cost drugs in late stages of disease. The cost of NSCLC treatment by NCCN therapy is higher than the cost of NSCLC treatment by Vietnam therapy. PCN66 Hospital Cost of Thromboembolic Events in Breast or Prostate Cancer Patients Scotte F 1, Martelli N 1, Vainchtock A 2, Borget I 3 1Hopital Européen Georges Européen, Paris, France, 2HEVA, Lyon, France, 3Institut Gustave Roussy, Villejuif, France . . . . Objectives: Venous thromboembolism (VTE) is a common complication for cancer patients, leading to hospitalizations that increase the cost of management of these patients. The objective of the present study was to evaluate the number and the cost of VTE-related hospitalizations for patients with breast or prostate cancer during the two first years of their oncologic treatment. Methods: Patients with breast cancer (BC) or prostate cancer (PC) diagnosed in 2010 who had at least one VTE-related hospitalization during the following two years were selected from the French national hospital database (PMSI), using the disease-specific ICD-10 codes. Hospital costs were estimated from the third-party payer perspective using the official diagnosis related group (DRG) tariffs for each year considered. Results: In 2010,62,365 patients newly diagnosed with BC and 45,551 with PC were admitted in French hospitals. Among them, 1,271 in the BC cohort (2.0%) and 997 in the PC cohort (2.2%) were hospitalized for VTE at least once during the two-year followup, leading to 1,604 stays for BC patients and 1,210 stays for PC patients. During a 2-years follow-up, 15.9% of BC patients and 14.4% of PC patients were re-hospitalized for VTE recurrence. The mean cost per stay was 3,261€ and 3,584€ for BC and PC respectively. Mean cost per patient was estimated at 3,302€ and 3,611€ for BC and PC patients hospitalized once for VTE-related events, and it increased to 5,545€ and 5,692€ for BC and PC patients who presented recurrences. Over a 2-year period, total hospital cost induced by VTE-related events reached 1.98 million and 1.43 million € for BC and PC, respectively. Conclusions: VTE-related hospitalizations in breast or prostate cancer patients lead to a significant economic burden that could be reduced by decreasing VTE recurrence. PCN67 Assessing the Economic Burden and Health Care Utilization of U. S. Medicare Patients Diagnosed with Melanoma Xie L 1, Kariburyo M F 1, Du J 1, Baser O 2 1STATinMED Research, Ann Arbor, MI, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . . Objectives: To examine the economic burden and health care utilization of melanoma patients in the U. S. Medicare population. Methods: A retrospective database analysis was performed using U. S. national Medicare claims from 01JAN2008 to 31DEC2012. Melanoma patients were identified using International Classification of Disease 9thRevision Clinical Modification (ICD-9-CM) diagnosis code 172. xx. The first diagnosis date was designated as the index date. A comparator group was created, consisting of patients with the same age, region, gender, index year, and matched baseline Charlson Comorbidity Index scores but without a melanoma diagnosis. A random index date was chosen for the comparator cohort to reduce selection bias. Patients were required to have continuous medical and pharmacy benefits 1 year pre- and post-index date. One-to-one propensity score matching (PSM) was performed to compare follow-up health care costs and utilizations between the cohorts, adjusting for demographic and clinical characteristics. Results: Eligible patients (N= 12,762) were identified for the melanoma and comparison cohorts. After 1: 1 PSM, a total of 8,015 patients were matched from each cohort and baseline characteristics were well-balanced. More melanoma patients had health care utilizations, including Medicare carrier (99.6% vs. 64.3%), Durable Medical Equipment (DME; 19.8% vs. 14.4%) and Home Health Agency (HHA; 6.7% vs. 4.1%) claims, outpatient visits (68.4% vs. 35.4%), inpatient (10.9% vs. 6.6%) and skilled nursing facility (SNF) stays (2.4% vs. 1.7%), but fewer melanoma patients had prescription clams (43.6% vs. 49.8%). Patients diagnosed with melanoma incurred higher expenditures, including carrier ($3,527 vs. $1,324), DME ($153 vs. $101), HHA ($325 vs. $203), outpatient ($8,816 vs. $3,151), inpatient ($1,725 vs. $1,154) and total health care costs ($15,692 vs. $6,990) (p< 0.0001). Similar SNF, hospice and pharmacy claim costs were observed in both cohorts. Conclusions: Patients diagnosed with melanoma had a higher burden of illness compared to the comparison cohort of non-melanoma patients. PCN68 Resource Use and Health Care Costs of Cervical Lesions and Cervical Cancer in Slovakia Ondrusova M 1, Psenkova M 1, Mlyncek M 2, Masak L 3, Hlavinkova L 4, Trnovec P 4 Ltd, Bratislava, Slovak Republic, 2Faculty Hospital and Constantine the Philosopher University, Nitra, Slovak Republic, 3Cancer Institute St. Elizabeth, Bratislava, Slovak Republic, 4GlaxoSmithKline Slovakia, Bratislava, Slovak Republic . . . . . . 1Pharm-In Objectives: The objective of this cost study was to measure the resource utilisation and direct costs associated with health care management of cervical abnormal cytology and cervical cancer in Slovakia and to provide a basis for cost-effectiveness evaluations. Methods: The cross-sectional survey was performed to obtain the information on the management of patients with cervical lesions (Low Grade Squamous Intraepithelial Lesion (LSIL), High Grade Squamous Intraepithelial Lesion (HSIL), Atypical Squamous Cells of Undetermined Significance (ASCUS), Cervical Intraepithelial Neoplasia (CIN) I-III) and cervical cancer (clinical stages IA1 to IVB) and to estimate the direct costs of the disease management. All types of health care used in the management of cervical lesions/cancer were evaluated (diagnostics, treatment and follow-up). Average costs per patient were assessed on a yearly basis and correspond to the prices in 2013. Results: Concerning the cervical lesions, the highest mean annual costs were identified in the CIN III (36.45%; € 1406.62), followed by CIN II (27.0%; € 1041.97), the most costly part of health care management being the treatment of lesions. The costs of health care management of cervical cancer depended on the clinical stage and the frequency of examinations during the year. The highest total costs (follow-up included) were identified in the stage III (19.33%; € 4633.35) and IVB (14.24%; € 3413.22). However, the highest mean annual costs of diagnostics were determined in stage IB (15.39%; € 542.12), IIA (15.46%; € 544.33), IVA and IVB (14.89%; € 524.33). The most expensive treatment (including surgery, radiotherapy, concomitant chemoradiotherapy and palliative chemotherapy) was identified in the stage IIB (20.89%; € 3616.38) and III (21.01%; € 3637.79). In the most advanced stages IVA and IVB, the treatment expenses were lower. Follow-up management was the most expensive in the stage IVB (15%; € 531.66). Conclusions: A626 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 In the management of cervical lesions/cervical cancer, the most expensive are the costs of treatment. PCN69 Economic Burden of Melanoma in Three European Countries: A Retrospective Observational Study Kontoudis I 1, Harries M 2, Mohr P 3, Grange F 4, Ehness R 5, Benjamin L 6, Siakpere O 7, Barth J 5, Stapelkamp C 8, Pfersch S 6, McLeod L 9, Kaye J A 10, Wolowacz S 11 1GlaxoSmithKline Vaccines, Rixensart, Belgium, 2NHS, London, UK, 3Dermatologic Center Buxtehude, Buxtehude, Germany, 4Reims University Hospital, Reims, France, 5GlaxoSmithKline GmbH & Co KG, Munich, Germany, 6GlaxoSmithKline France, Marly le Roi Cedex, France, 7GlaxoSmithKline UK, Middlesex, UK, 8GlaxoSmithKline, Uxbridge, UK, 9RTI Health Solutions, Research Triangle Park, NC, USA, 10RTI Health Solutions, Waltham, MA, USA, 11RTI Health Solutions, Manchester, UK . . . . . . . . . . . . . . Objectives: To estimate cost-of-illness data associated with treatment of patients with stage IIIB/IIIC melanoma with macroscopic lymph node involvement in France, Germany and the United Kingdom (UK), whose primary melanoma and regional lymph node metastases had been completely resected. Methods: This retrospective observational study enrolled patients aged ≥18 years, first diagnosed between 1 January 2009 and 31 December 2011. Data were extracted from medical records and via patient survey. Costs were calculated in Euros (€ ) (France, Germany) or pound sterling (£) (UK) by collecting resource use and multiplying by country-specific unit costs. Costs were estimated from the health care provider perspective. National annual costs were estimated from study results and national disease prevalence estimates based on European cancer registry and other data. Results: 49 centres provided data on 558 patients (55.7% male, 58.2% aged < 65 years at diagnosis, 53.6% stage IIIB disease). Mean follow-up was 27 months in France, 26 months in Germany and 22 months in the UK. Most patients received no adjuvant treatment in France (93.0%) and the UK (97.4%). Use of interferon in Germany was limited (high-dose 11.0%; intermediate-dose 4.9%; low-dose 15.2%; pegylated 1.8%). Mean total direct cost per patient during follow-up was € 23,582 in France, € 32,058 in Germany and £31,123 (€ 37,348) in the UK. The largest cost drivers were melanoma treatment (mean € 14,004 France, € 21,269 Germany and £24,385 UK) and hospitalisation/emergency treatment (mean € 6,634 France, € 6,950 Germany and £2,827 UK). Total mean indirect costs per patient were € 129 in France, € 4,441 in Germany and £1,427 (€ 1,712) in the UK. Preliminary estimates for annual national direct cost were €13.1 million in France, €30.2 million in Germany and £22.8 (€27.6) million in the UK. Conclusions: The economic burden of stage IIIB/IIIC melanoma with macroscopic lymph node involvement was substantial in all three countries. Indirect costs varied by country. PCN70 Cost and Burden of Non-Small Cell Lung Cancer’s in Portugal Borges M 1, Gouveia M 2, Alarcão J 1, Sousa R 1, Teixeira E 3, Barata F 4, Laranjeira E 1, Lopes F 1, Parente B 5, Pinheiro L 1, Vaz-Carneiro A 1, Costa J 1 1Center for Evidence Based Medicine, Faculty of Medicine, University of Lisbon, Lisbon, Portugal, 2Católica Lisbon School of Business and Economics, Lisbon, Portugal, 3Centro Hospitalar Lisboa Norte, Lisboa, Portugal, Lisbon, Portugal, 4Centro Hospitalar Universitário de Coimbra, Coimbra, Portugal, 5Hospital CUF Porto, Porto, Portugal . . . . . . . . . . . . Objectives: This study estimates the impact of Non-Small Cell Lung Cancer (NSCLC) on population health levels and its economic impact in Portugal in 2012. Methods: Data from a cohort of 581 patients with NSCLC completed with an expert panel was used to develop a model of cumulated prevalence with a 6 months cycle period and a time horizon of 5 years. The impact on health status was measured using the Disability Adjusted Life Years (DALYs). The economic impact analysis includes two components. The first estimates the direct costs generated by NSCLC including consumption of inpatient care and outpatient care (consultations, medication, diagnostic exams, transportation, etc). The second estimates indirect costs related to loss of productivity due to NSCLC. Results: A total of 3,180 deaths in Portugal in 2012 were caused by NSCLC, which corresponds to 2.0% of the total deaths in Portugal. The DALYs resulting from premature deaths caused by NSCLC in 2012 totaled 25,071 representing 4.5% of years lost generated by all deaths in the country. For 2012 it is estimated that 3,236 life years were lost due to disability. The total disease burden attributable to NSCLC is thus estimated at 28,307 DALY. The estimated direct cost generated by NSCLC was € 89 million and that total can be broken down into € 32 million for inpatient care and € 57 million for outpatient care. The indirect costs must also be taken into account and they added up to € 54 million. The economic burden of NSCLC is substantial, totaling € 143 million, about 0.09% of Portuguese GDP and 0.92% of all Portuguese health spending in 2012. Conclusions: NSCLC is an important cause of disease burden and costs in Portugal and should receive adequate attention from policy makers. PCN71 Metastatic Prostate Cancer and Skeletal Related Events, a Cost of Illness Study Geenen J W 1, Heine ter R 2, Frederix G W 1, Hövels A 3, Vulpen, van M 4, Kooistra A 2, Klerk de J M H 2, Bloemendal H J 2 1Utrecht University, Utrecht, The Netherlands, 2Meander Medical Center, Amersfoort, The Netherlands, 3Utrecht Institute for Pharmaceutical Sciences, Division Pharmacoepidemiology & Clinical Pharmacology, Utrecht University, Utrecht, The Netherlands, 4University Medical Centre Utrecht, Utrecht, The Netherlands . . . . . . . . . . . . . Objectives: To assess the intramural resource use and associated costs of treating patients with prostate cancer, metastatic to the bone, focusing on Skeletal Related Events (SREs). Secondary, to quantify the impact of SREs and different treatment strategies on total costs. Methods: A retrospective bottom-up cost of illness study performed at a large regional teaching hospital in the Netherlands. Results: A total of 136 patients were included, follow-up started at diagnosis of bone metastases and stopped at death. The mean total costs were EUR 17.931 per patient (median EUR 14.039), inpatient days were the most costly category at a mean of EUR 5955 (median EUR 3995). SREs that required hospitalization (n= 53) were, at median costs of EUR 2039 to EUR 9346, depending on care required, more costly than SREs without hospitalization (n= 165). These SREs had median costs of EUR 200 to EUR 1912, depending on care required. Conclusions: The impact of SREs on total costs could justify policy aimed at actively preventing SREs, e.g. with radionuclide therapy, possibly resulting in better quality of life and cost-reduction. Treatment of prostate cancer with bone metastases is not very costly compared to lung-and breast cancer with similar metastases. However, novel therapeutic options may dramatically increase treatment costs in the near future and proper head-to-head cost-effectiveness studies of all treatment modalities are therefore necessary. PCN72 Economic Impact Model of Breast Cancer Treatment at Early Stages in the Mexican Public Health Care Sector Guirant-Corpi L 1, Hernandez-Reyes F C 1, Kelly J 2, Muciño-Ortega E 1 S.A. de C.V., Ciudad de México, Mexico, 2Centro Médico Nacional Siglo XXI, Mexico City, Mexico . . . . . 1Pfizer Objectives: Nowadays breast cancer represents a great economic burden to public health care sector in Mexico. This economic burden is due by a lack of prevention campaigns and late diagnosis. The analysis aims to estimate the economic impact of standard treatment in Mexico for an optimistic scenario of 80% of all breast cancer population at early stages compared to the actual distribution (55%) of patients along the stages in Mexican setting. Methods: A hypothetical patient cohort of 8765 was evaluated; this estimation was obtained from breast cancer 2011 prevalence in Mexican population. The analysis was developed in order to estimate average cost from the optimistic (base-case) versus actual scenario through public health care perspective. Distribution of population in the actual case scenario was 16%, 39%, 30% and 14% at each stage, I-IV respectively. Direct medical costs were extracted from a published source where through a micro-costing technique the cost at each stage was obtained. Costs are expressed in 2014 USD ($1USD=$13MXN). It is assumed that the economical difference between both scenarios will be used for prevention campaigns. Results: The total cost of treatment for all patients on the actual scenario was $124,261,716 compared to $103,549,361 that was the cost calculated at the base-case scenario. Economic impact of optimistic scenario is 17% less than total cost of treatment at the current scenario. This difference, which represents $20,712,355, could be reallocated to diagnose 32,689 more patients at early stages. A sensitivity analysis with a best-case scenario was developed considering 100% of the patients diagnosed during the first two stages where economic benefits were 314% higher than base-case scenario. Conclusions: This analysis suggests that if more public policies focused on breast cancer prevention were implemented, then it may lead to a more optimal reallocation of resources since direct medical costs increase at late stages. PCN73 Burden of Prostate Cancer and Future Need for Health Care Services Purmonen T 1, Tyrväinen V 2, Kataja V 3 1Medfiles Ltd, Kuopio, Finland, 2Proper Ltd, Joensuu, Finland, 3Central Finland Health Care District, Jyväskylä, Finland . . . Objectives: Prostate cancer is the most common cancer with a current incidence of 0,18% among the 2,6million Finnish men. Demand on health care resource use is dependent on number of patients needing the service. Patient volumes are increasing throughout Europe due to ageing of the population. Despite the stabile or decreasing age-adjusted incidence rates, the absolute number of patients is growing. The population characteristics vary within geographical areas, and thus, using average parameters for the whole country would lead to biased estimates. In this study, burden of prostate cancer cases is estimated on municipality level throughout Finland. Methods: Number of new cancer cases in different hospital districts (N= 20) was extracted from the Finnish cancer registry. Official population statistics and forecast were used to identify the current and predicted age and sex distribution in all of the individual municipalities (N= 320). The data were combined with Tableau (8.0) software, where a map-based interface was constructed. This was also utilized to visualize the population changes and patient forecasts. Similar methodology has been previously utilized in different cancer types. Results: The number of new prostate cancer cases each year is estimated to be 1,5-fold by 2040. As a case example, a Finnish municipality with 135,000 inhabitants was chosen. There were 78 new prostate cancer cases among the 64,332 men in 2011, representing an incidence of 0,12%. According to our analysis, this university city with relatively young population would reach the current country average (0,18%) as late as 2040. Conclusions: The disease burden and population demographics are unevenly distributed across the country, and thus municipality level estimates are needed to inform local decision making and planning. Estimates on the absolute number of patients across relevant disease areas are required in order to prepare to the challenges health care systems are facing in the future. PCN74 The Burden of Non-Small Cells Lung Cancer (NSCLC) in First Line (1L) Treatment: Patterns of Care and Cost of Illness Piedade A , Goes L , Castro A P , Alves A F , Minowa E Evidências Credibilidade Científica, São Paulo, Brazil . . . . . . . Objectives: Several treatment options are available for 1L NSCLC. In Brazil, patients with NSCLC have not been systematically evaluated and different management strategies may be associated to different economical outcomes. Therefore, we aimed to evaluate the patterns of care and cost of illness of 1L NSCLC treatment according to Brazilian supplementary health system. Methods: Metastatic NSCLC patients receiving 1L treatment during year 2013 were eligible and selected from the private market administrative claims database (Evidencias database). Treatments, demographics, supportive drugs and exams were evaluated. Name and any other personal identification were not available at the database. The most reported treatments according to generic name in 1L therapy were defined as patterns of care. The cost of illness was calculated by a bottom-up approach. Exams and associated drugs A627 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 reported were also considered for costing and values were derived from Tables Simpro and CBHPM. Exchange rate used was 1.00USD = 2.20BRL. Results: A total of 233 patients with lung cancer were identified at the database from which 101 were eligible to the study. Fifteen different chemotherapy regimens were reported. The most common regimen was carboplatin with pemetrexed (29,7%), 20,8% received bevacizumab containing regimen and only 6% were in use of oral chemotherapy. About 87% of the patients received a total of 3 cycles of treatment. Costs per cycle of the schemes observed ranged from US$ 11.890,36 to US$ 1.712,95. Considering the number of cycles administered reported in the database, the average cost of management for one patient with metastatic NSCLC in 1L is US$19.001,79. Conclusions: Treatment of 1L metastatic NSCLC represents a significant burden to the Brazilian private payers. The preferences of chemotherapy regimens may change in the following years due to the incorporation of oral chemotherapy in the list of mandatory coverage treatments and procedures. PCN75 Real World Management and Costs in Unresectable Metastatic Melanoma (UMM) Patients Treated at the Antwerp University Hospital (UZA) Strens D 1, Specenier P 2, Peeters M 2 1Realidad, GRIMBERGEN, Belgium, 2University Hospital Antwerp, Edegem, Belgium . . . Objectives: To assess the management and associated lifetime costs in uMM patients as from the diagnosis of unresectable metastatic disease until death. Methods: We performed a retrospective patient chart review to obtain data on medical consumption related to the management of uMM. A complete registry of all patients diagnosed with melanoma at UZA between 2007 and May 2014 was compiled. Eligible for this retrospective chart review were patients with uMM with sufficient data available and who deceased before May 2014. Data on demographics, disease characteristics and management of uMM were collected. Direct costs were calculated by multiplying each item of resource use with its unit cost (2013, € ) using the Belgian public health care payer’s perspective (PHCP) and patient’s perspective. Average (bootstrap 95%CI) overall costs per patient were calculated. Results: 396 patients were registered. 44 eligible patients were included. 81.8% (n= 36) of patients were treated by systemic treatment (s) of which 13.6% (n= 6) received up to 4 different treatment lines. 14 patients received “new drugs” (ipilimumab: 13; vemurafenib: 4) 48/145 (40%) hospitalizations were for treatment administration. Mean overall cost/patient was € 45,297 (bootstrap 95% CI: 33,905- 57,550), of which € 44,346 (95%CI: 33,098-56,584) was reimbursed. The PHCP cost was driven by systemic treatments costs (51% of cost). Mean PHCP cost was € 93,537 (95 % CI: 81,710-105,346) for patients treated with “new drugs”, € 26,564 (95 % CI: 20,688-32,999) for patients treated with chemotherapy but no “new drugs” and € 7,394 (95 % CI: 5,490-9,472) for patients on best supportive care (BSC) only. Median overall survival was 6.1 months (9 months, 5.2 months, and 1 month for patients treated with “new drugs”, with chemotherapy only, and with BSC only, respectively). Conclusions: Management of uMM results in considerable costs for the PHCP, mainly driven by systemic treatment costs. PCN76 Burden of Disease of the MASTOCARCINOMA IN AUSTRIA Bauer M The Institute for Pharmaeconomic research, Vienna, Austria identify the direct costs, including costs for drugs and medical services, and indirect costs, including loss productivity, loss of work, cost for accommodation and transportation due to treatment and other costs of patient and caregivers. Results: Annual total cost per patient with asthma accounts for 6,618,682 VND, from which direct costs are 2.5 times more than indirect cost (4,738,682 vs 1,880,000 VND). The total cost of asthma has increased following the asthma severity with 4,960,278; 8,098,156; 10,759,234; 13,196,280 VND in mild, intermittent, moderate and severe asthma rate. In the structure of total cost, following the asthma severity, the percentage of direct costs increases and indirect costs decreases especially in the mild rate. Total economic burden of asthma was 23,165 billion VND, from which 71.5% (16,585 billion VND) are for direct costs and 29.5% (6,580 billion VND) for indirect cost. Conclusions: The economic impact of asthma (23,165 billion VND) on asthmatic patients, families and society is significant. According to the study, putting more public health efforts to better control asthma is the first necessary step to reduce the costs of asthma. PCN78 Economic Burden of Disease in France In 2012: A Top-Down Allocation of Health Care Expenditure by Disease Based on the French Health insurance Database (SNIIRAM) Gastaldi-Menager C 1, Pestel L 1, Drouin J 1, Fagot-Campagna A 1, Gissot C 2, Polton D 2 1CNAMTS (National Health Insurance), Paris Cedex 20, France, 2CNAMTS (National Health Insurance), paris cedex 20, France . . . . . . Objectives: The aim of this study is to assess health care expenditure by disease in order to provide detailed analyses of resource allocations for 2012, based on the French health care system database. Methods: Using information about 60 millions of individuals from the general scheme insurance database (85% of the French population), we identified all people who received care for each of 56 groups of diseases or medical events or treatments, which are frequent, severe and/or costly. Algorithms have been applied to each patient, using ICD-10 diagnoses for long-term chronic diseases or hospital stays, specific drugs or medical procedures. Costs of all reimbursed expenditures (outpatient/inpatient care, disability/sickness benefits) were extracted per individual for a top-down method allocated expenditure to each of the 56 diseases based on the average expenditure by disease calculated for individuals with only one disease. All expenditures were thereafter extrapolated to the whole population to fit national health account aggregates. Results: Among the 146 billion euros of expenditures (all insurance schemes), 22.1 billion (15%) were related to psychiatric disorders and treatments, 14 (10%) to cancer, 14.6 (10%) to cardiovascular diseases, 7.5 (5%) to diabetes, 6.6 (5%) to vascular-risk treatments, 6.1 (4%) to neurologic disorders, 3.5 (2%) to chronic respiratory diseases, 4.7 (3%) to inflammatory/rare diseases/HIV, 3.5 (2%) to treated chronic renal insufficiency, 1.4 (1%) to liver/pancreas diseases, 1.5 (1%) to chronic pain killer users and 4 (3%) to other long-term chronic diseases. Isolated hospitalizations represented 32.8 billion € (23%), basic care 14.6 (10%) and maternity 9.1 (6%). This analysis also provides detailed patterns of expenditures by diseases. Conclusions: Our study provides helpful information to policy makers by monitoring the performance of the health care system at a disease-based level. This tool will be used to forecast the impact of ageing and epidemiologic patterns on health expenditures. . Objectives: Breast cancer or rather called mastocarcinoma, is a malignant growth of the mammary gland. With an incidence rate of 5.105 people in 2010, breast cancer is the most frequent cancer disease in Austria. Annually about 1.500 cases of death are registered, therefore breast cancer is the leading cause for death among women. As a result, enormous costs arise for the health care system. Hence, the aim of the analysis was to evaluate the cost and disease burden of breast cancer. Methods: A burden of disease study, which evaluates all kinds of disease burden, like costs, quality adjusted life years (QALYs) and disability adjusted life years (DALYs) lost, was conducted. This analysis had an incidence based approach and projects costs over a time horizon of one year. Data were collected via literature review, published list prices and hospital records. Costs from published sources were used from the societal perspective. The direct costs include hospital, treatment and physician consultation costs. Indirect costs cover patient care giver costs as well as work absenteeism. Results: Each breast cancer patient causes direct costs of 6,423.91 EUR and indirect costs of 37,511.87 EUR. Due to the fact that not only costs are important for the health care systems, the mortality rate, QALYs and DALYs were identified. The mortality rate of breast cancer conducts 7.7 percent of all cancer deaths in Austria. Compared to the healthy population, a breast cancer patient has average utility decrements of 0.11. With regard to an incidence rate of 5.105 people in 2010, a breast cancer patient will lose 0.03 DALYs within one year. Conclusions: The analysis showed that the treatment of new diagnosed breast cancer patients causes about 225.5 million EUR. Moreover patients had a lower QALY of 561,55 and a loss of 153,15 DALYs within one year. PCN77 Economic Burden of Asthma in Vietnam: An Analysis from Patients’ Perspective Nguyen T T T 1, Nguyen N B T 2 of Medicine and Pharmacy in HCMC, Ho Chi Minh City, Vietnam, 2University of Medicine and Pharmacy in HCMC, HCMC, Vietnam . . . . . . 1University Objectives: Nowadays, health care costs of asthma are under pressure in all countries due to high prevalence, incidence and the chronic nature of disease. Estimating the economic burden of asthma from patients’ perspective is necessary not only to understand the value and structure, but also the economic influence of asthma to whole society to propose and implement the relevant medical. This is also the aim of this study. Methods: A tree-decision model has been developed to estimate the value and structure of the economic burden of asthma in Vietnam. The economic burden of asthma has been evaluated based on cost of asthma, including direct and indirect costs. A cohort of 227 asthma patients was selected to be surveyed to PCN79 The Burden of Health Care Costs Associated with Prostate Cancer in Ireland Burns R M 1, Leal J 1, Wolstenhome J 1, O’Neill C 2, Sullivan F J 3, Drummond F J 4, Sharp L 4 1University of Oxford, Oxford, UK, 2NUI Galway, Galway, Ireland, 3Prostate Cancer Institutue, Galway, Ireland, 4National Cancer Registry Ireland, Cork, Ireland . . . . . . . . . . Objectives: In 2010,3, 230 men were diagnosed with prostate cancer (PCa) in Ireland, yielding one of the highest incidence rates across Europe. The focus of this paper was to derive average costs for PCa diagnosis, treatment and follow-up, and to estimate the overall health care expenditure for PCa in 2010 in Ireland using a quasi, incidence-based, bottom-up approach. Methods: PCa (ICD10 C61) incidence and treatment data during 2007-2010 was obtained from the National Cancer Registry Ireland. Estimates of resource use for items not recorded by the NCRI were sourced from the literature and expert opinion. Costs associated with detection, diagnosis, treatment, treatment complications, follow-up until year 4 post-diagnosis and terminal care were estimated using various sources including study-specific survey data, Irish hospital costs (HSE Casemix) and costs published in the literature. A payer’s perspective using direct costs was adopted and costs were reported in 2010€ . Results: The overall expenditure associated with prostate specific antigen (PSA) testing for those diagnosed with PCa in 2010 was € 360,016 (average per patient cost (APPC) € 56 (confidence interval (CI): € 45, € 67)). Diagnosis including biopsies and staging was estimated at € 2,872,333 (APPC € 804 (CI: € 573, € 1,114)). Treatment costs (for the first year) varied considerably with active surveillance estimated at € 423,106 (APPC € 655 (CI: € 520, € 814)), external beam radiation therapy (in particular intensity-modulated radiation therapy) estimated at € 8,872,037 (APPC € 6,867 (CI: € 5,429, € 8,235) and radical prostatectomy estimated at € 3,549,864 (APPC € 7,327 (CI: € 5,797, € 8,787). The total burden of health care expenditure associated with PCa in 2010 was approximately € 46.9 million (CI: € 37.4m, € 56.6m). Conclusions: PCa expenditure is substantial and represents a sizeable proportion of the Irish health care budget (€ 11bn). PCa incidence has steadily increased since 1998 due to increasing PSA testing; should trends continue expenditure will increase and this will have implications for resource allocation in Ireland. PCN80 Health Care Costs and Utilization of U. S. Veteran Patients Diagnosed with Pancreatic Cancer Xie L 1, Kariburyo M F 1, Wang Y 1, Baser O 2 1STATinMED Research, Ann Arbor, MI, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . . A628 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To examine the economic burden and health care utilizations of patients diagnosed with pancreatic cancer in the U. S. veteran population. Methods: A retrospective database analysis was performed using Veterans Health Administration Medical SAS data from 01OCT2007 through 30SEP2012. Patients diagnosed with pancreatic cancer were identified using International Classification of Disease 9thRevision Clinical Modification (ICD-9-CM) code 157. The first diagnosis date was defined as the index date. A group of patients with similar age, region, gender and index year but without a pancreatic cancer diagnosis were identified as the comparison group and matched by baseline Charlson Comorbidity Index scores. One year of continuous health plan enrollment was required before and after the index date for both groups. Study outcomes, including health care costs and utilizations, were compared between pancreatic cancer and comparator groups using 1: 1 propensity score matching (PSM). Results: A total of 10,894 patients were identified for the pancreatic cancer and comparison cohorts. After applying a 1: 1 PSM, 3,671 patients were matched from each cohort, and the baseline characteristics were proportionate. Pancreatic cancer patients were more likely to have higher health care resource utilizations, including inpatient admissions (38.03% vs. 2.18%, p< 0.0001), emergency room (ER) (30.67% vs. 5.48%, p< 0.0001), physician office visits (98.39% vs. 57.91%, p< 0.0001) and prescription fills (83.55% vs. 62.82%, p< 0.0001). Risk-adjusted health care costs were also higher for pancreatic cancer patients, including inpatient ($18,079 vs. $442, p< 0.0001), ER ($330 vs. $46, p< 0.0001), physician office ($5,600 vs. $1,106, p< 0.0001) and pharmacy costs ($2,244 vs. $321, p< 0.0001), resulting in higher total costs ($26,503 vs. $1,977, p< 0.0001) relative to the comparator cohort. Conclusions: During a period of 12 months, VHA patients diagnosed with pancreatic cancer reported higher health care resource utilization and costs than their matched controls. PCN81 A Systematic Literature Review of the Economic Burden in Multiple Myeloma Rizzo M 1, Xu Y 2, Panjabi S 3, Iheanacho I 1 . . . . 1Evidera, London, England, 2Evidera, Lexington, MA, USA, 3Onyx Pharmaceuticals, Inc., an Amgen subsidiary, South San Francisco, CA, USA Objectives: To explore the resource use and costs for multiple myeloma (MM) in a systematic literature review. Methods: We searched MEDLINE, Embase, the Cochrane Library, and EconLit for English-language studies published from January 1, 2003 to May 16, 2013 that evaluated resource use and costs related to MM, including relapsed or relapsed and refractory MM (R/RRMM). These studies underwent data abstraction and qualitative synthesis. Results: Thirteen primary studies in the US (n= 5), UK (n= 1), US and UK (n= 1), The Netherlands (n= 2), France (n= 1), Italy (n= 1), Sweden (n= 1), and Switzerland (n= 1) met the inclusion criteria. Three studies were in newly-diagnosed MM patients, one in newly-diagnosed and previouslytreated patients, and nine in R/RRMM. The key cost drivers in R/RMM (as percentages of total costs) were medications (32% to 66%), hospitalisations (15% to 35%), and adverse events/complications/comorbidities (6% to 42%). Among treatment-related adverse events in MM, neutropenia was the main cost driver, accounting for at least two-thirds of total costs for managing such events, followed by thrombocytopenia, anaemia, and infections. In addition, metastatic bone disease significantly increased direct treatment costs (+$57,720 per patient; 2004 USD) and accounted for 17% (€ 6,937 per patient; 2002 € ) of total MM treatment costs. Also, patients with progressive disease incurred costs over three times those in patients without disease progression after initial treatment ($837 vs. $237 PPPM [2010 USD]). Conclusions: The primary drivers of MM management costs are medication, hospitalisations, and adverse events. Neutropenia, skeletal complications and progressive disease are also especially costly in MM. This evidence highlights the need for effective therapies with improved tolerability profiles that delay progression across a broad range of patients with and without comorbidities. PCN82 Exploring the Usefulness of Social Media and Patient Forums in Identifying indirect Costs of a Disease Chalkiadaki C , Martin A Evidera, London, UK . PCN83 Trends in Secondary Care Costs for Treatment of Head and Neck Cancer in England Keeping S T 1, Naylor N R 1, Tempest M J 2, Stephens S 2, Carroll S M 1 Pasteur MSD, Maidenhead, UK, 2Pharmerit Ltd, York, UK . . . . . . . . . 1Sanofi Objectives: Encompassing a group of cancers originating from the upper aerodigestive tract, head and neck cancers are ranked in the top ten for both incidence and mortality among all malignancies globally. This study aimed to estimate trends in total secondary care costs associated with the treatment of head and neck cancers in England from 2006/2007 to 2010/2011. Methods: Data on inpatient and outpatient activity associated with oropharyngeal, oral cavity and laryngeal cancer was extracted from the Hospital Episode Statistics (HES) database. After grouping inpatient episodes into spells, a single Healthcare Resource Group (HRG) was derived for each and then cross-referenced with the National Tariff 2010/11 to estimate the associated cost. For specific types of therapy, including chemotherapy and radiotherapy, HRG definitions were cross-referenced with the National Reference Costs for the latest available year and inflated using the PCI index. Outpatient costs were estimated by grouping consultations by treatment speciality. Results: The total cost of treatment for all cancers over the entire period was estimated to be around £309 million, at 2011 prices. Inpatient care covered by bundled HRGs accounted for over 90% of this cost, at £280 million. Total costs due to oropharyngeal cancer were slightly higher than those estimated for laryngeal and oral cancer, costing £115 million (37.06%), £96 million (31.15%) and £98 million (31.79%) respectively. There was, generally, an increasing trend in the secondary care burden of all three cancers. Annual costs and patient numbers increased the most for oropharyngeal cancer, with annual inpatient costs increasing from £16,576,046 in 2006/07 to £28,467,016 in 2010/11. Conclusions: This study indicates that there is a significant, and increasing, health and economic burden associated with head and neck cancers in England, highlighting the need for preventative programmes. PCN84 Economic Burden of Chemotherapy Related Toxicities in Third Line Metastatic Breast Cancer Patients Majethia U 1, Tremblay G 1, He Y P 1, Faria C 1, Mccutcheon S 2, Kopenhafer L 3, Forsythe A 1 Lake, NJ, USA, 2Eisai Europe, Hatfield, UK, 3Curo Consulting, Marlow, UK . . . . . . . . 1Eisai, Woodcliff Objectives: During chemotherapy, a majority of patients experience treatmentrelated adverse-events (TEAEs). However, the number of patients requiring treatment and/or hospitalization (T&H) for managing TEAEs is significantly lower. The objective of this study was to evaluate economic burden of TEAEs in third line Metastatic Breast Cancer (MBC) patients. Methods: A post-hoc analysis was conducted on Phase 3 clinical trial of eribulin vs. Treatment of Physician’s Choice (TPC) in MBC patients with two prior chemotherapy regimens including anthracycline and taxane. TPC included capecitabine, gemcitabine, vinorelbine, docetaxel and paclitaxel. Grade 3&4 TEAEs which were observed in ≥ 5% of patients were considered. Patients who required T&H for managing TEAEs were further studied. Duration and costs associated with T&H were obtained from a micro-costing study where UK-NHS tariffs for each cost unit were applied. TEAE clinical trial frequency data for each treatment were annualized prior to cost evaluation, based on treatment duration (113 days for eribulin and 68 days for TPC patients). Results: Only a small proportion of patients who experienced TEAEs required T&H. The frequencies of TEAEs experienced vs. TEAEs that required T&H for most commonly observed TEAEs were as follows: neutropenia (Eribulin: 45.2% vs. 14.5%; TPC: 21.1% vs. 5.3%), leukopenia (Eribulin: 13.9% vs 4.2%; TPC: 5.7% vs. 1.6%), peripheral-neuropathy (Eribulin: 8.2% vs. 3.8%; TPC: 2.0% vs. 0.4%) and asthenia/fatigue (Eribulin: 8.8% vs. 1.6%; TPC: 10.1% vs. 2.0%). Monthly TEAEs requiring T&H rates were Eribulin: 19.83%, TPC: 23.67%. Average annual costs for management of TEAEs were £2,621 for eribulin vs. £2,740 for TPC patients. Conclusions: Economic burden of toxicities in MBC patients may be lower than expected as very few TEAEs require T&H. This study may have led to omission of rare but costly AEs because of the patient selection criteria used. Using individual treatments as comparators instead of TPC may have brought additional clarity to the results. Further research is warranted to validate the findings. . Objectives: The objective of this study was to assess the breadth and usefulness of information available in social media and patient-specific forums on indirect costs of a disease. Methods: Internet searches of general social media and patient-specific forums were performed to identify sources of information on the indirect costs of breast cancer and schizophrenia. Results: Searches of general social media sites mainly resulted in posts related to available treatments, health awareness campaigns, non-patient opinions and news articles, as well as irrelevant, often malicious posts, such as direct insults to schizophrenic patients. Fewer, but lengthier, posts were identified on patient forums, which could be an effective way of identifying indirect costs of the diseases. Such costs include time spent off work by the patient and their caregivers, delays in restoring independence such as being able to drive, and non-medical items used by women on chemotherapy. Breast cancer appears to be better represented than schizophrenia, possibly because patients with mental health problems face social stigma and thus don’t communicate their struggle as openly. Topics of conversation across forums include adverse treatment events and their associated costs, as well as other costs incurred by the patients due to their deteriorating health and quality of life. Conclusions: This study looked at available information on the economic burden of breast cancer and schizophrenia on its patients and found that general social media appear to be less useful than patient forums. Depending on the disease being researched, different kinds of information become available and prove to be useful in drawing conclusions about the indirect costs of an illness. The breast cancer community appears to be better represented and more vocal about resource use associated with the disease or treatment than the schizophrenia community. The perspective of caregivers is less represented in breast cancer than schizophrenia. PCN85 The Cost of NSCLC Treatment in Three Countries: France, Germany and UK Mcguire A 1, Martin M 2, Lenz C 3, Sollano J 4 1London School of Economics, London, UK, 2OptumInsight, Uxbridge, Middlesex, UK, 3Pfizer, Berlin, Germany, 4Pfizer, New York, NY, USA . . . . Introduction: Lung cancer is a highly prevalent condition with non-small cell lung cancer (NSCLC) representing ~80%. Given its high prevalence and poor survival rates, it is important to understand costs associated with NSCLC treatment. Objectives: To carry out a study similar to the study by Ramsey (2008) in three European countries: France, Germany and UK. Methods: Three similar administrative databases were accessed: Hospital Episode Statistics (England), Gesundheitsforen Leipzig (Germany), French Hospital Discharge system (France), using ICD-9/10 codes and treatment/surgery algorithms to identify NSCLC patients. An incidence population of NSCLC patients was obtained using an index year (ranging from 2007-2008), ensuring the absence of prior lung cancer (12-months). Data were extracted on treatment information, patient characteristics and disease staging. Average NSCLC treatments were estimated by age and severity. For England 20,081 patients were identified, for France, 15,061, for Germany, 1,038. Results: In-patient length of stay was 8.9, 8.7 and 10.1 days for France, England, Germany respectively for the first year. Hospital in-patient costs in the first year amounted to € 11,667, € 11,363, € 5,985 for France, Germany and England respectively. In year two these were € 5,916 (France), €6,568 (Germany) and € 1,156 (England). Hospital outpatient costs were in year 1 and 2: € 2,313 - € 676 (France); € 1,925 - € 1,766 (Germany); € 1,209 - € 834 (England). Medicine costs in year 1 and 2 amounted to: € 3,542 - € 321 (France); € 4,488 - € 3,805 (Germany); € 8,593 yr1 only (England). Other costs reported A629 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 for France and Germany in year 1 and 2, were € 502 - € 126 for France; € 1,429 - € 1,156 for Germany. Total costs reached € 18,024 (yr1), € 7,039 (yr2) France; € 19,025 (yr1), € 13,295 (yr2) Germany; € 15,785 (yr1), € 1,990 (yr2) England. Two-year costs totalled € 25,063 (France); € 32,500 (Germany); € 17,777 (England). Subgroup analyses showed higher costs for elderly patients, those with non-metastatic disease and smokers. Conclusions: Considerable differences in average treatment costs were observed. In-patient costs dominate in the first year of treatment in all countries. The study highlights the costly nature of NSCLC. PCN86 Cost of Best Supportive Care for Non-Small Cell Lung Cancer Patients – A German Perspective Schmidt U 1, Lipp R 2, Drechsler M 1 Ingelheim Pharma GmbH & Co. KG, Ingelheim, Germany, 2GermanOncology, Hamburg, Germany . . . 1Boehringer Objectives: Best supportive care (BSC) is in general individually provided to patients. Thus, the scope of BSC and its costs can vary widely. Only limited information on BSC costs for patients with non-small cell lung cancer (NSCLC) exists. Aim of this research was to estimate annual BSC costs for NSCLC patients in Germany. Methods: To estimate BSC costs at first a literature search in PubMed with the key words “best supportive care”, “cost”, “non-small cell lung cancer” and “economic analysis” individual and combined search terms was performed. International publications of economic evaluations including data on single cost items which could be applied to the German health care system were included. Additionally, data on BSC from a NSCLC patient registry (n= 193) and prescription data derived from a randomised controlled trial (RCT) were used as further references. Cost- items were extracted from each reference and finally applied to the German inpatient and outpatient reimbursement system. Results: The literature research yielded 317 records of which 3 met the inclusion criteria (2= UK; 1= North America). An extrapolation of these evaluation results to the German health care system showed that BSC costs based on UK data ranged from 16,940€ to 45,426€ (North America: 31,352€ ). According to health economic data from an RCT annual cost for BSC amounted to 17,531€ , while data from a NSCLC patient registry added up to 28,070€ . The average annual costs for BSC were estimated at 27,864€ . Conclusions: Since BSC is individually delivered to patients, it leads to a high variance of annual BSC costs for NSCLC patients in Germany. Furthermore, international economic evaluations were extrapolated to the German health care system. Hence, results should be interpreted with caution as international treatment guidelines and reimbursement schemes are not fully applicable to Germany. Future analyses should be based on a German population only. PCN87 Mastectomy Due to Breast Cancer in Brazil: Geographic Distribution and Costs from the Public Health Care Perspective Valle P M 1, Mosegui G B G 2, Vianna C 1, Araujo R L 1 Estadual do Rio de Janeiro, Rio de Janeiro, Brazil, 2Universidade Federal Fluminense, Niterói, Brazil . . . . . . . . 1Universidade Objectives: Treatment for breast cancer is usually based on chemotherapy and radiotherapy, but in unsuccessful cases, mastectomy is required. In Brazil, mastectomies are performed as simple or radical, with lymphadenectomy. This study aims to relate geographic distribution, temporal trends and economic profile of this procedure in Brazil. Methods: Assessments about hospital admissions were performed to analyze costs and geographic distribution among mastectomies related with breast cancer realized in Brazil, from January 2008 to December 2012. The data used were extracted by Brazilian Hospital Information System (SIH/SUS) database, according to ICD-0416120032 (simple mastectomy) and ICD-0416120024 (radical with lymphadenectomy). Costs were estimated in 2014 Brazilian Real (BRL) and represents federal reimbursement values for hospitalizations (exams, drugs, medical procedures and fees). Results: In Brazil, the number of mastectomies related with breast cancer ranged from 8,687 in 2008 to 9,703 in 2012. In Southeast region were performed 22,977 procedures, which was the largest number per region, compared with 1,814 in North; 8,443 in South; 2,280 in Midwest and 10,538 in Northwest. In São Paulo; 10,111 procedures occurred in this period, while 8 were performed in Amapá. Total costs with mastectomies due to breast cancer in Brazil during this period were 44,219,235.66 BRL. The value per patient increased about 12,5% over the years with mean costs from 2008 to 2012 of 799.75BRL, 842.62BRL, 845.28BRL, 878.20BRL and 870.25BRL, respectively. In 2010, the mean mortality rate among simple and radical mastectomies was 0,22% and increased until 0,54% in 2012. Conclusions: Geographic distribution of mastectomy due to breast cancer in Brazil is concentrated in Southeast region, in comparison to other regions and from 2008 to 2012 there was no change in this pattern. Although costs elevated, mortality also increased in this period. PCN88 Estimation of Economic Losses Resulting from Diseases Associated with Smoking in Mexican Insured and Uninsured Population Muciño-Ortega E 1, Hernandez-Reyes F C 1, Reynales-Shigematsu L 2 S.A. de C.V., Ciudad de México, Mexico, 2Instituto Nacional de Salud Pública, Cuernavaca, Mexico . . . . 1Pfizer Objectives: Few data has been published in Mexico related to direct smoking costs from the consumer perspective. To estimate lost income and pocket costs related to smoking diseases from the perspective of Mexican families. Methods: The analysis incorporated patients with public, private and no-health insurance from 18-65 years. Through Mexican literature review, medical costs and prevalence of 3 major diseases associated with tobacco consumption were identified: acute myocardial infarction (AMI), chronic obstructive pulmonary disease (COPD) and lung cancer (LC) as well as absenteeism associated to each disease. Average income was extracted from 2012 national income survey. Scenarios evaluated were: 1) Publicly health-insured patients: from the 4th day of disability, the Instituto Mexicano del Seguro Social covers 60% of wages, 2) Private health-insured: economical loss of private beneficiaries is a 20% co-pay plus daily average income lost. 3) No-healthinsurance: Medical costs and absenteeism represent the economical loss. Costs are expressed in 2014 USD (1USD= 13MXN). Morbidity cases were extracted from published data by National Institute of Public Health and used to calculate a weighted average of economical losses for each scenario. Results: The distribution of people with smoking-related disease (AMI, COPD and LC) was 25.0%, 65.4% and 9.6%, respectively. Patients in scenario 1) 2) and 3) spend/lose an average of $446.6, $8,448.2 and $37,384.95 per year, respectively (medical costs are the drivers of the economic resources lost by not-insured population). Average illness expense derived from tobacco consumption regarding the proportion of people in each scenario was $13,917 yearly (local per capita GDP is $9,749). Regardless of health-insurance status, LC is the most expensive disease ($39,564.26), followed by AMI and COPD ($14,337.06 and $10,109.32, respectively). Conclusions: The study showed that in the long run smokers incur significant economic losses even if they have medical insurance. Costs increases to people that do not have any insurance. PCN89 Cost Comparison Among First Line Monoclonal Antibodies-Based Oncology Treatment Protocols Jakovljevic M 1, Gutzwiller F S 2, Schwenkglenks M 2, Milovanovic O 3, Rancic N 4, Varjacic M 3, Stojadinovic D 5, Dagovic A 6, Matter-Walstra K 2 1The Faculty of Medical Sciences University of Kragujevac, Kragujevac, Serbia and Montenegro, 2University of Basel, Basel, Switzerland, 3Faculty of Medical Sciences, University of Kragujevac, Kragujevac, Serbia and Montenegro, 4Military Medical Academy University of Defence Belgrade, Belgrade, Serbia and Montenegro, 5Urology Clinic, University Clinical Center Kragujevac, Kragujevac, Serbia and Montenegro, 6Oncology and Radiation Therapy Center, Clinical Center Kragujevac, Kragujevac, Serbia and Montenegro . . . . . . . . . . Objectives: To assess and compare the costs of first-line monoclonal antibodies (mABs) treatment protocols in breast cancer, non-Hodgkin lymphoma and colorectal carcinoma in South-Εastern Europe. Methods: A retrospective, bottom-up case series study design was implemented with one-year time horizon and payer’s perspective. The study sample size was 265 patients (breast cancer, N= 137, colorectal cancer, N= 44, and non-Hodgkin lymphoma, N= 84) while treatment protocols included adjuvant mAbs: trastuzumab (N= 137), bevacizumab (N= 28), cetuximab (N= 16) and rituximab (N= 84). ICD-10 related, direct medical and lost productivity costs (€ ) across treatment groups during 2010-2013. Results: The average length of observation was 128±97 days per patient. Total mean direct and indirect costs of care were: trastuzumab breast cancer group € 17,740; bevacizumab colorectal carcinoma group € 8,775; cetuximab colorectal carcinoma group € 27,181 and rituximab non-Hodgkin lymphoma group € 19,431. An average mAbs-treated patient incurred € 17,897 costs of medical care. The total combined budget of these 265 patients was € 4,742,775. Conclusions: The use of mAbs strongly correlated with high costs in first-line cancer medical care and dominated other cost domains. Cetuximab-based treatment protocol in colorectal carcinoma patients was substantially more expensive compared to trastuzumab (C50); bevacizumab (C20) and rituximab (C80) alternatives. Extremely high costs of mAbs are the key-issue for Eastern European policy makers by crossing the upper limits of affordability in middle-income economies. PCN90 Use Patterns and Costs of Isolated Limb Perfusion and Infusion in the Treatment of Regionally Metastatic Melanoma: A Retrospective Database Analysis Ma Q , Zhao Z , Barber B , Shilkrut M Amgen Inc, Thousand Oaks, CA, USA . . . . Objectives: Isolated limb perfusion and infusion (ILP/ILI) are therapies for regionally metastatic melanoma where high doses of anticancer drugs are delivered directly into the circulation of an affected limb, while minimizing systemic drug exposure. This procedure can lead to high response rates but without proven benefits to overall survival. It is recommended by ESMO and NCCN guidelines as a treatment option for patients with stage III unresectable metastatic melanoma. However, limited information is available on its use pattern and costs in the literature. This study was to examine patterns of ILP/ILI use and associated costs in patients with melanoma in the US. Methods: This is a retrospective, observational study using large administrative claims from the MarketScan® databases. Patients who underwent ILP/ILI (CPT-4: 36823) with diagnosis of melanoma (ICD-9-CM: 172. xx, V10.82) between 1/1/2002 and 3/31/2013 were included. Patient characteristics, use patterns, hospital length of stay, and costs (2013 US $) of ILP/ILI were assessed. Results: A total of 113 patients met the study criteria and were included in the analysis. The mean age was 62.1 years (standard deviation [SD] 14.1); 39.8% were male. The mean baseline Charlson’s comorbidity index was 0.24 and 36.4% of patients were Medicare beneficiaries. Overall, 86.4% of patients had melanoma in the lower limb, 12.7% in the upper limb, and 0.9% in both upper and lower limbs; 59.3% had lymph node metastasis and 56.8% had skin metastasis. Four patients (3.5%) underwent multiple ILP/ILI procedures. The mean (±SD) hospital length of stay was 5.6 (± 3.5) days and the mean (±SD) cost was $35,898 (± $26,492) per ILP/ILI procedure. Conclusions: The use of isolated limb perfusion and infusion was associated with relatively long hospital stay and high cost. The results of this study may provide source data for economic evaluations of treatment options for regionally metastatic melanoma. PCN91 A Guideline-Based Estimate of Health Care Resource Use and Cost of Metastatic Unresectable Osteosarcoma Cornelio N 1, Burudpakdee C 2 of North Carolina at Charlotte, Charlotte, NC, USA, 2MKTXS, Raritan, NJ, USA . . 1University Objectives: To estimate the resource use and costs a health plan can anticipate during the diagnosis, treatment, and surveillance of a patient with metastatic unre- A630 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 sectable osteosarcoma using national guideline recommendations. Methods: An economic disease model was developed based on recommendations from the 2013 NCCN Clinical Practice Guidelines in Oncology for bone cancer. The model quantified resource use for diagnosis, 12 months of treatment, and 12 months of surveillance of a metastatic unresectable osteosarcoma patient. Costs in 2014 dollar value were derived from publically available sources for reimbursement of CPT codes, HCPCS codes, and generic WAC prices for medications. Chemotherapy dosing was based on NCCN recommended treatment regimens. Results: The diagnostic cost was estimated to be $1,706 per patient. Treatment costs, consisting of stereotactic radiosurgery and chemotherapy with drug monitoring, varied widely across the four NCCN recommended regimens due to differences in the price of pharmacotherapy. The chemotherapy regimens were estimated to be the major cost components associated with this disease. Doxorubicin, cisplatin, and high-dose methotrexate cost $103,051 per patient; doxorubicin and cisplatin cost $17,549 per patient; doxorubicin, cisplatin, high-dose methotrexate, and ifosfamide cost $38,404 per patient; and cisplatin, ifosfamide, and epirubicin cost $38,936 per patient. Additionally, stereotactic radiosurgery was estimated at $2,755 per patient, and the cost of drug monitoring during the one year of chemotherapy averaged to $5,899 per patient. Additionally, one year of disease surveillance cost $4,264 per patient. Conclusions: A guideline-based disease model can assist health plans to better understand and anticipate the expected diagnosis, treatment, and surveillance resources and costs for unresectable metastatic osteosarcoma patients. PCN92 Resource Use and Health Care Costs of Metastatic Malignant Melanoma in Slovakia Ondrusova M 1, Psenkova M 1, Hlavata Z 2, Visnovska M 3, Urbancek S 4 Ltd, Bratislava, Slovak Republic, 2National Cancer Institute, Bratislava, Slovak Republic, 3East Slovak Cancer Institute, Kosice, Slovak Republic, 4F.D. Roosvelt University Hospital, Banska Bystrica, Slovak Republic . . . . . 1Pharm-In Objectives: The objective of this cost study was to measure the resource utilisation and the direct costs associated with health care management of metastatic malignant melanoma (mMM) in Slovakia and provide a basis for cost-effectiveness evaluations. Methods: The cross-sectional survey was performed and included 3 oncologists experienced in mMM management. The survey was performed to obtain the information on the management of patients with mMM and to estimate the direct costs of the disease. The studied population were 3 cohorts of mMM patients which are usually identified as the health states in the cost-effectiveness models: “Before progression“, “Disease progression“ and “Terminal care“. Costs of drugs were assesed separately from health states and rated particularly according to BRAF positivity. The cost data were assessed for the year 2013. All types of health care used in mMM management were evaluated (outpatient and inpatient visits, diagnostics, prescription drugs and medical examinations). Costs of adverse events (AEs) were set for one single event. Results: The most frequent treatment regimens used in the first treatment line of BRAF mutant and BRAF negative patients were identical - dacarbazin (94.9% of treated patients), fotemustin (4.5%) and ipilimumab (0.6%). Monthly costs of mMM management in addition to the active treatment in the state “Before progression“ count for 6.64% (€ 188.51/patient), during the “Disease progression“ it was 45.56% (€ 1 294.31/patient) and during the “Terminal state of patient“ 47.80% (€ 1 358.02/patient). Adverse event (AE) costs were evaluated for grade 3 and 4. The most costly AEs were neutropenia (€ 1 014.66), fever (€ 364.87) and rash (€ 230.35). Conclusions: In the management of mMM (excluding the active drug cost), the most expensive are the costs of hospitalization and symptomatic treatment. The most costly period is the “Terminal state“. PCN93 Cost-Benefit Assessment of the Electronic Health Records for Clinical Research (EHR4CR) European Project Beresniak A 1, Schmidt A 2, Proeve J 3, Bolanos E 4, Patel N 5, Ammour N 6, Sundgren M 7, Ericson M 8, De Moor G 9, Kalra D 10, Dupont D 1 Mining International, Geneva, Switzerland, 2F Hoffmann-La Roche Ltd, Basel, Switzerland, 3Bayer Healthcare, Leverkusen, Germany, 4Eli Lilly and Company, Alcobendas, Spain, 5Eli Lilly and Company (until December 2013), Windlesham, Surrey, UK, 6Sanofi-Aventis R&D, Chilly-Mazarin, France, 7AstraZeneca, Mölndal, Sweden, 8Amgen, Neuilly-sur-Seine, France, 9University of Ghent, Ghent, Belgium, 10The European Institute for Health Records (EuroRec), London, UK . . . . . . . . . . . 1Data Objectives: The EHR4CR 4-year research partnership between the European Union and the European Federation of Pharmaceutical Industries and Associations (EFPIA) has developed a platform for the trustworthy reuse of hospital electronic health records’ data for clinical research. A cost-benefit assessment (CBA) was conducted from the pharmaceutical industry perspective to assess the value of the first two EHR4CR clinical research scenarios (S): Protocol feasibility assessment (S1), and Patient identification and recruitment (S2), either used individually or sequentially within a clinical trial workflow, versus current practices. Methods: The EFPIA partners have conducted a resource utilization assessment to calculate the actual person-time and cost of performing S1 and S2 for one oncology clinical study (Phase II or Phase III) as reference case. Assuming that an estimated 50% reduction in actual person-time and cost under EHR4CR conditions would directly translate in accelerated time to market (TTM), potential benefits to global pharmaceutical industry were derived using global market values (2012) of oncology products1. Absolute cost-benefit analyses were conducted using Monte-Carlo simulations. Results: Compared to current practices, individual EHR4CR scenarios S1 and S2 have yielded efficiency gains of 134 days and 37 days respectively, and of 171 days when used sequentially. Should these efficiency gains from study design optimisation translate in faster TTM, corresponding estimated benefits for the global pharmaceutical oncology franchise could reach 160,45, and 205 Million € , respectively. Conclusions: This CBA is the first to assess the value of EHR4CR scenarios for oncology clinical trials. The results confirm that the EHR4CR platform could generate substantial added value for pharmaceutical industry should its efficiency gains translate in faster TTM. Further benefits are expected from the EHR4CR platform in other therapeutic areas. Disclosure: The EHR4CR project is mandated by the Innovative Medicines Initiative (co-funded by the European Commission and EFPIA). 1. Evaluate Pharma September 2013 PCN94 Cost-Effectiveness of Colonic Stents for the Management of Malignant Large Bowel Obstruction Goodall S 1, Church J 2 of Technology, Sydney, Sydney, Australia, 2University of Technology Sydney, Sydney, Australia . . 1University Objectives: The aim was to determine the cost-effectiveness of colonic stent insertion for the management of malignant bowel obstructions. Colonic stents are a minimally invasive alternative to open surgery for patients medically unfit for single stage surgery. Methods: Two economic models were developed. The first compared patients who received palliative or definitive stents and were not medically fit for re-anastomosis. The second compared patients who received stents as a bridge-tosurgery and were medically fit for a second stage of two-stage surgery, this included colostomy or Hartmann’s procedure. ResultsFor patients requiring palliation, the cost of colonic stent insertion was estimated to be $17,809 compared to $20,516 for palliative colostomy (a saving of $2,707). The benefits associated with both procedures were 0.099 QALYs and 0.089 QALYs gained, respectively, an incremental benefit of 0.01 QALYs per patient. For patients requiring a bridge-to-surgery, the cost of colonic stent insertion was estimated to be $29,729, compared to $30,169 for patients that received multi-stage surgery (either a colostomy or a Hartmann’s procedure). This represented a cost savings of $440. The estimated average patient would gain 0.510 QALYs compared to 0.458 QALYs in the multi-stage surgery group. This yields an incremental benefit of 0.052 QALYs per patient. The main drivers of both models were the technical and clinical success of the stent insertion, and length of hospital stay following the procedures. The probability of a resection with primary anastomosis after insertion of a stent and the cost of stenting were also drivers in the bridge-to-surgery model. Conclusions: In terms of cost-effectiveness, colonic stent insertion for malignant bowel obstruction in patients requiring palliation or a bridge-to-surgery dominated the current alternative surgical procedures. PCN95 A Multi-State Model of Metatstatic Colorectal Cancer van Rooijen E M 1, Coupé V M H 2, Koopman M 3, Punt C J A 4, Uyl-De Groot C A 5 for Medical Technology Assessment, Erasmus University, Rotterdam, The Netherlands, 2VU University Medical Centre, Amsterdam, The Netherlands, 3University medical centre Utrecht, Utrecht, The Netherlands, 4Academic Medical Centre Amsterdam, Amsterdam, The Netherlands, 5Institute for Medical Technology Assessment (iMTA), Erasmus University, Rotterdam, The Netherlands . . . . . . . . . . . 1Institute Objectives: The aim of this study is to develop and validate a decision-analytic model describing the current course of disease, including treatment, in metastatic colorectal cancer. This baseline model will serve as the comparator in analyses of the (cost-) effectiveness of new treatment strategies. Methods: An individual-based micro-simulation model was constructed based on the disease states a patient may experience after a diagnosis of metastatic colorectal cancer. The states include first-line second-line and third-line treatment, as well as states of progression of disease after first-, second- or third-line, finally a death state is included. Time spent in each disease state was predicted using log-logistic, log-normal or weibull survival models, each dependent on a number of patient characteristics. All survival models and patient characteristics were based on patient-level data, provided by the CAIRO trial (NCT00312000). Two oncologists evaluated the model for face validity, the model was further validated by comparing various model outcomes with the original data, the national cancer registry and a population based study. Results: There were no significant differences in patient and treatment characteristics, nor intermediate and overall survival estimates between the simulated and original patient-level data. External validation with national cancer registry data showed few differences in survival with the simulated data. Additionally the simulated survival did not significantly differ from the survival as recorded in a pilot oxaliplatin study of 119 patients who were observed in the same timeframe as the RCT. Conclusions: The micro-simulation decision model described in this article underwent an internal and external validation and can be used to evaluate new possibilities for research and treatment in metastatic colorectal cancer. PCN96 Economic Consequences of the Adaption of the 21 Gene Reverse Transcriptase-Polymerase Chain Reaction RT-PCR Assay from The Greek Third Payer Perspective Kikilias N 1, Siskou O 2, Kaitelidou D 3, Galanis P 3, Tsoulos N 4, Vafeiadis J 1, Liaropoulos L 3 Organization for Health Care Services Provision -EOPYY, Marousi, Greece, 2Center for Health Services Management and Evaluation, National and Kapodistrian University of Athens, Athens, Greece, 3National and Kapodistrian University of Athens, Athens, Greece, 4GENEKOR, Gerakas, Greece . . . . . . . 1National Objectives: The evaluation of the economic consequences of 21-gene RT-PCR assay OncotypeDX introduction to the reimbursement scheme of National Organization for Health Provision-EOPYY. Methods: A decision tree was developed concerning two treatment scenarios for the year 2013: a) chemotherapy admission according the common treatment practice without the application of the test vs b) chemotherapy admission depending on the results of the test. The sub-group of breast cancer patients appropriate for applying the test was determined according international guidelines and included early stage breast cancer women with hormone receptor positive and lumph node negative age ≤65 years. Cancer incidence was derived from ELSTAT and OECD base, while some assumptions were made concerning the age structure and disease stage of the population. The percentages of women assessed as high risk (score> 31) for recurrence were obtained from EOPYY data for 2013. The estimated cost for OncotypeDX test was set according the EOPYY reimbursement price (€ 2,848 for 2013). Cost of chemotherapy and other cost items (eg laboratory A631 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 tests) were based on official reimbursed prices. Results: Out of 4,934 newly diagnosed breast cancer women, 27.5% (1,357) were appropriate for the test application. Only 35% of the women undertaken the oncotype test (N= 1,357) were found as high risk for recurrence (N= 475). The average total cost of chemotherapy treatment was estimated to € 8,271 from which more than 80% refer to pharmaceuticals. The total treatment cost for women who didn’t undertake the test reached to € 10.9 mil., while the relevant cost for women who undertook the test was estimated to € 8 mil. Conclusions: The introduction of Oncotype DX® to the Greek health care system had as a result annual cost savings of almost € 3 million and avoidance of unnecessary chemotherapy treatment (and associated complications) to more than 880 women. PCN97 Cost Consequence Model Investigating the Impact of Bowel Cleansing on Prevention of Colorectal Cancer in a German Screening Population Fischbach W 1, Pohl J 2, Scola A M 3, Conway P 4 1Klinikum Aschaffenburg, Aschaffenburg, Germany, 2Horst Schmidt Klinik, Wiesbaden, Germany, 3McCann Complete Medical, Macclesfield, UK, 4Norgine Ltd, Harefield, UK . . . . . Objectives: The degree of benefit from colonoscopy in the prevention of colorectal cancer (CRC) is highly dependent on the quality of bowel cleansing. In a randomized study of patients undergoing screening colonoscopy in Germany (MODEC), 2L polyethylene glycol with electrolytes + ascorbate components (PEG+ASC) resulted in numerically higher overall polyp/adenoma detection rates (PDR/ADR) and significantly higher right-sided PDR/ADR than sodium picosulfate/magnesium citrate (NaPic/MgCit), together with better bowel cleansing. The objective of the model was to examine the socioeconomic impact of bowel cleansing quality on the effectiveness of CRC screening in the eligible German population. Methods: A costconsequence model was constructed to compare the total cost of colonoscopy and treatment of subsequent CRC over a 10-year period in a cohort of 10,000 patients aged ≥ 55 years receiving 2L PEG+ASC or NaPic/MgCit prior to colonoscopy. The rates of successful bowel cleansing, completed colonoscopies, and PDR/ADR were obtained from the MODEC study. Published rates of surveillance colonoscopy, associated costs and health care resource utilization in Germany were used, with costs inflated to 2013 prices. Results: The model predicts that the use of 2L PEG+ASC versus NaPic/MgCit increases the average per patient cost associated with colonoscopy by € 67. However, better bowel cleansing and numerically higher overall PDR/ ADR achieved using 2L PEG+ASC rather than NaPic/MgCit avoids progression to CRC in 166 patients, due to early detection, equating to an average per patient saving in CRC treatment costs of € 488. The model shows that the use of 2L PEG+ASC versus NaPic/MgCit at screening/surveillance colonoscopy leads to an average overall cost saving of € 420 per patient over 10 years. Conclusions: Modeling of long-term outcomes shows that using an effective bowel cleansing preparation in CRC screening may reduce the number of patients developing CRC, and may lead to reduced CRC treatment costs. PCN98 A Cost-Consequence Analysis of Human Papillomavirus Vaccination in Romania Preda A L 1, Moise M 1, Van Kriekinge G 2 . . . . 1GlaxoSmithKline, Bucharest, Romania, 2GlaxoSmithKline Vaccines, Wavre, Belgium Objectives: The objective of the study was to estimate the potential cost and epidemiological impact of a Human Papillomavirus (HPV) mass vaccination in Romania for the two available vaccines in the Romania: AS04 adjuvanted HPV16/18 vaccine (AS04V) and the HPV6/11/16/18 vaccine (QV). Methods: We applied, to the Romanian settings, a population steady state model previously published with a one year time horizon estimating the effect (cases and costs) of a vaccination programme. The number of cases and costs (in RON – Romanian National Currency) were collected from the hospitalization Diagnosis Related Group database for the year 2012. cervical cancer (CC) and genital warts (GW) were considered. Vaccine effectiveness was approximated by weighting vaccine-type and non-vaccine-type efficacy with HPV distribution reported for GW (literature) and CC (HPV Centre) for each vaccine. One way sensitivity analysis was conducted on key input parameters. Results: HPV vaccination would save17,706,490 RON with AS04V and 16,432,592 RON with QV. An additional 820 CC-related hospitalisations amounting to a cost difference of 1,273,898 RON was estimated in favour of the AS04V. A total of 205 cases of GW prevented and 153,395 RON associated costs were estimated in favour of the QV. The total cost difference amounted to 1,120,503 RON. Robustness of the results was confirmed by sensitivity analyses”. Conclusions: Implementing the AS04V would result in > 1 million RON saved versus the QV mainly due to a difference of extra 820 CC cases prevented that completely offsets the benefit associated with the prevention of GW. The observed difference is mainly due to higher protection associated with non-vaccine types for AS04V. PCN99 Association of Health Care Cost with Quality of Life for Various Types of Cancers Hayran M , Yuce D , Huseyin B , Esin E , Kilickap S , Erman M , Celik I Hacettepe University, Ankara, Turkey . . . . . . . Objectives: The new cancer treatment modalities are improving survival rates today and one of the main outcomes is the improved health related quality of life (QoL). Prolonged survival also increases the financial burden of cancer care on health care systems. In this study we aimed to analyze the associations between QoL and direct health care costs in different types of cancers. Methods: We evaluated QoL (EORTC QLQ-C30) and direct medical costs (DMC) in 350 patients with lung (Lng), breast (Br), hematological (Hem), head and neck (H&N), colorectal (CR), gastric (Gas), gynecological (Gy), and prostate (Pr) cancers. The DMC data of each patient in the following 3-month period after QoL assessment was obtained from the hospital finance department database. DMC per QoL point was calculated by DMC/QoL score. Results: Mean DMC per QoL was lowest in Pr, and highest in Hem cancers (ranged 60.1-195.1 TL/global QoL score) (Pr< Gy< CR≈Gas≈Br≈Lng≈H&N< Hem). QoL was lowest in Gy and highest in CR (ranged 53,1-65,2) (Gy< Lng< Pr< Br≈Hem≈H&NC). Total DMC ranged from 3124-13557 TL (Pr≈GY< Br< Gas≈CR< Lng≈H&N< Hem). Depending on the type of the cancer the association between DMC and QoL could be in different directions (the correlation between DMC and role functioning was positive in Gas, while it was negative in H&N cancer). Conclusions: For a fixed period of time the total DMC associated with the management of different types of cancers vary substantially. As expected the total cost does not however purchase equal amount of QoL for each type of cancer. For those cancers with higher DMC per QoL, we should consider implementing wider psychosocial support measures. Depending on the type of cancer DMC may reflect disease progression leading to decreased QoL, or it may reflect presence of an effective and aggressive management leading to increased QoL. PCN100 Cost-Effectiveness Model of Pertuzumab in Combination with Trastuzumab and Docetaxel Compared with Trastuzumab in Combination with Docetaxel for the 1st Line Treatment of HER2+ Metastatic Breast Cancer in Colombia Saenz Ariza S A Productos Roche, Bogota, Colombia . . Objectives: To evaluate the cost effectiveness of Pertuzumab plus Trastuzumab and docetaxel (PTD) vs. Trastuzumab and docetaxel (TD) for the first-line treatment in patients with HER2+ metastatic breast cancer in Colombia. Methods: For the evaluation of the cost-effectiveness a health economic area under the curve model was developed. The model considers three health states: progression-free survival, disease progression and death. The proportion of patients in each health state were derived using patient level data from the CLEOPATRA trial like efficacy and safety results, with the exception where overall survival (OS) utilized data from longer term clinical registries. The primary model outcome is the ICER cost per QALY gained in the first-line PTD vs. TD. The following main model input data assumptions were applied for the base case analysis: Time horizon: 15 years; model cycle length: weekly; 3) reference prices for drugs in Colombia, except for Pertuzumab which was supplied by the manufacturer; 4) treatment duration: actual treatment duration from the CLEOPATRA study extrapolated using an exponential function; and 5) discount rates: annual rate of 3.0% for both, future costs and health benefits. Results: The outcomes over a time horizon show an increase in mean OS time for patients assigned to the PTD group as compared to those in the TD of 0.72 years. Mean QALYs are also higher in the PTD group than in the TD group 0.58 QALYs. The addition of Pertuzumab leads to higher total average treatment costs of $143.529 dollars per patient compared to the TD group. These findings result in an ICER of $200.509 per life year gained and of $249.582 per QALY gained. Conclusions: When compared to commonly accepted cost-effectiveness thresholds, these results exceed commonly applied willingness-to-pay thresholds, but Pertuzumab becomes a therapeutic alternative that offers a better health outcome. PCN101 Cost-Effectiveness of Ipilimumab for Previously Untreated Patients with Advanced Metastatic Melanoma in Spain Aceituno S 1, Canal C 2, Paz S 3, Gonzalez P 2, Marquez-Rodas I 4 10, Castellón, Spain, 2BMS Spain, Madrid, Spain, 3Outcomes’10, Castellon, Spain, 4Instituto de Investigación Sanitaria Gregorio Marañón, Madrid, Spain . . . . . 1Outcomes Objectives: To assess the cost-effectiveness of ipilimumab compared to dacarbazine as first-line treatment in patients with advanced metastatic melanoma. Methods: A three-state Markov (progression-free, progression and death) with three-week cycles model using the Spanish Healthcare System perspective was developed over a lifetime horizon. The clinical profile of ipilimumab (3 mg/kg) was obtained from a pooled dataset of chemotherapy naive patients from four phase II and phase III studies, and from the CA184-024 trial for dacarbazine. Parametric extrapolation methods were used to project survival over lifetime. Costs included were: drug acquisition (ex-factory price -7,5% mandatory rebate) and administration, medical/terminal care, and adverse events management. Unit costs were derived from Spanish health care cost databases (Euros, 2013). For drugs with a double-pricing system (like ipilimumab), costs were based upon the official notified prices in Spain. Costs and benefits were discounted at 3%. Utility values were taken from the CA184-024 trial. Univariate and probabilistic sensitivity analyses (PSA) were performed. Results: The life years (LYs) and quality-adjusted life years (QALYs) gained with ipilimumab as first-line treatment over dacarbazine were 2.01 and 1.68, respectively. The incremental cost of using ipilimumab versus dacarbazine was 69,598 € . The incremental cost-effectiveness ratio (ICER) and the incremental cost-utility ratio (ICUR) were 34,566€ /LY gained and 41,459€ /QALY, respectively. PSA showed that ipilimumab is up to 100% and 90% likely to be cost-effective at the threshold established by the NICE for oncology drugs that meet ‘End-of-Life’ criteria (50,000-62,000€ ) for ICER and ICUR, respectively. Additionally, at the threshold acceptable in Spain (30,000-45,000€ ) the likelihood of ipilimumab being cost-effective is up to 94% and 66% for ICER and ICUR, respectively. Conclusions: Results suggest that ipilimumab is a cost-effective alternative for previously untreated patients with advanced metastatic melanomain Spain. PCN102 The Potential of (TARGETED) MR Colonography as a Screening Tool for Colorectal Cancer: A Cost-Effectiveness Analysis Greuter M J 1, Demirel E 1, Berkhof J 1, Fijneman R J 1, Stoker J 2, Meijer G A 1, Coupé V M 1 University Medical Center, Amsterdam, The Netherlands, 2Academic Medical Center, Amsterdam, The Netherlands . . . . . . . . . . . 1VU Objectives: MR colonography may have potential as a colorectal cancer (CRC) screening tool since it has comparable test characteristics as colonoscopy but is less invasive. Furthermore, innovators in the field of MR technology are striv- A632 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 ing to develop a targeted contrast agent that specifically detects adenomas at increased risk of progressing to CRC. This might even further raise the potential of MR colonography. We explored the potential of conventional and targeted MR colonography in terms of (cost-) effectiveness using the ASCCA model. Methods: Thirteen screening strategies were evaluated, differing in primary screening instrument and number of screening rounds. The strategies under consideration were conventional and targeted MR colonography, colonoscopy and CT colonography with two, three and four screening rounds at a ten year screening interval. Furthermore, eleven rounds of biennial faecal immunochemical test (FIT) screening were considered. Each strategy was evaluated assuming realistic and perfect participation rates. Incremental costs and effects were estimated from a societal perspective. Results: All screening strategies were cost-effective compared to no screening. For conventional MR colonography, the ICER ranged between € 1,271/ LYG to € 3,003/LYG for two to four screening rounds at 34% participation per round. For 62% and 100% participation, the ICER ranged from respectively € 1,576/LYG to € 3,777/LYG and € 1,971/LYG to € 4,577/LYG. However, conventional MR colonography screening was more expensive than other screening strategies at comparable LYG, for all participation rates. Targeted MR colonography was only slightly more effective than conventional MR colonography but considerably more costly, even under the most favourable assumptions regarding test characteristics and costs per test. Conclusions: This is the first study to evaluate the cost-effectiveness of MR colonography screening for CRC. Although conventional and targeted MR colonography are cost-effective compared to no screening, at present they cannot compete with more established screening tests because of the high costs per test. PCN103 Cost-Effectiveness Analysis of Abiraterone Acetate Treatment Compared with Cabacitaxel in the Republic of Panama, in Patients with Metastatic Castration-Resistant Prostate Cancer that Have Failed to Chemotherapy with Docetaxel Obando C A 1, Desanvicente-Celis Z 1, Gonzalez L 2, Muschett D 1, Gonzalez F 1, Goldberg P 1 . . . . . . . 1Janssen, Panama, Panama, 2Janssen, Raritan, NJ, USA Objectives: To assess the cost-effectiveness of Abiraterone Acetate plus Prednisone (A-P) compared with Cabazitaxel plus Prednisone (C-P) in Panama, in patients with Metastatic Castration-Resistant Prostate Cancer (mCRPC) that have failed to chemotherapy with Docetaxel. Methods: A three-health state cohort simulation Markov Model (progression-free, post-progression and death) was developed based on overall and progression free survival data. The time frame was 10 years. The perspective was that of the Public System of Health of Panama. The health outcomes of interest were Quality Adjusted Life Years (QALYs) and Life Years (LYs). Efficacy data was taken from clinical trials (COU-AA-301 for A-P and TROPIC for C-P). Utilities for health states and negative utilities for adverse events were estimated based on quality of life endpoints of the COU-AA-301 trial. The base year was 2012. All costs are presented in United States Dollars (USD). Costs and outcomes were discounted at 5%. Probabilistic sensitivity (PSA) analysis was performed to evaluate uncertainty surrounding the parameters. Results: A-P resulted in 0.79 QALYs and 1.35 LYs, per patient, respectively. C-P resulted in 0.71 QALYs and 1.28 LYs, per patient, respectively. Mean total costs per patient were: USD 76.179 for A-P and USD 86.286 for C-P. The results of the probabilistic sensitivity analysis showed that, when compared with C-Z, A-P was found dominant (associated with reduced costs and increased QALYs) in the majority of the iterations. A-P had a 73% probability of being cost effective, independent of the willingness to pay, when compared to C-P. When the willingness to pay increases, A-P is more likely to become cost effective. Conclusions: A-P can be considered dominant (cost-saving), when compared with C-P, in patients with Metastatic CastrationResistant Prostate Cancer that have failed to chemotherapy with Docetaxel, from the perspective of the Public System of Health of Panama. PCN104 Everolimus Plus Exemestane Compared to Exemestane and Fulvestrant for the Treatment of ER+ HER2- Metastastic Breast Cancer in the United Kingdom – A Societal Perspective Polanyi Z 1, Dale P 2, Taylor M 3, Lewis L 3, Glanville J 3, Vieira J 1, Chandiwana D 1 1Novartis Pharmaceuticals UK Limited, Camberley, UK, 2HEOR Solutions, London, UK, 3York Health Economics Consortium, York, UK . . . . . . . Objectives: This study evaluated the cost-effectiveness of everolimus plus exemestane (EVE+EXE) versus exemestane (EXE) and fulvestrant (FUL) in the treatment of postmenopausal women with ER+ HER2- metastatic breast cancer in the United Kingdom (UK) from a societal perspective. Methods: A partitioned survival model was developed to compare treatment with EVE+EXE versus EXE and FUL over a 10-year time horizon. Progression-free survival and overall survival for EVE+EXE and EXE were estimated from the BOLERO-2 trial. Log-logistics functions were used to extrapolate trial data beyond the follow-up period. In the absence of head-tohead evidence vs. FUL an indirect treatment comparison was conducted using a Bayesian fixed effect model. Background health state and terminal care resource use were derived from NICE Clinical Guideline 81. Drug costs were taken from the British National Formulary. Productivity loss, defined as working days lost due to disease, was included in the analysis. Utilities from published sources were combined with trial data to calculate quality-adjusted life years (QALYs) for the model health state. Results: EVE+EXE led to an incremental gain in life years of 0.20 vs. EXE and 0.19 vs. FUL. The incremental QALY gain was 0.31 vs. EXE and 0.27 vs. FUL. The cost of lost productivity was £66,163 in the EVE+EXE compared to £75,067 in the EXE arm and £73,434 in the FUL arm. The incremental cost per QALY was £27,644 vs. EXE and £14,030 vs. FUL. Probabilistic sensitivity analysis demonstrated that, at a threshold of £30,000 per QALY gained, EVE+EXE had a 51.6% likelihood of being cost-effective vs. EXE and 59.0% vs. FUL. Conclusions: Patients receiving EVE+EXE experienced an improvement in survival which translated into health gains in terms of both LYs and QALYs. EVE+EXE was associated with savings in productivity costs compared to both EXE and FUL. PCN105 Economic Evaluation of NAB-Paclitaxel Plus Gemcitabine Versus Gemcitabine Alone for The Management of Metastatic Pancreatic Cancer in Greece Fragoulakis V 1, Papakostas P 2, Pentheroudakis G 3, Dervenis C 4, Maniadakis N 1 1National School of Public Health, Athens, Greece, 2Hippokration Hospital, Athens, Greece, 3Medical School, University of Ioannina, Ioannina, Greece, 4Konstantopoulio hospital, Athens, Greece . . . . . Objectives: To estimate the cost-effectiveness of nab-paclitaxel+gemcitabine (Npg) versus gemcitabine (Gem) alone for the first-line treatment of metastatic pancreatic cancer in Greece from a National Health System perspective Methods: A Markov model was developed, included several stages such as: “pre-progression on firstline treatment”, “pre-progression off first-line treatment”, “post progression”, “four weeks to death” and “death”. Data from the MPACT trial were used to estimate overall survival (Life-Years- (LYs)) and adverse events. The prices of drugs used in the model are publicly available for all hospitals in Greece. Cost assigned in each health state reflected: drugs, adverse events, monitoring, administration and palliative care. Utility values were obtained from the international literature to estimate Quality-AdjustedLife-Years (QALYs). Costs and health gains were discounted at 3.5% per annum. A probabilistic sensitivity analysis was also conducted to construct confidence intervals (CI). Results: The mean number of QALYs was 0.71 (95%CI: 0.66–0.78) and 0.56 (95%CI: 0.52–0.60) for Npg and Gem, respectively, giving an incremental gain of 0.15 (95%CI: 0.08-0.25) QALYs in favour of Npg. The mean cost of therapy per patient was estimated at €15,628 (95%CI: €14,377- €17,027) and €8,284 (95%CI: €7,455-€9,112) for Npg and Gem, respectively. The incremental cost per LY gained with Npg was estimated at €37,007 and the incremental cost per QALY gained at €47,120. The probability for Npg to be cost-effective at a threshold three times the per capita income (€ 60,000 per QALY) was 82%. Conclusions: The severity of pancreatic cancer, in combination with the limited number of effective treatments, results in a high level of unmet need. If the societal willingness-to-pay threshold for an additional QALY is relatively higher for patients with short life expectancy or for diseases with a relatively higher burden, the combination of nab-paclitaxel+gemcitabine could be considered a cost-effective choice compared with gemcitabine alone in Greece. PCN106 Cost-Effectiveness Analysis of Panitumumab Plus Mfolfox6 Versus Bevacizumab Plus Mfolfox6 for First-Line Treatment of Patients with Wild-Type Ras Metastatic Colorectal Cancer Graham C N 1, Hechmati G 2, Hjelmgren J 2, De Liège F 3, Lanier J 3, Knoof A 3, Knox H 1, Barber B 4, de Pouvourville G 5 1RTI Health Solutions, Research Triangle Park, NC, USA, 2Amgen (Europe) GmbH, Zug, Switzerland, 3Amgen France SAS, Neuilly-sur-Seine, France, 4Amgen, Inc., Thousand Oaks, CA, USA, 5ESSEC Business School, Cergy-Pontoise, France . . . . . . . . . . Objectives: To compare the cost-effectiveness of panitumumab plus mFOLFOX6 (oxaliplatin, 5-fluorouracil and leucovorin) versus bevacizumab plus mFOLFOX6 as first-line treatment for patients with wild-type RASmetastatic colorectal cancer (mCRC). Methods: Using a French health collective perspective, a lifetime Markov model was constructed, with health states related to first-line therapy (progressionfree), disease progression with/without subsequent active treatment, resection of metastases, disease-free after successful resection, and death. Transitions to disease progression and death were estimated using parametric survival analyses of patientlevel progression-free (PFS) and overall (OS) survival from the only head-to-head clinical trial of panitumumab versus bevacizumab in mCRC (PEAK). Additional data from PEAK informed the amount of each drug consumed, duration of therapy, subsequent therapy use, and toxicities related to mCRC treatment. Literature and French public data sources were used to estimate unit costs associated with treatment, duration of subsequent active therapies, and survival post-resection. Patient-level data from panitumumab trials in the first-, second-, and third-line settings were used to determine utility weights. One-way and probabilistic sensitivity analyses were performed. Scenario analyses examined modelling of PFS and OS using observational survival data and PEAK hazard ratios. Results: Based on the better efficacy outcomes for patients with wild-type RASmCRC who received panitumumab plus mFOLFOX6 versus bevacizumab plus mFOLFOX6 in PEAK, the incremental cost per life-year gained was estimated to be €26,918, and the incremental cost per quality-adjusted life year (QALY) gained was estimated to be €36,577. Sensitivity and scenario analyses indicate the model is robust to alternative parameters and assumptions. Conclusions: Panitumumab plus mFOLFOX6 can be considered cost-effective in first-line treatment of patients with wild-type RAS mCRC. PCN107 Cost-Utility Analysis of Pazopanib VersUS Sunitinib as First-Line Treatment of Metastatic Renal Cell Carcinoma (MRCC) iN Spain Espinosa J 1, González-Larriba J,L 2, Maroto P 3, Méndez-Vidal M J 4, Díaz-Cerezo S 5 General Universitario de Ciudad Real, Ciudad Real, Spain, 2Hospital Clínico San Carlos, Madrid, Spain, 3Hospital Sant Pau, Barcelona, Spain, 4Hospital Reina Sofía, Córdoba, Spain, 5GSK España, Tres Cantos, Spain . . . . . . 1Hospital Objectives: To assess the cost-utility of pazopanib vs. sunitinib as first-line treatment of mRCC from the Spanish National Healthcare perspective. Methods: A published partitioned-survival analysis model was used to estimate the incremental cost-utility ratio (ICUR) of pazopanib vs. sunitinib. Progression free survival (PFS) and overall survival (OS) data from COMPARZ (NCT00720941), a non-inferiority head to head phase III-trial of sunitinib vs. pazopanib, were used to generate survival functions up to overall time horizon. Patients can be in one of three mutually exclusive health states over time: alive and no progression, alive with progression and dead. Utilities values were obtained from PISCES trial (NCT01064310) and adjusted taking into account the different treatment schedules (sunitinib-cycles: 4 weeks-on plus2 weeks-off; pazopanib 800 mg/day). Health resource utilization was collected using individual patient data from COMPARZ trial and their corresponding unit costs were retrieved from published Spanish tariffs. Base-case analysis considered: A633 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 5-year time horizon, PFS data assessed by independent review, drug doses adjusted by relative dose intensity reported in COMPARZ trial and a discount rate of 3% for costs and outcomes. Results were expressed as € 2014. Deterministic (10-year time horizon, discount rates 0 and 5%, PFS assessed by investigator, and plenty-doses) and probabilistic sensitivity analyses were conducted to determine the robustness of the results. Results: In the base case analysis, pazopanib showed as a dominant alternative, yielding more quality of life adjusted years (0.081) and less total costs (€ 6,671) vs. sunitinib. Base-case results were robust in the alternative scenarios examined via deterministic sensitivity analyses. In the probabilistic sensitivity analysis (PSA), a 67% of the simulations were plotted in the dominant quadrant of the cost-effectiveness plane. Conclusions: In the light of the present analysis, pazopanib should be considered as a dominant alternative vs. sunitinib in the firstline mRCC treatment from the Spanish National Healthcare perspective. PCN108 Economic Evaluation of the Use of Gefitinib for the Treatment of Locally Advanced or Metastatic NSCLC Polanco A C 1, Salazar A 1, Pizarro M 2, Carpio E 1, González L A 3 1AstraZeneca, Tlalpan, Mexico, 2Hospital Infantil de Mexico Federico Gomez, Mexico City, Mexico, 3Health Solutions Consulting, D. F., Mexico . . . . . . . Non-small-cell lung cancer (NSCLC) is the most common type of cancer representing 18.2% of all cancer deaths around the world and in Mexico the estimated mortality rate is 13.4 by 100,000 patients. Objectives: Evaluate gefitinib as first and second line treatment of locally advanced or metastatic NSCLC compared to available treatment alternatives in Mexico. Methods: A two-way analysis was performed: (1) For the first-line treatment in patients with Epidermal Growth Factor Receptor Inhibition in Mutation-Positive Non–Small-Cell Lung Cancer (EGFR M+ NSCLC) a cost-minimization analysis was used comparing gefitinib versus erlotinib (Kim ST, 2012), also a Markov model was developed to perform a cost-effectiveness analysis evaluating gefitinib versus carboplatin+paclitaxel (Mok TS, 2009), with efficacy measure Progression-free survival (PFS); and (2) For patients in a secondline NSCLC treatment, regardless of EGFR mutation, a cost-minimization analysis was conducted comparing gefitinib versus docetaxel and pemetrexed (Hanna N, 2004) (Kim ES, 2008). The costs were obtained from institutional sources. An exchange rate of $13.12 MXN per USD was used. Sensitivity analyses were performed in order to test the robustness of the model. Results: For first-line treatment, gefitinib was a cost-saving alternative respect to erlotinib, obtaining a cost differential of $9,710 USD in favor of gefitinib. To the same patients gefitinib compared to carboplatin plus paclitaxel generated an additional cost of $2,361 USD per patient, with additional PFS of 0.37 years and an ICER of $7,023. For secondline treatment gefitinib had a lower cost compared to pemetrexed and docetaxel, generating a saving per patient of $927 USD and $21,346 USD respectively. Robustness of results was confirmed by additional deterministic and probabilistic sensitivity analysis. Conclusions: The use of gefitinib for the treatment of locally advanced or metastatic NSCLC is a cost-saving alternative compared to erlotinib, pemetrexed and docetaxel, and also cost-effective compared to carboplatin plus paclitaxel. PCN109 Cost-Effectiveness of Ofatumumab Plus Chlorambucil in First Line Chronic Lymphocytic Leukemia in Canada Herring W 1, Pearson I 2, Purser M 1, Nakhaipour H R 3, Haiderali A 4, Wolowacz S 2, Jayasundara K 3 1RTI Health Solutions, Research Triangle Park, NC, USA, 2RTI Health Solutions, Manchester, UK, 3GlaxoSmithKline, Mississauga, ON, Canada, 4GlaxoSmithKline, Collegeville, PA, USA . . . . . . . . Objectives: This study aimed to estimate the cost-effectiveness of Ofatumumab plus Chlorambucil (OChl) compared with Chlorambucil (Chl) for patients with Chronic Lymphocytic Leukemia for whom fludarabine-based therapies are considered inappropriate, from the perspective of the publicly funded health care system in Canada. Methods: A semi-Markov based decision model was developed with a lifetime time horizon. The model comprised two distinct phases. The preprogression phase was based on the overall response rates (ORR), progression free survival (PFS) and overall survival (OS) observed in the COMPLEMENT-1 trial. The postprogression phase was based on Canadian treatment practices, treatment patterns identified in clinical guidelines and published literature. The incremental cost per qualityadjusted life year (QALY) gained was computed using model-estimated first- and subsequent-line treatment costs, general disease management costs, and QALYs based on health-state preference utility weights. Results: The discounted, lifetime health and economic outcomes estimated by the model showed that first-line treatment with OChl in comparison with Chl in the target population led to an increase in QALYs (0.41) and an increase in total costs (CAD $27,850), resulting in an incremental cost-effectiveness ratio (ICER) of CAD $68,672/QALY gained. Various scenario analyses indicated that the cost-effectiveness results were sensitive to the time horizon, the method used to assess response, and the extrapolation of OS treatment effect beyond the trial period. One way and probabilistic sensitivity analyses aligned with the results of the base-case analysis. Conclusions: The base-case results indicate that the improved ORR, PFS, and OS for OChl in comparison with Chl translate to improved long-term health outcomes. The analysis found that the ICER for OChl versus Chl in the target population was CAD $68,672/QALY gained. A variety of sensitivity and scenario analyses confirmed that the model’s cost-effectiveness estimates were robust. PCN110 Cost-Effectiveness Analysis of Panitumumab+Mfolfox over Bevacizumab+Mfolfox as a First-Line Treatment for Metastatic Colorectal Cancer Patients with Wild-Type Ras in Greece Kourlaba G 1, Boukovinas I 2, Saridaki Z 3, Papagiannopoulou V 4, Tritaki G 4, Maniadakis N 5 and Kapodistrian University of Athens School of Medicine, Athens, Greece, 2Bioclinic Thessaloniki – Oncology Unit, Thesaloniki, Greece, 3Laboratory of Tumor Cell Biology School of Medicine-University of Crete, Herakleion, Crete, Greece, 4AMGEN Hellas, Marousi, Greece, 5National School of Public Health, Athens, Greece . 1National . . . . . Objectives: To conduct a cost-effectiveness analysis of panitumumab plus mFOLFOX6 versus bevacizumab plus mFOLFOX6 as first-line treatment (FLT) of metastatic colorectal cancer (mCRC) patients with wild-type RASin the Greek health care setting. Methods: An existing Markov model consisting of seven health states was adapted from the public third-party-payer perspective. Both efficacy and safety data considered in the model were extracted from the PEAK trial and other published studies. Utility values were also extracted from the literature. Direct medical costs consisting of drug-acquisition costs for FLT, administration costs, subsequent therapy costs and other medical costs were incorporated into the model and reflect the year 2014. Primary outcomes were patient survival (life-years), quality-adjusted life years (QALYs) and the incremental cost-effectiveness ratio (ICER) per QALY gained. Probabilistic sensitivity analysis (PSA) was conducted to account for uncertainty and variation in the parameters of the model. Results: The analysis showed that panitumumab plus mFOLFOX6 produced greater discounted survival and quality adjusted survival by 0.87 LYs and 0.65 QALY benefit in relation to bevacizumab plus mFOLFOX6. The total lifetime cost was € 75,200 and € 52,736 for panitumumab and bevacizumab plus mFOLFOX6, respectively. This difference was mainly attributed to the higher acquisition cost of panitumumab compared to bevacizumab during the pre-progression health state (€ 32,223 and € 14,730 respectively). Incremental analysis showed that panitumumab plus mFOLFOX6 was more effective and more costly than bevacizumab plus mFOLFOX6 resulting in an ICER equal to € 34,644 per QALY gained. PSA revealed that the probability of panitumumab plus mFOLFOX6 being cost-effective over bevacizumab plus mFOLFOX6 was 81.5% at the predetermined threshold of € 51,000 per QALY gained (3 times the GDP per capita of Greece). Conclusions: The results suggest that panitumumab plus mFOLFOX6 may be a cost-effective alternative relative to bevacizumab plus mFOLFOX6 as FLT of mCRC patients with wild-type RASin Greece. PCN111 Cost-Effectiveness and Cost-Utility of Granulocyte ColonyStimulating Factors in the Primary Prophylaxis of Chemotherapy Induced Febrile Neutropenia (FN) in Breast Cancer Patients in Greece: A Comparative Analysis Kourlaba G 1, Palaka E 2, Papagiannopoulou V 2, Maniadakis N 3 1National and Kapodistrian University of Athens School of Medicine, Athens, Greece, 2AMGEN Hellas, Marousi, Greece, 3National School of Public Health, Athens, Greece . . . . Objectives: To conduct an economic evaluation comparing pegfilgrastim with filgrastim or lenograstim used either in an 11-day regimen or in a 6-day regimen for the prophylaxis of febrile neutropenia (FN) in breast cancer patients, in the Greek health care setting. Methods: A cost-effectiveness model was locally adapted from the public third-party-payer perspective. Efficacy and utility values extracted from published studies were considered in the model. The analysis was conducted for a 6-cycle horizon, to reflect the common clinical practice in Greece. Drug acquisition costs, administration costs and FN management reimbursed costs were considered (in € 2014). The outcomes of the model were the incremental cost per additional FN event avoided and per QALY gained (ICER) of pegfilgrastim to its comparators. The ICERs were evaluated at the predetermined willingness-to-pay threshold of € 34,000/ QALY gained. Results: The incremental cost per additional FN event avoided with pegfilgrastim ranged between € 11,015 and € 27,079 compared to 11-day regimens of originator and a biosimilar filgrastim respectively, while pegfilgrastim was found to be dominant compared to the 11-day regimen of lenograstim. Comparing pegfilgrastim with the 6-day regimen of filgrastim and lenograstim, it was found that the ICER per additional FN event avoided ranged between € 9,538 and € 15,207 in case of lenograstim and biosimilar filgrastim respectively. Similarly, cost-utility analysis revealed that pegfilgrastim was cost-effective over 11-day and 6-day regimens of originator filgrastim with ICERs of € 11,065 and € 19,942/QALY gained, respectively. Compared to lenograstim, pegfilgrastim was found to be dominant over the 11-day regimen and cost-effective over the 6-day regimen (ICER: € 15,546). Conclusions: Our findings suggests that pegfilgrastim for the prophylaxis of chemotherapyinduced FN in breast cancer patients is associated with greater health benefit and lower cost over 11-day use of lenograstim, while it is a cost-effective option over either the 6-day or the 11-day regimen of biosimilar filgrastim, in Greece. PCN112 Cost-Effectiveness of Vismodegib VersUS Standard of Care Therapy in the Treatment of Locally-Advanced or Symptomatic Metastatic Basal Cell Carcinoma in Hungary – A Global Cost-Effectiveness Model Adaptation Mikudina B 1, Péter T 1, Nagy B1, Horváth K2 1Healthware Consulting Ltd., Budapest, Hungary, 2Roche Hungary, Budaörs, Hungary . . Objectives: Hungarian adaptation of global cost-effectiveness models of vismodegib vs. standard of care (SOC) in the treatment of locally advanced or symptomatic metastatic basal cell carcinoma (laBCC and mBCC). Methods: Global Markovmodels were developed to compare the cost-effectiveness of vismodegib vs. SOC in patients with laBCC or mBCC. The model inputs were based on the pivotal phase II clinical study (ERIVANCE). Health state utility values were based on a time trade off study. To support the reimbursement dossier submission, the adaptation of the global cost-effectiveness models was conducted. The costs and resource use were recalculated based on a questionnaire survey with Hungarian health care professionals. In the model there were two treatment arms, vismodegib and SOC. The model had three states, progression-free, progressed and death. For progression-free survival (PFS) and overall survival (OS) the results of the phase II clinical trial were used in the vismodegib arm of the model. Originally on the SOC arm the model calculated with mortality data of the general population, due to lack of relevant data on the PFS and OS of patients with advanced BCC. Therefore, a research (Delphipanel survey) was conducted to estimate the OS of patients with laBCC and mBCC, treated with SOC. Results: According to the Delphi-panel survey the median OS for patients with laBCC and mBCC was 48 months and 24 months, respectively, on the SOC arm. The average time spent in progression-free health state is longer with vis- A634 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 modegib therapy than with SOC for both, laBCC and mBCC patients. Conclusions: Vismodegib could provide an effective treatment for this therapeutic area with high rate of unmet need. During the adaptation process Delphi-panel surveys seemed to be an appropriate method to earn consensus statement to ensure estimation and help interpretation. PCN113 Potential Monetary Value of Human Papillomavirus Vaccination on Human Papillomavirus-Related Cancers and Genital Warts in the United Kingdom sitivity analysis (PSA) was run with thousand repetitions and a one-way sensitivity analysis was calculated showing its results in a tornado chart. Results: The model showed that everolimus + exemestane results in 0.74 progression free years gained with an incremental cost of $18.6 million (MM) resulting in an incremental cost-effectiveness ratio (ICER) of $26 MM. The PSA showed that the ICER is within the range recommended by WHO (1-3 GDPs per capita) in 71% of cases (Currently the GDP per capita in Chile is $10 MM). Conclusions: This analysis showed that using everolimus plus exemestane in patients with ER +, HER2- advanced breast cancer who have failed on NSAIs is a cost-effective option according to WHO recommendations. Van Kriekinge G 1, Starkie-Camejo H 2, Li X 1, Demarteau N 1 . . . . 1GlaxoSmithKline Vaccines, Wavre, Belgium, 2GlaxoSmithKline, Uxbridge, UK Objectives: The United Kingdom (UK) runs a successful human papillomavirus (HPV) girls vaccination programme. Debate is ongoing on the value of including boys in the programme. This study aims at quantifying the potential value associated with genital warts (GW) and HPV-related cancer prevention in UK males and females based on a willingness-to-pay threshold of £20,000 per quality-adjusted life-years (QALY) gained, representing the potential value a government places on the prevention of these diseases. Methods: A static vaccine steady-state (VSS) population model, stratified by age, with a 1-year time horizon, replicated the incidence of GW and HPV-related cancers in females (cervical (CC), anal (AC), vulvar (VuC), vaginal (VaC), oropharyngeal (OP)) and males (penile (PC), AC and OP) pre-vaccination and at VSS. Data were retrieved from UK cancer registries, sexually transmitted diseases reports and HPVCentre. Costs and utilities were identified from the literature. The VSS vaccine effectiveness for GW and HPVrelated cancers was estimated combining efficacies (AS04-adjuvanted HPV-16/18 vaccine for cancers; HPV-6/11/16/18 vaccine for GW) weighted by vaccine-types (HPV-6/11/16/18) and non-vaccine types (HPV-31/33/35/39/45/51/52/56/58/59) HPV distribution. Costs and QALYs were discounted at 1.5%. Per-course vaccine costeffective price (vCE-p) was determined by increasing vaccine course price until £20,000 per incremental QALY gained at VSS was reached. Sensitivity analyses on key variables were performed. Results: The vCE-p in women (men) was: CC £790, OP £20 (£57), AC £123 (£77), VaC £37, VuC £58, (PC £40), GW £26 (£27). Total value of cancer prevention in women (men) was £1,027 (£173), a proportion of 6: 1. The value of CC alone is 4.5 times larger than the total value of cancer prevention in men. Sensitivity analyses showed results were robust while influenced by potential herd protection. Conclusions: The vCE-p was estimated to be up to 6 times higher in women than in men due to the higher burden and frequency of HPV-related cancers in women. PCN114 Cost-Effectiveness of Aprepitant in Egyptian Patients Receiving Highly Emetogenic Therapy from the Third Party Payer Perspective PCN116 Cost-Effectiveness of 2-DOSE AS04-Adjuvanted Human Papillomavirus 16/18 Vaccination Schedule in Slovakia Hlavinkova L 1, Li X 2, Van Kriekinge G 2, Trnovec P 1 1GlaxoSmithKline Slovakia, Bratislava, Slovak Republic, 2GlaxoSmithKline Vaccines, Wavre, Belgium . . . . Objectives: Slovakia is a country with high incidence and mortality of cervical cancer (CC). Despite the improvements in screening (22.9% coverage rate), the CC incidence has increased over the past 30 years in Slovakia. Human Papillomavirus (HPV) vaccination could help to reduce this CC burden. The objective of this analysis was to assess the cost-effectiveness of adding the AS04-adjuvanted HPV-16/18 vaccine (AS04V), using a 2-dose administration schedule, to the current CC screening programme in Slovakia. Methods: A previously published Markov cohort model, reproducing the natural history of HPV infection, the impact of screening and vaccination, was adapted to the Slovakian settings. Local data on health care costs of pre-cancer lesions and CC, obtained from the expert panel, were used. Transition probabilities and utilities were estimated from published data. Costs were from a health care payer perspective. The incremental CC cases avoided, cost, quality-adjusted life-years (QALYs) and resulting cost-effectiveness ratio (ICER) of AS04V added to the current CC screening programme versus the current CC screening in Slovakia was estimated. The base case assumes a 100% vaccination coverage among 12-year-old girls (N= 24,859). A discount rate of 5% was used. Univariate sensitivity analyses were carried out on key parameters. Results: Compared to screening alone, adding AS04V to the current screening programme was estimated to reduce the lifetime CC cases by 328 at an ICER of 11,621 € / QALY gained. Compared to the official cut-off of 19,320€ /QALY gained, it can be considered as cost-effective. Undiscounted analysis shows that AS04V generates more QALYs with similar cost versus screening alone (ICER= 5€ /QALY gained). Parameters most driving the results were discount rate, vaccine efficacy and duration of protection. Conclusions: AS04V vaccination of 12-year-old girls in a 2-dose schedule was estimated to be a cost-effective CC prevention strategy in Slovakia. Helal M 1, Elsisi G 2 . . 1CAPA, Cairo, Egypt, 2Central Administration for Pharmaceutical Affairs, Cairo, Egypt Objectives: to evaluate the cost-effectiveness of aprepitant as add-on therapy to the standard Egyptian regimen in patients receiving highly emetogenic therapy. Methods: A decision tree model was developed based on the Egyptian clinical practice, and was derived from published sources. This decision analytical model was constructed to assess the costs and consequences associated with aprepitant containing regimen compared with standard therapy for ChemotherapyInduced Nausea and Vomiting. The clinical parameters were derived from a randomized trial previously published. The utility of the health states was derived using the available published data. Direct medical costs were obtained from the third party payer tariff in Egypt. Deterministic sensitivity analyses were conducted. All costs (in 2014 EGP) and outcomes were discounted at 3.5% annually. Results: The total quality-adjusted life-years (QALYs) of adding aprepitant to the standard regimen was estimated to be 0.0082, whereas that of the standard regimen was estimated to be 0.0072 (with a net difference of 0.001QALYs). The total costs for aprepitant plus standard regimen and standard regimen alone were EGP 414.25 and EGP 346.62 respectively (with a net difference of EGP 67.63). Thus the incremental cost-effectiveness ratio (ICER) for aprepitant was EGP 66,004/QALY gained. The probability of complete protection and incomplete response of both arms were found to have the greatest effect on the results. Conclusions: The present study concludes that adding aprepitant to the standard regimen is cost effective based on the threshold stated by world health organization (3xGDP/capita) for patients with severe vomiting after chemotherapy. PCN115 Cost Effectiveness Analysis of Everolimus + Exemestane for Patients with Advanced Breast Cancer with Positive Estrogen Receptor (ER +), HER2-, Refractory to Non-Steroidal Aromatase Inhibitors (NSAIS) in Chile Ratto B 1, Torres Ulloa R 2, Cerda Veneros H 2, Anaya P 3 1Novartis Pharmaceuticals, Buenos Aires, Argentina, 2INSTITUTO NACIONAL DEL CÁNCER, Santiago, Chile, 3Novartis Pharmaceuticals, Mexico City, Mexico . . . . Objectives: To evaluate the cost-effectiveness of everolimus plus exemestane in patients with ER+, HER2- advanced breast cancer, who have failed on NSAIs. Methods: A Markov model was developed with monthly cycles and a time horizon of five years. The model compares progression free survival (PFS) of exemestane + everolimus (EVE+EXE) to exemestane monotherapy (EXE). Transition probabilities for PFS of EVE+EXE and EXE were based on BOLERO-2 study and calculated using a fitted Weibull distribution. The R-squared values for the Weibull fits were 0.998 and 0.990 for EVE+EXE and EXE alone respectively. The Weibull parameters used in the model were: 0.067 and 1.118 for EVE+EXE and 0.191 and 1.006 for EXE. Costs considered included drugs and cost of treating neutropenia (other AEs are not covered by the National Formulary). The analysis was designed from the perspective of the Chilean Public Healthcare. Results are shown in 2014 Chilean pesos. A 5% discount rate for costs and efficacies was applied. A probabilistic sen- PCN117 A Cost Effectiveness Analysis of Everolimus Plus Exemestane Compared to Chemotherapy Agents for the Treatment of ER+ HER2- Metastastic Breast Cancer in the United Kingdom Polanyi Z 1, Dale P 2, Taylor M 3, Lewis L 3, Glanville J 3, Vieira J 1, Chandiwana D 1 1Novartis Pharmaceuticals UK Limited, Camberley, UK, 2HEOR Solutions, London, UK, 3York Health Economics Consortium, York, UK . . . . . . . Objectives: To evaluate the cost-effectiveness of everolimus plus exemestane (EVE+EXE) versus chemotherapy agents [docetaxel (DOC), vinorelbine (VIN), doxorubicin (DOX) and capecitabine (CAPE)] for the treatment of hormone receptor positive (HR+) HER2 negative (HER2-) advanced or metastatic breast cancer in the United Kingdom (UK). Methods: A partitioned survival model was developed to compare treatment with EVE+EXE versus DOC, VIN, DOX and CAPE in patients with ER+ HER2- metastatic breast cancer over a 10-year time horizon from a UK NHS perspective. Progression-free survival and overall survival for EVE+EXE were taken from the BOLERO-2 trial. Log-logistic functions were used to extrapolate trial data beyond the follow-up period. In the absence of head-to-head evidence comparing EVE+EXE versus chemotherapy a naïve chained comparison was conducted with the link between EVE+EXE established via tamoxifen using the Bucher method. A class effect was assumed for the four chemotherapy agents. Background health state and terminal care resource use were derived from NICE Clinical Guideline 81. Drug costs were taken from the British National Formulary. Utilities for stable and progressive states were obtained from the literature (Lloyd et al. 2006). Results: Over a ten year time horizon, EVE+EXE led to a life expectancy of 3.55 years, compared to 1.88 for chemotherapy agents (DOC, VIN, DOX and CAPE). EVE+EXE resulted in 2.06 QALYs, compared to 0.95 for chemotherapy agents. Total costs were £48,085 for EVE+EXE compared to £31,835 vs. DOC, £25,021 vs. VIN, £23,743 vs. DOX and £21,851 vs. CAPE. The incremental costs per QALY were £14,550 vs. DOC, £20,653 vs. VIN, £21,797 vs. DOX and £23,491 vs. CAPE. Results were most sensitive to changes in PFS for chemotherapy and disease related costs. Conclusions: Everolimus in combination with exemestane is a cost effective option compared with commonly used chemotherapeutic agents (docetaxel, vinorelbine, doxorubicin and capecitabine) in UK clinical practice. PCN118 Cost-Effectiveness Analysis of Bevacizumab- PaclitaxelCarboplatin (PC) Versus PC in First-Line Therapy of Advanced NonSmall Cell Lung Cancer from Patients’ Perspective in Vietnam Tran T T H , Nguyen T T T University of Medicine and Pharmacy in HCMC, Ho Chi Minh City, Vietnam . . . . . . Objectives: Bevacizumab in combination with carboplatin/paclitaxel (BCP) was approved to be the first-line therapy of advanced NSCLC due to its high clinical efficacy. However, economic effectiveness of BCP has been controversial. This study aimed to estimate the cost-effectiveness of BCP versus PC in treatment of advanced NSCLC patients from patients’ perspective in Vietnam. Methods: A A635 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Markov model was developed to estimate the health outcome (QALY) and total treatment costs with Markov cycle of 21 days and lifetime horizon. The effectiveness data was retrieved from the randomized clinical trial ECOG 4599. Direct costs, including cost of drugs, administration, medical services, hospital bed day and adverse drug reaction management were estimated based on treatment guideline of NCCN for NSCLC. Indirect costs, including loss of earnings, cost for meal, transportation, accomodation of patients and their caregivers due to treatment, were estimated based on survey of a cohort of 87 patients with NSCLC in HCMC Oncology Hospital. Both cost and effectiveness were discounted 3% annually. Results: Adding bevacizumab to PC regimen in first-line therapy of advanced NSCLC patients resulted in incremental QALY gained of 2.26 month compared with PC regimen (7.88 versus 5.62). The total treatment cost with BCP was 3 times higher than PC (2,499 millions vs 761.7 millions VND, respectively). ICER of BCP versus PC was 768,732,924 VND, which is 3.35 times higher than the Willingness-To-Pay of Vietnam in 2013 (229,242,416 VND). A probability sensitivity analysis demonstrated the patient’s weight and bevacizumab’s price as the most affecting factors to the ICER of BCP vs PC. Conclusions: Conducted analysis showed that combination of bevacizumab and PC regimen in first-line therapy of NSCLC was not cost-effective compared with PC regimen. Support from the manufactures, suppliers and insurance organizations are neccessary to raise its economic effectiveness in treatment of advanced NSCLC. PCN119 Cost-Effectiveness Simulation of Colonography Versus Colonoscopy in Germany: Is Laxative-Free Colonography CostEffective? Kriza C 1, Jahn B 2, Hassan C 3, Kolominsky-Rabas P L 1 1Centre for Health Technology Assessment (HTA) and Public Health (IZPH), Friedrich-AlexanderUniversity Erlangen-Nürnberg, Erlangen, Germany, 2UMIT - University for Health Sciences, Medical Informatics and Technology, Hall in Tyrol, Austria, 3Nuovo Regina Margherita Hospital, Rome, Italy . . . . . Objectives: Colorectal cancer (CRC) screening using computerised tomographic colonography (CTC), also referred to as virtual colonography, has attracted considerable attention due to its positive impact on high screening uptake rates, especially with a laxative-free preparation before screening. A decision analysis model was constructed in order to evaluate the clinical and economic consequences of performing three different screening tests versus a no screening scenario in a population at average risk of colorectal cancer in Germany: colonoscopy, conventional CT-colonography and laxative-free CT colonography. Methods: A state-transition microsimulation was developed for the evaluation of the different screening strategies using TreeAge Pro Healthcare 2014. A hypothetic population of 100,000 German asymptomatic adults aged between 50 and 100 years was used for the basis of the model. The simulation of the screening strategies was undertaken by assessing the number of screening patients diagnosed with CRC on the basis of the sensitivity and specificity of each strategy and the related uptake of each screening method. Sensitivity analysis will be applied to test the impact of parameter uncertainty on model outcomes and recommendations. Results: Initial results of the simulation show that laxative-free colonography was found to be the most costly screening option, with a total cost of EUR 4,115 per screening patient in the simulation model. Colonoscopy was found to be the least costly screening method, with total equivalent costs of EUR 2,132. The most effective screening was modeled for laxative-free colonography. The ICER of laxative-free colonography compared to colonoscopy was simulated at 5,221 EUR per life year saved. Conclusions: Our simulation has shown that using data from new research indicating the possibility of less costly use of CTC than previously used for modeling, laxative-free CTC screening has the potential to become a cost-effective alternative screening method for CRC due to its advantage related to improvements in screening uptake. PCN120 Cost-Effectiveness Analysis of Abiraterone Acetate Treatment Compared With Cabacitaxel in Costa Rica, in Patients with Metastatic Castration-Resistant Prostate Cancer That Have Failed to Chemotherapy with Docetaxel Obando C A 1, Desanvicente-Celis Z 1, Gonzalez L 2, Muschett D 1, Gonzalez F 1, Goldberg P 1 . . . . . . . 1Janssen, Panama, Panama, 2Janssen, Raritan, NJ, USA Objectives: To assess the cost-effectiveness of Abiraterone Acetate plus Prednisone (A-P) compared with Cabazitaxel plus Prednisone (C-P) in Costa Rica, in patients with Metastatic Castration-Resistant Prostate Cancer (mCRPC) that have failed to chemotherapy with Docetaxel. Methods: A three-health state cohort simulation Markov Model (progression-free, post-progression and death) was developed based on overall and progression free survival data. The time frame was 10 years. The perspective was that of the Public System of Health of Costa Rica. The health outcomes of interest were Quality Adjusted Life Years (QALYs) and Life Years (LYs). Efficacy data was taken from clinical trials (COU-AA-301 for A-P and TROPIC for C-P). Utilities for health states and negative utilities for adverse events were estimated based on quality of life endpoints of the COU-AA-301 trial. The base year was 2012. All costs are presented in Costa Rican currency (Colones - CRC). Costs and outcomes were discounted at 5%. Probabilistic sensitivity (PSA) analysis was performed to evaluate uncertainty surrounding the parameters. Results: A-P resulted in 0.79 QALYs and 1.35 LYs, per patient, respectively. C-P resulted in 0.71 QALYs and 1.28 LYs, per patient, respectively. Mean total costs per patient were: CRC 33.881.184 for A-P and CRC 41.981.207 for C-P. The results of the probabilistic sensitivity analysis showed that, when compared with C-Z, A-P was found dominant (associated with reduced costs and increased QALYs) in the majority of the iterations. A-P had an 89% probability of being cost effective, independent of the willingness to pay, when compared to C-P. Conclusions: A-P can be considered dominant (cost-saving), when compared with C-P, in patients with Metastatic Castration-Resistant Prostate Cancer that have failed to chemotherapy with Docetaxel, from the perspective of the Public System of Health of Costa Rica. PCN121 Cost-Effectiveness Analysis of Abiraterone Acetate Treatment Compared with Cabacitaxel in Dominican Republic, in Patients With Metastatic Castration-Resistant Prostate Cancer that Have Failed to Chemotherapy with Docetaxel Obando C A 1, Desanvicente-Celis Z 1, Gonzalez L 2, Muschett D 1, Gonzalez F 1, Goldberg P 1 . . . . . . . 1Janssen, Panama, Panama, 2Janssen, Raritan, NJ, USA Objectives: To assess the cost-effectiveness of Abiraterone Acetate plus Prednisone (A-P) compared with Cabazitaxel plus Prednisone (C-P) in Dominican Republic, in patients with Metastatic Castration-Resistant Prostate Cancer (mCRPC) that have failed to chemotherapy with Docetaxel. Methods: A three-health state cohort simulation Markov Model (progression-free, post-progression and death) was developed based on overall and progression free survival data. The time frame was 10 years. The perspective was that of the Public System of Health of Dominican Republic. The health outcomes of interest were Quality Adjusted Life Years (QALYs) and Life Years (LYs). Efficacy data was taken from clinical trials (COU-AA-301 for A-P and TROPIC for C-P). Utilities for health states and negative utilities for adverse events were estimated based on quality of life endpoints of the COU-AA-301 trial. The base year was 2012. All costs are presented in Dominican currency (Dominican Pesos - RD$). Costs and outcomes were discounted at 5%. Probabilistic sensitivity (PSA) analysis was performed to evaluate uncertainty surrounding the parameters. Results: A-P resulted in 0.79 QALYs and 1.35 LYs, per patient, respectively. C-P resulted in 0.71 QALYs and 1.28 LYs, per patient, respectively. Mean total costs per patient were: RD$ 2.204.289 for A-P and RD$ 2.732.365 for C-P. The results of the probabilistic sensitivity analysis showed that, when compared with C-Z, A-P was found dominant (associated with reduced costs and increased QALYs) in the majority of the iterations. A-P had a 75% probability of being cost effective, independent of the willingness to pay, when compared to C-P. Conclusions: A-P can be considered cost-saving (dominant), when compared with C-P, in patients with Metastatic Castration-Resistant Prostate Cancer that have failed to chemotherapy with Docetaxel, from the perspective of the Public System of Health of Dominican Republic. PCN122 Clinical and Economic Analysis of Effectiveness of Everolimus in the Treatment Of HR+, HER2- Advanced Breast Cancer in Russia Krasnova L , Vorobiev P , Holownia M Russian Society for Pharmacoeconomics and Outcomes Research, Moscow, Russia . . . Objectives: Modern therapeutic approaches in treatment of advanced breast cancer can achieve clinically significant regression of symptoms, prolong life and improve its quality. Aim of this study was to conduct clinical and economic analysis of application of everolimus in the treatment of hormone-receptor-positive (HR+), human epidermal growth factor receptor-2-negative (HER2-) advanced breast cancer in postmenopausal women. Methods: An epidemiological and pharmacoeconomic evaluation of HR+, HER2- advanced breast cancer in postmenopausal women with using a survey of experts from different regions of Russia and modeling method. Calculating the cost of drugs and medical services was conducted according to experts and standard of medical care. Filling of a Markov model was conducted without using and with using of everolimus in the treatment of the patients within 5 years. Calculated indicators were: the impact of the disease on budget, the cost of one additional year of life. Results: From 13 regions of Russia 8 experts refused to provide information on the questionnaire, which may indicate the unwillingness to disclose information on epidemiology and tactics of treatment of disease. The burden of breast cancer for 5 years without the use of everolimus in the treatment regimens of patients with postmenopausal HR+, HER2- advanced breast cancer and with using it were: in Moscow 118.668.419€ and 137.596.651€ ; St. Petersburg-36.730.318€ and 38.133.492€ ; Republic of Khakassia-18.854.270€ and 19.812.467€ ; Omsk region-32.428.540€ and 33.603.456€ ; Primorsky Krai-39.176.077€ and 40.877.880€ . The use of of everolimus with exemestane in the treatment of advanced breast cancer increases by 1.5-2 times life expectancy and its “cost-effectiveness” indicator is 2 times lower comparing to exemestane monotherapy and chemotherapy. Sensitivity analysis using the results from 5 regions of Russia showed unidirectional comparison. Conclusions: The use of everolimus with exemestane is the dominant technology of treatment HR+, HER2- advanced breast cancer in postmenopausal patients compared with traditional technology of application of chemotherapy drugs or exemestane alone. PCN123 A Cost-Effectiveness Analysis of EGFR-TK Mutation Status-Guided 1st- and 2nd-Line Treatment of Stage III/IV Non-Small Cell Lung Cancer in the UK Patel K , Montouchet C , Cheynel J , Ruff L Covance Inc., London, UK . . . . Objectives: Lung cancers are the most common malignant tumours, accounting for 1.38 million annual deaths worldwide. Non-small cell lung cancer (NSCLC), the predominant tumour subtype, is associated with significant deteriorations in both survival and quality of life. Epidermal growth factor receptor tyrosine kinase (EGFR-TK) has emerged as a drug therapy target. The National Institute for Health and Care Excellence (NICE) recommends erlotinib – an EGFR-TK inhibitor – for first-line treatment of NSCLC in EGFR-TK mutation-positive patients, and secondline treatment in all patients irrespective of EGFR-TK mutations. We developed a model to assess the cost-effectiveness of an EGFR-TK mutation status-guided treatment strategy for stage III/IV NSCLC, compared with a strategy not dependent on mutational status. Methods: A Markov model was developed from the perspective of the UK National Health Service (NHS) over a lifetime horizon. This compared a current scenario (in which a cohort of NSCLC patients received doublet chemotherapy at first-line therapy, followed either by erlotinib or docetaxel at second-line) to a revised scenario (in which all EGFR-TK mutation-positive patients received erlotinib at first-line followed by second-line docetaxel, and all mutationnegative patients received doublet chemotherapy followed by either docetaxel or A636 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 erlotinib). Efficacy data were based on the TORCH and TAX317 randomised controlled trials. Cost data were obtained from NHS Reference Costs, British National Formulary list prices and other publically-available sources. Results: In the basecase analysis, the estimated incremental cost-effectiveness ratio exceeded the NICE willingness-to-pay threshold of £20,000 per quality-adjusted life year gained. Univariate and probabilistic sensitivity analyses suggested the results were robust to parameter changes, showing greatest sensitivity to variation in overall survival parameters. Conclusions: Our model suggests that, from the perspective of the UK NHS, an EGFR-TK mutation status-guided treatment strategy across first- and second-line treatment of NSCLC is not cost-effective compared with a strategy not dependent on mutational status. PCN124 Comparative Cost-Effectiveness Study of Modern Radiation Therapies in Hungary for Localized Prostate Cancer Zemplényi A 1, Kalo Z 2, Mangel L 1, Endrei D 1, Boncz I 3 1University of Pécs, Pécs, Hungary, 2Eötvös Loránd University (ELTE), Budapest, Hungary, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . Objectives: The introduction of innovative medical devices with high investment and operational costs is often delayed in countries with severe resource constraints. Cost-effectiveness analysis can help decision-makers to understand the economic value of such technologies. The purpose of our study was to compare the cost-effectiveness of two modern radiation therapy techniques, the stereotactic body radiation therapy (SBRT) and intensity-modulated radiation therapy (IMRT) compared to the 3-dimensional conventional radiation therapy (3DCRT) for treatment of low- to intermediate-risk prostate cancer in Hungary. Methods: A Markov model was constructed with the following disease states of a 65-year-old patient with organ confined prostate cancer: no evidence of disease after radiation therapy, hormone therapy, chemotherapy, death. Transition probabilities were calculated based on the international literature for SBRT, IMRT and 3DCRT. Utility values for each health state were obtained from publically available secondary sources. Costs in the model were calculated based on the Hungarian Health Insurance Fund rates, and were converted to EUR by applying actual exchange rates (1 EUR = 305 HUF). Analysis was conducted from payer perspective for 65-year-old patients over 10 years time horizon. Results: Based on preliminary calculations the expected mean cost of patients undergoing SBRT, IMRT and 3DCRT were 2,201 EUR, 5,704 EUR and 11.549 EUR respectively. Expected QALYs were 6.00 for SBRT, 5.8 for IMRT and 3.9 for 3DCRT. Compared to 3DCRT, both IMRT and SBRT were less costly and resulted in more health gain. Conclusions: The modern SBRT and IMRT are not only cost-effective compared to the conventional 3DCRT but also provide a great cost saving potential for the Hungarian health care system and may improve access to radiation and quality of life for patients. Appropriate financial incentives in the DRG system should support the uptake of cost-effective hospital technologies in Hungary. PCN125 Systematic Critical Review of Economic Evaluations of Rituximab, Added to Conventional Chemotherapy Regimen in the Treatment of Patients with Chronic Lymphocytic Leukemic Refractory Koury C D N 1, Nunes A A 2, Nita M 1 -Fundação de Ensino e Pesquisas Econômicas, Brasilia, Brazil, 2University of São Paulo USP, Ribeirão Preto, Brazil . . . . . . 1FIPE Objectives: To review the cost-effectiveness studies of chronic lymphocytic leukemia (CLL) treatment, in combination and in comparison with fludarabine and cyclophosphamide chemotherapy (R-FC) in refractory patients or patients who had been previously treated. Methods: Search and analysis of scientific evidence: the basics of The Cochrane Library, Centre for Reviews and Dissemination (CRD), Embase, Lilacs, Database of the Brazilian Network for Technology Assessment (SISREBRATS), and MEDLINE via PubMed were searched. Aiming to meet economic evaluations (AVE), or evaluations of health technologies (ATS), comparing schemas cyclophosphamide and fludarabine (CF) and the same plus Rituximab (R-FC). Studies were only selected in second-line treatment for CLL. Results: Two economic evaluations studied the treatment of patients with refractory or relapsing disease (R-FC vs FC). In the study, 24% had improvement in progression-free survival outcome (p < 0.05) in the R-FC, with more patients achieving partial or complete response in this group (61% vs 49%, p < 0.05). There was no statistically significant difference in overall survival. The Rituximab caused more adverse effects, but values of statistical tests for these outcomes are not presented. In a technology assessment conducted by NICE, even with reservations, the drug was recommended in view of the British health care system. Conclusions: There is significant uncertainty in the relevant outcomes for stages of refractory or relapsing disease. Few clinical trials evaluating the effectiveness of Rituximab in patients with CLL, which demonstrate no impact on overall survival, were found. In addition to the significant increase in costs for managing the disease. PCN126 What is the Most Cost-Effective Strategy for Treating Chronic Myeloid Leukemia After Imatinib Loses Patent Exclusivity in Europe? Padula W V 1, Conti R 2, Larson R 1 of Chicago, Chicago, IL, USA, 2University of Chicago, Chicago, GA, USA . . . . 1University Objectives: To analyze the cost-effectiveness of treating all chronic-phase chronic myeloid leukemia (CML) with imatinib initially compared to physician-choice between imatinib or the second-generation tyrosine kinase inhibitors (TKIs) dasatinib or nilotinib. Imatinib will lose patent exclusivity between 2015-2016 and its price is expected to drop 60-90% within one year throughout Europe. Methods: A Markov model simulating “step-therapy” compared to “physician-choice” in treating CML in 2015 through 5 years. The model assumes a European societal perspective. In both approaches, if initial treatment fails, patients are switched to a second-generation TKI. Patients are assumed to switch if they fail to meet efficacy endpoints: complete cytogenetic response (CCyR) or major molecular response (MMR). The model assumes stabilized prices of second-generation TKIs, but discounts the price of imatinib: 100% for first 6-months; 60-80% for second 6-months; and 10-30% thereafter. For each drug, tolerance, efficacy and the probabilities of treatment choice, switching and failure were drawn from published clinical trials. Quality-adjusted life years (QALYs) were based on U. K. preference weights (Szabo et al. 2010). According to Hoyle et al. (2011), direct medical costs per patient were: £20,244 for imatinib; and ~£30,000 for dasatinib and nilotinib. Additional costs included patient monitoring and allogeneic transplantation. Costs and QALYs were discounted at 3% (British Pounds Sterling (£); 2013). Sensitivity analyses tested parameters for impact on results at a willingness-to-pay of £50,000/QALY. Results: Step-therapy costs less and offers clinically-equivalent utility (£62,388; 2.864 QALYs) compared to physician-choice (£71,268; 2.879 QALYs), at an ICER of £592,000/QALY. The results are robust to changes based on univariate analyses of each parameter. Multivariate probabilistic sensitivity analyses found step-therapy cost-effective in 99.9% of 10,000 Monte Carlo simulations. Conclusions: When imatinib loses patient protection between 2015-2016 throughout Europe, it will be the cost-effective initial treatment strategy for CML compared to second-generation TKIs. PCN127 Literature Review of Decision-Analytical Models Used in the Economic Evaluation of Empirical/Targeted Antifungal Treatments for Invasive Fungal Infections Buyukkaramikli N C 1, Gaultney J 2, Leunis A 1, Severens J L 1 for Medical Techonology Assessment (iMTA), Rotterdam, The Netherlands, 2Mapi Group, Houten, The Netherlands . . . . . . 1Institute Background: Invasive fungal infections (IFIs) are an important cause of morbidity and mortality in immunocompromised patients. Based on the pathogen identification status, either empirical (without diagnosis) or targeted (with diagnosis) antifungal therapy is administered to symptomatic patients (e.g. with fever). Several antifungal agents are available and their cost-effectiveness is often evaluated using decision analytic models (DAMs). Objectives: The objective was to review all published DAMs used in economic evaluations of empirical/targeted antifungal treatments for IFIs. This approach is novel as previous reviews were either pathogen or agent-specific. Methods: A review was conducted in MEDLINE/EMBASE to identify all economic evaluations that included DAMs published until 1-1-2014. Previous reviews were checked for additional studies. Non-English and studies of prophylactic treatment were excluded. Data extracted included: population, indication, comparators, model structure, time horizon, outcomes, events, year, country, and sponsorship. Results: Overall, 24 published economic evaluations including a DAM were identified. 54% (n= 13) were for targeted treatments and the remaining (n= 11) for empirical treatments. 62% of the DAMs on targeted treatments (n= 8) focused on invasive pulmonary aspergillosis and the remaining 38% (n= 5) on invasive candiasis/candidemia. The majority (73%, n= 8) of DAMs evaluating empirical treatments focused on patients with persistent fever/febrile neutropenia. Lipid formulation amphotericin-B was a comparator in 46% (n= 11) of the studies, followed by caspofungin in 42% (n= 10) and voriconazole in 42% (n= 10). 92% of the DAMs (n= 22) included only a decision tree, whereas the remaining 8% (n= 2) embedded a lifetime Markov model. The majority (54%, n= 13) had a hospital perspective and time horizon of less than 12 weeks (54%, n= 14). Only one study utilized real-world data. Conclusions: There are major differences in the modeling approach, time horizon, comparator (s), treatment sequences and outcomes of published economic evaluations in IFI. A list of minimal, consensus-based methodological and structural requirements for DAMs on antifungal treatments of IFIs, elicited from key experts is needed. PCN128 Expansion of the Norwegian HPV Vaccination Program Sæterdal I , Juvet L , Jimenez E , Couto E , Klemp M , Torkilseng E B Norwegian Knowledge Centre for the Health Services, Oslo, Norway . . . . . . . Objectives: To evaluate the cost-effectiveness of expanding the Norwegian HPV vaccination program to catch-up females and 12 years old boys. Methods: We systematically searched the literature for randomized clinical trials (RCTs) that examined the effect of HPV vaccines on cancer mortality and incidence, precancerous stages and serious adverse events. We assessed selected publications for potential risk of bias, and the overall quality of the evidence for each outcome using GRADE. We adapted a published economic model to the Norwegian setting with respect to incidence of HPV-related outcomes, costs and quality adjusted life years (QALYs) lost from HPV-related diseases. The cost utility analysis reported results in Euros/ QALY gained in both a public health budget and a societal perspective. Results: We included 46 publications reporting on 13 RCTs for young women, and 3 on 2 RCT for boys (maximum follow-up period: three-four years). We found a borderline protective effect of HPV catch-up vaccination on all CIN2+, with a pooled risk ratio (RR) of 0.80 (95% CI: 0.62-1.02) for a follow-up period of 4 years. HPV catch-up vaccination was associated with a reduction in VIN2+ and VaIN2+ lesions, and genital warts. No difference in risk of serious adverse events was seen in vaccinated participants versus unvaccinated women (pooled RR of 0.99 (0.91-1.08)). We are currently reviewing the studies on boys. From a public health budget perspective, catch-up vaccination led to higher costs and health gains and an ICER= 70371€ . From a societal perspective, the incremental costs were lower, resulting in an ICER= 67365€. Conclusions: This systematic review indicates that a HPV catch-up vaccination could be beneficial and cost-effective for young women. The long-term effect of such a vaccination, and its effect on cancer incidence and mortality is still unclear. PCN129 Cost-Effectiveness of Radical Prostatectomy, Radiation Therapy and Active Surveillance for the Treatment of Localized Prostate Cancer – A Claims Data Analysis Brandes A 1, Koerber F 1, Schwarzkopf L 1, Hunger M 1, Waidelich R 2, Rogowski W 1 1Helmholtz Zentrum München, Neuherberg, Germany, 2University of Munich, Munich, Germany . . . . . . A637 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Standard treatment for localized prostate cancer is radical prostatectomy (PE) or radiation therapy (RT) which frequently cause erectile dysfunction (ED) and incontinence (IC). As tumor progression often is slow, active surveillance (AS) has been proposed as an alternative treatment strategy. This study compares the cost-effectiveness of the three treatment strategies in a German context. Methods: Based on claims data of a German sickness fund we analyzed men diagnosed with prostate cancer (ICD-10 code C61) in 2008. Life years gained and complication rates of ED and IC as well as costs of inpatient and outpatient treatment, pharmaceuticals, physical therapy, medical aids and copayments were tracked for 2.5 years after the initial treatment. An excess-cost analysis was applied. Strategies were compared in an age-matched and comorbidity-adjusted approach. Results: The baseline study sample included 25,376 individuals. Exclusion of metastases, other cancer diagnoses and treatment strategies resulted in 910 men with PE, 292 with RT and 124 with AS. After matching 107 men remained in the AS group and 214 each in the PE and RT groups with a mean age of 70 years. Risk of long-term ED (PE: 0.112, RT: 0.009, AS: 0.056) and IC (PE: 0.313, RT: 0.009, AS: 0.084) was highest in the PE group. Compared to RT and AS, PE was associated with more life years gained during the cause of the study. Due to high inpatient costs of the initial surgery PE had ca. € 11,000 higher total per capita costs than RT and AS. Conclusions: The analysis indicates that PE is associated with better prognosis and higher overall costs compared to RT and AS. 2.5 years follow-up might, however, not be enough to detect prostate cancer-specific deaths. PCN130 Critical Review of Cost-Effectiveness Analyses (CEA) of Prevention Strategies Against Diseases Associated with Human Papillomavirus (HPV) Infection Gervais F 1, Jiang Y 1, Largeron N 2, Nikoglou T 2 1Amaris, London, UK, 2Sanofi Pasteur MSD, Lyon, France . . . . Objectives: It is estimated that almost all cervical cancers are associated with HPV infection. In most industrialised countries, cervical screening and vaccination with a bivalent or quadrivalent vaccine are recommended to prevent the disease. The current study aimed to critically review the results of CEAs that have assessed the trade-off between screening and vaccination. Methods: A systematic literature review was conducted in order to explore the cost-effectiveness of HPV vaccination strategies with or without different screening strategies within the geographical context of Western Europe, North America and Australia. Modelling approach, disease considered, vaccination/screening settings and costs were compared. Results: A total of 1,188 citations were identified and 20 studies were included in the review. Heterogeneity was seen across studies in terms of modelling approach, disease and prevention strategies considered. Inclusion of more HPV-related diseases significantly improves cost-effectiveness. The strategies combining screening and vaccination were found to be cost-effective when compared to vaccination or screening alone. In terms of screening strategy, HPV DNA testing with cytological triage showed a trend to be the optimal strategy in vaccinated girls. However the gain in benefits reduced as the interval between screenings is reduced. Delaying the starting age of screening could be cost saving, with a limited increase in risk of cancer. An increasing vaccine valence seemed to counterbalance the detrimental effect of delayed/less frequent screening while the total costs of cervical disease prevention/treatment may be maintained or decreased. Lastly, vaccine price seemed to affect the incremental cost-effectiveness ratio proportionally. Conclusions: Despite heterogeneity in methodology across studies, similar trend of cost-effectiveness of competing prevention strategies was witnessed. In light of the trial results of the new nonavalent HPV vaccine, which provides protection against five additional types of the virus, the optimal prevention strategy needs to be reassessed within local context. PCN131 Cost-Effectiveness Analysis of Fulvestrant in the Treatment of Metastatic Breast Cancer in Second-Line Chemotherapy Krysanov I 1, Krysanova V 2 1Postgraduate Medical Institute, Moscow National University of Food Production, Moscow, Russia, 2I. M. Sechenov First Moscow State Medical University, Moscow, Russia . . Objectives: To conduct a pharmacoeconomic evaluation of the application of fulvestrant compared with docetaxel and paclitaxel in the treatment of metastatic breast cancer in second-line chemotherapy. Methods: Literature review of clinical effectiveness and safety of use of fulvestrant was conducted. Assess of the quality of research and level of evidence obtained in these results was performed. Direct medical costs consisted of the cost of the drug, the cost of patient management and correction of side effects. Duration of therapy, its effectiveness and side effects were obtained from relevant studies on clinical effectiveness (CONFIRM 2013, S. Jones et al. 2005). The cost of certain hematologic side effects have been taken from the study Belousov DU et al, 2012. To estimate the duration of hospital stay in the development of not hematological side effects, conducted a survey of experts. After calculating the total medical costs on compared regimens was conducted cost - effectiveness analysis with the calculation of CER. ResultsAccording to studies CONFIRM, 2013 and S. Jones et al. 2005., in patients taking fulvestrant PFS and OS were to 6.5 and 26.4 months, docetaxel - 5,7 and 15.4 months, paclitaxel - 3,6 and 12.7 months. The total cost of treatment were maximal for the docetaxel - 17685 USD, significantly lower for fulvestrant - 11803 USD and the minimal for paclitaxel - 7205 USD Costeffectiveness analysis showed that in spite of the average cost of treatment, taking into account its effectiveness in PFS and OS, the best CER was shown for fulvestrant, followed by paclitaxel and docetaxel. The sensitivity analysis showed that the simulation results are resistant to increase of the prices for fulvestrant i up to 12%. Conclusions: The use of fulvestrant for the treatment of metastatic breast cancer in second-line chemotherapy is more cost effective than the appointment of docetaxel and paclitaxel. PCN132 Cost-Effectiveness Evaluation of Brentuximab Vedotin for Refractory/Relapsed Hodgkin Lymphoma: A Comparative Analysis of the Results of Mexico and Venezuela Meza-Torres B 1, Gay J G 1, Jakouloff D E 2 1Ti Salud, Mexico City, Mexico, 2Takeda Pharmaceuticals, Zurich, Switzerland . . . . . Objectives: Brentuximab vedotin is an orphan drug currently indicated for treatment of patients with refractory/relapsed hodgkin lymphoma CD30+ following prior Auto Stem Cell Transplant (ASCT) or following two prior chemotherapy regimens. This is a group of patients with a reported median survival of 12 months, with no defined standard of care and for whom clinical trials are single armed due to lack of appropriate comparators and scarcity of patients. Hence, an indirect comparison was performed to determine the cost-effectiveness of brentuximab vedotin in different countries. Methods: A three state Markov model was developed. Effectiveness of brentuximab vedotin was obtained from the clinical trial of Gopal 2012. Effectiveness for the control group was obtained from 3 clinical trials evaluating survival of post-ASCT patients where data was disaggregated based on the patients´ response to prior ASCT/chemotherapy. The assumption was that only patients with ASCT/chemotherapy failure would serve as controls. The treatments received by the control group were based on the review of Martinez 2013, where 64% received chemotherapy, 29% AlloSCT and 8% AutoSCT. Simulations were run for the Mexican and Venezuelan contexts. Direct medical costs were obtained from the local public sectors and WHO-CHOICE. Results: For the base case scenario of both countries the ICERs (USD/LYG) were respectively $38,614.34 (Mex) and $57,854.07 (Ven), which compares favorably against accepted ICERs in the orphan drugs field. In the univariate sensitivity analysis the model was mainly sensitive to the costs of brentuximab, AutoSCT and AlloSCT. Conclusions: Brentuximab vedotin is a cost-effective alternative for both countries, especially in the space of orphan drugs. The low costs of AutoSCT and AlloSCT in Venezuela relative to its GDP were what mainly accounted for higher ICERs. Differences in chemotherapy usage and costs did not alter the model. As a limitation, local epidemiology was not accounted for due to lack of data. PCN133 Economic Evaluetion of Fulvestrant 500 MG (F500) Versus Original Nonsteroidal Aromatase Inhibitors in Patient with Advanced Breast Canser in Russia (2 LINE THERAPY) Krysanov I 1, Tyapkina M 2 Medical Institute, Moscow National University of Food Production, Moscow, Russia, 21st Moscow State Medical University named after I.M. Sechenov, Moscow, Russia . . 1Postgraduate Objectives: to perform cost-effectiveness analysis fulvestrant 500mg (F500) for the treatment of first progression or recurrence of advanced breast cancer in postmenopausal patients compared with anastrozole 1mg (ANAS1), letrozole 2.5mg (LET2,5), exemestane 25mg (EXE25) and exemestane 25mg+everolimus 10mg (EXE25+EVE10). Methods: the data on efficacy and safety of 2-line hormonal therapy of breast cancer were derived from a network meta-analysis and clinical data publication for overall survival (OS), progression free survival (PFS) and serious adverse events (SAE). We considered the direct costs on second and third line hormonal therapy and resource utilization. Data on resource usage, were based on expert opinion and open sources. 1-way sensitivity analyses were conducted. Results: in terms of OS F500 (mean 23.33 month) was as effective as ANAS1 (22.12) and more effective than LET2.5 (17.44) and EXE25 (18.31). The highest incremental cost-effectiveness ratio (ICER) estimated for F500 versus ANAS1 was 84,592 USD per year with incremental effectiveness 1.21 month. The lowest ICER estimated for F500 versus LET2.5 was 22,873 USD per year with incremental effectiveness 5.90 month. The ICER for F500 versus EXE25 was 25,890 USD per year. In terms of PFS EXE25+EVE10 was more effective and costly, than F500. The CER for F500 was 1,714 USD per year versus 4,215 USD for EXE25+EVE10. A series of one-way sensitivity analyses showed this result is robust to variations in costs of drugs, physician examination, and variation in costs associated with SAE. Conclusions: the use of F500 is more effective than LET2.5 and EXE25, and at least as efficacious as ANAS1 in terms of OS among postmenopausal women with advanced breast cancer after failure on 1-line endocrine therapy. In terms of PFS F500 less efficacious than EXE25+EVE10, however substantially cheaper. From perspective of federal health care system, the cost of LYG for F500 is less than the willingness to pay threshold. PCN134 Will Governments be Able to Afford a Cancer Cure Under Current Health Economic Evaluation Methods? Jamali E 1, Focsa S 2 1PAREXEL, Sommerville, NJ, USA, 2PAREXEL, London, UK . . Objectives: Cancer accounts for around 1.3 million deaths and € 50 billion in health care expenditure in the European Union. Balancing increasing treatment costs and prevalence will be increasingly difficult for governments to manage. Advances in immunotherapies provide hope for a cancer cure; however its cost might be out of reach for governments under current health economic evaluation methods which will be the aim of this research. Methods: The years of life lost (YLL) in the UK due to cancer were obtained from the Institute of Health Metrics and Evaluation (IHME) database and multiplied by the NICE cost effectiveness threshold of £20,000 per Quality Added life Year (QALY), this gave a first estimate of the potential cost of a cancer cure that would be within an acceptable cost effectiveness threshold. This cost was then modified to take into account the quality of life (QoL) of the general population, QALY discounting, cancer onset age, and other demographics. YLL due to disability in cancer were not included in the calculation. Results: It is estimated that 32.4% of the total YLL per year in the UK (5,615,310) are a consequence of cancer. The cost of saving these YYL at £20,000 per QALY was estimated to be around £12 billion for all cancers per year, meaning an extra £425 in taxes would have to be generated A638 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 from each taxpayer. Conclusions: A cancer cure evaluated under current health economic evaluation methods would cause a budget impact that would be unaffordable for governments due to the high prices that could be achieved while remaining cost effective. Although these types of technologies therapies are not currently available, payers might want to explore new methods of evaluation, as exploring the possibility of calculating costs based on quality adjusted lifetimes rather than years or increasing discount rates on QALYs for immunotherapies. PCN135 Economic Impact of the Inclusion of Pertuzumab for the Treatment of Metastatic Breast Cancer HER2 + Nazco G 1, Gutierrez F 1, Bullejos M 1, Viña M 2, Gonzalez I 1, Valcarcel C 3 1University Hospital of Canary Islands, La Laguna, Spain, 2University Hospital of Nuestra Señora de la Candelaria, Santa Cruz de Tenerife, Spain, 3Canary Islands Foundation for Health and Research, Santa Cruz de Tenerife, Spain . . . . . . Objectives: To analyze the economic impact of the incorporation of pertuzumab for the treatment of metastatic breast cancer HER2+ in a University Hospital according to real data of our patients. Methods: Cross-sectional study where the patients with breast cancer were analyzed in our hospital during one year (April 2013 to April 2014). The demographic data of the patients (age and weight) and antineoplastic treatments used were obtained from the computer program Hospiwin®. The dose and efficacy data have been obtained from the phase III CLEOPATRA trial. This clinical trial compares docetaxel + trastuzumab vs docetaxel + trastuzumab + pertuzumab with progression-free survival (PFS) 12.4 (CI 10.4-13.5) vs. 18.5 (CI 16.621.6) months respectively. Costs of the drugs were included in the economic model developed in Excel® data base. The time horizon was one year and the perspective of medical leadership of the hospital was used. Results: During the study period 371 patients were treated for breast cancer and 75 patients (20.2 %) were HER2+. The mean weight of 71.5 kg (SD = 17.1) and men BMI of 29.3 were obtained. The annual cost of docetaxel + trastuzumab + pertuzumab was 69,245.32 € vs 29,837.4 € (CI (in the docetaxel + trastuzumab treatment group. The cost per PFS per year was 44,964 € (CI 38,469-50,177 € ) in the docetaxel + trastuzumab + pertuzumab group vs 29,837 (CI 26,640-35,948 € ) in the docetaxel + trastuzumab treatment group. The incremental cost effectiveness ratio (ICER) was 15.127 € /PFS per year. Conclusions: The addition of pertuzumab to treatment with docetaxel / trastuzumab for metastatic breast cancer has shown an increase in SLP. However, the economic impact of this new drug, requires careful selection of patients who could benefit. Health authorities will have to consider whether pertuzumab is cost-effective in terms of their willgness to pay. PCN136 Cost Effectiveness of Sunitinib as First-Line Targeted Therapy for Metastatic Renal Cell Carcinoma in China Shi Q 1, Yin H 2, Xuan J 3, Wu Y 4, Cheng G 5 1Pfizer China, Shanghai, China, 2Philadelphia College of Osteopathic Medicine-Georgia Campus, Suwanee, GA, USA, 3Fudan University, School of Public Health, Shanghai, China, 4Foshan food and drug inspection testing center, Foshan City, China, 5Jinan University, Guangzhou, China . . . . . Objectives: Multitargeted receptor tyrosine kinase inhibitors are more effective alternatives to interferon-α and monoclonal antibodies in patients with metastatic renal cell carcinoma (mRCC). However, studies on the economic and humanistic outcomes associated with these treatments are sparse in the Chinese setting. This study evaluated the clinical and economic consequences of sunitinib compared with sorafenib and interferon-α from the third-party payer’s perspective in China. Methods: A Markov model was developed to simulate disease progression and determine cost and outcomes over patient’s lifetime. The time horizon of analysis was patients’ lifetime with a maximum of five years in cycles of six weeks. The model was used to conduct a cost-utility analysis on sunitinib compared to interferon-α and sorafinib. Costs of physician, anti-cancer medications, hospitalization, laboratory, and palliative care were estimated. Outcomes were measured in progression-free life years (PFLYs), life years (LYs) and quality-adjusted life years (QALYs). A 3.5% discount rate was applied to both costs and QALYs gained. Results: In the base case, the total cost of the sunitinib arm was RMB217,038.50, the progression-free life year was 1.57, life year was 2.55, and QALY was 1.70. The incremental cost per PFLY between sunitnib and IFN-α was –RMB78,562.10 and RMB 22,501.03 between sunitinib and sorafenib. The incremental cost per life year between sunitnib and IFN-α was –RMB168,633.00 and RMB 21,022.38 between sunitinib and sorafenib. The incremental cost per QALY between sunitnib and IFN-α was – RMB184,825.00 and RMB 29,493.42 between sunitinib and sorafenib. Conclusions: This economic study used the final clinical results of the pivotal sunitinib trial that provides more accurate modeling results than previous studies based on extrapolation. It was found that sunitinib was dominant compared to IFN-α . Sunitinib was cost effective compared to sorafenib based on the threshold recommended by the World Health Organization. PCN137 An Evidence-Based Model Design to Inform the Cost-Effectiveness Evaluation of Primary Endocrine Therapy And Surgery for Older Women with Primary Breast Cancer Mousa R 1, Chen L C 1, Cheung K L 2 of Nottingham, Nottingham, UK, 2University of Nottingham, Derby, UK . . . . . 1University Objectives: Despite the lack of evidence-based information on their clinical and cost-effectiveness, surgery and primary endocrine therapy (PET) are the most commonly used initial treatment strategies for older women with primary breast cancer in the United Kingdom (UK). To evaluate the cost-effectiveness of PET and surgery, a decision analytical modelling is necessary. This systematic review aimed to summarise the modelling methodologies from the literature to inform the model design in older women. Methods: An electronic database search was conducted using NHS Economic Evaluation Database, Cochrane Library, Ovid Medline, PubMed, and EMBASE to identify full economic evaluations that compared different treatment strategies in postmenopausal women with primary breast cancer. Quality and modelling methodologies of included studies were assessed and summarised. Results: All the 31 included studies assessed surgery and none assessed PET as the initial treatment. Most included economic studies used a Markov model with life-time horizon and 1-year cycle length. Nine studies which included sub-group analysis for older women (over 65 years old) used similar economic models and transition states with younger women (50 to 65 years old). The key disease-related health states were disease-free, recurrence, and death. Recurrence was mostly separated into loco-regional and distant recurrence. Conclusions: This systematic review can inform the design of an economic model comparing PET with surgery as initial treatment in older women based on the following assumptions: (1) health states are applicable across age groups; (2) transition states for modelling surgery in the literature are transferable to model the same treatment for older women; (3) metastasis transition states including progression, progression-free, and death can be used to model the PET pathway. Future study will validate this model by using a longitudinal dataset of older women with primary breast cancer, and synthesize data from different data sources to populate this economic model. PCN138 Cost Effectiveness of Cetuximab in 1st-Line Treatment of RAS WildType Metastatic Colorectal Cancer in Scotland: A Summary of the Submission to the Scottish Medicines Consortium Jarrett J 1, Ovcinnikova O 1, Hnoosh A 2, Harty G 2, Byrne B 3, von Hohnhorst P 4 Serono Ltd., UK, UK, 3Merck Serono Ltd., FELTHAM, UK, 4Merck Serono, Darmstadt, Germany . . . . . . 1MAPI, London, UK, 2Merck Objectives: Colorectal cancer is the third most common cancer in Scotland, with nearly 4,000 cases reported in 2011 and 5.2% increase in incidence in the previous 10 years. Recent studies have shown that a subgroup of metastatic colorectal cancer (mCRC) patients with wild-type (wt) RAS (KRAS and NRAS exons 2,3, 4) expressing tumours are likely to have enhanced response to anti-EGFR treatment compared to patients with mutant RAS exons (2,3, 4). RAS biomarkers aid identification of the patient group that is likely to benefit the most from anti-EGFR treatment such as cetuximab and therefore allow more efficient use of NHS Scotland resources. A New Product Assessment Form was submitted to the Scottish Medicines Consortium with the aim of demonstrating the latest improved outcomes in RAS wt mCRC patients (versus KRAS wt) treated with cetuximab in combination with chemotherapy and its cost effectiveness compared to currently available treatments. Methods: A state-transition Markov cohort model was developed to simulate patient outcomes and costs for first and subsequent lines of treatment including the long-term survival after a successful curative resection of liver metastases. Results: The model estimated an incremental 0.28 life-years gained (LYG) with cetuximab + FOLFIRI compared to FOLFIRI alone and an incremental 0.32 LYG with cetuximab + FOLFOX compared to FOLFOX alone. The model was most sensitive to length of treatment with cetuximab. Conclusions: The incremental cost effectiveness ratios imputed in the model are close to the traditional willingness to pay threshold adopted by the SMC. This analysis demonstrates that cetuximab in combination with FOLFIRI or FOLFOX in mCRC RAS wt patients is a cost-effective treatment compared with chemotherapy alone, specifically when taking into consideration that cetuximab qualifies as an end of life medicine (following SMC criteria) which raises the value of such intervention. PCN140 Cost-Effectiveness Analysis of Bevacizumab, Fotemustine and Extended-Dose Temozolomide in Patients with Recurrent Glioblastoma in Spain Garcia Lopez J L 1, Rodriguez Barrios J M 2, Puig-Junoy J 2, Carrato Mena A 1 Cajal University Hospital, Alcala University, Madrid, Spain, 2Pompeu Fabra University, Barcelona, Spain . . . . . . 1Ramon Y The treatment of glioblastoma after first-line treatment progression is not clearly established in Spain. Most accepted alternatives are nitrosoureas (fotemustine, F), extended-dose temozolomide (eT) or bevacizumab (B). Without clear standards of care, increased clinical and health policy uncertainty among decision makers should be clarified. So, economic evaluation might reduce those uncertainties. Objectives: To analyze the cost-effectiveness of bevacizumab, extended-dose temozolomide and fotemustine in patients with either recurrent or progressive glioblastoma after standard therapy, compared to standard clinical practice (SCP). Methods: A costeffectiveness markov model was conducted from a payer perspective (time horizon 1 year, 3%, discount rate, 2012, € ). Our model got three health states: alive without progression, alive with toxicity and progression as absorbing state. We subsequently performed a deterministic and probabilistic analysis of sensitivity. Main efficacy outcome was progression-free survival at six months (6m PFS). Toxicity data was based on relevant phase II studies. Health state utility values were estimated based on published values from an HTA report by Garside et al, 2007. Costs were obtained from a Spanish University Hospital. Results: Cost/effectiveness ratios were: SCP (based on carmustine) 2,368.45 € /year to obtain 6m PFS with stable health state utility value, F 4,112.97 € /year, B 15,122.49 € /year and eT 5,470.05 € /year. Incremental cost-effectiveness ratios were: F 7,404.12 €/year € /year to obtain 6m PFS with stable health state utility value, B 40,371.8 €/year and eT 45,853.51 €/year. Tornado diagram and CEAC showed our results robustness. Conclusions: ICER analysis shows fotemustine to be the dominant option in the treatment of patients with recurrent or progressive glioblastoma. PCN141 Pharmacoeconomic Analysis of Axitinib as Second-Line Treatment for Metastatic Renal Cell Carcinoma Kulikov A , Komarov I First Moscow State Medical University named after I. M. Sechenov, Moscow, Russia, . . A639 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: To identify the dominant scheme of mRCC second-line target treatment (compare two alternatives – Axitinib and Everolimus). Methods: Based on the Markov model, the cost-effectiveness analysis was realized. Overall survival, annual survival rate, time to progression of disease and direct cost of mRCC treatment was evaluated. Costs analysis included: costs of two target therapy lines (Sunitinib as a first-line in combination with Axitinib or Everolimus treatment); cost of 3 and 4 grade side effects compensation; cost of diagnosis and inpatient care; cost of disease progression; cost of palliative and best supportive care. Results: During the pharmacoeconomic analysis of Axitinib use as a second-line therapy for mRCC, it was found that this target therapy regimen would significantly increase the time to progression and the overall survival which amounted to 22,75 months, with an annual survival rate of 68%, 38% and 17% of patients following the first, second and third year of treatment, respectively. Despite the high cost of this treatment regimen, reaching 51.327 EUR at the horizon of ten year study, the treatment regimen including Axitinib will be characterized by the lowest values of the cost-effectiveness ratio, reflecting the costs incurred by the health care system for patient’s life saving, and incremental cost-effectiveness ratio, which was less than willingness to pay threshold in Russia. Conclusions: It is shown, Axitinib use as second-line of target therapy in patients with mRCC is the most preferable treatment regimen then Everolimus from the pharmacoeconomic point of view. PCN142 Cost-Effectiveness Analysis of Hydralazine and Magnesium Valproate LP Associated With Treatment for Adult Patients with Metastatic Recurrent or Persistent Cervical Cancer in Mexico Soto H 1, Sanchez K 2, Escobar Juárez Y 2, Constanzo A 2, Fernandez Z 3, Melendez C 3 1Iteliness Consulting, Mexico City, Mexico, 2HS Estudios Farmacoeconómicos, Mexico City, Mexico, 3Psicofarma S.A. de C.V., Mexico City, Mexico . . . . . . Objectives: Demonstrate through an economic evaluation of cost-effectiveness that using Hydralazine LP magnesium valproate (Transkrip ®) associated with first-line chemotherapy in the treatment of persistent or recurrent metastatic stage IVB cervical cancer, type not candidates for surgery or radiotherapy is more effective than the alternatives available in Mexican health institutions: cisplatin with topotecan (CT), cisplatin with paclitaxel (CP), cisplatin with vinorelbine (CV), carboplatin with paclitaxel (CaP), paclitaxel (P), cisplatin and cisplatin with gemcitabine © (CG). Methods: A cost-effectiveness analysis was developed, using a Markov model with a time horizon of 2 years divided into 24 monthly cycles, the measure of effectiveness was determined as the years gained free survival (PFS), being an advanced cancer is significant, only were measured direct medical costs, an analysis of incremental cost-effectiveness was performed. To test the robustness of the model a deterministic and probabilistic sensitivity analysis was performed. Results: The cost per patient using therapy with hydralazine LP magnesium valproate (Transkrip ®) is $ 142,109.93, gaining 0.8846 years in Progression free survival this treatment was more effective but more expensive, paclitaxel had a cost of $ 26,988.64 with 0.3631 years in PFS, this therapy is less expensive but more effective than all comparators. The ICER of (Transkrip ®) was $ 220.757 pesos per year in PFS versus paclitaxel. Conclusions: The economic results of treatment of patients with metastatic cervical cancer, through an epigenetic therapy with hydralazine LP magnesium valproate (Transkrip ®) over a time horizon of two years showed that hydralazine LP magnesium valproate (Transkrip ®) is a cost effective alternative respect to comparators showed a greater response in years of progression-free survival and ICER below of 2 GDP. PCN143 Prioritization of Future Outcomes Research Studies in Chronic Myeloid Leukemia: Value of Information Analysis Rochau U 1, Kühne F 2, Jahn B 2, Kurzthaler C 2, Corro Ramos I 3, Chhatwal J 4, Stollenwerk B 5, Goldhaber-Fiebert J D 6, Siebert U 7 1UMIT - University for Health Sciences, Medical Informatics and Technology/ ONCOTYROL Center for Personalized Cancer Medicine, Hall in Tyrol/ Innsbruck, Austria, 2UMIT - University for Health Sciences, Medical Informatics and Technology, Hall in Tyrol, Austria, 3Erasmus University Rotterdam, Rotterdam, The Netherlands, 4MD Anderson Cancer Center, Houston, TX, USA, 5Helmholtz Center Munich, Neuherberg, Germany, 6Western University, London, WA, USA, 7Medical Informatics and Technology, and Director of the Division for Health Technology Assessment and Bioinformatics, ONCOTYROL, Hall i. T, Austria . . . . . . . . . . Objectives: Value-of-Information analysis can help to guide decision about future research priorities: If and what further research is needed? Our aim was to guide decision regarding future outcomes research on parameters related to different regimens for chronic myeloid leukemia (CML). Methods: We updated a previously developed state-transition Markov model of CML, which evaluates seven treatment regimens including tyrosine kinase inhibitors, chemotherapy and stem cell transplantation (SCT). We derived model parameters from published trials data, Austrian clinical, epidemiological, and economic data. We performed a cohort simulation over a lifetime horizon, adopted a societal perspective, and discounted costs and benefits at 3% annually. We calculated the expected value of perfect information (EVPI), partial perfect information (EVPPI), and the population EVPI (PEVPI). Additionally, we examined the expected value of sample information (EVSI) for different trial sizes. Results: Three strategies are on the efficiency frontier: imatinibàchemotherapy/SCT, nilotinibàchemotherapy/SCT (140,000 € /QALY) and nilotinibàdasatinibàchemotherapy/SCT (176,000 € /QALY). The EVPI for eliminating all uncertainty resulted in a curve with two peaks. One peak is around a WTP threshold of 150,000 €/QALY (EVPI 4,600 € ) and another peak is at 180,000 € /QALY (EVPI 7,700 € ). The PEVPI for Austria assuming a 10-year technology horizon was 2.5 million € (WTP 150,000 € /QALY) and 4.5 million € (WTP 180,000 € /QALY). EVPPI identified four parameters most responsible for decision uncertainty: duration of first-line therapy, probability of progressing from chronic phase to accelerated phase, probability of receiving a SCT, and the health-utility after SCT. The EVSI commented on the optimal study size for these parameters given the cost of obtaining information. Conclusions: Acquiring additional evidence could prove valuable for determining optimal treat- ment regimens for chronic myeloid leukemia. If further research were funded, studies should examine a combination of natural history, treatment, and quality of life parameters, especially the effectiveness of first-line TKI treatment. PCN144 Decision Analysis on the Cost-Effectiveness of Sequential Treatment Strategies for Patients with Chronic Myeloid Leukemia in the United States Rochau U 1, Kluibenschaedl M 1, Stenehjem D 2, Kuo K L 2, Oderda G 2, Brixner D 3, Siebert U 4 - University for Health Sciences, Medical Informatics and Technology/ ONCOTYROL - Center for Personalized Cancer Medicine, Hall in Tyrol/ Innsbruck, Austria, 2University of Utah, Salt Lake City, UT, USA, 3UMIT - University for Health Sciences, Medical Informatics and Technology/ ONCOTYROL - Center for Personalized Cancer Medicine/ University of Utah, Hall in Tyrol/ Salt Lake City, UT, Austria, 4Medical Informatics and Technology, and Director of the Division for Health Technology Assessment and Bioinformatics, ONCOTYROL, Hall i. T, Austria . . . . . . . . 1UMIT Objectives: The first goal was to adapt an existing Austrian decision-analytic model for chronic myeloid leukemia (CML) treatment to the US-American health care context. Secondly, we updated the model with new data and further treatment strategies to identify the most effective and most cost-effective strategy for the treatment of CML patients with different sequential tyrosine kinase inhibitors (TKIs). Methods: We evaluated 18 different treatment strategies within the US-American setting in terms of survival, quality-adjusted survival and costs. For model parameters, data from literature, a US-American expert survey, the Utah Cancer Registry, and economic data from a US-American database were used. Evaluated treatment strategies included imatinib, dasatinib, nilotinib, bosutinib, ponatinib, stem-cell transplantation and chemotherapy. The Markov state-transition model was analyzed as a cohort simulation over a lifelong time horizon, a third-party payer perspective was adopted and a discount rate of 3% was used. Additionally, several deterministic and probabilistic sensitivity analyses were conducted. Results: Imatinib without second-line TKI resulted in an incremental cost-utility ratio (ICUR) of $148,700/QALY gained (incremental cost-effectiveness ratio (ICER) of $128,800/Lys) compared to baseline strategy ‘chemotherapy’. Imatinib with second-line nilotinib yielded an ICUR of $217,100/QALY gained (ICER $242,200/ LY) compared to imatinib without second-line TKI. Imatinib followed by secondline bosutinib had an ICUR of $331,300/QALY gained (ICER $265,100/LY) compared to imatinib followed by second-line nilotinib. Imatinib with second-line dasatinib produced an ICUR of $343,200/QALY gained (ICER $279,600/LY) compared to imatinib with second-line bosutinib. All remaining strategies were excluded due to dominance. ICURs and ICERs obtained from the probabilistic sensitivity analysis deviated up to 6.5% (2.5%) compared to base-case ICURs (ICERs). Conclusions: Based on our analysis and current treatment guidelines, we recommend imatinib followed by second-line nilotinib as the most cost-effective treatment strategy. Our model results may support clinicians and patients in CML treatment decision making. PCN145 The Cost-Effectiveness of Brentuximab Vedotin in Hodgkin Lymphoma in Sweden Engstrom A Takeda Pharma Sweden, Solna, Sweden . Objectives: To assess the cost-effectiveness of using brentuximab vedotin (BV) in treating relapsed or refractory Hodgkin Lymphoma compared to standard chemotherapy and allogeneic stem cell transplant in the Swedish health care setting. Brentuximab vedotin is a novel antibody drug conjugate targeting CD-30 and is indicated for treating relapsed/refractory Hodgkin Lymphoma. Methods: A Markov model with a lifetime horizon was constructed to compare BV to chemotherapy or allogeneic stemcell transplant (alloSCT). The analysis had a societal perspective and included lost productivity using a human capital approach. The model uses comparators relevant to Sweden and all epidemiological and cost parameters were based on Swedish sources. Both costs and effects were discounted at 3% according to Swedish guidelines. Clinical effectiveness for BV was based on pivotal clinical trial results and published data from the literature for the comparators relevant to the reimbursement authorities and to enable long term modelling. Outcomes were measured in QALYs. Uncertainty was addressed both through probabilistic sensitivity analysis and one-way analyses of central variables. Results: Brentuximab vedotin dominated alloSCT (i. e a lower treatment cost and a better health outcome) and the ICER when compared to chemotherapy was SEK 419 000 (€ 47 000). One-way sensitivity analyses showed that the results were stable when central variables were varied. The probabilistic analysis also showed that brentuximab vedotin had a high probability of being the most cost-effective treatment at the accepted threshold values for all scenarios. Conclusions: The ICERs calculated were all below commonly accepted willingness to pay for a QALY in Sweden for both comparator scenarios. Brentuximab vedotin is a cost effective treatment option for relapsed/ refractory Hodgkin Lymphoma in the Swedish health care setting. PCN146 Economic Evaluation of Axitinib for Second Line Treatment in Adult Patients with Advanced Renal Cell Carcinoma – the Portuguese Case Miguel L S 1, Luz R 2 - ISEG/UL, Lisboa, Portugal, 2Centro Hospitalar Lisboa Central, Lisboa, Portugal . . . 1CISEP Objectives: This study estimated the cost-utility of axitinib after sunitinib failure in adult patients with renal cell carcinoma. Total costs and quality adjusted life years accrued with axitinib was compared to everolimus, the only drug for second line treatment financed by the Portuguese National Health Service. Methods: A 4-week cycle Markov model with three health states (progression free, post progression, and death) was adapted to the Portuguese setting. In the absence of head-to-head clinical trials and the unfeasibility of a standard indirect comparison, relative efficacy was based on a previous simulated treatment comparison. Axitinib trial data on quality of life (utility) was used for the progression free stage and A640 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 assumed equal for everolimus, while utilities for the post progression stages were obtained from the literature. Resource use was determined by a panel of five experienced experts to reflect Portuguese clinical practice. Official unit costs were used, following the Portuguese National Health Service perspective. The model adopted a lifetime frame (15 years) with a 5%discount rate. Results: Axitinib allowed an increment of 0.20 years of progression free survival, 0.53 years of overall survival, and 0.32 quality adjusted life years compared to everolimus. Despite having a similar daily cost, the use of axitinib implied an incremental cost of 9,100€ , mainly due to the increase in progression free survival, that matches second line treatment duration. Consequently the cost per quality adjusted life year was 28,598€ . Sensitivity analyses showed that results were robust to model parameters specification, with the main uncertainty source being clinical efficacy. Conclusions: Axitinib increased progression free and overall survival, which allowed patients to benefit from more quality adjusted life years at a cost increase. Overall, it was possible to advocate that axitinib is cost-effective, as the cost per QALY is below commonly accepted thresholds. PCN147 Economic Evaluation of Paclitaxel Albumin, Paclitaxel, and Docetaxel as a Second Line Treatment for Metastatic Breast Cancer Gharaibeh M , Malone D C University of Arizona, Tucson, AZ, USA . . . Objectives: Clinical studies have shown that docetaxel to be superior to paclitaxel in overall survival (OS) and progression free survival (PFS) (median OS: 1.28 vs 1.06 year; median PFS: 0.47 vs 0.30 year) for the treatment of patients with metastatic breast cancer progressing after an anthracycline-based regimen. Other studies have shown paclitaxel-albumin extended OS by 9.7 weeks, and TTP by 4 weeks. An economic evaluation based on these two clinical trials was performed to compare paclitaxel albumin, paclitaxel, and docetaxel as a second line treatment for metastatic breast cancer. Methods: A Markov model was conducted using three health states: PFS, progressed, and death to estimate overall survival, cost, life year gain (LYG) and quality adjusted life year (QALY). Efficacy data for the treatments were obtained from the published literature. In the absence of head-to-head trials, comparative efficacy and safety of taxanes were estimated using indirect comparisons. A 3% discount rate for cost and outcomes was used. Cost of chemotherapy, administering, monitoring the disease, loss of productivity, and adverse drug reactions for patients on treatment were included from the US societal perspective. Results: Compared to docetaxel, paclitaxel albumin was found to be less expensive ($36,241 vs $73,510) and more effective in term of QALYs (0.782 vs 0.710). The incremental cost effectiveness ratio (ICER) for paclitaxel albumin compared to paclitaxel was $77,670/ QALY. The probabilistic sensitivity analysis showed that paclitaxel albumin has 70% probability of being cost effective at $100,000/QALY threshold value. Conclusions: Paclitaxel-albumin is an attractive treatment option for the treatment of metastatic breast cancer in patients who have failed 1st-line treatment for metastatic disease. The primary analysis comparing paclitaxel albumin to docetaxel demonstrated that paclitaxel albumin dominated docetaxel because it was less costly and more effective. PCN148 Cost Effectiveness Analysis of Targeted Intraoperative Radiotherapy Alone (TARGIT-A) in Early Breast Cancer Patients Vaidya A 1, Vaidya P 2, Both B 3, Brew-Graves C 4, Vaidya J 4 University, Maastricht, The Netherlands, 2O-Zone HEOR Consultancy, Maastricht, The Netherlands, 3Carl-Zeiss Meditec AG, Oberkochen, Germany, 4University College London, London, UK . . . . . 1Maastricht Objectives: Whole-breast external beam radiotherapy (EBRT) is normally given over 3-6 weeks after lumpectomy in early breast cancer patients to reduce recurrence and mortality. An individualised risk-adapted approach to adjuvant radiotherapy has been tested in the randomised TARGIT-A trial which tested the efficacy of one dose of radiation to tumour bed during lumpectomy. The objective of the present study was to assess the cost effectiveness of TARGIT-A in these patients. Methods: A model based economic evaluation compared single dose TARGIT-A with current practice of EBRT in UK. A state transition Markov model approach was used to simulate the treatment outcomes in a time horizon of 20 years post-surgery. The primary outcome of interest was quality adjusted life years gained (QALY) and analysis was conducted from the health care payer’s perspective. To address decision uncertainty, probabilistic sensitivity analysis was performed. A discount rate of 3.5% was applied to future costs and effects. Results: In the Base Case Analysis TARGIT-A was a dominant strategy yielding higher QALYs at a lower cost than EBRT. Discounted EBRT and IORT costs for the time horizon of 20 years were £ 20,926 and £ 14,461 respectively. Discounted incremental QALY gained by use of IORT was 0.0069. Model results were robust to parameter uncertainty and probabilistic results were similar to the deterministic results. Application of the net monetary benefit (NMB) framework revealed higher NMB for TARGIT-A in all Monte Carlo simulations. Cost effectiveness acceptability curves show that TARGIT-A is cost effective at various willingness to pay thresholds. Conclusions: TARGIT-A is a cost effective strategy to treat early breast cancer patients in the UK. Implementation of this one-off radiation treatment within a risk-adapted approach could improve quality of life by sparing them from the protracted course of EBRT, improve compliance, prevent unnecessary mastectomies and save valuable NHS resources. PCN149 Early Cost-Effectiveness Modeling for Tumor Infiltrating Lymphocytes (TIL) -Treatment Versus Ipilimumab in Metastatic Melanoma Patients Retèl V P 1, Steuten L M G 2, Mewes J C 2, van Harten W H 2 Cancer Institute, Amsterdam, The Netherlands,, 2University of Twente, Enschede, The Netherlands . . 1Netherlands . . . . . . . Objectives: Metastatic melanoma has a poor prognosis with 10 year survival being < 5%. Standard therapy is the effective but costly Ipilimumab. An emerging 1st line treatment is Tumor Infiltrating Lymphocytes (TIL), with response rates > 50% and expected survival rates of 25%-42% versus 45% (1yr) and 23,5% (2yr) for Ipilimumab. TIL is highly personalized, however complex and requests substantial upfront investments from the hospital in expensive lab-equipment, staff expertise and training, as well as extremely tight hospital logistics. Therefore, an early health economic modelling study, supporting a Coverage with Evidence Development (CED) program, was performed. Methods: We used a Markov decision model to estimate the expected costs and outcomes (quality adjusted life years; QALYs) for TIL versus Ipilimumab in metastatic melanoma patients from a societal perspective over a life long time horizon. Three mutually exclusive health states (stable disease, progressive disease and death) were modelled, divided in first and second line treatment. Technical failures and non-compliance were incorporated to reflect the dynamic nature of the technology. To inform further research prioritization, Value of Information (VOI) analysis was performed. Results: TIL is expected to yield more QALYs compared to Ipilimumab (0.99 vs 0.52 respectively) at lower total costs (€ 83,588 vs € 87,834 respectively). Based on current information TIL has a probability of 88% for being cost effective at a cost/QALY threshold of €30,000. Expected Value of Perfect Information (EVPI) amounted to € 1,2 million. Partial EVPI (EVPPI) was highest for survival data (€ 550,000). Expected Value of Sample information was estimated € 355,000 for an optimal sample size of n= 50. Conclusions: TIL is expected to improve QALYs compared to Ipilimumab at lower incremental cost and has the highest probability of being cost-effective. To reduce decision uncertainty, a future clinical trial to investigate survival seems most valuable, and should preferably be undertaken as part of a CED program. PCN150 A Cost Effectivness Analysis of Everolimus Compared with Axitinib in the Treatment of Metastatic Renal Cell Carcinoma in the United Kingdom Chandiwana D 1, Perrin A 2, Sherman S 2 1Novartis Pharmaceuticals UK Limited, Camberley, UK, 2Analytica LA-SER International, Inc, New York, NY 10018, NY, USA . . . Objectives: This study assessed the cost-effectiveness of everolimus versus axitinib for the treatment of advanced metastatic renal cell carcinoma (mRCC) in the United Kingdom (UK). Methods: A Markov model was developed with three health states: stable disease, disease progression and death. The model time horizon was 12 years and a UK NHS perspective was considered. There are no head to head studies comparing everolimus with axitinib, thus evidence from a weighted adjusted indirect analysis based on the RECORD-1 and AXIS trials was used to compare progression-free survival (PFS) for everolimus versus axitinib. Survival distributions for PFS were fitted to the post-matched population and fit statistics were generated. As overall survival (OS) data were not available from the AXIS trial at the time of the indirect analysis, the model assumed that the OS for axitinib was equivalent to that of everolimus, based on OS from the RECORD-1 trial. The Weibull survival distribution was used for both PFS and OS. Quality of life data were derived from the Swinburn et al. study and drug costs were obtained from the British National Formulary. Results: Everolimus resulted in a progression-free life expectancy of 0.60 years compared to 0.57 with axitinib. Everolimus resulted in 0.65 QALYs compared to 0.63 QALYs for axitinib. Active drug costs were £8,105 for everolimus and £25,723 for axitinib. Total costs were higher for axitinib (£42,533) compared to everolimus (£24,387). The cost difference reflects the higher treatment costs per month and longer treatment duration for axitinib compared to everolimus. Therefore, the incremental cost of axitinib compared with axitinib was -£18,146, highlighting that everolimus is less expensive. The incremental cost per QALY gained was -£1,048,954. Conclusions: This cost-effectiveness analysis demonstrates that everolimus likely dominates axitinib, i.e. it is more effective and less expensive compared with axitinib in the treatment of mRCC. PCN151 Cost-Minimization Analysis of Trastuzumab Intravenous Versus Trastuzumab Subcutaneous for the Treatment of Patients With HER2+ Early Breast Cancer And Metastatic Breast Cancer in Greece Mylonas C 1, Kourlaba G 2, Fountzilas G 3, Skroumpelos A 4, Maniadakis N 1 School of Public Health, Athens, Greece, 2Collaborative Center for Clinical Epidemiology and Outcomes Research (CLEO), Athens, Greece, 3Aristotle University of Thessaloniki School of Medicine, Thessaloniki, Greece, 4Roche (Hellas) S.A., Athens, Greece . . . . . 1National Objectives: To conduct an economic evaluation comparing Herceptin subcutaneous formulation (Herceptin-SC) with -Herceptin intravenous formulation (Herceptin-IV), in the treatment of patients with human epidermal growth factor receptor 2-positive (HER2+) early and metastatic breast cancer (EBC-MBC), in the Greek health care setting. Methods: A cost-minimization model was developed to compare the total cost of care, from the hospital perspective, for new and existing patients, over 18 cycles therapy course. Total cost of therapy reflects drug acquisition cost, consumables dispensed, hospital overheads, physician and other staff time. Costing data were obtained from official Government sources (in 2014) and resource utilization data from a local validation of an international time and motion study. Due to the short time horizon of the study, costs were not discounted. Results: The mean total cost of therapy per patient on Herceptin-IV was estimated at € 24,163 compared to € 23,042 per patient receiving Herceptin-SC. Drug acquisition costs accounted for € 22,630 and € 22,579 of total therapy costs for Herceptin-IV and Herceptin-SC, respectively. Following drug acquisition costs, the administration cost was € 518 and € 161 for Herceptin-IV and Herceptin-SC, respectively. Moreover, the central venous access device cost was € 290 and € 0 of the total costs of Herceptin IV and Herceptin SC, respectively. Finally, overhead costs made up approximately € 725 of the total cost for Herceptin-IV and € 302 for Herceptin-SC. Sensitivity analysis showed that the results of the model were sensitive to drug acquisition costs and patient weight. Conclusions: The cost of treatment with Herceptin-SC is A641 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 lower than that with Herceptin-IV in the management of patients with HER2+ EBC and MBC. Hence, the substitution of Herceptin-IV with Herceptin-SC can produce valuable savings for the Greek health care system, especially in the current economic environment where hospitals’ pharmaceutical budget has significantly been reduced. PCN152 Cost-Minimization Analysis of Bevacizumab VersUS Cetuximab in First-Line Treatment for Metastatic Colorectal Cancer in Kras Wild-Type Patients in the Supplementary Health Care System in Brazil Tsuchiya C T , Kim H S J , Maximo M F M , Ramos L A Roche Brazil, São Paulo, Brazil . . . . . . . . . . Objectives: Due to increasing costs in cancer management, there is a crescent need to rationally allocate resources in health care systems. Recently, a head-tohead phase III study (CALGB80405) showed no significant difference in OS and PFS for first line (1L) mCRC in KRAS wild-type (wt) patients amongst bevacizumab (Bev) and cetuximab (Cet) – the most commonly used biologics in this setting. Since benefit of both drugs is comparable, the aim of the study was comparing treatment costs of Bev vs. Cet in 1L KRAS wt mCRC. Methods: A cost-minimization analysis was conducted under payer perspective in Brazilian Supplementary Healthcare System. Backbone chemotherapy regimens (mFOLFOX6 and FOLFIRI) were based on CALGB80405 trial. Direct medical costs regarding drug acquisition, material and procedures/service fees were included. Adverse events management costs were excluded. The resource usage data was taken from the literature and drug labels. Costs were taken from CMED price list and UNIMED reimbursement lists. A univariate sensitivity analysis was conducted varying parameters from ±20% range. Results were reported in Brazilian Reais (BRL). Results: The average monthly cost per patient was lower with Bev: BRL23’945 (Bev+mFOLFOX6) vs. BRL30’017 (Cet+mFOLFOX6) – reduction of 20.2% - and BRL23’008 (Bev+FOLFIRI) vs. BRL29’075 (Cet+FOLFIRI) – reduction of 20.9%. Average monthly cost per patient according to mFOLFOX6/FOLFIRI usage proportion reported on CALGB80405 was BRL23’699 (Bev) and BRL29’766 (Cet); considering PFS data presented in the trial, the average total treatment cost was estimated as BRL256’899 (Bev) and BRL311’060 (Cet). The sensitivity analysis showed that model was more influenced by Cet price, Bev price and patient height. Conclusions: Bev is a cost-saving choice for 1L KRAS wt mCRC in combination with chemotherapy, potentially achieving around 20% of reduction in monthly direct treatment costs compared to Cet, mainly because of Cet higher total acquisition costs and weekly administration schedule, resulting in additional resource consumption. PCN153 Economic Impact of Using Subcutaneous Trastuzumab Gutierrez F 1, Nazco G 1, Viña M 2, Bullejos M 1, Gonzalez I 1, Valcarcel C 3 1University Hospital of Canary Islands, La Laguna, Spain, 2University Hospital of Nuestra Señora de la Candelaria, Santa Cruz de Tenerife, Spain, 3Canary Islands Foundation for Health and Research, Santa Cruz de Tenerife, Spain . . . . . . Objectives: To analyze the economic impact of the incorporation of trastuzumab subcutaneous (TSC) in a University Hospital according to real data of our patients. Methods: Retrospective cost minimization study that included patients diagnosed with breast cancer treated with trastuzumab intravenous (TIV) from april 2013 to april 2014. The demographic data of the patients (age and weight) and antineoplastic treatments used were obtained from the computer program Hospiwin®. An economic model was developed in Excel® data base, based on the dose used in previous clinical trials: IV loading dose of 8mg/kg and after 6mg/kg/3 weeks and SC fixed dose of 600 mg/3 weeks. The time horizon was one year and the perspective of medical leadership of the hospital was used. The Spain cost of TSC is not aproved yet. Two posibilities was analyzed: The cost of filing 600mg of TSC equal to the cost of a 68kg patient with TIV (situation A) and the cost of a 63kg patient with TIV (situation B). A sensitivity analisys included the cost of using an oncology chair (168€ /treatment) was performed. Results: During the study period 371 patients were treated for breast cancer. Of these 75 were treated with TIV (20.2%), with an average weight of 71.5 kg (SD= 17.1) and a cost of 990,996.88€ /per year. If all patients had been treated with TSC: Situation A the total spending would be 829,965.4€ ; situation B the total spending would be 768,938.5€ . So the savings would be 161,031.4€ (19.4%) and 222,058.3€ (28.8%) respectively. If the cost of oncology chair (not necessary for the TSC) it´s included, the savings would be 253,549.4€ and 314,576.3 respectively. Conclusions: In this study we wanted to show how TSC saved costs in all of the situations analyzed. The TSC is a therapeutic innovation that helps promote the systems health´s sustainability. PCN154 Pharmacoeconomic Analysis of Oral Capecitabine and Tegafur for Colorectal Cancer Treatment in Russia Gerasimova K 1, Avxentyeva M 1, Rebrova O 2 Russian Presidential Academy of National Economy and Public Administration, Moscow, Russia, 2Pirogov Russian National Research Medical University, Moscow, Russia . . . 1The Objectives: To conduct a pharmacoeconomic analysis of oral drugs, tegafur vs capecitabine, for advanced colorectal cancer (CRC) in adult patients. Methods: Indirect comparison and network meta-analysis of clinical efficacy and safety of tegafur vs capecitabine and tegafur + calcium folinate vs capecitabine were performed. Cost-minimization analysis (CMA) with calculation of cost minimization difference was used for economic evaluation of studied drugs. Results: There was no statistically significant difference in the full and partial objective tumor response between oral tegafur (both in monotherapy or in combination with calcium folinate) and capecitabine for advanced CRC treatment in an indirect comparison and network meta-analysis. Capecitabine vs tegafur + calcium folinate has less 3-4th grade stomatitis but there was no difference in the incidence of diarrhea and 3-4th grade nausea/vomiting. There was no difference in safety between tegafur and capecitabine monotherapy in terms of incidence of diarrhea, vomiting, stomatitis/mucositis. The hand-foot syndrome occurrred in less than 5% in case of tegafur. Tegafur (in monotherapy or in combination with calcium folinate) is less costly than capecitabine. The difference in costs in favor of tegafur monotherapy amounted to € 1,956.97 per 1 patient per 6 months or € 3,778.53 per year; of tegafur + calcium folinate - € 2,168.12 and € 4,220.06 per 1 patient per 6 and 12 months, respectively. Conclusions: Tegafur is a cost-saving option compared with capecitabine with similar efficacy and safety. PCN155 Cost-Effectiveness Analysis of Bendamustin-Rituximab Compared to Chop-Rituximab in the Treatment of Indolent Follicular NonHodgkin Lymhoma in the Czech Republic Mlcoch T 1, Klimes J 2, Kruntoradova K 2, Mandelikova M 1, Dolezal T 2 OUTCOMES, s.r.o., Prague, Czech Republic, 2VALUE OUTCOMES, Prague, Czech Republic . . . . . 1VALUE Objectives: There is new RCT phase 3 clinical evidence that bendamustinrituximab (B-R) is more effective in terms of progression free survival compared to the standard of care CHOP-rituximab (CHOP-R) in indolent non-Hodgkin lymphoma (iNHL). Based on this RCT, we performed a cost-utility analysis of B-R compared to CHOP-R in the treatment of follicular iNHL (stage III and IV) in the Czech Republic. Methods: We developed a life-time Markov cohort model with 28-day cycle length and 5 health states, i.e. on treatment, rituximab maintenance (R-M), stable disease, progression and death. Additionally, we modeled adverse effects of treatment and four sub-states during progression (observation, imunochemotherapy, R-M, post R-M). Transition probabilities and utilities were derived from published literature. Resource use (costs) was calculated from health care payer’s perspective in cooperation with major Czech hemato-oncologic experts. Costs and outcomes were discounted by 3.5%. Probabilistic sensitivity analysis (PSA) with 1000 iterations using a willingness to pay (WTP) threshold equal to 3 times GDP per capita (40 100 EUR) in the Czech Republic was performed. Results: Over a life-time horizon, B-R compared to CHOP-R brings additional 1.21 QALY (7.47 vs. 6.26) and 1.31 LYG (9.74 vs. 8.43). The incremental total costs were 1,368 EUR (total life time costs for B-R and CHOP-R were 43,080 EUR and 41,712 EUR, respectively). ICERs thus equal to 1,133 EUR/QALY and 1,044 EUR/LYG. The results of the PSA show that B-R is costeffective in 100% iterations under the WTP threshold; and simultaneously in 99.3% iterations is cost-effective while using threshold equal to 7,300 EUR. Conclusions: B-R proved that it is a highly cost-effective therapy in patients with follicular iNHL. The higher costs of initial bendamustin treatment are in the long-term horizon offset by substantial savings of progression costs. There is 100% probability of B-R being cost-effective at the selected WTP threshold. PCN156 ‘DE NOVO’ Quantification of Genotype-Directed Therapy with Afatinib in Metastatic Lung Cancer Zaim R 1, Tran L 2, Groen H J M 3, Uyl-de Groot C A 1 University, Institute for Medical Technology Assessment, Rotterdam, The Netherlands, 2Boehringer Ingelheim BV, Alkmaar, The Netherlands, 3University Medical Center Groningen, Groningen, The Netherlands . . . . . . . 1Erasmus Objectives: The inhibition of epidermal growth factor receptor (EGFR) signaling pathway by innovative therapeutics presents promising upshots in oncology. Our study aims to quantify first-line treatment with afatinib, an irreversible tyrosine kinase inhibitor, compared to pemetrexed+cisplatin (pem+cis), for patients with metastatic lung adenocarcinoma harboring common EGFR mutations (DEL19 or L858R) in the Netherlands. Methods: An area under the curve partitioned survival model, constructed to quantify lifetime consequences of therapy with afatinib versus pem+cis, was amended to the Netherlands. The updated (2014) LUX-Lung 3 trial results and data from public sources were used to populate the model. Study outcomes were expressed in quality-adjusted life years (QALY), incremental cost-utility ratios (ICUR) and net monetary benefits (NMB). The analyses were conducted from health care and societal perspectives. Uncertainty assessment was performed using one-way and probabilistic sensitivity analyses (PSA). Results: Metastatic lung adenocarcinoma patients with common EGFR mutations (89%) had higher overall survival when treated with afatinib compared to pem+cis (HR: 0.78, p=0.10). The corresponding base-case ICUR was <€20,000/QALY gained. For the subgroup of patients harboring DEL19 mutations (49%), treatment with afatinib resulted in cost-savings. Although NMB calculations were favorable for the genotype-directed therapy, inclusion of the entire patient population (all EGFR mutations) resulted in higher incremental costs. PSA results of lung adenocarcinoma patients with common EGFR mutations showed that afatinib is >95% cost-effective compared to pem+cis at a €80,000 threshold. Conclusions: This study shows that genotype-directed therapy with afatinib improved survival in metastatic lung adenocarcinoma and translated itself as value-for-money, particularly for the DEL19 subgroup, in the Netherlands. Further research is encouraged to compare afatinib with reversible EGFR inhibitors in this setting. PCN157 Model-Based Cost–Utility Analysis of Erythropoiesis-Stimulating Agents for the Treatment of Cancer-Treatment Induced Anaemia in the UK NHS Snowsill T 1, Huxley N 1, Hoyle M 1, Crathorne L 1, Haasova M 1, Briscoe S 1, Coelho H 1, Medina-Lara A 1, Mujica Mota R 1, Napier M 2, Hyde C 1 1University of Exeter, Exeter, UK, 2Royal Devon & Exeter NHS FT, Exeter, UK . . . . . . . . . . . Objectives: To assess the cost–utility of erythropoiesis-stimulating agents (ESAs) in conjunction with red blood cell transfusions (RBCTs) in patients with cancertreatment induced anaemia (CIA). Methods: A cost–utility analysis from an NHS and personal social services perspective was conducted by developing an ad hoc economic model. A lifetime time horizon was used and outcomes were discounted at 3.5% per annum. All ESAs were assumed to have the same clinical effectiveness. Haemoglobin (Hb) levels were assumed to drive health-related quality of life (HRQoL), with haemoglobin linearly mapped to utility. This was used to calculate A642 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 incremental quality-adjusted life years (QALYs) while ESAs were administered and during a Hb “normalisation period” following cancer treatment. Incremental long-term QALYs were accrued solely through extrapolated overall survival. Shortterm mortality and HRQoL associated with adverse events and RBCTs were not modelled. Costs included: ESA acquisition (list prices, British National Formulary) and administration, RBCT, additional blood tests with ESA therapy, and adverse event costs. Results: All ESAs except epoetin beta and darbepoetin alfa were cost-effective versus using RBCT only at an upper cost-effectiveness threshold of £30,000/QALY. Incremental cost-effectiveness ratios (ICERs) ranged from £19,400/ QALY (biosimilar epoetin alfa) to £35,000/QALY (epoetin beta). Probabilistic sensitivity analysis showed that biosimilar epoetin alfa was cost-effective at the lower cost-effectiveness threshold of £20,000/QALY in 50.9% of simulations. In 19.5% of simulations it was clinically effective but not cost-effective and in 31.4% of simulations it was dominated by RBCT only. Additional sensitivity analyses demonstrated that overall survival was one of the most influential and uncertain parameters. When the survival advantage of ESAs (not statistically significant) was removed, the ICERs for all ESAs were over £100,000/QALY. Conclusions: There is substantial uncertainty regarding the impact of ESA therapy on overall survival, which leads to significant uncertainty about the cost-effectiveness of ESAs in CIA. PCN158 Qaly Weightings Based on the Burden of Illness Applied to a Uk Cost-Effectiveness Analysis of Nab-Paclitaxel + Gemcitabine Versus Gemcitabine Alone for the Treatment of Metastatic Pancreatic Cancer (comparison cohort) who were the same age, race, and gender were identified and matched. A random index date was chosen to minimize selection bias. Patients in both cohorts were required to be at least age 18 years, with continuous medical and pharmacy benefits 1-year pre- and 1-year post-index date. One-to-one propensity score matching (PSM) was used to compare health care costs and utilizations during the follow-up period, between the diseased and comparison cohorts, and adjusted for baseline demographic and clinical characteristics. Results: After risk adjustment by PSM, a total of 19,079 patients in each cohort were matched. Significantly more breast cancer patients had inpatient admissions (23.77% vs. 12.56%, p< 0.0001) and long-term care (7.77% vs. 6.60%, p< 0.0001), other service (99.88% vs. 87.86%, p< 0.0001) and pharmacy visits (77.80% vs. 68.85%, p< 0.0001), compared to those without breast cancer. Breast cancer patients also incurred significantly higher inpatient ($2,141 vs. $1,537, p< 0.0001), long-term care ($7,471 vs. $5,335, p< 0.0001), other service visit ($23,592 vs. $14,780, p< 0.0001) and pharmacy costs ($3,379 vs. $2,787, p< 0.0001) compared to those in the comparison cohort. Conclusions: Breast cancer patients in the Medicaid program incurred substantially higher health care resource utilization and costs compared to those without the disease. PCN161 Nab-Paclitaxel or Docetaxel as Alternatives to Solvent-Based Paclitaxel in Metastatic Breast Cancer (Mbc): A Cost Utility Analysis from a Chinese Health Care Perspective Dranitsaris G 1, Yu B 2, Qing Z 2, King J 3, Zhang A 3, Kaura S 3 Pharma Consulting, Toronto, ON, Canada, 2Fudan University Shanghai Cancer Center, Shanghai, China, 3Celgene Corporation, Summit, NJ, USA . . . . . . 1Augmentium Objectives: To demonstrate the impact of QALY weightings based on the burden-ofillness (BoI) of pancreatic cancer on the incremental cost-effectiveness ratio (ICER) of nab-paclitaxel plus gemcitabine (NPG) versus gemcitabine (G). Methods: A markov model using data from the MPACT trial plus resource use data and costs from NHS Scotland have been submitted to the Scottish Medicines Consortium. The base case ICER was £52,885/QALY based on a cost of £8,232 and a QALY gain of 0.156 (SMC DAD). QALY weightings up to a maximum of 2.5 distributed across six modifier factors, including BoI, have been proposed (NICE consultation on Value Based Assessment), with BoI measured according to proportional QALY shortfall associated with the condition. The estimated 98% loss of healthy life (proportional QALY shortfall) in pancreatic cancer (Hutchings 2014) represents an almost complete loss of life, and thus a very high BoI. A BoI weighting of 2.5 (maximum weighting allocated entirely to BoI, or BoI FULL) and an alternative BoI weighting of 1.417 (maximum weighting shared equally between six modifiers, so 1/6thof 2.5, or BoI PARTIAL) were therefore applied to the QALY gain of NPG versus G. Results: The BoI FULL weighting gives an adjusted QALY gain for NPG versus G of 0.39 and a corresponding ICER of £21,108/QALY. The BoI PARTIAL weighting gives an adjusted QALY gain for NPG versus G of 0.221 and a corresponding ICER of £37,249/QALY. Conclusions: Various ways of accounting for disease severity can be considered and made workable by HTAs, including QALY weightings according to proportional QALY shortfall. The adjusted QALY gain and corresponding ICERs of NPG versus G in pancreatic cancer show that the value of medicines for life-threatening ‘end-of-life’ conditions with a high relative shortfall can be reflected by an appropriate system of QALY weightings. Objectives: Paclitaxel and docetaxel are used for the treatment of MBC in China. However, one important drawback, particularly with docetaxel, is the potential for dose-limiting toxicity. To improve the side effect profile and efficacy of paclitaxel, an albumin-bound formulation (nab-paclitaxel) is currently available in China (Abraxane®). Clinical trials have demonstrated that nab-paclitaxel is safer and more effective than both docetaxel and paclitaxel. To provide economic data for China, a cost utility analysis comparing nab-paclitaxel to docetaxel, both as alternatives to paclitaxel was conducted. Methods: Clinical data was obtained from a meta analysis of randomized trials comparing either nab-paclitaxel (260 mg/m2 q3wk) or branded docetaxel (100 mg/m2 q3wk) to solvent-based branded paclitaxel (175 mg/m2 q3wk). Health care resource use for the delivery of chemotherapy and the management of grade 3/4 toxicity was collected from a time and motion study in three Chinese cancer centers and from a survey of clinicians. Using the Time Trade-off technique, treatment preferences and utility estimates were obtained from interviewing 28 cancer patients from two centres in China. All costs were reported in 2014 $U. S. Results: Nab-paclitaxel had the most favourable safety profile characterized with the lowest incidence of grade 3/4 neutropenia, febrile neutropenia, anemia and stomatitis. This translated into lower costs for managing the grade 3/4 side effects of nab-paclitaxel relative to both docetaxel and paclitaxel ($21 vs. $166 vs. $81). In the preference assessment, 22 of 28 (78.6%) patients selected nab-paclitaxel as their preferred agent. As an alternative to paclitaxel, the cost per quality adjusted life year (QALY) gained was more favourable with nab-paclitaxel than docetaxel ($57,900 vs. $130,600 respectively). Conclusions: Nab-paclitaxel is an economically attractive alternative to paclitaxel and docetaxel in MBC, providing a substantially lower cost per QALY. Additionally in the patient preference survey, 78.6% of patients selected nab-paclitaxel as their preferred agent. PCN159 Economic Evaluation of Lapatinib in Her-2-Positive Metastatic Breast Cancer Patients in Egypt PCN162 The Cost-Effectiveness of Second-Line Crizotinib in Eml4-Alk Rearranged Advanced Non-Small Cell Lung Cancer Elsisi G 1, Mady E 1, Abo Taleb A A 2 for Pharmaceutical Affairs, Cairo, Egypt, 2WHO, Cairo, Egypt Djalalov S 1, Graham D M 2, Beca J 1, Hoch J S 3, Tsao M S 4, Leighl N 4 Hospital, Toronto, ON, Canada, 2Princess Margaret Hospital, Toronto, ON, Canada, 3Canadian Centre for Applied Research in Cancer Control (ARCC), Toronto, ON, Canada, 4Ontario Cancer Institute, Toronto, ON, Canada Cowell W 1, Gladwell D 2, Parnaby A 3 1Celgene UK, Uxbridge, UK, 2BresMed Health Solutions LTD, Sheffield, UK, 3Celgene International Sarl, Boudry, Switzerland . . . . . . . 1Central Administration Objectives: The objective of the current analysis was to assess the cost-effectiveness of lapatinib plus capecitabine versus capecitabine alone in human epidermal growth factor receptor-2-positive metastatic breast cancer patients from the third party payer perspective over a time horizon of ten years. Methods: A half cycle corrected Markov chain model comprising 3 health states (stable, progression and death) was developed to estimate the projected clinical and economic implications of Lapatinib. Transition probabilities were estimated based on the results from the EGF100151 clinical trial of Lapatinib. Health state utilities and major adverse events were obtained from published sources. Direct medical costs were obtained from the third party payer list. Costs (in 2013 EGP) and effects were discounted at 3.5% annually. One way sensitivity analyses were conducted. Results: The economic evaluation of lapatinib plus capecitabine as combination therapy resulted in additional cost of 1,597,796 EGP, with an incremental positive effect of 5.7 quality adjusted life years (QALY) or an incremental cost-effectiveness ratio (ICER) of 277,169 EGP/QALY gained. The overall survival of the two arms was found to have the greatest impact on the results. Conclusions: Compared with our willingness-to-pay threshold stated by world health organization for middle and lower income countries, the addition of lapatinib to capecitabine is not clearly cost-effective; and most likely to result in an ICER higher than the threshold limit. PCN160 Health Care Utilization and Costs of Breast Cancer in the Medicaid Program Li L 1, Shrestha S 1, Baser O 2, Wang L 1 Research, Plano, TX, USA, 2STATinMED Research and The University of Michigan, Ann Arbor, MI, USA . . . . 1STATinMED Objectives: To evaluate health care resource utilization and costs among patients diagnosed with breast cancer in the Medicaid program. Methods: Patients diagnosed with breast cancer (International Classification of Disease, 9thRevision, Clinical Modification [ICD-9-CM] diagnosis code 174, 233.0, 238.3, 239.3) were identified using Medicaid data from January 1, 2008 through December 31, 2010. The initial diagnosis date was designated as the index date. Patients without breast cancer . . . . . . . . . 1St. Michael’s Objectives: Targeted therapy with ALK inhibitor crizotinib offers significant improvement in clinical outcome for treatment of EML4–ALK fusion positive nonsmall cell lung cancer (NSCLC) patients. We estimated the cost-effectiveness of companion EML4-ALK genetic testing in combination with crizotinib treatment in the second-line setting for advanced NSCLC in Ontario. Methods: We performed a cost-effectiveness analysis using a Markov model from a Ministry of Health perspective and a lifetime horizon. Transition probabilities and mortality rates were calculated based on the data of a recent second-line randomized trial of crizotinib versus chemotherapy (Shaw et al. New Engl J Med 2013). Costs were obtained from OCCI database, public labs and Princess Margaret Hospital. All parameters were varied separately in one-way and selected two-way sensitivity analyses. Various scenarios to assess the impact of model assumptions about testing and treatment were conducted. Results: The use of pemetrexed and docetaxel in ALK-rearranged NSCLC, based on our preliminary model, could yield as much as 0.539 QALY and 0.429 QALY respectively, assuming no crossover from chemotherapy to crizotinib. Average costs per patient based on the preliminary model are estimated at CAD $19,388 for pemetrexed and $$33,226for docetaxel, with incremental cost-effectiveness ratios of $333,595/QALY and $125,812/QALY gained respectively. The results of the one-way sensitivity analysis indicated that the primary drivers of the ICER were the utilities and cost of crizotinib treatment. The model was least sensitive to IHC and FISH genetic test costs, re-biopsy cost, probability of progression while on pemetrexed treatment and probability of re-biopsy. Conclusions: EML4–ALK genetic testing in combination with crizotinib treatment for all NSCLC patients eligible for chemotherapy is not economically attractive in the current setting. Lower drug costs would be required to make this strategy economically feasible. PCN163 Cost-Effectiveness of Ipilimumab in Previously Untreated Patients for Advanced Melanoma in Sweden Barzey V 1, Asukai Y 1, Gueron B 2, Holmberg C 3, Kotapati S 4 . . . . . A643 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1IMS Health, London, UK, 2Bristol-Myers Squibb, Rueil Malmasion, France, 3Bristol-Myers Squibb, Bromma, Sweden, 4Bristol-Myers Squibb Pharmaceuticals, Wallingford, CT, USA Objectives: Ipilimumab was the first compound to substantially prolong survival in advanced melanoma. Evaluate the cost-effectiveness of ipilimumab in untreated advanced melanoma compared to dacarbazine and vemurafenib from a Swedish national payer perspective (TLV/NLT). Methods: A three-state Markov model with stable disease, progression and death was developed, estimating costs and benefits over a lifetime horizon. Given a lack of head-to-head data and a connected evidence network to allow for a robust NMA, the comparison used two sources of data unadjusted for study characteristics. Ipilimumab survival data were based on a pooled sample of treatment-naïve patients from clinical trials (n= 78) and real-world settings (n= 181), the clinical data package used for EMA submission. Parametric extrapolation methods were applied to dacarbazine data from CA184024. Mixture modelling was used to extrapolate vemurafenib data from BRIM-III. Resource use was taken from a survey of Swedish oncologists (n= 5). EORTC-8D utility data from an untreated population were used because they match the population of interest. Costs were obtained from official Swedish price lists. Survival and utility assumptions were varied in scenario analyses. Results: Ipilimumab was associated with a 0.93 QALY gain and an ICER of SEK782,000 (€ 84,000) versus dacarbazine. Ipilimumab dominated vemurafenib with a 0.76 QALY gain and a SEK109,000 (€15,000) cost-saving. The unadjusted comparison was the most conservative among alternative methods of clinical comparisons explored. The ICER versus dacarbazine was SEK521,000 (€ 56,000) using a published survival algorithm (7) and SEK532,000 (€ 57,000) using a covariate-adjusted survival regression based on the 78-patient dataset. A real-world scenario using the patient shares of dacarbazine and vemurafenib was also deemed cost-effective. Conclusions: As in the previously treated setting, ipilimumab produces large (> 0.5 year) survival and quality-adjusted survival gains relative to current treatments. TLV/ NLT considered ipilimumab a costeffective treatment for advanced melanoma based on these results. PCN164 Cost-Utility Analysis of Trastuzumab in Treatment Of Metastatic Her2-Positive Breast Cancer in Vietnam Nguyen T T C , Nguyen T T T University of Medicine and Pharmacy in HCMC, Ho Chi Minh City, Vietnam . . . . . . Objectives: Trastuzumab, a targeted therapy, has been widely used in treatment of HER2-positive breast cancer because of its proved effectiveness and safety by many studies but its economic impact with low-income countries like Viet Nam has not been assessed yet. The aim of this study is to evaluate the cost-effectiveness of trastuzumab in combination of standard therapy versus standard therapy in treatment of metastatic HER2-positive breast cancer. Methods: A Markov model was developed with 3 states (stable disease, progressive disease and death) to simulate a hypothetical cohort of 1,000 metastatic HER2-positive breast cancer women of an average age of 53 year old with the same entry criteria as in the M77001 study group. Chemotherapy with ACD regimen (Doxorubicin, Cyclophosphamide, Docetaxel) was compared with ACD regimen plus trastuzumab. The cycle length of model was 1 month and time horizon was lifetime. Both cost and Quality-adjusted life-years (QALYs) were discounted annually with 3% discount rate. Probabilistic sensitivity analysis was also conducted. Results: Combination of trastuzumab and standard therapy compared with standard therapy in treatment of metastatic HER2-positive breast cancer resulted in addition of 170.9 million VND (658,8 vs 487,9 million VND) and 0.81 QALY (1.77 vs 0.96). The incremental cost-effectiveness ratio resulted in VND 208,736,442.49/QALY, which was less than willing to Pay (WTP) of VietNam in 2013 (VND 253,503,360.00). Therefore, using trastuzumab in treatment of metastatic HER2-positive breast cancer women has been considered to be costeffective. Sensitivity analyisis showed that the most affecting factors on the costutility of trastuzumab are trastuzumab’s price and patient’s weight. Conclusions: Trastuzumab in combination of standard therapy is cost-effective in treatment of metastatic HER2-positive breast cancer women in Vietnam. Trastuzumab’s price and patient’s weight are the most affecting factors on the cost-effectiveness of trastuzumab. PCN165 What’s the optimal visual inspection screening intervals for cervical cancer screening in real practice of rural china? A costutility modeling study Li X 1, Goggins W 1, Zhao F H 2, Qiao Y L 3 1The Chinese University of Hong Kong, Hong Kong SAR, China, 2Cancer Institute and Hospital, Chinese Academy of Medical Sciences, Beijing, China, 3Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China . . . . . . Objectives: Chinese government initiated a nation-wide cervical screening program, covering 10 million rural women in 221 counties all over the country. The objectives of the present study were to compare costs, health outcomes, and costeffectiveness of visual inspection with acetic acid (VIA) screening strategies in rural China, and to identify optimal screening intervals for policy makers. Methods: Markov simulation model was developed to synthesize the evidence of screening and treatment practices in rural China, and applied to predict the long-term costs and effectiveness for hypothetical cohorts over 20 years of screening. Model was validated by calibrating prediction with observation data on age-specific cervical cancer mortality and incidence in China. Costs were considered from a societal perspective while health effects were mainly expressed as quality-adjusted life years (QALY). Both cost and utility were collected on-site and discounted at 5% per year. Results: All completing alternatives showed certain benefits due to the decreased number of women developing cervical cancer. A trend for shorter screening intervals to have greater benefit was found. Under different screening intervals, mortality and incidence were expected to be reduced by 6.67-31.74% and 5.12-23.60%, respectively. Comparing to no screening (status quo), ten-year VIA screening was identified as the most cost-effective option, followed by VIA screening every five-, three- and one year, with corresponding incremental cost-utility ratio (ICUR) ranged from 11,921 to 17,215 CNY (1889 to 2728 US dollars, 2012) per QALY saved. All of the ICURs were much less than China’s GDP per capita (6247 US dollars, 2012). Conclusions: VIA screening at different intervals were all very costeffective options for 35-59 years old women in rural China. It is also noted that the cost-effective manner of aselected strategy largely depends on the local economic status and the performance of such organized program. PCN166 Economic Evaluation of Home Parenteral Nutriton in Cancer Patients; The French Context Hashim M S , Povero M , Aliano A , Pradelli L AdRes HE&OR, Turin, Italy . . . . . Objectives: This study aims to estimate incremental cost and utility of Home Parenteral Nutrition (HPN) in a heterogeneous group of cancer patients from the French public purchaser perspective, as compared with the same patients receiving no HPN. Methods: Two economic models, from public French perspective, were defined: a 28 day cost-utility model and a lifetime state transition model; both models were based on ecological data from an extensive literature search. Four health states were used in the second model: Home, two hospitalization states and death. Transition to death state was based on survival analysis obtained from published summary statistics from two different studies. Functional Assessment Cancer Therapy-general (FACT-g) scores, reported in a recently published French observational study, were used to compute utility weights in the intervention group by applying relevant published algorithms. In the control groups, they were computed by decrementing the baseline utility weight of the intervention group. All costs: nutrition costs, resources consumption costs, complication costs and hospitalization costs were adapted from published French studies. Both Deterministic and Probabilistic Sensitivity Analyses were performed to test uncertainty. Results: The cost-utility ratio of HPN is estimated in 508,059 and 182,584 Euro per QALY gain, in the 28 day cost-utility model and the lifetime state transition model, respectively. DSA showed that survival in the control group and cost of the nutrition were the most influential parameters on the cost-utility ratio in both models. The probability for cost-effectiveness, considering a willingness to pay (WTP) 87,000 Euro (3XGDP per capita in France) for a QALY, was below 1% in both models. Conclusions: Final judgment on HPN cost-effectiveness is difficult, even if it seems to be not costeffective according to standard WTP. The high cost-utility ratio, which declines with increasing survival benefits, should urge clinicians and policy makers to control the sources of ineffectiveness. PCN167 Cost-Effectiveness Analysis of Ugt1a1 Genotyping Before Colorectal Cancer Treatment with Irinotecan Butzke B 1, Oduncu F 2, Heinemann V 2, Pfeufer A 1, Giessen C 2, Stollenwerk B 1, Rogowski W 3 1Helmholtz Center Munich, Neuherberg, Germany, 2Ludwig-Maximilians University, Munich, Germany, 3Helmholtz Zentrum München, Neuherberg, Germany . . . . . . . Objectives: Irinotecan is an anti-cancer agent that is used for the treatment of metastatic colorectal cancer. Although it prolongs survival, it can cause severe toxicity (e.g. diarrhea and neutropenia) in patients who carry the UGT1A1*28 allele. This study evaluates the cost-effectiveness of UGT1A1 genotyping prior to irinotecanbased chemotherapy from the perspective of the German statutory health insurance. Methods: We develop a decision-analytic Markov model to analyze costs and QALYs during a time horizon of six months (two-week cycles). No testing was compared to (1) change of chemotherapy to an irinotecan-free regimen, (2) dose reduction of irinotecan-based chemotherapy and (3) administration of a prophylactic G-CSF growth factor for patients with a UGT1A1*28 variant. Probability, utility and cost parameters used in this study were extracted from published literature. Uncertainty was assessed by deterministic and probabilistic sensitivity analyses. Results: Strategy (2) was the cheapest strategy associated with costs of about € 12,600 and effects of approx. 0.32 QALYs. All other three strategies were absolutely dominated. Compared to no testing, strategy (2) resulted in only marginal increases of QALYs (0.0003) but reduced costs by about € 1,500 per patient. Strategy (1) resulted in smaller health gains (0.0002 QALYs) and smaller cost savings (about € 60). Strategy (3) yielded approximately the same QALY gains as strategy (2) but at higher costs. In the probabilistic analysis, strategy (2) was the optimal strategy in 52% of simulations at a threshold of € 50,000 per QALY. Uncertainty for this strategy originated primarily from the utility weights and the costs of chemotherapy. Conclusions: Our analysis suggests that UGT1A1 genotyping and subsequent reduction of irinotecan-based chemotherapy has a substantial cost-saving potential. Due to the promising results, further research, for example in the form of a managed entry agreement would be desirable to validate these findings. PCN168 Cost-Effectiveness Analysis of Testing for Brca Mutations in Women Diagnosed with Ovarian Cancer and their Female FirstDegree Relatives: A Uk Health Service Perspective Dyer M 1, Vereecken W 1, Worrall J 1, George A 2, Rahman N 2 1AstraZeneca UK Ltd., Luton, UK, 2Institute of Cancer Research, London, UK . . . . . Objectives: Mutations in BRCA1 and BRCA2 are associated with an increased risk of breast and ovarian cancer. If a mutation is detected in women with ovarian cancer their unaffected relatives can potentially undergo gene testing and cancer risk-reducing surgery. Current UK practice is for any such relative to have access to testing. Guidelines also recommend that gene testing should be offered to individuals with BRCA mutation carrier probability of ≥ 10%, although this is not routinely implemented. In particular, many eligible women with ovarian cancer are not offered BRCA testing. Our aim is to evaluate the long-term cost-effectiveness in the UK of providing BRCA testing to women with ovarian cancer and to the unaffected female first-degree relatives of those with BRCA mutations. Methods: A Markov model with a lifetime horizon was developed to reflect the clinical and economic A644 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 outcomes following BRCA testing in women with ovarian cancer and their female first-degree relatives. Two strategies are being compared: no testing versus BRCA testing. Estimates of cancer incidence and mortality, uptake and impact of riskreducing surgery and costs of BRCA testing, cancer treatment and palliative care were based on literature review. Outcomes are expressed as quality-adjusted life years (QALYs). One-way sensitivity analyses are conducted for key model parameters. Results: We first evaluated the cost-effectiveness of gene testing in relatives of ovarian cancer patients with BRCA mutations. Results showed this was associated with an ICER below the UK cost-effectiveness threshold of £20,000 per QALY gained compared with no testing. Sensitivity analyses showed the results were robust. Conclusions: We demonstrate that gene testing in unaffected female first-degree relatives of women with ovarian cancer due to BRCA mutations is cost effective. The final results will consider the cost effectiveness of offering BRCA testing to all eligible ovarian cancer cases and their unaffected female first-degree relatives. PCN169 Burden of Renal Impairment: Relative Health Care Resource use in Prostate Cancer Patients with Bone Metastases between March and April 2014 and retrieved from an extra boost of ONCOVIEW database. ONCOVIEW is a continuous syndicated study on cancer treatment in the hospital setting, based on the collection of patient questionnaires. Patients inclusion criteria were the presence of an mCRC diagnosis, 3rd or later actual therapy line and no participation in a phase II or III clinical study. Information collected included patient demographic characteristics, mCRC characteristics (TNM Classification, Karnofsky performance status scale and mutation analyses) and treatments (actual and previous schedules, dosages and durations). Furthermore, an evaluation of the “Rechallenge” occurrence, in other words the use in 3rd or later line of treatment of drugs previously used, has been performed. Results: 261 patients diaries have been collected: 218 out of 261 patients were in third line of treatment, while 43 patients were in 4th or later treatment line. The most administered schema among third line patients was Capecitabine alone (63 patients), while the most used schema in fourth line was a combination of Fluorouracil and Folinic Acid (7 patients). About 40% of molecules administered in 3rd line and 67% of molecules administered in 4th line were used in previous lines. Conclusions: Results from the present study underline the unmet medical need in 3rd or later line of treatment of mCRC patients and the need for additional evidence-based treatment options. 1Amgen Qian Y 1, Arellano J 1, Gatta F 2, Burke T P 3, Holbrook T 3 Inc., Thousand Oaks, CA, USA, 2Amgen (Europe) GmbH, Zug, Switzerland, 3Adelphi Real World, Bollington, UK PCN172 Burden of Drug Waste in Oncology: Optimization of Resource Use Objectives: Existing evidence suggests that around 49% of patients with bone metastases from solid tumors show evidence of renal impairment (eGFR< 60ml/ min/1.73m2) following diagnosis of bone metastases with approximately 80% of them developing chronic kidney disease. The objective of this analysis is to assess the economic and clinical burden of renal impairment in prostate cancer patients with bone metastases. Methods: Patients with a diagnosis of prostate cancer and bone metastases from the Adelphi Real World Disease Specific Programme USA 2012 were included in the analyses. Propensity Score Matching was used; patients with evidence of renal impairment were matched with those without on a 1:1 basis, controlling for: age, BMI, smoking status, employment status, and relevant comorbidities. Outcomes included number of hospitalizations and length of stay in the past 12 months prior to the point of data collection. A Wilcoxon sign-rank test was used to quantify the impact of renal impairment. Results: 109 patients per group were included in the analyses (total 218). The renal impairment group was estimated to have an increased risk of inpatient visits of 63% (p= 0.036) compared to the control group (0.78 vs 0.48 inpatient visits per patient per year). Additionally, the renal impairment group had a mean of 2.43 (p= 0.027) more inpatient days per year than the control group (5.00 vs 2.56 inpatient days per patient per year). It was also observed that the patients in the renal impairment group were less likely to have received chemotherapy (37% vs 47% received chemotherapy). Conclusions: Findings suggest an increase in health care utilization in the hospital setting in prostate cancer patients with bone metastases and renal impairment. In addition, compromised renal function in those patients may potentially have restricted the use of nephrotoxic chemotherapy agents. Castro A P , Alves A F , Piedade A , Clark L G O, Bueno C C , Minowa E Evidências Credibilidade Científica, São Paulo, Brazil . . . . . . PCN170 Estimating the Voi of Pivotal Studies Towards Predictive Biomarkers of High Dose Alkylating Chemotherapy in Triple Negative Breast Cancer Miquel Cases A 1, Retèl V P 1, van Harten W H 2, Steuten L M G 2 1Netherlands Cancer Institute, Amsterdam, The Netherlands, 2University of Twente, Enschede, The Netherlands . . . . . . . . Objectives: To estimate the expected benefits from a pivotal randomised controlled trial of predictive biomarkers for high dose alkylating chemotherapy (HDAC) in triple negative breast cancer (TNBC) and to inform decisions about the design and priority of further studies. Methods: A markov decision model compared treating 40- years old TNBC women with HDAC based on four predictive biomarker strategies: 1) BRCA1-like by MLPA testing; 2) BRCA1-like by aCGH testing, 3) strategy 1 followed by XIST and 53BP1 testing; and 4) strategy 2 followed by XIST and 53BP1 testing, versus treating all patients with standard chemotherapy. A Dutch societal perspective and a 20-year time horizon were used. Input data came from literature and expert opinions. We assessed four primary outcomes: the expected value of (partial) perfect information (EV(P) PI), the expected value of sample information (EVSI) and the expected net benefit of sampling (ENBS) for the ongoing pivotal TNM trial (NCT01057069) and/or potential future studies. Results: The population EVPI was € 663 million (M). The EVPPI suggests prioritizing further research towards effectiveness parameters, specifically prevalence and positive predictive value of the biomarkers; response rates in biomarker negative patients and TNBC unclassified patients, which are estimated to collectively have a value of information of circa € 630M. The value of further research on transition probabilities is estimated at € 41M, followed by utilities at € 34M and costs at € 34M. Further information on transition probabilities could be gathered from the TNM trial and that of effectiveness parameters and costs from accompanying studies to this trial, altogether estimated to have an ENBS of € 657 M. Conclusions: Further research on predictive biomarkers for HDAC should focus on gathering transition probability data from the current TNM trial, and on accompanying studies to derive data on other effectiveness parameters and costs. PCN171 Real World Data in Oncology: Third- and Fourth-Line Treatments Administered in Metastatic Colon-Rectal Cancer (MCRC) Heiman F , Ripellino C , Visentin E CSD Medical Research S.r.l., Milan, Italy . . . Objectives: The objective of this study was to assess the oncologists’ real clinical practice in the management of mCRC patients, with a focus on the 3rd, 4th and later lines of therapy in Italy. Methods: Data presented in this study were collected from medical records obtained by Italian oncologists on mCRC patients . . . . . . . . . . Objectives: Minimizing waste of the use of drugs allows optimization of available resources in a scarce environment. Grouping patients may be an alternative to reduce drug waste in oncology. The aim of this study is to evaluate the economic impact of streamlined form of dispensation and percentage of drug wastage of the total drug expenditure in supplementary health system. Methods: Patients receiving antineoplasic treatment for stomach, colon, rectosigmoid, rectum, lung and breast cancer were eligible and selected retrospectively from the private market administrative claims database (Evidencias database). Name and any other personal identification were not available at the database. Prescription and date of administration were collected from 3 selected private institutions considering large to small size in terms of patients. Waste of drug was calculated and it was defined as unavoidable or inappropriate clearance of partially drug use. All analyses were performed according to regimen and disease. Saving costs were calculated assuming minimization of waste by optimizing fully drug among group patients. Costs were derived from Simpro table. Exchange rate used was 1.00USD = 2.20BRL. Results: Seventeen drugs were identified among reported chemotherapy regimens in which 11 were analyzed due to potential of saving costs. From these, only 6 drugs could be rationalized. The optimization of drug dispensing would lead to a year savings of US$ 83.587,88, US$ 17.592,22 and US$ 8.225,24 to a large, medium and small clinic, respectively. Calculated drug wastage represented from 2% to 8% of the total drug expenditure, regarding on the antineoplastic used. Five of the 11 drugs did not cause savings due to small number of patients receiving those treatments. Conclusions: Grouping patients for drug wastage minimization is an effective way to reduce costs. Furthermore, savings can be increased by gathering patients of different diseases. PCN173 Resources Utilization for the Investigation of Pulmonary Nodules in a University Hospital Center in Quebec, Canada Gouault-Laliberté A 1, Bergeron C 2, Lachaine J 1 of Montreal, Montreal, QC, Canada, 2CHUM Hotel-Dieu, Montreall, QC, Canada . . . 1University Objectives: Lung cancer is the leading cause of death among cancer patients; therefore, the detection of a pulmonary nodule cannot be ignored. With the increasing prevalence of lung nodule detection, the investigation requires a large number of health care resources. The objective of this study was to measure the health care resources used for the investigation of pulmonary nodules. Methods: A retrospective medical chart review was conducted at the CHUM-Hotel-Dieu in Montreal, Canada. Eligible patients were selected consecutively using the electronic appointment book of the pulmonary clinic, from January 1 st 2011 to May 23rd 2012. Inclusion criteria were: 40 year-old and over, presenting a pulmonary nodule ranging from 0.8 to 3.0 cm with no prior history of cancer in the last 5 years and no history of lung cancer. Patient’s demographics, nodule characteristics, medical information and resources utilization were extracted for each eligible patient. Results: A total of 47 patients (23 women and 24 men, mean age = 64) were included in the analysis. The mean nodule size was 1.8 cm. Thirteen patients (28%) had a benign nodule and 34 (72%) had a malignant nodule. The most frequent non-invasive procedures were Thorax CT-Scan, PETScan and Chest X-ray performed at least once in respectively 96%, 85% and 77% of patients. The minimally invasive procedures (bronchoscopy and transthoracic needle biopsy) and the invasive procedures (thoracoscopy and thoracotomy) were mostly performed in patients who were eventually diagnosed with a lung cancer. On average, patients with a benign nodule underwent 0.77 minimally invasive or invasive procedures vs. 1.94 for patients with a malignant nodule (p= 0.028). Conclusions: A significant amount of health care resources are deployed for the investigation of pulmonary nodules. This study tends to demonstrate that minimally invasive and invasive procedures are mostly deployed for the diagnosis of malignant nodules. PCN174 Impact on Hospitalization Derived from the Use of Denosumab for the Prevention of Skeletal-Related Events in Patients with Bone Metastases Secondary to Breast Cancer in Germany Diel I 1, Ikenberg R 2, Cristino J 3, Gatta F 3, Qian Y 4, Arellano J 4 1Praxisklinik am Rosengarten, Mannheim, Germany, 2Amgen GmbH, Munich, Germany, 3Amgen (Europe) GmbH, Zug, Switzerland, 4Amgen Inc., Thousand Oaks, CA, USA . . . . . . A645 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Skeletal-related events (SREs) defined as pathologic fracture, radiation to bone, surgery to bone and spinal cord compression, are common consequences of bone metastasis. Prior studies have shown that SREs increase the utilization of health care resources, including hospitalizations. We estimated the decrease in hospitalizations when a novel superior therapy is used (denosumab), in substitution of zoledronic acid (zol), to treat patients with bone metastases secondary to breast cancer in Germany. Methods: An analysis was run for predicting the number of SREs avoided and the reduction in number of hospitalizations attributable to treatment with denosumab. The number of breast cancer patients was collected from a German registry using the International Classification of Disease codes. Epidemiological data were then utilized to derive the number of hospitalizations due to bone metastases. The number of patients treated for SREs prevention was obtained from market research and applied to either treatment. The total number of SREs observed was based in the SREs rates seen in a phase 3 clinical trial. German data from a multinational retrospective chart review was used to quantify the inpatient hospitalization rates and length of stay associated with each type of SRE. Results: Approximately 9,500 patients per year with breast cancer and bone metastases were estimated to be treated with denosumab or zol in Germany. The resulting 2,100 SREs prevented per year led to a reduction of approximately 700 hospitalizations per year. The total number of days per inpatient stay avoided by using denosumab instead of zol was approximately 5,000 per year. Conclusions: The superior efficacy of denosumab compared to zol reduces the disease burden by decreasing the number of SREs and consequently the number of hospitalizations and inpatient days. PCN175 Clinical Implementation of Genomic Sequencing in Pediatric Oncology: Identification and Valuation of Resources and Costs Associated with Next-Generation Sequencing Oberg J A , Sireci A N , Mansukhani M M , Nagy P L , Glade Bender J L, Kung A L Columbia University, New York, NY, USA . . . . . . . . . . . Objectives: Beyond understanding the pure cost of genomic sequencing, the real costs associated with implementing next-generation sequencing (NGS) into clinical practice are currently unknown. To elucidate the real costs and to provide a potential benchmark for reimbursement, a pilot study was conducted to identify and valuate the resources related to conducting clinical cancer wholeexome (cWES), transcriptome, and targeted panel sequencing in a cohort of pediatric cancer patients. Methods: A cost model was calculated using 25 pediatric cancer patients who underwent clinical genomic sequencing at Columbia University Medical Center. Our institutional workflow developed by the Precision in Pediatric Sequencing (PIPseq) Program in the Division of Pediatric Oncology and the Personalized Genomics Medicine Laboratory in the Department of Pathology guided the identification of resources and costs associated with NGS. Results: 17 pediatric patients received cWES testing and 11 patients received transcriptome testing using Illumina HiSeq 2500 technology. 7 patients received targeted cancer panel testing using Illumina MiSeq technology. The total cost per case per test [cWES (tumor/normal), $4,459; transcriptome (tumor), $1,764; targeted panel (tumor), $383) was calculated from summing the total variable cost (reagent cost, pathologist time) with the fixed cost per case (annual machine cost, annual maintenance, tech labor cost, informatics cost, space for NGS hardware, server time, NGS analysis lease, and data storage). Clinical utility was demonstrated by identifying a potentially actionable mutation in 24% of participants. Conclusions: Since the reimbursement landscape for clinical genomic sequencing is currently unknown, a comprehensive cost calculation reflecting resource utilization across the whole sequencing workflow including costs associated with directed therapy based on molecular profiling results is necessary. These data serve as a starting point toward identifying and valuating resources associated with NGS and serve as a first step toward demystifying reimbursement for clinical genomic sequencing in Pediatric Oncology. PCN176 Amnog Benefit Assessment for Oncologic and Orphan Drugs in Germany – Implications for Price Discounts Theidel U , Wahlers K , Mittendorf T Herescon GmbH, Hannover, Germany . . . Objectives: With the start of AMNOG in 2011, industry is demanded to submit and defend evidence as well as negotiate discounts for new drugs with an additional patient benefit in Germany. The National Association of Statutory Health Insurances (SpiBu) negotiates with a discount on the ex-factory price on top of a mandatory discount. Evaluating final prices, questions arise regarding factors might influence discounts and impact the German market especially for oncologic and orphan drugs. Methods: A database containing detailed information for all past assessments was used to explore potential impact factors for discounts. Starting with an analysis of background information (discount, areas etc.) descriptive statistics were employed. Furthermore, we analyzed past assessments with respect to: number and size of target population, results of the benefit assessment, change in ex-factory price, magnitude of discount, incorporation of quality of life evidence, and acceptance of comparator therapy among others. Results: Until June 2014 24 price negotiations were completed for oncologic and/or orphan drugs (neoplasm: 16, orphan drugs [w/o oncologics]: 8). It is inconclusive, if the scale of benefit influences the price discount. An additional benefit was granted by the G-BA in 21 of 24 cases. The magnitude of the discount tends to be higher in groups with smaller target populations and quality of life evidence has no influence on the negotiated discount. Discount rates might be smaller if the companies do not deviate from recommendations given by the G-BA. In addition to discounts, some companies needed to lower their ex-factory price. Conclusions: Whereas binding and strict rules are in place for the benefit assessment itself, there is no algorithm for prediction of levels of price discounts. To decrease uncertainty for scenario planning a set algorithm or procedure would be preferable. Further research is needed to evaluate those criteria. Cancer – Patient-Reported Outcomes & Patient Preference Studies PCN177 Targeted Literature Review of Medication Event Monitoring Systems to Evaluate Adherence in Observational Real-World Studies Hanson K A , Payne K A United BioSource Corporation, Dorval, QC, Canada . . . . Objectives: To identify and review methods employed to evaluate medication adherence in studies of oral antineoplastic agents, with particular interest in the opportunities and challenges associated with medication event monitoring systems (MEMS) implemented in observational studies. Methods: A targeted literature review was conducted to identify studies that have measured adherence with antineoplastic agents. Our review included studies that were published between January 1990 and May 2014. Key data abstracted from each study included patient characteristics, study design and duration, cancer type, treatment, and adherence methodology and results. Based on preliminary results, a second targeted review was conducted to evaluate the literature on the risk of the Hawthorne effect in observational studies utilizing MEMS in any therapeutic area. Results: We identified 69 studies that evaluated adherence to oral chemotherapy; 6 studies were interventional and excluded from further review. Of the remaining 63 studies, 28 (44%) were prospective, 25 (40%) were retrospective, and 10 (16%) were cross-sectional. A total of 15 studies used MEMS to evaluate medication adherence. Among observational studies that utilized MEMS and evaluated the Hawthorne effect (n= 3), mixed results were observed. In two studies, patients reported their behavior was affected by their awareness of being evaluated. This was demonstrated by a significant decrease in adherence between months 1-3 in one study but not measured in the second. The third study showed no change in adherence scores over time and concluded there was no Hawthorne effect. Potential ways to minimize the Hawthorne effect include: study duration > 3 months, blinding patients and physicians to results of MEMS downloads, and use of a patient-completed ‘debriefing form’ to assess behavior modifications. Conclusions: MEMS have been utilized in observational studies evaluating oral antineoplastic agents. The Hawthorne effect may be present with MEMS caps, but can be minimized and is not prohibitive to study conduct. PCN178 A Systematic Review of Health State Utility Values for Advanced Ovarian Cancer Al-Dakkak I 1, Borrill J 2, Murphy E 1, Posnett J 1, Zhang Y 1 International, London, UK, 2AstraZeneca, Macclesfield, UK . . . . . 1PAREXEL Objectives: Identifying appropriate utility values to inform cost-effectiveness analysis is a common problem. The aim of this study was to review health-state utility values (HSUVs) for patients with advanced ovarian cancer and make recommendations about their use in the economic evaluation of a targeted maintenance therapy for platinum-sensitive recurrent (PSR) ovarian cancer. Methods: A systematic search of Embase®, MEDLINE®, and MEDLINE®in Process was conducted in June 2013 for studies reporting direct (standard gamble (SG) or time trade off (TTO)) or indirect (EQ-5D, SF-6D, or HUI-3) utility values for patients with advanced ovarian cancer. HTA agency websites were also searched. Study design, country, HSUV elicitation method, health state (HS) description, and who valued the HS were extracted. Mean (SD) utility scores, or medians (ranges), if means were unavailable, were recorded for each HS. Results: A total of 10 publications were found, representing five primary sources of utility values. Two were derived from trial-based patient-reported EQ-5D profiles; one derived profiles from patients with ovarian cancer and utility values from a sample of the general population using a SG; two derived HSUVs from a sample of women without cancer using a TTO. These studies reported utilities for 18 different health states. Where comparisons were possible, utility values differed widely: clinical remission 0.83-0.977; progression-free after recurrence 0.50-0.715; progressive disease 0.40-0.725. None of the studies reported values for patients receiving maintenance therapy. Conclusions: There is limited health-state utility data for advanced ovarian cancer and wide variations in sample size, methods of elicitation, populations used to provide utility values, and in health state descriptions. Further research is required to provide robust estimates to populate an economic model for a targeted maintenance therapy for PSR ovarian cancer. Given the limitations of the current evidence base additional methods, such as mapping algorithms should be considered. PCN179 Health-State Utility Values in Breast And Prostate Cancer Measured using the EQ-5D: A Systematic Review of the Literature Hughes R , Mitchell C R , Bishop R S , Fotheringham I Oxford PharmaGenesis Ltd, Tubney, Oxford, UK . . . . . Objectives: In cost–effectiveness analyses (CEA), a paucity of health-related quality of life (HRQoL) data often necessitates use of utility values from populations which may be ill-matched with the disease modelled. Use of the most pertinent data increases model precision and the accuracy of CEA. This systematic literature review aimed to identify utility values derived from the EQ-5D in patients with breast cancer (BC) or prostate cancer (PC). Methods: A systematic search was conducted using Medline, Embase and Cochrane databases. Eligible studies for inclusion comprised those reporting EQ-5D utility values in patients with BC or PC at any stage, undergoing or not receiving treatment. Results: 31 studies reported relevant data (BC, 17; PC, 14). Utility values for metastatic BC (6 studies) ranged from 0.55–0.75 and were lower for patients receiving palliative chemotherapy (CT) or terminal care (0.51–60), while for stage 0–III disease (9 studies), values ranged from 0.74–0.88. In early-stage BC there was little change during follow-up; for stage II/III BC, rapid and sustained recovery was observed following high-dose (HD) CT and was maintained long-term post-HDCT and following adjuvant CT. Values for metastatic PC (9 studies) ranged from 0.63–0.85 and were lower for patients with bone metastases, worse performance status or undergoing palliative care. For localised PC (2 studies), A646 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 values ranged from 0.81–0.90. Pain, severity of disease, urinary and bowel function and performance status affected utility values in patients with PC. Conclusions: For BC and PC, disease progression, exposure to CT and worsening performance status were associated with decreases in utility values. For corresponding disease stages, utility values tended to be lower for BC than PC, although heterogeneity of data across study populations makes comparisons challenging. No studies reported utility values according to response to treatment. Further research is warranted to improve the evidence available for CEA. PCN180 Health State Utility Valuation in Radio-Iodine Refractory Differentiated Thyroid Cancer (RR-DTC) Kerr C 1, Fordham B 1, de Freitas H M 1, Pelletier C L 2, Lloyd A 1 1ICON Plc, Oxford, UK, 2Eisai Inc, Woodcliff Lake, NJ, USA . . . . . . . Objectives: The study is designed to capture health related quality of life (HRQL) weights for radioiodine refractory differentiated thyroid cancer (RR-DTC) health states. Current treatment options for RR-DTC are limited, with generally poor prognosis. As new treatments emerge for RR-DTC, associated cost-effectiveness evaluations require appropriate preference-weighted HRQL values. Methods: Vignette descriptions for RR-DTC treatment response and adverse event (AE) health states were informed by qualitative work conducted with RR-DTC patients in the US and interviews with 6 clinicians and nurses in the UK and US with RR-DTC treatment experience. Health states included: stable disease, treatment response, progressive disease, stable + grade III diarrhea, stable + grade III fatigue, stable + grade III hand foot syndrome (HFS), stable + grade I-II alopecia. The vignettes were reviewed by the UK and US clinical experts and piloted with UK general public participants in cognitive debrief interviews (n= 5). All vignettes were valued by a UK general public sample (n= 100) using a visual analogue scale (VAS) rating and time trade off (TTO) interview. Data were analysed using descriptive and regression methods. Results: The mean TTO health utilities for RR-DTC states ranged from treatment response (0.86; 95% confidence intervals (CI) 0.83,0. 89); through stable disease (0.80; CI 0.77, 0.84); to progressive disease (0.50; CI 0.45, 0.56). AEs had a significant effect also (stable + grade I-II alopecia (0.75; CI 0.71, 0.79), + grade III fatigue (0.72; CI 0.67, 0.77), + grade III HFS (0.52; CI 0.46, 0.58), and + grade III diarrhea (0.42; CI 0.36-0.48). Conclusions: TTO utilities from this vignette study show clear differentiation between RR-DTC states. The order and magnitude of HRQL impact demonstrated by the utility values reflected clinical opinion and elicited VAS scores. The values reported in this study are suitable for use in cost-effectiveness evaluations for new treatments in RR-DTC. PCN181 French Utility Elicitation in Previously Treated European Patients with Indolent Non-Hodgkin Lymphoma (INHL) Bec M 1, Cognet M 1, Taieb V 1, Pacou M 2, Gauthier A 1 . . . . . 1Amaris, London, UK, 2Amaris, Paris, France Objectives: Since October 2013, the Haute Autorité de Santé (HAS) requires costeffectiveness evidence to assess innovative health technologies. For cost-utility assessments, the HAS strongly recommends using French utility values. To date, the EQ-5D-3L and the HUI3 are the only instruments with a set of preferences values obtained from a representative sample of the French population. In order to inform a cost-utility model assessing pharmaceutical treatments of previouslytreated iNHL, this study aimed to review available utility values in France and to collect such estimates if necessary. Methods: First, a reproducible MEDLINE search was undertaken to identify studies documenting utility values of previously treated iNHL and CLL in France. Then, a web-survey including socio-demographic and clinical questions as well as the EQ-5D-3L was conducted. Given the difficulty to recruit patients from the target population, the questionnaire was conducted in France, the United Kingdom, Germany, Italy and Spain. In line with the HAS guidelines, French tariffs from Chevalier et al. were applied to the collected EQ-5D data. Mean utility values were generated by health state: progression free and progressive disease. Results: Only one French cost-utility model conducted by Deconinck E. et al was identified as relevant, but the study used English utility inputs previously published by Wild et al. Results from the EQ-5D-3L questionnaire conducted, illustrated that quality of life was substantially higher in patients with stable disease versus patients with progression disease. Utilities were calculated with the following scoring function, U (E) = 1–u1-u2-u3-u4-u5–N3. Values between u1 to u5 depending of the 3 levels of the 5 dimensions: u1= 0; 0.15; 0.37, u2= 0; 0.21;. 32; u3= 0; 0.16; 0.19; u4= 0; 0.11;. 026; u5= 0; 0.09; 0.20 and N3= 0.17. Conclusions: EQ-5D is a standardised and validated generic instrument which can be used to elicit utility in iNHL patients. To our knowledge, these French utilities of previously-treated iNHL patients are the first to be published. PCN182 Generating Health State Utility Values from Fact-Ovarian Data Collected in a Phase Ii Maintenance Study in Platinum Sensitive Recurrent Ovarian Cancer (Study 19): A Comparison of Mapping Algorithms Hettle R 1, Borrill J 2, Suri G 1, Wulff J 1 Consulting, London, UK, 2AstraZeneca, Macclesfield, UK . . . . 1PAREXEL Objectives: Where direct or indirect estimates of health state utility values (HSUVs) are not available, mapping algorithms can be used to generate HSUVs from health related quality of life data. In a phase II randomised study of olaparib maintenance therapy in platinum-sensitive recurrent ovarian cancer, the Functional Assessment of Cancer Therapy Ovarian (FACT-O) questionnaire was used. Although no FACT-O mapping algorithms are currently available, several algorithms using FACT-General (G) domains of FACT-O have been published. In this analysis, we applied FACT-G mapping algorithms to the FACT-O data collected in the olaparib study and compared the HSUVs generated. Methods: FACT-O data were collected at scheduled visits, and on treatment discontinuation. Three algorithms mapping FACT-G to EuroQol (EQ-5D) [(Cheung, 2009), Ordinary Least Squares (OLS) and Tobit (Longworth, 2014)] and one from FACT-G to Time-TradeOff (Dobrez, 2007) were applied to data from the phase II study. The agreement between HSUVs was assessed using concordance correlation coefficients (CCCs), and paired t-tests for mean HSUVs. Results: HSUVs were generated for 93% of patients in the study. Mean predicted HSUVs using OLS and Tobit were statistically consistent (p-value= 0.947), whilst Cheung and Dobrez HSUVs were different from other algorithms (p-values < 0.05). The CCCs comparing OLS to Tobit and OLS to Cheung were 0.915 and 0.851, respectively. The CCCs comparing Dobrez to the EQ-5D algorithms were 0.629 (OLS), 0.619 (Tobit) and 0.783 (Cheung). The lowest and highest mean predicted HSUVs were estimated using OLS and Dobrez, respectively. Conclusions: HSUVs can be estimated from FACT-O using FACT-G mapping algorithms. Comparable HSUVs were generated using OLS and Tobit algorithms, whilst Cheung and Dobrez generated distinct HSUVs profiles. Without trial data directly comparing EQ-5D to FACT-O, it is difficult to identify the optimal mapping algorithm. Instead, a range of plausible mean HSUVs can be derived for use in cost-utility analyses. PCN183 Obtaining Indirect Utilities with the Sf-6d and the Porpus-U in Prostate Cancer Patients Avila M M 1, Pardo Y 1, Castells M 2, Ferrer F 3, Boladeras A 3, Pera J 3, Prada P 4, Guix B 5, de Paula B 6, Hernandez H 7, Pont A 1, Alonso J 1, Garin O 1, Ferrer M 1 1IMIM (Hospital del Mar Medical Research Institute), Barcelona, Spain, 2Hospital Universitari de Bellvitge, L’Hospitalet de Llobregat, Spain, 3Institut Català d’Oncologia, L’Hospitalet de Llobregat, Spain, 4Hospital Universitario Central de Asturias, Oviedo, Spain, 5Fundación IMOR, Barcelona, Spain, 6Instituto Oncológico de Guipúzcoa, Gipuzkoa, Spain, 7Hospital Meixoeiro- Complejo CHUVI, Vigo, Spain . . . . . . . . . . . . . . . Objectives: To compare indirect utilities for prostate cancer patients obtained with a generic (SF-6D) and a disease-specific instrument (Patient Oriented Prostate Utility Scale, PORPUS-U). Methods: This was a cross-sectional study of 480 prostate cancer patients enrolled in two similar prospective cohorts. The first one included men diagnosed in 2003-2005 with localized prostate cancer (stage T1 or T2) treated with radical prostatectomy, external radiotherapy, or interstitial radiotherapy at 10 hospitals. The second cohort included patients with stage T2 or T3, treated with external radiotherapy alone or combined with brachytherapy, recruited in 2003–2006 at 6 hospitals. Annual computer-assisted telephone interviews carried out in both cohorts included several questionnaires: the SF-36v2 (from which is derived the SF-6D), the PORPUS-U, and the Expanded Prostate Cancer Index Composite (EPIC), measuring urinary, bowel, sexual, and hormonal domains. ANOVA tests were performed to compare the means of utilities among severity groups of severity defined by EPIC items (severe, small or no relevant problem). Effect sizes between extreme groups were calculated to estimate the magnitude of differences. Results: Mean age was 66.8 years (SD= 6.4) at prostate cancer diagnosis, 20.4% were treated with radical prostatectomy, 33.3% with brachytherapy, 26.7% with external radiotherapy, and 19.6% with combined radiotherapy. The utilities indirectly obtained ranged 0.83-0.99 with PORPUS-U and 0.61-0.84 with SF-6D. Both instruments showed significant differences according to problem severity of all domains measured with EPIC (p< 0.001). Utilities for patients without problems were higher than patients with severe problems. The effect sizes between the extreme groups with PORPUS-U and SF-6D were: 1.23 and 1.24 for urinary; 1.03 and 0.75 for bowel; 0.98 and 0.96 for sexual; and 0.94 and 2.17 for hormonal domains. Conclusions: Our results suggest that both the generic index SF-6D and the disease-specific index PORPUS-U discriminated adequately the problems related to prostate cancer and their treatments. PCN184 Patient Satisfaction Regarding their Treatment and Disease Decisions in Infra-Centimetric Breast Cancer Dalenc F 1, Pau D 2, Chauvet M P 3, Belkacemi Y 4 1Institut Claudius Regaud, TOULOUSE, France, 2Roche, Boulogne Billancourt, France, 3CLCC Oscar Lambret, LILLE, France, 4Hôpital Henri Mondor, CRETEIL, France . . . . . Objectives: The primary objective of this French prospective multicenter non interventional ODISSEE study was to describe daily practice management of infra-centimetric Breast Cancer (BC). One secondary objective was to describe circumstances of diagnosis and disease announcement. Patients had to answer at inclusion a questionnaire regarding satisfaction of the management of their disease and upcoming treatment. Methods: From May 2009 to March 2010, 616 women with infiltrating, unifocal pT1ab, pN0 BC who underwent surgery were recruited by 116 centers. Follow-up period is 10 years. Clinical data, treatments and outcome were collected in routine visits. Regarding patient’s questionnaire at inclusion, seven topics (disease, treatment planned/organization/duration/ side effects, complementary exams needed and disease evolution) were to be discussed with their physician at disease discovery and patient’s satisfaction was evaluated. Results: 546 (89%) patients answered the questionnaire. All 7 topics were discussed with 52% of patients and none of them for 1% of patients. Disease and treatment planned was discussed for 98% of patients, for whom 67% were fully satisfied and 2% were not satisfied. While 60% of patients aged more than 50 years old felt fully involved in their treatment choice, only 44% of patients aged less than 50 years old felt involved. Overall 12% of patients have considered that they have not been involved in their treatment choice. When time between disease discovery/physician visit/surgery is short (< 10 days), patient satisfaction is high (> 85%) and when time is greater than 1 month, patient satisfaction falls below 40%. Physican specialty (surgeon/oncologist/gyneco-obstetrician) or location of disease management (public/private) does not show significant difference in patient’s satisfaction. Conclusions: 57% of patients felt involved in the choice of their treatment, and 65% of patients were globally satisfied of their disease management. The shorter is the management of infra-centimetric breast cancer, the better is the satisfaction. A647 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PCN185 Public Preferences for Genetic Screening for Colorectal Cancer: A Discrete Choice Experiment Veldwijk J 1, Lambooij M S 1, Bredenoord A 2, van Kranen H 3, Dekker E 4, Kallenberg F 4, Smit H A 2, de Wit G A 1 1National Institute for Public Health and the Environment, Bilthoven, The Netherlands, 2University Medical Center Utrecht, Utrecht, The Netherlands, 3National Institute of Public Health and the Environment, Bilthoven, The Netherlands, 4Academic Medical Center, Amsterdam, The Netherlands . . . . . . . . . . . Objectives: To explore individual’s preferences concerning genetic screening for colorectal cancer (CRC) within a population-based CRC screening program and to calculate the initial uptake of genetic screening for different scenarios based on the three major categories of genetic CRC; familial adenomatous polyposis (FAP), hereditary non-polyposis colorectal cancer or Lynch syndrome and other familial non-polyposis colorectal cancer (FCC). Methods: By means of ongoing data collection, a Discrete Choice Experiment (DCE) questionnaire was sent to a representative sample of the Dutch population aged 55-65years. The DCE included nine D-efficient designed choice tasks. Panel-mixed-logit models were used to estimate the relative importance of the four included genetic screening attributes; risk of being genetically predisposed, risk of developing CRC, frequency of follow-up colonoscopies, survival rate. The potential uptake was calculated for the three screening scenarios. Results: When the chance of being genetically predisposed increased, respondents were more likely to participate in genetic screening. Respondents preferred biannual and annual colonoscopies over having a follow-up colonoscopy every 5 years. Increasing predicted CRC survival rates were associated with increased willingness to participate in screening. The risk of developing CRC as a result of being genetically predisposed did not impact the willingness of respondents to participate in screening. The frequency of follow-up colonoscopies was relatively most important for respondents when deciding about participation in genetic screening. The calculated potential uptake rates were high with respectively 91.3% for the FCC, 91.9% for the Lynch and 91.0% for the FAP screening scenario. Conclusions: Individuals are willing to participate in genetic screening for CRC. This decision is mostly driven by the frequency of follow-up colonoscopies, CRC survival rate and the chance of being genetically predisposed. This interest of the general public in genetic screening should be taken into account when discussing about the possibility of introducing genetic screening in population-based cancer screening programs. PCN186 Brio: A European Prospective Observational Study to Assess the Burden of Disease and Treatment in Metastatic Breast Cancer (Mbc) Patients Treated with Oral Vinorelbine (Nvboral) or Intravenous Vinorelbine (Ivvino) Barni S 1, Mouysset J L 2, Sediva M 3, Zamagni C 4, Garau I 5 Italiana di Oncologia Medica, Treviglio, BG, Italy, 2Clinique Provençale, Aix en Provence, France, 3Bulovka University Hospital, Prague, Czech 4 Republic, Oncologia Medica Addarii Policlinico S. Orsola Malpighi, Bologna, Italy, 5Son Llatzer, Palma de Mallorca, Spain . . . . . . 1Ospedale Treviglio-Caravaggio; Associazione Objectives: MBC is an incurable disease for which NVBOral is a standard treatment available in 45 countries. The burden of disease, treatment patterns and motivation to choose across treatments and their consequences are seldom described in the real world. Methods: A prospective observational study was conducted to assess the global burden of disease and treatment with NVBOral or ivVino in Czech Republic, France, Germany, Italy, Poland and Spain. Patients and disease characteristics at inclusion, events, tolerability, hospitalisations observed during one cycle of treatment, and reasons to choose either route of administration are described. Eligible patients received NVBOral either alone or combined with capecitabine, or ivVino either alone or combined with capecitabine or trastuzumab. Results: 184 patients were registered in 6 European countries: 128 with NVBOral and 56 with ivVino. Patients’ characteristics differed significantly only in term of age (64 years for NVBOral and 58 years for ivVino, p= .001) and time between diagnosis and inclusion (104 and 71 months respectively, p= .002). NVBOral was mainly given to patients with HER2-negative (90%), estrogen-positive (80%) and progesterone-positive (62%) disease (ivVino: 66%, 66% and 51% respectively). Patients in the two cohorts had similar performance status and were similarly distributed across chemotherapy lines. NVBOral was mainly chosen for patients’ convenience (93%) and upon patients’ request (21%) (ivVino: 30% and 12% respectively). The two treatments were similarly safe, with grade 3/4 neutropenia in 2% with NVBOral and 6% with ivVino, nausea and vomiting in 2% and none respectively. NVBOral was given without interruption, delay, or dose adjustment in 85% of patients (ivVino: 78%). Hospitalisation was required during the cycle observed for 14% and 26% of patients respectively. Conclusions: In this European prospective observational study, NVBOral appeared a safe, patientconvenient and easy to manage treatment for MBC patients. NVBOral is used particularly after hormonal therapy. PCN187 Health Related Quality of Life and Patient Satisfaction in Prostate Cancer Patients Treated Through Radical Prostatectomy Solano-Moreno H 1, Ramirez-Muñoz O R 1, Balderas-Peña L M A 2, Flores-Larios E A 3, Ramírez-Conchas R E 2, Sat-Muñoz D 4, Salcedo-Rocha A L 5, García de Alba-García J E 5 1UMAE HE Centro Medico Nacional de Occidente Instituto Mexicano del Seguro Social, Guadalajara, Mexico, 2UMAE Hospital de Especialidades Centro Médico Nacional de Occidente IMSS, Guadalajara, Jalisco, Mexico, 3Universidad de Guadalajara, Guadalajara, Jalisco, Mexico, 4UMAE Hospital de Gineco-Obstetricia Centro Médico Nacional de Occidente IMSS, Guadalajara, Jalisco, Mexico, 5Delegación Jalisco IMSS, Guadalajara, Jalisco, Mexico . . . . . . . . . . . . . . . Objectives: To determine HRQoL and patient satisfaction scores in prostate cancer patients treated through radical prostatectomy with a curative intent. Methods: Were selected prostate cancer patients treated through radical prostatectomy. One month after of the surgery we applied EORTC questionnaires QLQ-C30, PR25, and IN-PATSAT32. The scores for each questionnaire were calculated according formulas and instruction in the EORTC Scoring Manual. We calculated mean and standard deviation for each score. Results: Global Health Status/QoL has a mean score 74.80 (±20.74); the functional scales: physical 87.4 (±18.08); emotional 80.3 (±17.23); role 87.6 (±24.44), social 89.4 (±20.09). Symptoms scales: fatigue 15.6 (±15.48); insomnia, 27.3 (±30.00), pain 12 (±17.27), constipation 18.4 (±27.63), financial difficulties 17 (±29.38) the others symptom scales showed mean scores under 10. For scales in PR25 questionnaire, the scores were: sexual activity 72.7 (±20.68), sexual function 59.0 (±31.45); urinary symptoms 32.5 (±21.57), symptoms related with hormonal therapy 17.5 (±14.13), incontinence support 29.2 (±38.24). About scores for IN-PATSAT32, for doctors: Interpersonal skills 88.30 (±16.72); technical skills 87.23 (±19.95); capacity to bring information 89.01 (±17.29); availability 88.56 (±16.85); for nurse: interpersonal skills 87.41 (±16.74) technical skills 86.52 (±18.51); capacity to bring information 84.40 (±19.00); availability 79.26 (±20.89); other personnel kindness 80.85 (±20.47), access 76.86 (±23.16), waiting time 82.18 (±21.45), exchange of information 77.66 (±22.86), comfort/cleanness 80.85 (±21.62) and general satisfaction 84.57 (±18.46). Conclusions: The functional scales for QLQ-C30 showed a good quality of life, however symptom scales related to sexual function showed a diminished QoL, satisfaction with doctors and nurses is good, but not for other personnel. PCN188 Health Related Quality of Life (Hrqol) in Multiple Myeloma Patients Treated in a Tertiary Referral Hospital Miranda-Ruvalcaba C 1, Rubio-Jurado B 1, Balderas-Peña L M A 2, Albores-Arguijo R C 1, Garces-Ruiz O M 1, Borjas-Gutierrez C 1, Aguilar-Lopez L B 1, Vega-Ruiz A 1, Alcantara-Cadillo R R 2 1UMAE HE Centro Medico Nacional de Occidente, INSTITUTO MEXICANO DEL SEGURO SOCIAL, Guadalajara, Mexico, 2UMAE Hospital de Especialidades Centro Médico Nacional de Occidente IMSS, Guadalajara, Jalisco, Mexico . . . . . . . . . . . . . . . Objectives: To determine the HRQoL scores in multiple myeloma patients who are receiving treatment in a tertiary referral hospital. Methods: Outpatients attended consecutively at the hematology department were enrolled, at any stage of treatment. We interviewed through EORTC questionnaires: QLQ30, MY20 and INPATSAT32. The medians and percentiles were calculated. Questionnaire scores were calculated according to formulas and instructions of the EORTC scoring manual. Results: 23 patients were analyzed; the median of age was 62.21 years (interquartile interval 59.02-69.88). 69.56% were men. Global health status/ QoL showed a median score of 66.67 (P25= 62.5, P75= 83.33), emotional functioning 83.33 (P25= 58.33, P75= 91.66), pain scale 33.33 (P25= 16.66, P75= 50.00), insomnia 33.33 (P25= 0, P75= 33.33), fatigue 33.33 (P25= 16.66, P75= 44.44). The median for the body image scale was 100.00 (P25= 50.00, P75= 100.00) and future perspective was 77.78 (P25= 55.55, P75= 97.22), the other medians: symptoms 27.78 (P25= 13.88, P75= 38.88), treatment-related side effects 20.00 (P25= 8.33, P75= 36.66). In the satisfaction questionnaire, the scores for the doctor showed: technical skills 75.00 (P25= 54.16, P75= 97.91), information provided 41.67 (P25= 25.00, P75= 58.33) for nursing: technical skills 75.00 (P25= 58.33, P75= 89.58) and information provided 75.00 (P25= 50.00, P75= 91.66), waiting times 62.50 (P25= 50.00, P75= 96.87) and in overall satisfaction average was 75.00 (P25= 50.00, P75= 93.75). Conclusions: Multiple myeloma patients treated in this hospital show favorable results on scores of quality of life in terms of symptoms of the disease and treatment-related side effects, especially in the fatigue. These results can be compared satisfactory accordingly with other reports from the international literature. PCN189 Health Related Quality of Life and Patient Satisfaction in Colorectal Cancer Patients Treated Through Radical Surgery in Curative Intent in a Colo-Proctology Clinical Department Balderas-Peña L M A 1, Sat-Muñoz D 2, Palomares-Chacon U R 3, Flores-Larios E A 4, Ramírez-Conchas R E 1, Salcedo-Rocha A L 5, García de Alba-García JE5 1UMAE Hospital de Especialidades Centro Médico Nacional de Occidente IMSS, Guadalajara, Jalisco, Mexico, 2UMAE Hospital de Gineco-Obstetricia Centro Médico Nacional de Occidente IMSS, Guadalajara, Jalisco, Mexico, 3UMAE HE Centro Medico Nacional de Occidente Instituto Mexicano del Seguro Social, Guadalajara, Mexico, 4Universidad de Guadalajara, Guadalajara, Jalisco, Mexico, 5Delegación Jalisco IMSS, Guadalajara, Jalisco, Mexico . . . . . . . . . . . . Objectives: To determine HRQoL and patient satisfaction scores in colorectal cancer patients treated through radical surgery with a curative intent. Methods: Were selected colorectal cancer patients treated through radical surgery. One month after of the surgery we applied EORTC questionnaires QLQ-C30, CR29, and IN-PATSAT32. The scores for each questionnaire were calculated according formulas and instruction in the EORTC Scoring Manual. We calculated median and percentilar values 25 and 75 for each score. Results: Were studies 83 colorectal cancer patients: 34 women and 49 men, the results for EORTC questionnaires are: Global Health Status/QoL has a median score 83.33 (70.83-100); the functional scales: physical 93.33 (83.33-100); emotional 83.33 (75-91.67); role 100 (83.33-100), social 100 (83.33-100). Symptoms scales: fatigue 11.11 (0.00-33.33); insomnia, 0.00 (0.00-0.00), pain 0.00 (0.00-16.67), constipation 0.00 (0.00-33.33), financial difficulties 0.00 (0.00-33.33). For scales in CR29 questionnaire, the scores were: body image 100 (83.3-100), anxiety 66.67 (66.67-100); sexual interest in women 0.00 (0.00-33.33), sexual interest male 33.33 (0.00-33.33), urinary frequency 16.67 (0.00-33.33), blood stool 0.00 (0.00-16.67), stool frequency 0.00 (0.00-16.66), urinary incontinence 0.00 (0.00-0.00). About scores for IN-PATSAT32, for doctors: Interpersonal skills 100 (91.66-100); technical skills 100 (95.83-100); capacity to bring information 100 (91.66-100); availability 100 (100100); for nurse: interpersonal skills 100 (91.66-100) technical skills 100 (83.33-100); capacity to bring information 91.67 (70.83-100); availability 87.50 (75.00-100); other personnel kindness 91.67 (83.33-100), access 87.50 (62.5-100), waiting time 87.50 (62.5-100), exchange of information 100 (75-100), comfort/cleanness 75.00 (75.00-100) and general satisfaction 75.00 (75.00-100). Conclusions: The functional scales for QLQ-C30 showed a good quality of life, however symptom scales related to sexual A648 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 function showed a diminished QoL, satisfaction with doctors and nurses is good, but not for other hospital and doctors office conditions. PCN190 Investigating the Framing-Effects of Risk Attributes in Discrete Choice Experiments: A Pilot Study Vass C M , Rigby D , Campbell S , Payne K University of Manchester, Manchester, UK . . . . . Objectives: To understand how the communication of risk in a discrete choice experiment (DCE) affects respondents’ choices. Methods: An online pilot DCE was designed to understand the preferences of female members of the public (recruited via an internet panel provider) for a breast screening programme described by three attributes (probability of detecting a cancer, risk of unnecessary treatment, and outof-pocket cost) each with four levels. Two versions were used that presented the risk attributes (probability of detecting a cancer and risk of unnecessary treatment per 100 women screened) as: (A) a percentage or (B) a percentage and risk image (icon array). The DCE was blocked into four surveys, each containing 10 choice sets. The design, generated using Ngene, included an internal validity test through the inclusion of a dominant choice set. The DCE data were analysed using conditional logit models. Results: 62 women completed the DCE (31 for each version A and B); all were currently eligible for screening under the current NHS programme. All coefficients, but no interactions, were significant and had the expected signs. Of the respondents who received the percentages only version, almost 20% failed the validity test (compared to only 3% of those who received the risk image). Probability of detecting a cancer was the most important attribute. Willingness-to-pay (WTP) for an additional cancer detected was £175 for respondents presented with the risk image, compared to only £152 in the percentages only group. Similarly, WTP to avoid an unnecessary treatment was £59 for respondents presented with the risk image compared to only £19 in the percentages only group. Conclusions: This pilot study highlights the impact attribute framing can have on respondents’ choices in a DCE. The use of risk images also resulted in fewer “irrational” responses implying respondents had a greater understanding of the task. PCN191 Results and implications of using a new eq-5d value set for costutility analyses in sweden. An application using enzalutamide (xtandi®) versUS best supportive care for treatment of metastatic castration resistant prostate cancer (MCRPC) receive information about it (p < 0,001). 95.74% of women to arouse the attention on breast cancer prevention advertisements. Conclusions: Results show that higher graduates are less appear on screening contrary to lower. Rate of self-examination is higher if family history of breast cancer known. Based on results, it is important to reach women who have not yet participated in preventive performances. The reduced activity of people living in cities should be improved, used of advantages of city. Important to organize a training program, especially for the correct application of the method and breast self-examination with usage of questionnaire. PCN193 Knowledge of Human Papillomavirus among University Students in Hungary Vajda R 1, Kálmán D 1, Pakai A 2, Boncz I 3, Ágoston I 1, Molics B 1, Csákvári T 2, Danku N 1, Horváthné Kívés Z 1 1University of Pécs, Pécs, Hungary, 2University of Pécs, Zalaegerszeg, Hungary, 3Faculty of Health Sciences, University of Pécs, Pécs, Hungary . . . . . . . . . Objectives: The main objective of our study was to assess the knowledge of HPV among the female university students in Hungary and also learn their attitudes about the vaccine. Methods: The quantitative cross-sectional questionnaire survey was carried out among female students attending the University of Pécs Faculty of Health Sciences. 180 questionnaires were distributed, of which 165 proved to be evaluable. χ 2-test and t-tests were performed as a statistical method besides 95% probability (p< 0.05). The data analysis was performed with SPSS 20.0 programs. Results: 92% of the women knew the meaning of acronym HPV. 57 % of respondent women knew the cause of infection. 79.5% of the women knew about the virus causing lip and oral cavity cancer. 37.8% of the participants in the sample said that “only women”, and 62.2% of them said that “both men and women” were affected by the infection. It was known by more women who elderly (χ 2=5,034, p=0,024) and living in marriage or in partnership (χ 2=7,415, p=0,006). To sum up the analysis on the issues of HPV, those respondents were considered to be informed who responded well for 5 questions of 6 ones. This rate was 21.4% that is 32 participants of 149 women. 97,3% of respondent women had heard about the vaccination against HPV. 15 women of the respondents had HPV vaccination, among them there were significantly more single, divorced and widow persons. Conclusions: Overall the awareness of human papillomavirus of the students responding is low (21,4%). The against HPV vaccination does not happen because of the deficiencies in knowledge therefore this program is the most important task. Ghatnekar O 1, Nørgaard K 1, Skaltsa K 2 Pharma a/s, Kastrup, Denmark, 2Quintiles Consulting, Barcelona, Spain PCN194 Patients’ Preferences for Bone Metastases Treatments in Turkey Objectives: The guideline on economic evaluations from the Swedish price and reimbursement authority (TLV) states that experience-based valuation of QALY weights are preferred before a hypothetical valuation. The UK hypothetical EQ-5D valuation set [Dolan 1997] has been used so far, but since 2013 an experiencebased “tariff” exists for the general population in Sweden [Burström 2013]. This study explores the implications on ICERs by applying the two different value sets. Methods: FACT-P mapped EQ-5D responses from patients with mCRPC in the AFFIRM trial [Skaltsa 2013; Scher 2012] were converted into utility weights using both the hypothetical and experience-based value sets. A Markov cohort cost-utility model (with stable, progressed and dead health states) analyzing enzalutamide (Xtandi®) vs. best supportive care (BSC) was used for studying the implications of applying the different utility weights. Results: As the experience-based value set had a more compressed event space, and was therefore less sensitive to QoL changes, the “on treatment” utility for Xtandi was 0.03 vs. 0.06 (hypothetical). The stable and progressed health states attained utility weights of: 0.688 and 0.603, respectively (hypothetical); 0.826 and 0.784, respectively (experienced). The survival gain with Xtandi vs. BSC in combination with the lower hypothetical health state utility weights resulted in a QALY gain of 0.66 vs. 0.75 with the experienced weights. With an incremental cost of € 51,100 (€ 1= 9SEK) the resulting ICER was €77,600 and €68,200 using the hypothetical and experience-based value sets, respectively. Conclusions: The Swedish experience-based EQ-5D value set generates a greater absolute utility but with a more compressed event space compared to the UK hypothetical value set. In terms of ICERs, this tends to favour technologies that extend survival compared to QoL improving technologies, although Xtandi provides both. These implications exert great challenge on Swedish decision makers on how the Swedish value set should be implemented. Gatta F 1, Qian Y 2, Ertugrul G 3, Hauber A B 4, Gonzalez J M 4, Posner J 4, Oksuzoglu B 5, Arellano J 2 1Amgen (Europe) GmbH, Zug, Switzerland, 2Amgen Inc., Thousand Oaks, CA, USA, 3Amgen Turkey, Istanbul, Turkey, 4RTI Health Solutions, Research Triangle Park, NC, USA, 5Ankara Oncology Training and Research Hospital, Ankara, Turkey . . . 1Astellas PCN192 Possibilities of Breast Cancer Prevention Karácsony I 1, Giczinger D 2, Ferenczy M 1, Oláh A 3, Boncz I 4, Germán Z 3, Pakai A 5 1University of Pécs, Szombathely, Hungary, 2Health Visiting Services, Nyőgér, Hungary, 3University of Pécs, Pécs, Hungary, 4Faculty of Health Sciences, University of Pécs, Pécs, Hungary, 5University of Pécs, Zalaegerszeg, Hungary . . . . . . . . . . . . . . . . Objectives: To assess patient preference for the currently available bone-targeted agents (BTAs) used to prevent skeletal-related events (SREs; commonly defined as pathologic fracture, radiation to bone, surgery to bone and spinal cord compression) in Turkey. Methods: Adult patients with a self-reported physician diagnosis of bone metastases secondary to a solid tumor were recruited in several hospitals to complete a pencil and paper discrete-choice experiment survey consisting of a series of 10 choices between pairs of hypothetical medication profiles. Each profile was defined using five attributes with several levels (based on prescribing information): time until first SRE (10,18 and 28 months); time until worsening of pain (3, 6 and 10 months); annual risk of osteonecrosis of the jaw (ONJ; 0, 1 and 5%); annual risk of renal impairment (0, 4 and 10%); and mode of administration (daily oral tablet, or subcutaneous injection, 15-minutes or 120-minute intravenous infusion every 4 weeks). Twelve versions of the 10-questions were sequentially administrated across participants. A main-effects random parameters logit model was estimated. Results: A total of 91 patients were included in the analysis and provided demographic information. Among the attributes included in the survey, annual risk of renal impairment, time until worsening of pain and delaying SREs were the three most important attributes, with better levels of outcomes preferred to worse levels. Daily oral administration was the preferred mode of administration and there was no statistically significant difference between injection and infusion of different durations. Annual risk of ONJ was judged by patients to be the least important attribute. Conclusions: When considering treatment choices, patients in Turkey focused mainly on the risk of renal impairment, the delay of pain worsening and delaying SREs. . Objectives: Aims to gain knowledge about the rate of breast screening among Hungarian women. In addition are they familiar with the concept of self-examination, with procedure and if so, whether it is used and how often. Also would like to develop an educational program based on results, which would target risk groups and self-examination and process would receive priority in prevention activities of breast cancer. Methods: A descriptive, cross-sectional, prospective and quantitative study made with women living in and around Sárvár, Hungary. Inclusion criteria was considered the group of women over 45 years. A non-random, convenience sampling was applied with 100 women, between January and June, 2013. Data collected with a questionnaire of 31 questions. Descriptive statistics with frequency range, Chi2-test was performed with Microsoft Excel 2007. Results: There is a significant difference between the appearance of screening and the distance from medical centre (p < 0.05). A higher proportion of women with higher education recognized therapeutic options for breast cancer than those with lower education (p < 0.05). For women where breast cancer was diagnosed in family history, 80% of them regularly attend screening. The women participate in less complex programs for the prevention, because they did not PCN195 What Matters to Patients and their Caregivers: Using Social Media and Patient Forums to Obtain Valuable Information from a Patient and Carer Perspective Chalkiadaki C , Martin A Evidera, London, UK . . Objectives: The aim of this study was to assess whether social media and diseasespecific patient forums can be valuable sources of information on what matters to patients and caregivers about their disease and its management. Methods: We investigated the accessibility of such information in breast cancer and schizophrenia. General social media websites and forums dedicated to sufferers of the two diseases were examined. Results: Breast cancer was more frequently mentioned in general social media websites than schizophrenia, with more charities, large organisations and fundraising events dedicated to breast cancer. Searches for breast cancer and schizophrenia on general social media sites result in hundreds of results, but these are mainly posted by advertisers or pharmaceutical companies, or contain non-patient opinions or news articles. Patients tend not to discuss their condition openly on general social media sites, whereas disease-specific forums contain daily entries on the patient experience. While searches within forums lead to fewer results, these are A649 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 much more accessible and informative, and require less filtering through irrelevant posts. For example, posts from breast cancer patients typically discuss side effects of investigations and treatments and how to manage them, as well as offering emotional support. The limitations of using this type of information include the lack of a mechanism to confirm that contributors really do have the relevant disease, although the number of “factitious” patients is likely to be small and outweighed by “true” patients and caregivers. Conclusions: This study showed that more valuable information can be found in patient-to-patient correspondence in forum threads than in general social media. Breast cancer patients appear to be more vocal in patient forums than schizophrenia patients, possibly due to the stigma attached to mental health problems. Little was found on what matters to breast cancer caregivers, since patients tend to present their own personal experience. PCN196 Patient and Disease Characteristics are Important Determinants of Health-Related Quality of Life of Patients with Metastatic Renal Cell Carcinoma Results from a Population-Based Registry de Groot 1, Redekop W1, Oosterwijk E2, Kiemeney L2, Uyl-de Groot C 1 1Erasmus University Rotterdam, Rotterdam, The Netherlands, 2Radboud University Medical Centre, Nijmegen, The Netherlands . Objectives: Limited data are available on the health-related quality of life (HRQOL) of patients with metastatic renal cell carcinoma (mRCC) in daily practice. The aim of this study was to estimate HRQOL of patients with mRCC in daily practice, and to assess the influence of patient- and disease characteristics, such as comorbidities and adverse events on HRQOL. Methods: Patients with mRCC were selected from a Dutch population-based registry (PERCEPTION). In this RCC registry, data were collected on patient- and disease characteristics, treatments and adverse events. To measure HRQOL, a generic questionnaire (EQ-5D) was used. Multiple linear regression was used to determine which patient- and disease characteristics were associated with HRQOL. Results: In this study, 100/465 (21.5%) patients presented with or progressed to mRCC. Average age at diagnosis was 62.9 years (range: 40-82) and 77% was male. Patients with mRCC reported an average EQ-5D utility score of 0.73 (95%CI: 0.64-0.82) at diagnosis. Two to six months after diagnosis, the average EQ-5D utility score was 0.75 (95%CI: 0.66-0.84) (P= 0.319). Multiple linear regression showed that presence of comorbidities (> 1), number of metastatic sites (> 1), radiotherapy and presence of grade 3 or grade 4 toxicity were significantly associated with a lower EQ-5D utility score. Although presence of bone metastases significantly lowered the EQ-5D utility score, this factor was not significant after adjustment for other factors. Conclusions: This is one of the first studies that provides insights into the HRQOL of patients with mRCC in daily practice. In contrast to previous studies, this study also identified patient- and disease characteristics that influence HRQOL. Presence of comorbidities, number of metastatic sites, radiotherapy and presence of severe toxicity related to targeted therapies were significantly associated with HRQOL. This information is helpful in correcting EQ-5D utility scores for these characteristics, and will be useful in future cost-effectiveness analyses. PCN197 Development of a Conceptual Model for Pediatric Oncology Results from a Review of Qualitative Research Literature and Clinician Interviews Wells T1, Abetz-Webb L2, Evans C1, Theodore-Oklota C3 Outcomes, Boston, MA, USA, 2Patient-Centred Outcomes Assessments LTD, Macclesfield, Cheshire, UK, 3Genentech, South San Francisco, CA, USA 1Endpoint Objectives: Childhood cancer’s profound effects should be assessed appropriately within clinical trials: however, a conceptual model is required to help design appropriate measurement strategies. The aim of this study was to develop an initial pediatric oncology conceptual model. Methods: Key databases were searched for articles focused on qualitative research with children (ages 0-18) with cancer and/or their caregivers. Four patient/parent internet forums were reviewed. Telephone interviews with five pediatric oncology clinicians were performed. Data were analyzed using grounded theory methods. Results: 112 qualitative studies were reviewed. Few studies presented data by specific cancer type/stage or child age; blogs/clinician interviews provided insights into these areas. Across cancer types, but especially for hematological cancers, pain, fatigue, and “feeling unwell”/fever were emphasized, as well as impacts on daily functioning. For brain tumors, headaches, sudden lack of coordination/balance, blurred vision, seizures, vomiting, dizziness and cognition-related impacts on mood and language were highlighted. For solid tumors, tumor-location pain often led to related mobility and/or functioning problems; clinicians emphasized lymphadenopathy and weight loss. Adolescents appeared to be most impacted by cancer, due to their grasp of the cancer’s gravity, self-image issues, loss of autonomy, school absences and social life limitations. Though patients/parents/clinicians found it difficult to distinguish between treatment- and cancer-related symptoms, treatments caused several problems, such as neutropenia, changes in appearance, irritability, nausea/vomiting, fatigue and pain. Not emphasized by clinicians but reported frequently in literature/blogs, mucositis and changes in tastes concerned parents/children. Clinicians strongly emphasized shortterm/long-term/working memory loss and that anxiety contributed to the occurrence/ severity of other side-effects/impacts. Conclusions: Cancer’s impact on children is multifaceted and complicated. Issues identified in this review that must be considered when designing outcomes strategies are: the child’s age, cancer type and stage, delineating disease and treatment symptoms/impacts, and short/long term side effects impacts on nutrition and development. PCN198 Misuse, Abuse, and Diversion of instanyl® (Fentanyl Nasal Spray) in France Blin P 1, Dureau C 1, Lamarque S 1, Bernard M A 1, Robinson P 1, Lassalle R 1, Grolleau A 1, Droz C 1, Moore N 2 1INSERM CIC Bordeaux CIC1401, Univ. Bordeaux, Bordeaux, France, 2INSERM CIC Bordeaux CIC1401, Univ. Bordeaux, INSERM U657, CHU Bordeaux, Bordeaux, France . . . . . . . . . . Objectives: Instanyl® (fentanyl nasal spray) received European market authorisation in July 2009 for the management of breakthrough pain in adults already receiving maintenance opioid therapy for chronic cancer pain, with precise instructions on indications for use and dosage. The study objectives were to evaluate patientreported misuse, abuse, and diversion of Instanyl® in real-life in France. Methods: Cross-sectional observational study of patients with an Instanyl® dispensation from a non-hospital pharmacy. An anonymous self-administered questionnaire was distributed to patients at the time of drug dispensation between 27 July 2011 and 12 November 2012. The questionnaire collected data on indication, contraindications, Instanyl® use, and previous completion of the questionnaire. Results: Among the 272 eligible questionnaires (at least one item completed in addition to age, gender, time since first prescription, and absence of previous completion of the questionnaire), all patients were adult and 95% declared misuse. Among the 160 patients who declared having cancer, 94% declared misuse: 76% declared at least one indication/contraindication misuse and 86% at least one posology misuse. Widening the definition of use for breakthrough pain to use for both breakthrough and chronic pain in cancer patients, reduced the indication/contraindication misuse (63%), but when posology misuse was also considered this did not markedly change overall misuse (93%). Abuse of Instanyl® (using the drug for emotional reasons, relaxation, or sleep disorders) concerned 21 patients (15 with cancer, and 6 without); diversion (passing the drug to another person) concerned 2 patients (1 with cancer and 1 without). Conclusions: Misuse of Instanyl® was widespread. Nearly half reported not to have cancer, and among those who did, only a few used this drug correctly. There seems to be a communication deficit as to the proper prescribing of this drug, and its proper use when prescribed. PCN199 Development of a Patient-Led End of Study Questionnaire to Evaluate the Experience of Clinical Trial Participation Brohan E 1, Bonner N 1, Turnbull A 1, Khan S 2, Dewit O 3, Thomas G 2, Manson S 2 1Adelphi Values Ltd, Bollington, UK, 2GlaxoSmithKline, Uxbridge, Middlesex, UK, 3GlaxoSmithKline, Cambridge, UK . . . . . . . Objectives: Obtaining feedback from participants in clinical trials using a structured method is a valuable opportunity to identify strengths and weaknesses of trial design and conduct, highlight improvement opportunities and provide patient insight to help future recruitment. This study aimed to develop a questionnaire to provide a structured approach to evaluate patients’ experience of clinical trial participation. Methods: A draft questionnaire assessing patients’ experiences before, during and after the trial was developed in collaboration with oncology patient-advocates before being qualitatively reviewed by individuals who have participated in a clinical trial or who have provided support to trial participants. A literature review informed these draft concepts. Three interviewer-led focus groups involving clinical trial participants (on-going or completed within the last 12 months) (n≤ 9/group) were conducted to evaluate the concepts, and elicit further concepts for questionnaire inclusion. An updated questionnaire was developed based on combined insights from the focus groups and literature review. Content validity of the revised measure was considered using cognitive debriefing interviews (n= 12). This testing aimed to identify the relevance and clarity of the instrument. Results: A literature review confirmed the relevance of the concepts assessed by the draft questionnaire. Aspects of the draft questionnaire included the trial enrolment and consent process, logistics of study participation and dissemination of results. The focus group participants provided further evidence on the relevance of the concepts included in the draft instrument and the importance of capturing the clinical trial experience from the patient perspective. Cognitive debriefing results are also presented. Conclusions: The results provide evidence to support the content validity of the post-trial questionnaire. Assessing the clinical trial experience from the patient perspective using a robust questionnaire may offer potential to improve trial design and ensure subjects stay engaged throughout the trial process. PCN200 Patient-Reported Outcomes Assessed Using the Breast-Q Instrument in Women Undergoing Breast Reconstruction Post-Mastectomy: A Systematic Literature Review Lee L J , Milburn C , Macarios D LifeCell Corporation, Bridgewater, NJ, USA . . . . Objectives: Assessing the effects of breast reconstruction (BR) on patient-reported outcomes (PRO) is important as BR becomes increasingly common after mastectomy. A systematic literature review was conducted to assess the clinical use of a wellvalidated PRO instrument, the BREAST-Q (BQ) in women undergoing BR. Methods: Searches using PubMed, ScienceDirect, Cochrane Library, and references in plastic surgery journals were conducted from when BQ was developed in 2008 to May 2014 using the term, ‘breast-q OR breastq’. Non-English, reviews, letters, protocols and single-patient case reports were excluded. BQ modules unrelated to reconstruction (e.g., Augmentation or Reduction) were also excluded. Each BQ scale score ranged from 0-100 with a higher score representing better PRO. Results: After applying exclusion criteria, 35 studies were retrieved. Majority of studies were conducted in North America (54%, 19/35) or Europe (29%, 10/35). Study sample size ranged from N= 14 to 7,110 with mean age ranging 46-55 years. When compared to mastectomy alone, BR was associated with higher PROs (p< 0.05) however, adjuvant radiotherapy was associated with significantly lower PROs than without radiotherapy (p< 0.05). Among PRO scales, sexual well-being was consistently low (range: 30-84) and the scores of physical well-being and satisfaction with surgeons were consistently high (ranges: 68-89 and ≥ 70, respectively). Majority of the studies administered BQ as a single administration post-BR. Only 4 studies reported the change from pre- to post-BR (Range of change score = -19 to 36). From 3 of 4 studies, the scale with the largest Cohen’s d effect size (ES) was psychosocial well-being (ES range: 1.523.83). Conclusions: Our findings suggest that while BQ is widely used, study type and method of administration are disparate. Further assessment of the BQ A650 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 especially in comparative and longitudinal studies is recommended in order to enhance use of BQ in clinical decision making around BR modalities. PCN201 Patients’ Priorities in the Treatment of Neuroendocrine Tumors: An Analytic Hierarchy Process Mühlbacher A C , Juhnke C , Kaczynski A Hochschule Neubrandenburg, Neubrandenburg, Germany . . . . Objectives: Neuroendocrine tumors (NET) are relatively rare, usually slow-growing malignant tumors. So far there is no data on the patient preferences/priorities regarding the therapy of NET. This empirical study aimed at the elicitation of patient priorities in the drug treatment of NET. Methods: Qualitative patient interviews (N= 9) were conducted. To elicit patient’s perspective regarding various treatment aspects of NET a self-administered questionnaire using Analytic Hierarchy Process (AHP) was developed. The data collection was carried out using paper questionnaires supported by an item response system in a group discussion. To evaluate the patient-relevant outcomes, the eigenvector method was applied. Results: N= 24 patients, experts and relatives participated in the AHP survey. In the AHP all respondents had clear priorities for all considered attributes. The attribute “overall survival” was the most significant feature of a drug therapy for all respondents. As in the qualitative interviews, “efficacy attributes” dominated the side effects in the AHP as well. The evaluation of all participants thus showed the attributes “overall survival” (Wglobal: 0.418), “progression free survival” (Wglobal: 0.172) and “response to treatment” (Wglobal: 0.161) to be most relevant. “Occurrence of abdominal pain” (Wglobal: 0.051) was ranked last, with “tiredness/fatigue” and “risk of a hypoglycemia” (Wglobal: 0.034) on a shared seventh place. Conclusions: The results thus provide evidence about how much influence a treatment capacity has on therapeutic decision. Using the AHP major aspects of drug therapy from the perspective of those affected were captured, and positive and negative therapeutic properties could be related against each other. Based on the assessment of the patient’s perspective further investigation must elicit patient preferences for NET drug therapy. In the context of a discrete choice experiment or another choice -based method of preference measurement, the results obtained here can be validated and the therapeutic features weighted according to their preferability. PCN202 What Relapsed/Refractory Cll/Mcl Treatment Outcomes Do German Patients And Physicians Find Most Important? Results from Qualitative Interviews Gaudig M 1, Ireland S 2, Musingarimi P 2, Jackson C 3, Tweats E 3, Eriksson J A 4, Landfeldt E 4 1Janssen, Neuss, Germany, 2Janssen, High Wycombe, UK, 3Adelphi Research UK, Bollington, UK, 4OptumInsight, Stockholm, Sweden . . . . . . . . Objectives: Despite the availability of a wide number of treatments for relapsed/ refractory (r/r) chronic lymphocytic leukemia (CLL) and r/r mantle cell lymphoma (MCL), no standard of care has emerged. There are no studies evaluating preferences for treatment outcomes for r/r CLL and r/r MCL. This study was designed to elicit preferences for r/r MCL and r/r CLL treatment outcomes among patients, the general public and physicians experienced in treating CLL/MCL in Germany. Methods: Interviews (90 minutes) in German of 6 CLL/6 MCL hematologists, 6 r/r CLL and 5 r/r MCL patients were conducted (total 23 interviews). Participants were asked to state their most important treatment outcomes. Transcripts were translated to English and analyzed by counting the number of times each outcome was mentioned. We present here results of patient and physician preferences. Results: r/r CLL patients mention overall survival (OS; 5 counts), mode of treatment administration (4), quality of life (QOL) aspects (4) and progression free survival (PFS) disease control (4). A tolerable side effect (SE) profile/controlling disease symptoms was mentioned by 3 patients. Other treatment outcomes were infections, nausea (2 each), fatigue, weight loss, pain, fever, polyneuropathy and long treatment intervals (1 each). CLL physicians mentioned OS (4), QOL (4) and PFS (4). r/r MCL patients mentioned efficacy benefits such as cure (4) and OS (2), PFS (1); various QOL aspects (5) and a tolerable SE profile/controlling disease (3). Other treatment outcomes were longterm organ damage (2), hair loss, nausea and night sweat (1 each). MCL physicians mentioned OS (6), QOL (5) and a tolerable SE profile (4). Conclusions: Extending life, disease control, maintaining QOL and avoiding SE are important r/r MCL/CLL treatment outcomes to German patients and physicians. PCN203 Patients’ Preferences in Late Stage Treatment of Non-Small-Cell Lung Cancer: A Discrete-Choice Experiment Mühlbacher A C 1, Bethge S 2 1University of Applied Sciences Neubrandenburg, Neubrandenburg, Germany,, 2Hochschule Neubrandenburg, Neubrandenburg, Germany . . . Objectives: Lung cancer is a major cause of cancer-related deaths and thus represents a global health problem. To date, decisions on which treatment to use are often driven by health care professionals’ opinions. The perspective of patients with metastatic non-small cell lung cancer (NSCLC) on the importance of different treatment criteria and the ranking of these decision criteria are rarely taken into consideration. Aim of the study is the evaluation of patients’ preferences for different treatment characteristics of NSCLC patients. Methods: The literature review and 10 qualitative interviews revealed seven patient-relevant treatment attributes. A Discrete-Choice Experiment (DCE) was used to rank the patient-relevant treatment characteristics. The DCE was conducted using a fractional factorial design (Ngene) and the statistical data analysis used random effect logit and GLLAMM latent class models for subgroup identification. Results: In total N= 211 patients with metastatic NSCLC participated in the computer-assisted personal interviews. The estimation revealed a clear dominance for “progression-free survival” (coef.: 1.087) and “tumor-associated symptoms (cough, shortness of breath and pain)” (coef.: 1.090), followed by the side effects: “nausea and vomiting” (coef.: 0.605), “rash” (coef.: 0.432), “diarrhea” (coef.: 0.427) and “tiredness and fatigue” (coef.: 0.423). The “mode of administration” was less important for participants (coef.: 0.141). Conclusions: “Progression-free survival” and “tumor-associated symptoms” were identified as key patient-relevant characteristics in this study. The sole consideration of the “progression-free survival” as foundation for decisions is not sufficient from the patients’ perspective and multiple criteria are important. Subgroup analysis revealed that the importance of “progression-free survival” increases with increased therapy experience. Basically, the results give insight into how much a deciding factor affects the treatment decision from the perspective of patients. In addition, the results of this survey can provide a basis for patient-oriented evaluation of treatment options in NSCLC. PCN204 Carer Perceived Burden as a Predictor of Health-Related Quality of Life: The Case of Colorectal Cancer Hanly P 1, Maguire R 1, Hyland P 1, Sharp L 2 College of Ireland, Dublin, Ireland, 2National Cancer Registry Ireland, Cork, Ireland . . . . 1National Objectives: This study aimed to (i) investigate the impact of subjective and objective factors on colorectal cancer carer physical and mental health, and (ii) identify key subjective burden predictors of these two domains. Methods: 228 colorectal cancer survivors diagnosed October 2007–September 2009 nominated an informal carer. Carers were posted a questionnaire which included questions on socio-demographic characteristics, relationship with the care recipient, the caregiver reaction assessment (CRA) scale and the SF-12v2. Multivariate linear regression was used to assess whether five CRA domains (family support, finances, schedule and health, esteem) predicted carer mental or physical health, controlling for age and other confounders. Results: 153 carers (82% female) completed the questionnaire (response rate = 68%). Carers’ mean physical component score (PCS) was 48.56 (SD= 10.38) and mean mental component score (MCS) was 49.22 (SD= 9.7). The most negatively affected CRA domain was disrupted schedule (mean= 3.0), followed by financial problems (mean= 2.4), health problems (mean= 2.3) and lack of family support (mean= 2.0). Multiple regression analysis showed health burden was the strongest predictor (β = -.54, p <.001) of carer PCS, followed by having a comorbid condition (β = -.34, p <.001), age (β = -.33, p <.001) and schedule burden (β =.25, p =.008). MCS was significantly predicated by financial problems (β = -.20, p =.02), age (β = -.16, p =.05) and esteem (β = -.16, p =.05). Conclusions: Our results demonstrate the need to recognise the different aspects of the impact of caring on caregivers (i.e. physical and mental), and that different domains of subjective caregiver burden impact differently on each of these. PCN205 Impact of Brain Metastases on Quality of Life and Estimated Life Expectancy in Patients with Advanced Non-Small Cell Lung Cancer Roughley A 1, Damonte E 2, Taylor-Stokes G 1, Rider A 1, Munk V C 2 Real World, Macclesfield, UK, 2Roche, Basel, Switzerland . . . . . . 1Adelphi The majority of patients with Non-Small Cell Lung Cancer (NSCLC) are diagnosed with advanced disease (Stage IV). The site of the metastasis as well as the underlying disease influences the outcome and the patient’s quality of life. Objectives: To evaluate the impact of brain metastases compared with other metastatic sites on health-related quality of life (EQ-5D) and physician-perceived life expectancy in stage IV NSCLC patients. Methods: Data were drawn from the Adelphi NSCLC Disease-Specific Programme (DSP®), a cross-sectional survey of 120 pulmonologists and oncologists and their NSCLC patients conducted between July and September 2010 in France and Germany. Each physician completed detailed record forms on 10 advanced patients being actively treated for NSCLC. Patients were invited to complete an equivalent patient self-completion questionnaire (PSC) which included the EQ-5D. Analysis was conducted on patients with only one metastatic site, either brain, contralateral lung, adrenal gland, bone or liver. Mann-Whitney tests were used to assess the differences between metastatic sites. Results: 498 patients with one metastatic site were identified of whom 325 (65%) completed a PSC. The higher the EQ-5D score the better the health state, EQ-5D was significantly lower for patients with brain metastases (mean 0.52, n= 29) compared with contralateral lung metastases (0.69, n= 111, p= 0.0196); adrenal glands (0.83, n= 43, p= 0.0001) and liver (0.71, n= 46, p= 0.0191). No significant difference was observed between brain and bone metastases (0.53, n= 92, p= 0.8219). Estimated life expectancy was significantly shorter for brain metastases (25.3 weeks) compared with contralateral lung (50.5 weeks), bone (49.4 weeks), adrenal glands (48.7 weeks) and liver (44.9 weeks) (all p< 0.01). Conclusions: The development of brain metastases in patients with advanced NSCLC is associated with a significant reduction in quality of life and estimated life expectancy compared with other metastatic sites. PCN206 Skeletal-Related Events (SRES) Impact Significantly the HealthRelated Quality of Life (HRQOL) of Chemo-Naive Men With Metastatic Castration Resistant Prostate Cancer (MCRPC) Ivanescu C 1, Phung D 2, Loriot Y 3, Saad F 4, Mansbach H 5, Beer T M 6, Tombal B 7, Holmstrom S 8 1Quintiles Consulting, Hoofddorp, The Netherlands, 2Astellas Pharma Global Development, Leiden, The Netherlands, 3Institut Gustave Roussy, University of Paris Sud, Villejuif, France, 4CHUM, Montreal, QC, Canada, 5Medivation, Inc, San Francisco, CA, USA, 6Oregon Health & Science University, Portland, OR, USA, 7Cliniques Universitaires Saint-Luc, Brussel, Belgium, 8HEOR, Astellas Pharma Global Development, Leiden, The Netherlands . . . . . . . . . Objectives: Men with mCRPC are at risk of experiencing SREs, defined as pathologic bone fracture (“fracture”), spinal cord compression (“compression”), and the need for radiotherapy or surgery to bone (“radiotherapy/surgery”). We examined the patient and clinical relevance of SREs for HRQoL in men with mCRPC. Methods: We analysed data from patients experiencing any type of SRE (n= 587; irrespective of treatment) in PREVAIL - a phase 3 trial of enzalutamide (n= 872) vs. placebo (n= 845) in asymptomatic/mildly symptomatic chemo-naïve mCRPC patients. For patients with multiple SREs, only the first event was included. HRQoL was assessed using the FACT-P and EQ-5D tools. Impact of first SRE on HRQoL was evaluated as follows: A651 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 1) estimate each patient’s longitudinal HRQoL trajectory before the first SRE using repeated measures analysis; 2) calculate how SREs affect HRQoL changes. Results: We found statistically significant declines in utility scores after all three types of SREs with an adjusted mean change of -0.06 (95% CI [-0.10, -0.02]) for “radiotherapy/ surgery” (N= 107), -0.20 (95% CI [-0.36, -0.04]) for “fractures” (N= 31) and -0.24 (95% CI [-0.39, -0.08]) for “compression” (N= 23). “Fractures” and “compression” affected a number of FACT-P domains and total score to a clinically meaningful and statistically significant extent. Compression had the broadest impact, affecting 7 out of 9 FACT-P domains and induced a mean decrease in the FACT-P total score of -16.96 (95% CI [-26.47, -7.44]). Radiotherapy/surgery was associated with a statistically significant decline in physical (-1.28, 95% CI [-2.06, -0.50]) and functional well being (-1.51, 95% CI [-2.37, -0.64]), and improvement in social well being (1.26, 95% CI [0.60, 1.91]). Full results will be presented. Conclusions: The presence of SREs was significantly associated with poorer HRQoL in this patient population. Any therapy reducing or delaying the occurrence of SREs may slow down the observed HRQoL decline in mCRPC patients. PCN207 Sustainable Measurement of Response Shift in Prostate Cancer Patients: Adjusting Health Related Quality of Life with the Then-Test ten Ham R M 1, Wilson L S 2, Broering J M 2, Cooperberg M R 3, Carroll P 3 University, Utrecht, The Netherlands, 2University of California San Francisco, San Francisco, CA, USA, 3University of California, San Francisco, San Francisco, CA, USA . . . . . . . . . 1Utrecht Objectives: Patients diagnosed with prostate cancer (PCa) have similar survival rates across treatments, making treatments choices based on health related quality of life (HRQoL) more important. Objective was to increase HRQoL usage, sustainability and reliability measurement over time for use in cost–effectiveness analyses by investigating the occurrence of response shift (RS) in PCa patients. Never before has RS been measured over this long a time in PCa patients, not limited by few available treatment options. Methods: A prospective cross sectional cohort study was started in January of 2012 with 1,720 CAPSURE patients using the SF-36 and UCLA-Prostate Cancer Index (PCI). In January of 2012 patient were asked to fill out questionnaires asking about their perceived HRQoL at time of diagnosis (then score) and current HRQoL (post-score). These scores were matched to previous collected scores at baseline (pre-score), ranging from 3 months to 20 years. RS (then-pre score), True Change (TC= post-then score) and Felt Change (FC= post-then) were calculated and compared with t-tests in different questionnaire domains. Linear regression was used to explore relations between scores and patient characteristics. Results: RS and FC are found to be negative overall for SF-36 (RS:-3.6/-14.5 TC:-8.1/1.8 FC:-3.0/-22.7) and PCI (RS:-4.0/-30.2 TC:-20.2/0.07 FC:-3.6/-49.8), showing significant difference (P< 0.05). Significant difference was also seen over the whole time range between post-scores and RS-adjusted-post-scores for SF-36 and PCI. A difference was found between recurrence and non-recurrence in both questionnaires (ranging 0/-4.0), although not significant. RS over time did not show change. Conclusions: Mean negative RS scores were found in this population using SF-36 and PCI, indicating over reporting in retrospective collected data. It is recommended physicians and researchers adjust found HRQoL scores with RS values found in this study, to increase usage, sustainability and reliability of retrospective collected HRQoLscores. Further research is needed to investigate RS dependency on other variables or characteristics. of life (QoL) and the ability to work (WA) too. Till now in the Slovak Republic there was not realised the study oriented on the both mentioned categories. Methods: 102 patients with LC were studied. The average of age was 50.3 y., weight – 78.3 kg, height – 175.1, duration of illness – 2.4 y., symptoms of illness before diagnosis – 1.2 y. Metastases were present in 72 patients. 4 patients were mentioned as strong pessimists, 10 predominant pessimists, 9 neither pessimists nor optimists, 61 predominant optimists, and 18 strong optimists. QoL and the ability to work was evaluated by means of the numeric scale from 0 to 10 (0-the worst, 10- the best) by patients themselves. Results: The QoL was evaluated in these domains: in the time of good health – 8-21, in the time of diagnosis – 2.82, in the current time – 3.71. The WA had these results: in time of good health – 8.89, in the time of diagnosis – 5.08, and in the current time – 2.17. The impact of treatment on the QoL was 2.49 and on the patients families QoL it was 2.75. The willingness to pay for perfect cure was 92.35 € per month (the average salary in Slovakia in 2013 was 824 € ). Conclusions: LC has a great impact on QoL and on the WA too. There was a strong correlation between QoL and WA, although the WA has the later onset as QoL. Our research confirmed the importance of early diagnosis and high effective treatment of this disease. PCN210 Validation of the Proposed Reduced Quality of Life Questionnaire to the Eortc QLQ-C30 in Cubans Patients with Cancer Viada C Center of Molecular Immunology, Havana, Cuba . Objectives: To validate the short version of the QLQ-C30 obtained for patients with non-small-cell lung cancer in patients with head and neck, prostate, breast or cervix cancer. Methods: We analysed data of 636 patients distributed: 237 diagnosed with head and neck cancer, 146 diagnosed with breast cancer, 140 diagnosed with cervix cancer and 113 diagnosed with prostate cancer. The analysis followed a 4-step approach. First, we conducted a Mokken nonparametric item response analysis to ascertain the QLQ-C30 dimensionality and separate several scale if appropriate. Second, we conducted a parametric Samejima’s graded response model (GRM) to assess the item characteristics and information for each scale. Third, we did a confirmatory factor analysis (CFA) to test the scales unidimensionality and to obtain standardised factor loadings to suggest a reduced version of the QLQ. Finally, we assessed the discriminative validity of the reduced version by using receiveroperator curve (ROC) analysis. Results: Mokken analysis of the QLQ-C30 resulted in a unidimensional scale, with an overall scalability defined a medium scale. The unconstrained GRM showed that most items presented appropriate difficulty and discrimination parameters. The CFA supported an underlying unidimensional latent structure for the whole QLQ-C30 (CFI = 0.98; RMSEA = 0.05) with modification indexes pointing to important redundancy of information. The selection of items with standardized factor loadings > 0.70 lead to a 6-item QLQ that showed good discriminative validity against independent criteria of quality of life (ROC area = 0.76; 95% CI = 0.72 to 0.80) as compared with the values for the whole scale (ROC area = 0.70; 95% CI = 0.66 to 0.74). Conclusions: The EORTC reduced scale was validated in this study; it presents good psychometric properties and includes a unidimensional structure of patient-perceived quality of life. PCN211 Shared Decision-Making in Women with Early Stage Breast Cancer and Implications for Long-Term Health-Related Quality of Life Scott A M 1, Jhanwar S M 1, Pusic A 2, McCarthy C M 1 Sloan Kettering Cancer Center, New York, NY, USA, 2Memorial Sloan-Kettering Cancer Center, New York, NY, USA . . . . . . . PCN208 Quality of Life in Patients with Multiple Myeloma in Slovakia 1Memorial Misinova M 1, Gajdosikova E 1, Gerlichova K 1, Matisakova I 2, Bielik J 1 1Trencin University, Trencin, Slovak Republic, 2Trencin University of Alexander Dubcek, Trencin, Slovak Republic Objectives: Surgery for breast cancer has a substantial impact on a woman’s health-related quality of life (HR-QOL). The NIH and EORTC advocate treatment with breast conserving therapy (BCT) for women with early stage breast cancer. The aim of this study was to understand the shared surgical decision-making process from the patients’ perspective by implementing qualitative methods. Methods: All participants were recruited and consented from a single center. Inclusion criteria included women who selected BCT over mastectomy. Utilizing an interview guide, women were asked to share their experience with all aspects of decision-making related to breast cancer treatment. Interviews were audio-taped, transcribed, and coded with NVIVO8. Qualitative data were further analyzed to identify factors influencing decision-making regarding BCT. A comparative matrix analysis was conducted to further evaluate women’s appraisal of their surgical decision-making process and how this impacted their long-term HR-QOL. Results: Nineteen patients were included in the analysis. The mean age was 58.3 (+ 12.2) years, 52.9% were married, 84.2% were Caucasian, 68.4% were currently employed and 31.6% had a family history of breast cancer, 10.4% diagnosed with Stage 0, 52.6% stage I and 36.8% stage II breast cancer. Factors contributing to decision-making were dichotomized into satisfied (n= 11) or dissatisfied (n= 8). Satisfied patients were further categorized as either (i) positive outlook n= 2; (ii) acceptance of choice n= 9; Dissatisfied patients were further categorized as (i) experiencing regret n= 4; (ii) fear of recurrence n= 4. Conclusions: As decision-making needs vary by individual women, a personalized decision-making approach is an essential factor to improve HR-QOL among women with early stage breast cancer. Additional prospective quantitative studies of the preoperative decision-making and post-operative HR-QOL are necessary, as these findings may compliment existing outcomes research. . . . . . Objectives: Multiple myeloma is common in older adults and its incidence increases after the age of 60 (under the age of 40, it occurs rarely, < 2%). In SK approx. 400 new cases have been diagnosed and in the treatment there is in average 1500 patients. Research aim was to find out the impact of the disease on particular items of Quality of patients´ lives. Methods: Sample: research consisted of pacienti diagnosed with the disease of multiple myeloma. We distributed 120 questionnaires and compiled 82 questionnaires (68,33%). There were 36 men (43,90%) and 46 women (56,10%). Their age was between 40 and more. We utilised standardised questionnaire of quality of life, Quality of Life-BREF (WHOQOL-BREF). Its first part included 24 items in four domains (physical health, psychological health, social relationships, and environment) and two items of overall evaluation. Numerical scales were signed in the answers in the following way: the least auspicious answer had the smallest value and the most auspicious answer had the biggest numerical value (range 1 – 5). The results of respondents were compared to population norms of the WHOQOL-BREF domains. Results: In majority of items, the answers of the respondents do not differ very much in comparison with the average score of WHOQOL-BREF. Two items mostly differ from the average score most: domain of physical health – pain, where the average of 2.5 lowered in comparison to population norm of 4.03. The domain psychological health – negative emotions: where the average of 2.00 was lower in comparison to the population norm of 3.47. Conclusions: Prompt diagnostics of multiple myeloma significantly increases the possibility of the treatment to be successful. It also prolongs and improves quality of patient life. Research finding points out the reality that MM does not have the same impact on all domains of quality of life. PCN209 Quality of Life in Patients With Lung Cancer in the Slovak Republic CANCER – Health Care Use & Policy Studies 1Trencin Bielik J 1, Matisakova I 2, Kucerkova P 2, Mastiliakova D 1, Melus V 2 University, Trencin, Slovak Republic, 2Trencin University of Alexander Dubcek, Trencin, Slovak Republic PCN212 Evidence-Based Medicine as a Driver of Improving Colorectal Cancer Screening in Ukraine Objectives: The current incidence and mortality of lung cancer (LC) in adult patients in Slovakia ranges from about 57.3 resp. 50.3 in men and 10.6 resp. 7.6 in women per 100000 inhabitants. The lung cancer has a great impact on quality 1State . . . . . Rubtsova I 1, Melnyk Y 2 Expert Center of the Ministry of Health of Ukraine, Kyiv, Ukraine, 2The State Expert Center of the Ministry of Health of Ukraine, Kyiv, Ukraine . . A652 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: Colorectal cancer (CRR) is an important public health problem. The human and financial costs of this disease have prompted considerable research efforts to evaluate the ability of screening tests to detect the colorectal cancer at an early curable stage. It is now established that screening by faecal occult blood test (FOBT) in average-risk populations can detect asymptomatic colorectal cancers and precancerous lesions (high-risk adenomas). Methods: We have reviewed the evidence about the quality, accessibility, and cost of screening using the FOBT to reduce colorectal cancer mortality. Such databases as Medline, PubMed, EMBASE were used. Results: Faecal occult blood test screening benefits include reduction in CRR mortality, possible reduction in cancer incidence through detection and removal of colorectal adenomas and potentially, treatment of early colorectal cancers may involve less invasive surgery. Thus, implementation of biennial faecal occult blood test screening is an efficient use of health resources. Conclusions: The main objective for the improvement of cancer patients’ care is the introduction of national colorectal cancer screening programs based on evidence and available to people across the country. Introduction of clinical and cost-effectiveness screening methods in Ukraine and coordination of Ukrainian and world practices will provide timely and effective medical care for patients with CRR. In order to integrate these approaches into clinical practice based on evidence the clinical protocols are being developed. PCN213 A Comprehensive Assessment of Early Trial Evidence in Primary Breast Cancer: How Decisions Change Over Time Dequen P , Cooper N J , Abrams K R University of Leicester, Leicester, UK . . . . . Background: There is ongoing debate over the extension of accelerated approval and conditional market authorisation initiatives, particularly for new cancer drugs in the US and Europe. This raises concerns for the benefitrisk assessments of both regulators and payers when relying on early trial evidence. Methodological issues relating to immature clinical data and short-term follow-up, as well as the potential use of surrogate endpoints, may undermine these assessments and the resulting decisions made under conditions of uncertainty. In the context of health technology assessments (HTA) in the UK, we considered the impact on the probabilities of being the most effective and cost-effective treatment when reviewing available evidence along the drug development and assessment processes. Objectives: To conduct a retrospective analysis of four NICE technology appraisals in primary breast cancer for HER-2 positive women alongside a simulation study to evaluate the relative effectiveness and cost-effectiveness of selected drugs over time. Methods: We extracted data from TA107, TA108, TA109 and TA112 to analyse ‘snapshots’ of evidence at different time-points. Based on this example, we simulated individual patient characteristics and progression-free survival (PFS) times according to a predefined recruitment timeline. For both applied and simulated datasets, we combined a network-meta-analysis and three-stage Markov economic model in WinBUGS to predict the joint impact of growing evidence networks on HTA results. The model allowed for different assumptions regarding the extrapolation of and the correlation between PFS and overall survival. Results: The magnitudes of credible intervals depended heavily upon the level of restriction used for follow-up. In addition, the performance of PFS as a surrogate endpoint strongly relied on the assumptions made on the correlation between the two endpoints. Conclusions: Our analysis highlighted the inherent limitations and risks of early assessments with a substantial increase in the uncertainty of treatment acceptability at thresholds of £20K and £30K. PCN214 Expert Elicitation Used for Early Technology Assessment to Inform on Cost-Effectiveness of Next Generation Sequencing Retèl V P 1, Joosten S E 1, van Harten W H 2 1Netherlands Cancer Institute, Amsterdam, The Netherlands, 2University of Twente, Enschede, The Netherlands . . . . . . Objectives: Next Generation Sequencing (NGS) promises to find mutations (targets) in individual cancer patients, to subsequently assign targeted therapy. Currently, the technology itself is still in development and the effects on disease development, prognosis, or choice of therapy are still unclear. Besides information for the current patient, additional (secondary) information for future patients can also become available. To accelerate the reimbursement process and have a NGS-panel available for patients in the earliest possible stage, early Technology Assessment (TA) is ongoing. In this project, we report on expert elicitation by means of scenario drafting, to provide qualitative and quantitative data to fill the evidence gaps in early cost-effectiveness modeling. Methods: The following steps in a multi-parameter framework can be distinguished: 1) Identifying (dynamic) aspects having impact on adoption; 2) Brainstorm on possible scenarios (informal interviews with NGS experts); 3) Scenario construction; 4) Validation of scenarios (semi-structured questionnaires to European NGS-experts); 5) Quantification into parameters for cost-effectiveness modeling. Results: Based on the interviews and questionnaires (n= 29), the most likely scenarios as patients interest in NGS (likelihood 66,5% ±28,1); organizational readiness (84,4 ±18,5); advantage of including RNA, and the demand from the clinic (needs of medical staff) were identified as drivers for NGS development. Possible barrier scenarios were: number of actionable targets (55,2 ±23,7); demonstrating clinical utility (50,4 ±31,4); current evidence generation (65,5 ±27,9); consensus on panel for reimbursement (40,3% ±24,1); and competition within the field (64,2 ±21,9). Clustering into parameters for cost-effectiveness modeling resulted in: “failures”, “compliance”, “uptake”, and “future information”. Conclusions: Although there are many issues to overcome to adopt NGS, the likelihood of NGS incorporation in clinical practice is very high. The “additional future information” is the most interesting but complex variable to identify and to incorporate in cost-effectiveness modeling. PCN215 Next Generation Sequencing Technology: Health Technology Assessment, Market Access Trends and Potential Impacts on The Future of Companion Diagnostic Testing Faulkner E C 1, Spinner D S 2, Ransom J F 3, Paul A 4, Chawla A S 5, Doyle J J 6, Shaw W H 2, Fitzgerald J T 2 1Institute for Pharmacogenomics and Individualized Therapy, Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, NC, USA, 2Quintiles, Durham, NC, USA, 3Quintiles Global Consulting, Hawthorne, NY, USA, 4Quintiles Global Consulting, Durham, NC, USA, 5Quintiles Consulting, Durham, NC, USA, 6Quintiles, Hawthorne, NY, USA . . . . . . . . . . . . . . . Objectives: Next Generation Sequencing (NGS) offers a potentially powerful platform for extremely sensitive, high-throughput, multiplex, quantitative detection of nucleic acid biomarkers. While NGS currently represents a small portion of global clinical molecular diagnostic testing, new funding and reimbursement initiatives promise to accelerate its clinical utilization. Given increasing numbers of predictive/ prognostic biomarkers but limited tissue and need for less invasive sample acuisition, NGS has the potential to transform personalized medicine (PM) and companion diagnostics. The current study characterized global NGS availability and reimbursement trends. Health technology assessments (HTAs) for NGS and other relevant multiplex/ gene panel tests were also studied for evolving evidence requirements. Methods: Key health care provision, HTA agency, and payer websites in the EU, US, Australia and Canada were reviewed to identify NGS funding and reimbursement initiatives, and relevant HTAs. In addition, a limited number of stakeholder interviews were conducted to help further characterize the evolving global NGS landscape. Results: A number of NGS funding and reimbursement initiatives were identified, especially France, Germany, UK, US and Australia. Initiatives have been mainly centered on funding of pilot clinical utility demonstrations through research and clinical use. In Germany and US, specific initiatives are underway to develop specific NGS reimbursement codes and payment rates. A number of HTAs for NGS and other multiplex/ gene panel test platforms were identified, primarily for oncology, cardiovascular, infectious disease, inherited disease, and neuropsychiatry applications. Key HTA concerns include test clinical utility, cost-effectiveness, realworld reproducibility and equity of access given potential cost. Conclusions: Payers and providers increasingly recognize NGS as enabling expanded adoption of PM approaches. As PM expands with increasing numbers of clinically actionable biomarkers, ensuring that test evidence development is aligned with expectations, and expectations with reality are key steps. Further, developing reimbursement/ funding mechanisms to support testing uptake will be critical in all markets. PCN216 How is Research and Development Innovation Evolving? Focus on Oncology and Cardiovascular Disease Coulton K , Savi L HERON Commercialisation, London, UK . . By 2030, 50% of all deaths are expected to be caused by oncology, or cardiovascular diseases. Research and development therefore continues to evolve in these areas to provide a better answer to complex medical needs. As a consequence, the definition of innovation is evolving, going from therapeutic drugs with a new mode of action to novel biotechnologies, biological therapeutics, and vaccination. Objectives: We aim to investigate how therapies for oncology and cardiovascular diseases are expected to evolve in future into technologies that transform the concept of innovation. Methods: Data were extracted from Citeline to assess the scale of therapies in development including: vaccines, stem cells, antibodies, reformulation of combination therapy, gene therapy and RNAi. Results: In oncology, across the selected novel therapies, there are over 670 trials in preclinical to phase 3 of development. Vaccines have a higher rate of development with over 200 products in preclinical development to phase 3. Other areas of development include antibodies, and gene therapy, with over 30 therapies in phase 2 trials alone. Meanwhile, cardiovascular disease has fewer ongoing ‘novel therapy’ trials compared to oncology (around 75% fewer trials). The main areas of development include reformulation of fixed dose combinations, and stem cell research (a total of 60 trials, and 53 trials in preclinical to phase 3 respectively). The trend in innovation for cardiovascular disease instead tends to be focussed on integrated technology and medical devices. Conclusions: While we are fairly acquainted with the current evaluation methods and relevance of clinical and economic evidence of traditional therapies, the emergence of new technologies creates uncertainties around how they will be assessed by payers. For example, demonstrating the value of vaccines (which avoid illness) is difficult to express. Industry must now consider factors outside of the current remit to prepare for successful market access. PCN217 5-Year Survival is not an Appropriate Indicator for Cancer Control in the Population: Revisiting the Issue Based on UK Data Wen Y 1, Li S Q 1, Pan X F 1, Wang Y Y 2, Luo H J 3, Zhao Z M 1, Zhao Y 1, Chang H 1, Xue Q P 1, Yang CX1 1Sichuan University, Chengdu, China, 2epartment of Epidemiology and Statistics, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, Beijing, China, 3China CDC, Beijing, China . . . . . . . . . . . . . . . . . Objectives: It is controversial to use 5-year survival estimates to assess the progress in cancer prevention and control. The study aimed to analyze associations between 5-year survival, and two standard measures of cancer burden, incidence and mortality, based on publicly available population cancer statistics in England and Wales. Methods: Sex-specific mortality and incidence of 14 types of common cancer between 1976 to 1995 were obtained from a national database on the UK Office for National Statistics website. Sex-specific 5-year survival data retrieved from the Cancer Research UK website databases. The relationships between 5-year survival, and incidence and mortality were estimated based on both Pearson and Spearman correlation coefficients. Results: From 1976 to 1995, all male and female cancer types showed increased 5-year survival, ranged from 0.2% (pancreas and lung cancers) to 16.6% (prostate cancer) for males, and from A653 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 0.2% (pancreas cancer) to 16.6% leukemia for females. Inconsistent changes were noticed for either incidence or mortality of different cancers. There were no statistically significant correlations between change in 5-year survival, and change in incidence and mortality for either male or female cancers during this period. When other time periods were considered for analysis, statistically significant correlations were noticed between change in 5-year survival and incidence of male cancers from 1981 to 1995 (Pearson r= 0.68; P= 0.03) and from 1986 to 1995 (Pearson r= 0.78; P= 0.007). However, when the two mostly overdiagnosed cancers (prostate and breast) were excluded, there were no correlations between change in 5-year survival, and change in incidence and mortality for the three time periods. Conclusions: Our study shows no reliable relationships between changes in 5-year survival and incidence or mortality. The increase in 5-year survival might not represent progress in cancer control, but instead indicate improved diagnosis and treatment in clinical practice. PCN218 Assessment of Lung Cancer Treatment by Disease Phase Using National Cancer Registry Data Linked with Treatment Pathway cost Data in Austria Zauner G 1, Popper N 1, Breitenecker F 2 GmbH, Vienna, Austria, 2Vienna University of Technology, Vienna, Austria . . . 1dwh Objectives: Pilot study for evaluation of record linking and potential analysis of combination of epidemiologic cancer registry data and personalized pathway of care identified by billing data using lung cancer as test illness for Austrian datasets in the years 2006 till 2011. Methods: Data sets are linked by anonymized distinct social insurance numbers of approximately 85 percent of all detected incident lung cancer patients in Austria in the years 2006 till 2011 with billing data on single person level. The calculated dataset deals as starting point for the analysis of the remaining lifetime distribution depending on age and TNM – state. Analysis of care classified by eight main categories given in hospitals coded by the MEL-system in combination with treatment using defined drugs is analysed and reported for further interpretation by the experts. An exploratory data analysis on the socio-economic standard and regional differences in Austrian federal states complete the pilot study. Results: Data for the whole Austrian health insured population, including incident cases for lung cancer over a time span of two years were analysed for the following chronic diseases detected by predefined rules on hospital diagnoses and drug prescription in the one year preceding patient history: COPD, diabetes and psychiatric illnesses. For the detected 6616 patients the probability of surviving is calculated and visualized by Kaplan-Meier-Curves. Exemplarily the one year survival is given: 75,44% and the 3-years life expectancy: ~1% Further results combining different influence factors on survival including data from the cancer registry information table are performed. Conclusions: Combining data collected by the national epidemiological cancer registry system and personalized pathways of patient treatment in intramural and extramural care provides a broad basis for analysis concerning real world pathway comparison of cancer patients. This work can unlock potential in respect of defining clinical national cancer registries. PCN219 An Italian Oncology Research to Evaluate Adherence to Clinical Guidelines for Cancer Treatment: The Right Program Sgarbi S 1, Simoni L 1, Ori A 1, Fiori G G 2, Maiello E 3, Barni S 4 1Medidata srl, Modena, Italy, 2MEDIDATA SRL, MODENA, Italy, 3IRCCS Casa Sollievo della Sofferenza; Associazione Italiana di Oncologia Medica, Foggia, Italy, 4Ospedale TreviglioCaravaggio; Associazione Italiana di Oncologia Medica, Treviglio, BG, Italy . . . . . . . Objectives: Clinical oncology societies develop and regularly update evidencebased guidelines in order to achieve more reliable and updated tools for patient management. In 2004 AIOM (Italian Association of Medical Oncology) created the RIGHT program: Research for the Identification of the most effective and hIGHly accepted clinical guidelines for cancer Treatment. It aims to evaluate the concordance between AIOM breast (BC), colorectal (CRC) and lung (LC) cancer guidelines and clinical practice in Italian cancer centers. Methods: the RIGHT program is composed by three retrospective observational studies, one for each guideline, conducted in a sample of 35 (BC) +37 (CRC) +53 (LC) Italian centers for cancer care representative of 230 AIOM centers. Site sampling from AIOM database was stratified by geographic distribution (North, Center, South). Indicators were identified to verify the concordance between AIOM guidelines and clinical practice about staging and treatment. Patients were included if they had their first visit at the site after guideline emission. Patients were then followed-up for at least 6 months. Results: Patients enrolled for the breast, colorectal and lung cancer guideline evaluation were 324, 326, and 708, respectively. Adherence was on average 69% for BC, 76% for CRC and 69% for LC. For CRC adherence was 78% for colon, 69% for rectal cancer and 83% for advanced disease. For BC, the lowest degree of compliance (0%) was observed for the follow-up indicator in asymptomatic patients. For LC, on average 67%, 46% and 81% of stage I-II-IIIA, IIIB and IV patients respectively received recommended care according to defined lung cancer indicators Conclusions: The RIGHT program showed that guidelines adherence is generally high, with very few cases of low adhererence. Guidelines adherence monitoring and update represent crucial activity to get more useful instrument to plan health care interventions. PCN220 Which is More Important for Doctors in a Low-Middle Income Country: a National Guideline or the Medical Literature? A Guideline Adherence Survey of Trastuzumab use for Breast Cancer in iran Ansaripour A 1, Uyl-de Groot C 1, Foroozanfar M 2, Rahimi Moghadam S 3, Redekop W 1 1Erasmus University Rotterdam, Rotterdam, The Netherlands, 2Gilan University of Medical Sciences, Lahijan, Iran, 3Iran Social Security Organization, Tehran, Iran . . . . . Objectives: Most national standard therapeutic guidelines recommend a 52-week trastuzumab regimen for breast cancer treatment. In contrast, the Iranian national guideline (published by the Ministry of Health) recommends a nine-week regimen. We assessed the differences between current routine practice amongst Iranian specialists and the guidelines for trastuzumab treatment and HER2 receptor testing in breast cancer. Methods: 128 Iranian hematology oncologists and radiotherapy oncologists were asked to complete an online anonymous questionnaire. Concurrently, a 3-year retrospective claims database analysis was conducted using data from the Social Security Organization, a health insurer which covers approximately 50% of the Iranian population, to enable comparisons with the questionnaire results. Results: With a 41% (52/128) response rate, doctors reported a relatively high absolute adherence (86%) to the guideline for HER2 receptor testing but a low rate of absolute adherence (6%) to the guideline for duration of trastuzumab treatment. Doctors indicated that the planned duration was 9 weeks in only 32% of patients; in most cases, the plan was 52-week treatment. Patients with a 9-week treatment plan received trastuzumab for 8.6 weeks on average while patients with 52-week plans received treatment for 29.2 weeks. The general trends found in the survey were confirmed in the claims database analysis of 830 HER2 positive patients. Conclusions: When it comes to trastuzumab use, Iranian doctors appear to rely more on the medical literature than on national guidelines developed by the Ministry of Health. Policymakers, doctors and the general public should try to reach some consensus about the optimal way to treat eligible patients. This is particularly necessary in low-middle income countries, whose limited budget cannot easily accommodate all of the innovative technologies to come. Inventive reimbursement policies may form part of the solution as long as the experiences from other countries are examined. PCN221 Time Savings With Transtuzumab Subcutaneous (SC) Injection VersUS Trastuzumab Intravenous (IV) Infusion: A Time and Motion Study in 3 Russian Centers De Cock E 1, Pan Y I 2, Tao S 2, Baidin P 3 1United BioSource Corporation, Barcelona, Spain, 2UBC: An Express Scripts Company, Dorval, QC, Canada, 3F. Hoffmann-La Roche Ltd., Moscow, Russia . . . . . Objectives: Trastuzumab (TRA) subcutaneous (SC) injection is an alternative to intravenous (IV) administration for the treatment of HER2+ early breast cancer (EBC). The objective was to quantify health care professional (HCP) time and patient chair time related to TRA treatment to estimate potential time and cost savings with a transition from IV to SC. Methods: A multi-centre, prospective time-and-motion study was run in three Russian centres participating in the SafeHer clinical trial (MO28048). Case report forms listing pre-specified tasks for IV, SC, and pharmacy management in chronological order and tailored to site practices were used for data collection. Trained observers recorded patient chair time and durations that HCPs were actively completing the tasks. A random effects regression model was run for each task to generate mean and 95% confidence intervals. IV vs. SC process time was calculated as the sum of the mean task times. HCP and chair time were translated to cost using Russian salary data, and 354 roubles (p) per infusion chair hour. Results: Mean reduction in HCP time per patient session was 18.6min (-48%) (IV: 38.7min vs. SC: 20.1min; centre range: 8.6-31.1min), of which 61% of time reduced was achieved in the treatment room. Per treatment session (total 18 sessions), the estimated time saving was 5.6 hours (range across centres: 2.6-9.3hours). Reduction in mean chair time was 59.5min (-89%) (IV: 67.1min vs. SC: 7.6min; centre range: 29.8-97.3min). The monetary value of HCP and patient chair time saved was 1,175p and 6,314p, respectively, for 18 treatment sessions. Conclusions: Transition from IV to SC TRA leads to substantial reductions in administration chair time, active HCP time and associated costs. This allows more time to be used for other patient care activities, increasing the number of patients who could be treated and thus increasing the overall efficiency of treatment centers. PCN222 Process Mapping to Capture Breast Cancer Patients’ Journey in Greek Public Oncological Hospitals Kaitelidou D 1, Kalogeropoulou M 2, Katostaras T 2, Minogiannis P 2, Skitsou A 3, Siskou O 2, Liaropoulos L 1 1National and Kapodistrian University of Athens, Athens, Greece, 2Center for Health Services Management and Evaluation, National and Kapodistrian University of Athens, Athens, Greece, 3Frederick University, Nicosia, Cyprus . . . . . . . Objectives: The present study aimed to examine how breast cancer patients proceed through the public Greek health care system by using process mapping to identify constraints and bottlenecks and unnecessary process steps. Methods: The sample of the study were 86 patients diagnosed with breast cancer at two large public oncological hospitals. A process mapping study, by applying PERT (Program Evaluation Review Technique) analysis, was conducted. Results: The total time spent for obtaining the treatment (per os or chemotherapy) is higher up to 7 times compared with the time spent in activities contributing directly towards the patient’s outcome. The average time needed to complete the procedure (e.g. physician consultation, waiting times, prescribing, obtaining pharmaceuticals) in Hospital A was 78.65 minutes (SD= 37.43) while waiting time was 51 minutes (64.9%). In Hospital B the respective time was 101 min (SD= 50.54) and 83.1% of that time was the waiting time. Similarly, the average time required for chemotherapy at Hospital A was 4 hours and 27 minutes (SD= 106.61) while the waiting time was 2 hours and 15 minutes (50.8%). In Hospital B, the average time for chemotherapy was 3 hours 35 minutes (SD= 69.49) and 52.2% of that time was the waiting time. Among the main reasons for the delays were the lack of electronic records and poor function of the Central Information Prescribing System. Conclusions: Waiting time represented the higher percentage of the total time needed to complete the process for obtaining the treatment. As it represents one of the major causes for patient dissatisfaction, applying process mapping is a critical step for health care organizations to improve the beneficial time and the overall quality of the services offered. A654 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 PCN223 A Comparison of Market Access Evaluations for new Oncology Therapies in France, Germany and the UK: An Analysis Using the Prismaccess Database currently used agents, strategies such as cost-sharing to encourage payers to think beyond the price tag, and promoting familiarity with novel agents among regional and local payers will help optimize the market access opportunity. Droeschel D 1, de Paz B 2, Houzelot D 2, Walzer S 3 1Riedlingen University - SRH FernHochschule Riedlingen, Riedlingen, Germany, 2PRIORITIS Market Access, Paris, France, 3MArS Market Access & Pricing Strategy GmbH, Weil am Rhein, Germany PCN226 Pricing and Reimbursement Analysis of Bacillus Calmette-Guérin (BCG) Immunotherapy for Bladder Cancer Objectives: In recent years (2011-2014) various new oncology therapies were launched and evaluated by the different market access authorities. The international Prismaccess database includes all evaluations and decisions by the respective authorities in France, Germany and the UK. Methods: All decisions for new oncology therapies which were evaluated by the authorities in France, Germany and UK were systematically searched for. A comparison was executed with a focus on reimbursement decision, basis of decision, acceptance of submitted clinical endpoints, study designs, comparator, quality of life and indirect treatment comparison (ITC). Results: In total there were 23 new oncology therapies being evaluated in the three countries. In France 10 decisions were positive (ASMR I-III), further 6 of minor improvement (ASMR IV), 20 were positive in Germany (n= 3 ‘significant’; n= 6 ‘considerable’; n= 9 ‘minor’; n= 2 ‘non-quantifiable’ added benefit) and 4 were positive in England and 5 in Scotland. In 2 cases, respectively the assessment was positive (different magnitude) or negative in all countries. 26% (n= 6) it was similar in at least three countries. (n= 5 positive decisions; n= 1 negative decision). In case overall survival was the primary endpoint the likelihood was higher in all countries for a positive decision. Key differences in terms of decisions were given in acceptance of ITCs, comparator as standard of care and ratings for cost-effectiveness. Conclusions: Using the Prismaccess database the analysis shows that there might be key differences in terms of evaluation criteria between the three countries analysed. In Germany a key focus is given on the appropriate comparator(s) and patient-relevant endpoints. In the UK and Scotland cost-effectiveness might trump a positive benefit assessment. In France the key drivers are not only the severity of the pathology (for tumours, 25% of SMR are not substantial), but also efficacy/adverse events ratio, Effective amount, Comparator choice and Therapeutic strategy. Bekcic S 1, Mitrovic I 2, Baltezarevic D 3, Radojevic V 4, Samardzic J 5, Milenkovic V6 of Virology, Vaccines and Sera, Belgrade, Serbia and Montenegro, 2Institute of Virology, Vaccines and Sera “Torlak”, Belgrade, Serbia and Montenegro, 3Pharmacoeconomics Section of the Pharmaceutical Association of Serbia, Belgrade, Serbia and Montenegro, 4Clinical Centre, Belgrade, Serbia and Montenegro, 5University of Belgrade, Medical Faculty, Belgrade, Serbia and Montenegro, 6University of Belgrade, Faculty of Pharmacy, Belgrade, Serbia and Montenegro . . . . PCN224 Current Status of Reimbursement Decisions for Orphan Drugs or Cancer Drugs and Implementation for Access Schemes in Korea Kim Y , Na Y , Yim E , Kim J , You M Y Health Insurance Review & Assessment Service, Seoul, South Korea . . . . . . Objectives: Korean government has been making efforts to improve the access of orphan drugs or cancer drugs to patients since it is difficult for these drugs to be reimbursed due to its high price or lack of clinical evidence. We aim to investigate the current status of reimbursement and reviewed related schemes in Korea. Methods: Appraisal results for orphan or cancer drugs during 7 years (2007~2013) were included and recommendation rate, final listing rate and order of entry among 8 countries were analyzed. Results: Total 331 were recommended to be reimbursed in overall 467 appraisal results (71%), whereas 74 was recommended to be reimbursed among 121 results for orphan or cancer drugs (61%), indicating that it was less likely to be recommended for those drugs. Fifty eight orphan or cancer drugs (48%) were finally listed through NHIC negotiation process. For cancer drugs, recommendation and listing rate seem to increase from 47% to 64% and 32% to 48% (2008~2010 vs 2011~2013), respectively. Those drugs have been reimbursed in the 4.86th place among 8 countries including Korea on the average. Besides, 15 drugs considered as rule of rescue in those drugs have been listed for reimbursement. Conclusions: We identified that orphan or cancer drugs has been more accessible to patients as time goes. As the benefit enhancement plan for four major diseases (2013) and the Risk Sharing scheme (2013) have been implemented, it is expected for the coverage for those drugs in Korean National Health Insurance to be widened through these schemes. PCN225 Market Access Levers and Barriers for Key Oncology Agents in the EU5: Surveyed Oncologist and Interviewed Payer Insights Cox J , Nawaz K Decision Resources Group, London, UK . . Objectives: Health technology assessment (HTA), pricing and reimbursement (P&R) processes, and cost-containment strategies in France, Germany, Italy, Spain, and the UK, are increasingly stringent. This study explored the resulting impact on high-cost oncology brands, and carved out specific market access levers and barriers. Methods: Across the EU5, 500 medical or hematological oncologists were surveyed regarding their current and expected prescribing patterns, and 30 payers who influence reimbursement at national or regional level were interviewed. Results: Some 68-83% of surveyed oncologists in France, Italy, Spain, and the UK, and 44% in Germany report that the average time taken by their health care authority to review newly approved cancer treatments and settle reimbursement terms delays availability for prescribing by ≥ 6 months. Thereafter, country-specific prescribing restrictions impede uptake; e.g., 18% of German hematologists surveyed report that their indicative prescribing budget prevents use of ponatinib in > 20% of their chronic myeloid leukemia patients, while 30% of Italian medical oncologists say the national oncology drugs register monitoring use of costly agents severely restricts prescribing of erlotinib and gefitinib for non-small-cell lung cancer. Interviewed payers stress, however, that well-designed pivotal trials considering increasing focus on added benefit over direct comparators will help optimize HTA and P&R terms, with those in Italy and the UK, especially, advocating cost-sharing schemes to secure market entry. Furthermore, demonstrable downstream cost savings, locally targeted marketing campaigns, and manufacturer estimates of patient population size to aid regional/local budget planning are specified as uptake levers. Conclusions: HTA and P&R demands and tightening budgets negatively impact prescribing of costly oncology brands in the EU5. However, clear demonstration of robust benefits over . . . . . 1Institute Objectives: The objective of the analysis was to assess the pricing and reimbursement possibilities, as well as the budget impact, of a new medicine for bladder cancer immunotherapy. Methods: We investigated the number of medicines for bladder cancer immunotherapy that were registered and listed in the period between March 2011 and March 2014, in Serbia. We also made pharmacoeconomics analysis that would be a part of the Health Insurance Fund submission file. Results: According to the European Association of Urology (EAU) Guidelines for non-muscle invasive bladder cancer (NMIBC) there are several bladder preservation strategies available: intravesical immunotherapy, intravesical chemotherapy, device-assisted therapy and combination therapy. In Serbia, there are three medicines listed for chemotherapy (doxorubicin, epirubicin and mitomycin) and one for intravesical immunotherapy (BCG). BCG has become the standard of care for high-grade NMIBC and carcinoma in-situ (CIS) and is superior to intravesical chemotherapy in reducing recurrences, in preventing or delaying progression of the disease. Radical cystectomy should be considered after BCG treatment failure, when BCG is contraindicated or not available. BCG is reimbursed in Serbia (wholesale price: 79,18€ ) but due to continuous shortages of registered BCG from May 2012, non-registered BCG was also listed (price: not defined). Still there was no import, due to worldwide shortages. The absence of the treatment encouraged the Serbian Institute of Virology, Vaccines and Sera “Torlak” to develop BCG for immunotherapy. As the price proposal for new medicine would be 55,43€ , expenses per patient per year would be 498,87€ and total costs for 520 patients would be 259.412,40€ , it is projected that total savings would be 111.150,00€ per year. Market share of domestic BCG would be 0,29% of total B list (hospital medicines) budget. Conclusions: The future aim is to develop, register and list domestic BCG that would provide lower costs per patient, high quality, availability and the continuous immunotherapy. PCN227 The Expanding Value Footprint of Oncology Treatments Rejon-Parrilla J C 1, Hernández-Villafuerte K 1, Shah K 1, Mestre-Ferrandiz J 1, Garrison L 2, Towse A 1 1Office of Health Economics, London, UK, 2University of Washington, Seattle, WA, USA . . . . . . . Objectives: To provide a better understanding of: how changes in the use of an oncology medicine can affect its aggregate value; how different HTA systems have assessed these value expansions; and whether there is a link between value expansions and use. Methods: We examine all oncology medicines approved by the EMA between 2003 and 2005 – giving a sample of 10 medicines. Our framework sets out seven possible value expansions beyond an initial approved indication: different cancer type; different disease stage; different treatment line/stage; different treatment regimen; orphan designation; patient sub-population; and new route of administration. We then assessed how HAS (France), NICE (England and Wales), and Aetna (US) have recognised these value expansions. Finally, we analysed IMS data (2004-2013) on prices, volumes and sales for the five of the medicines. Results: Seven of the 10 medicines in the sample have additional value expansions following initial indication. Many are now used for indications that are very different from their original indication. Most of the HAS assessments resulted in the drug being reimbursed but the rewards to the manufacturers were in many cases relatively low because few of the recommended drugs were given low “improvement in medical service” (ASMR) levels. The majority of NICE appraisals (63%) resulted in the drug/ indication not being recommended for use in the NHS. Generally, the UK had lower prices, volumes and sales than France and the US (with some exceptions). The comparisons between France and the US were a little more equivocal. There is a mixed picture in terms of the correlation between NICE/HAS recommendations and sales in the UK/France. We observe a link between expansions in licensed indications and changes in sales. Conclusions: Health systems and policy makers need to recognise how product life-cycle considerations affect the value of medicines, and in particular, oncology medicines. PCN228 Impact of Introducing Costs/Qaly Threshold on Access to Oncology Medicines in Slovakia Psenkova M 1, Mackovicova S 1, Tomek D 2 1Pharm-In Ltd, Bratislava, Slovak Republic, 2Slovak Medical University, Bratislava, Slovak Republic . . . Objectives: In December 2011, a threshold value of costs/QALY was introduced to Slovak legislation. The aim of this work is to assess its impact on inclusion of oncology drugs to the reimbursement system and their availability in clinical practice. Methods: We evaluated the inclusion of oncology drugs to the reimbursement system based on data from the website of the Ministry of Health SR. We analysed the consumption of drugs based on the National Health Information Centre database. We acquired information about registered oncology drugs on the EMA website. We determined the level of availability of oncology drugs in clinical practice via a qualitative survey among oncologists. Results: The success rate of including new oncology drugs in Slovakia was high in 2000-2011. From the 62 oncology drugs registered in this period by the EMA, 48 were included to the Reimbursement List. Innovative oncology drugs were included relatively quickly A655 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 to reimbursements, causing a significant growth in costs. Expenses for oncology in 2003-2011 increased by 718% (from € 15.4 million to € 125.7 million). In December 2011, a cost/QALY threshold was introduced to legislation, creating a barrier to the inclusion of oncology drugs to the Reimbursement list. Following adoption of this legislation, of the 12 drugs registered by the EMA, only 3 oncology drugs were included to the List during 2012-2014. Uncategorized drugs are available only for a limited number of patients by way of individual exceptions, or by participation in clinical trials. Conclusions: In Slovakia, the willingness to pay for an additional unit of health was anchored in legislation, which greatly limited the availability of innovative oncological treatment. The health system in Slovakia needs to introduce efficient and transparent mechanisms that enable the treatment of oncology patients in line with the latest medical findings, while keeping expenses for treatment within economic possibilities. PCN229 Innovation May Drive Streamlined Access to New Biopharmaceuticals Across Some Emea Markets Gardiner R B , White R Access Partnership, London, UK . . . Objectives: Across geographies, approval, pricing and reimbursement of pharmaceuticals take place under varying timelines with different outcomes. There are some countries that may obtain access to new pharmaceuticals through early access schemes. Breakthrough and innovative products that are thought to have a profound impact on current standard of care are often eligible for quicker routes to access. This research sought to investigate how these schemes worked, where they were prevalent, and the outcomes of such schemes. Methods: The research was conducted through in-depth interviews with payers and clinicians across 10 EMEA markets. Results: Of the 10 markets studied, 5 countries were identified to have either easier or quicker routes to access for new biopharmaceuticals (e.g., ATU in France and Algeria, law 648/96 in Italy, non-formulary access in Saudi Arabia, and the “white list” in Norway). Most often, these routes were reserved for products with orphan indications or products that were believed to significantly impact current standard of care. If pharmaceutical companies opt for the streamlined route, there are often significant restrictions imposed on the product, as well as a reduction in volume. If companies elect for the standard route to approval and reimbursement, the review process is often more rigorous, however, the decision is likely more permanent and the volume of product is larger. Frequently, if products opt for the faster route to access, this will serve as additional evidence for getting the product reimbursed at a later date for use in a wider population. Conclusions: New pharmaceutical products that are likely to dramatically change the treatment landscape or are active in orphan diseases should take advantage of these schemes. Physicians grasp at the opportunity to use efficacious products as early as possible and companies need to leverage the opportunity for streamlined access to products. PCN230 Health Economic Impact of Volume Doubling Time as Biomarker in Lung Cancer Diagnosis Brinkhof S 1, Groen H J M 2, Siesling S 3, IJzerman M J 4 1University of Twente, MIRA Institute for Biomedical Technology & Technical Medicine, Enschede, The Netherlands, 2University Medical Center Groningen, Groningen, The Netherlands, 3Comprehensive Cancer Centre the Netherlands (IKNL), Utrecht, The Netherlands, 4MIRA Institute for Biomedical Technology & Technical Medicine and University of Twente, Enschede, The Netherlands . . . . . . . Objectives: Lung cancer has a continuously bad prognosis in terms of survival and quality of life, usually because of late detection of malignancies. Given an expected increase in the incidence, overall mortality will increase. Early detection and efficient diagnostic planning may offer additional gain in survival. The diagnostic pathway for patients with suspected lung cancer is characterized by a cascade of different imaging and diagnostic modalities. The main objective of this study is to estimate the health economic impact of diagnostic procedures and the expected gain by volume-doubling time on low-dose CT as biomarker in suspected lung cancer (NELSON protocol). Methods: A state-transition model is created to simulate the pathway of lung cancer diagnostic procedures, including x-ray, diagnostic CT, PET-CT, bronchoscopy, mediastinoscopy and more. Hospital registries and data from the National Cancer Registry were used to estimate the amount of diagnostic procedures in a cohort of lung cancer patients. Systematic literature search was performed to estimate the diagnostic performance of different modalities. Patient cohort is defined and the pre-test probability for malignancy is estimated through the Swensen criteria. Probabilistic sensitivity analysis is performed using Monte Carlo Simulations. Results: Diagnostic procedures for patients with suspected lung cancer can count up to almost € 3000 per patient. Pathway was modeled in a microsimulated cohort through Swensen criteria, leading to a mean chance of malignancy of 40%. Costly steps in the pathway include cervical mediastinoscopy and mutation analysis. Inclusion of NELSON protocol can lead to a reduction in costs. Decision making per patient can reduce overuse of diagnostic modalities. Conclusions: The diagnostic procedure for suspected lung cancer patients is a costly pathway and can be improved with use of the NELSON screening protocol or personalized selection of diagnostic procedures. PCN231 How Successful Have Pediatric Investigation Plans Been In Stimulating Research For Pediatric Cancers? Miller K L University of North Carolina at Chapel Hill, Chapel Hill, NC, USA . . Objectives: The European Pediatric Medicine Regulation was created in 2007 to encourage further drug development for pediatric diseases, by requiring pharmaceutical companies to submit pediatric investigation plans (PIPs) when submitting the marketing application for a new drug. The objective of this study was to determine how successful this legislation had been in stimulating research in pediat- ric cancers. Methods: Current oncology PIPs were manually extracted from the EMA website for the study period 2007-2014. Primary variables included: indication, applicant, decision, decision date, and date of expected completion. Indications for approved PIPs were classified into five categories: brain tumors, diagnostics, leukemias/lymphomas, side effects, and solid tumors. Results: A total of 105 PIPs were found; 36 of which were waivers, and 69 of which were approved. The study found that 39% (27) of approved PIPs were indicated for solid tumors, including melanomas and malignant tumors; 30% (21) of approved PIPs were indicated for leukemias or lymphomas; 17% (12) of approved PIPs were indicated for chemotherapy-induced side effects, such as anti-nausea and neuotropenia medications; 12% (8) of approved PIPs were indicated for brain tumors; and one oncology diagnostic PIP was also approved. The ramp-up of the PIP program was significant. PIPs approved in 2013 and the first half of 2014 accounted for 54% (37) of all PIPs approved during the study period. Conclusions: Approved PIPs covered a wide range of pediatric cancers, and the number of approved PIPs increased significantly over time. While the rampup of the PIP program indicated that it was successful in promoting research in this area, serious concerns remained regarding the feasibility of the program. For example, there were currently four trials planned for completion between 20152020 for extremely rare high grade gliomas. This limitation may compromise the integrity of the PIP program. PCN232 Treatment Patterns And Outcomes Of Patients Diagnosed With Ovarian Cancer In The Netherlands: A Registry Study Houben E 1, van Haalen H G M 2, Sparreboom W 2, Overbeek J A 1, Ezendam N 3, Pijnenborg H 4, van Herk-Sukel M P P 1 1PHARMO Institute for Drug Outcomes Research, Utrecht, The Netherlands, 2AstraZeneca, Zoetermeer, The Netherlands, 3Comprehensive Cancer Centre the Netherlands, Eindhoven, The Netherlands, 4Tweesteden Ziekenhuis, Tilburg, The Netherlands . . . . . . . . . . . . Objectives: Little information is available on the patterns of chemotherapy regimens administered for the treatment of ovarian cancer (OC) in the Netherlands. The objective for this study was to describe current chemotherapy patterns for OC in the Netherlands and to evaluate survival outcomes following subsequent lines of chemotherapy. Methods: Data from the Eindhoven Cancer Registry, including data on all newly diagnosed cancer patients, was linked to the PHARMO Database Network including, among other things, information on in- and outpatient drug use. Patients diagnosed with primary OC between January 2000 and December 2010 were selected. First and subsequent chemotherapy regimens were defined as the start of a different (combination of) chemotherapeutic agent (s) or a gap > 42 days (or > 91 or > 183 days in sensitivity analysis) between two treatment cycles. Results: Of the included OC patients, 70% received chemotherapy as initial treatment. Detailed chemotherapy data were available for 261 patients, who had a mean age of 63 (SD 12) years. Pathological tumor stage was known for 77% of patients, of whom 11%, 55% and 17% of patients had disease stage II, III, and IV, respectively. In first line chemotherapy, 76% of patients received platinum/taxane doublet chemotherapy. Of the 161 patients receiving second line chemotherapy, platinum-containing chemotherapy was received by 63% of patients (101 of whom 13 received platinum monotherapy). In third line chemotherapy this was 51% (53 patients). At least eight lines of chemotherapy were identified in 12 cases. Median survival as from diagnosis was 46 months. After second line chemotherapy, median survival remained 14 months. Conclusions: This study provides detailed information on the type of chemotherapy regimens administered to OC patients at initial diagnosis and during follow-up and the survival following the various chemotherapy regimens. PCN233 The FDA Black Box Warning Does Reduce the Use of Erythropoietin Stimulating Agents and Increases Blood Transfusions in Insured, Low Income Cancer Patients Noxon V 1, Bennett C 2, Wu J 3 1South Carolina College of Pharmacy – USC Campus, Columbia, SC, USA, 2University of South Carolina College of Pharmacy, Columbia, SC, USA, 3University of South Carolina, Greenville, SC, USA . . . Objectives: Erythropoietin stimulating agents (ESAs) are useful drugs for treating chemotherapy related anemia to reduce the number of blood transfusions. However, there were unrecognized toxicities of ESAs. These toxicities were finally recognized in 2007 when the FDA issued a black box warning for ESAs. The objective of this study is to determine the effect of the FDA black box warning on ESA use patterns and associated outcomes in insured, low-income cancer patients in South Carolina. Methods: The merged South Carolina Central Cancer Registry-Medicaid dataset was used to determine the trend of ESA use from 2001-2010. Female Breast, Colorectal and Non-Small Cell Lung cancer patients were identified from the registry. Of those, their chemotherapy status was identified along with ESA use from Medicaid medical claims. The major outcome measures were claims for use of ESAs after chemotherapy and the blood transfusion rate. Logistic regression was used as a quantitative method to determine if the likelihood of receiving ESA treatment was reduced after FDA black box warning. Results: Among 1,645 patients treated with chemotherapy from 2002-2010, the proportion of chemotherapy patients receiving ESAs decreased from 56.47 % before the black box warning to 23.16% after black box warning (p < 0.001). The blood transfusion rate per year during 2002-2007 remained around 10-15% and increased to 31% in 2009. The likelihood of ESA use was reduced by 63% after black box warning issued by FDA after adjusting for demographic and clinical variables. Conclusions: The black box warning may have been effective in reducing overall ESA utilization in cancer patients taking chemotherapy. PCN234 Treatment Patterns and Costs Of Neoadjuvant Systemic Therapies (NAT) For Early Breast Cancer (EBC): A Retrospective Claims Analysis Chen Y J 1, Santos E 2, Schabert V 1, Antao V P 2, De A P 1, Portera C C 2, Wang Y 1, Kamath T 2 Health, Alexandria, VA, USA, 2Genentech, Inc., South San Francisco, CA, USA . 1IMS . . . . . . . . . . . A656 VA L U E I N H E A LT H 1 7 ( 2 0 1 4 ) A 3 2 3 – A 6 8 6 Objectives: NAT for eBC have potential benefits in reducing tumor size, permitting breast conservation, identifying effective adjuvant regimens, and providing prognostic information. This study investigated the characteristics of eBC patients, real-world utilization patterns of NAT, and health care costs from diagnosis to primary surgery (neoadjuvant phase) using a US claims database. Methods: A cohort from the IMS PharMetrics Plus database included female patients, aged 18+, with the first (index) breast cancer (BC) diagnosis (ICD-9-CM 174. x, 233.0) between July 2006 and September 2012, primary surgery (mastectomy or lumpectomy) after index, continuous enrollment from 180 days before index (pre-index) to 90 days after surgery, no pre-index diagnosis for BC or other primary cancer, and no secondary malignancy from pre-index to surgery. Systemic therapies used by this cohort in neoadjuvant phase were assumed as NAT. Patients with eBC trastuzumab use were presumed HER2+. Results: Of 57,032 eligible eBC patients (median age= 56), 2,011 (3.5%) received NAT. Patients who received NAT had primary surgery in a median of 166 days after index diagnosis vs. 21 days for patients who did not receive NAT. Among patients receiving NAT, 485 (24.1%) had trastuzumab, with TCH (docetaxel, carboplatin, trastuzumab) and ACTH (doxorubicin, cyclophosphamide, a taxane, trastuzumab) most frequently used (49.5% and 26.2%, respectively). FEC regimen (5-flurouracil, epirubicin, cyclophosphamide) was used by only 5.6% trastuzumab users. From 2007 to 2011, there was a 46% increase in proportion of trastuzumab use in NAT users (19.6% to 28.6%). Among eBC patients receiving NAT, trastuzumab users had a higher monthly health care cost in neoadjuvant phase ($17,425 vs. $11,422) than those without trastuzumab use; however, the out-of-pocket spending by patients ($389 vs. $370) was similar. Conclusions: Based on these real-world data, neoadjuvant use of systemic therapies was infrequent. Among the patients with HER2+ eBC, TCH and ACTH were the most frequently used neoadjuvant regimens, consistent with their use in the adjuvant setting. PCN235 Off-Label Use of Anticancer Drugs In South Korea Na Y , Choi Y J , Bae S, C H O MR, KIM A R , Kang I H , Chung I W , Yun M S , Lim S H , Kang K S The Health Insurance Review and Assessment Service, Seoul, South Korea . . . . . . . . . . . . . . . . . . Objectives: The use of off-label medication is restricted by government in some countries, because using off-label anticancer drugs has some concerns about its efficacy and toxicity. HIRA (Health Insurance Review & Assessment Service) has the process for the assessment and control for the off-label use of anticancer drugs in South Korea. We would introduce the controlling for the off-label use of anticancer drugs and evaluate the trend of off-label use in anticancer drugs. Methods: Since Dec. 2006, HIRA has permitted off-label uses for which there is adequate evidence for the efficacy, toxicity, and cost effectiveness. We collected the patient’s medical record data (briefly recorded response rate, major adverse effects. etc) which updated every year from hospital. We defined 37 cancers (36 cancers, other cancer) and preparative regimens of Hematopoietic stem cell transplantation. We calculated the number of approved off-label regimens by year and cancer type, and the regimen of the most widely used. Results: From Dec. 2006 to 2013, total 203 off-label regimens were approved their use in 63 hospitals (number of cumulative cases: 16,596). From 2006 to 2012, the number of approved off-label regimen was increased (1, 3, 3, 14, 39.37, 67, respectively). In 2013, only 38 regimens were approved. Compared with the other cancer, non-Hodgkin lymphoma (30 regimens, 15%), ovarian cancer (10 regimens, 5%), and CLL (10 regimens, 5%) have many off-label regimens. The regimens: “< 3 weekly> S-1 + cisplatin for gastric cancer” was the most widely used regimens (59 hospitals, number of cumulative cases: 2,283). Conclusions: The use of off-label anticancer drugs has been increased in South Korea since 2006. These results suggest that the development of new drugs and the more clinical trials should be needed in cancer disease. PCN236 Using Innovative Modeling Analytics with Real World Data to Develop a National Breast Cancer Screening Program in the Kingdom of Saudi Arabia Zamakhshary M F 1, Hassanain M 1, Farhat G N 1, Higashi M 2, Kruzikas D 2, Basulaiman M 1, Saeedi M Y 1, Tahan F 1 1Saudi Ministry of Health, Riyadh, Saudi Arabia,, 2GE Healthcare, Barrington, IL, USA . . . . . . . . . . . Objectives: To develop a national breast cancer (BC) screening program through phased regional expansion using an advanced investment decision support simulation model informed by real world data in the Kingdom of Saudi Arabia (KSA). Methods: An agent-based modeling simulation (ABMS) tool will represent the KSA female population, BC burden and existing health system and project impact of infrastructure investment options. Data are drawn from existing sources including census, registries and health surveys; phased studies will generate real world data on the outcomes of new clinical interventions. In Phase 1, a mobile BC screening program was deployed in Riyadh with three mobile clinics equipped with appropriate technology and medical staffing with the goal to screen at least 10,000 women during 2012-2014 and establish a care pathway for accurate diagnosis. The modeling and phased studies will guide the national program development by evaluating the impact of investments on BC screening rates, outcomes and economic value. Results: The Phase 1 Riyadh program screened 12,877 females and established a care pathway model leading to 83 confirmed BC diagnoses (rate: 6.4 per 1000). Data visualization plotting breast cancer disease prevalence and mammogram installed base identified areas of high need and low resources in the regions of Riyadh, Hafr Al-Baten, Eastern, Al-Ahsa and Al-Jouf. The ABMS model to be developed will evaluate the impact of investment scenarios encompassing expansion of existing facilities and manpower, development of new radiology centers, and implementation of additional mobile programs. Conclusions: The Riyadh program revealed higher rate of breast cancer in the region than previously reported, emphasizing the need to ensure access for accurate diagnosis and create a national program. Data visualization readily identified regions for prioritized expansion. Real world data will continue to inform the ABMS model to identify investments required to establish a national breast cancer program across KSA. PCN237 Differential Pharmaceutical Pricing: Are Prices Co-Related With GDP? Iacobucci W1, Mehta P2, Marinoni G2, Ando G2, Dall T1 1IHS, Washington, DC, USA, 2IHS, London, UK Objectives: To assess co-relation of GDP per capita (purchasing power parity) on pharmaceutical pricing. Methods: Based on empirical research, 18 drugs were selected and grouped into seven therapeutic categories: (1) Blood Based Disorders; (2) Cardiovascular Disorders; (3) Inflammatory Disorders; (4) Oncology; (5) Respiratory Disorders (only fluticasone); (6) Diabetes; and (7) Viral Diseases Price per unit (mg, IU, and U; at ex-factory level) data was collected from IHS PharmOnline International (POLI) Database across 41 countries (Australia, Austria, Belgium, Brazil, Bulgaria, Canada, China, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, India, Ireland, Italy, Japan, Latvia, Lithuania, Luxembourg, Morocco, The Netherlands, New Zealand, Norway, Poland, Portugal, Romania, Russia, Slovakia, Slovenia, South Africa, Spain, Sweden, Switzerland, Turkey, UK, and United States) from 2007 to 2012. Prices were converted into Euros on a yearly exchange rate basis and adjusted for inflation. Additionally, GDP data was collected from World Bank for the same period. We fit the regression equation for the log price per unit (dependent variable), log GDP per capita, generic status, strength, percentage of population aged 65 and above, an indicator for the US market, and year (independent variables) as follows: Y (Price per Unit) = α + ∑ β i * Xi + ε . Results: eltrombopag (-0.977+.131, n= 160), filgrastim (-4.08+. 347, n= 1420), etanercept (-.253+. 227, n= 870), adalimumab (2.52+. 128, n= 306), cetuximab (-.325+.166, n= 203), pazopanib (-3.78+.164, n= 111), fluticasone (.559+.559, n= 108), sitagliptin (-4.00+.215, n= 413), Stocrin (-12.75+.803, n= 556) and Truvada (-5.08+.157, n= 168) had statistically significant GDP (PPP) coefficients at the 0.01 level, whereas Tasigna, bevacizumab, dabigatran, rivaroxaban, exenatide, liraglutide, saxagliptin, and interferon alpha were not significant at 0.01 level. Conclusions: Our model finds varying degrees of co-relation between GDP per capita (PPP) and price per unit. Nonetheless, sitagliptin, cetuximab, filgrastim, Stocrin, Truvada, and adalimumab exhibited highest co-relation; they are thus most differentially priced. PCN238 Health Care Resource Utilization (HCRU) In Hospitalized Febrile Neutropenia (FN) Patients Treated With Chemotherapy For Solid Tumors (ST) And Hematological Malignancies (HM) In Bulgaria Gercheva L 1, Goranov S 2, Raynov J 3, Mihaylova Z 4, Mihaylov G 5, Karanikolov S 6, Petrova R 7 1Multiprofile Hospital for Active Treatment “Sveta Marina” EAD Varna, Varna, Bulgaria, 2University Multiprofile Hospital for Active Treatment “Sveti Georgi” EAD, Plovdiv, Bulgaria, 3Military Medical Academy-Multiprofile Hopsital for Active Treatment, Sofia, Sofia, Bulgaria, 4Military Medical Academy Multiprofile Hospital for Active Treatment Sofia, Sofia, Bulgaria, 5Specialized Hospital for Active Trea