Research article Unclaimed prescriptions after automated prescription transmittals to pharmacies • Anders Ekedahl and Niclas Månsson Pharm World Sci 2004; 26: 26–31. © 2004 Kluwer Academic Publishers. Printed in the Netherlands. B.E.A. Ekedahl (correspondence, e-mail: anders.ekedahl@apoteket.se): Department of Community Medicine, Lund University, Sweden and The National Corporation of Swedish Pharmacies, Apoteket Vipan, Nobelvägen 36, SE-214 33 Malmö, Sweden P.N. Månsson: Medical Products Agency, Uppsala, Sweden Key words Community pharmacy Electronic prescriptions Initial non-compliance Non-redemption rate Primary non-compliance Unclaimed prescriptions Sweden Abstract Objective: To assess the extent of and the reasons for unclaimed prescriptions, primary non-compliance, after automated transmittals to pharmacies. Methods: Cross-sectional study in 3 health care districts (population 240,000) in the southernmost of Sweden on unclaimed electronic prescriptions (e-prescriptions), transmitted to 21 pharmacies during the period of 3 months (2000) and semi-structured interview study with patients not claiming their e-prescriptions transmitted to 4 pharmacies during a period of 3 weeks (2001). Main outcome measure: Proportion of unclaimed e-prescriptions of total number of (dispensed and unclaimed) e-prescriptions and reasons for primary non-compliance. Results: In total, 2171 (2.4%) e-prescriptions remained unclaimed at the pharmacies. The peak non-redemption rate was observed for those 15–24 years old (5.5%). The lowest rate was observed for those 65–74 years old. Men had a higher non-redemption rate than women. The highest rate for men was observed 25–34 years old (6.6%). Drugs for the musculoskeletal system (ATC group M) had higher non-redemption rates than expected (3.9%), and antibiotics (ATC-group J) lower (1.7%). Adolescents and young adults, 15–24 years old, had high non-compliance rates for drugs for the musculoskeletal system (14%) and anti-asthmatic drugs (11%). Of 78 interviewed patients, not claiming their e-prescriptions, 61% reported no need to have the prescription dispensed. However, unintentional non-compliance was reported by 28%, most of them were not aware that a prescription had been transmitted to the pharmacy. Thirteen percent reported that non-redemption had given further medical problems or made obtaining a new prescription necessary. Conclusion: Primary non-compliance was generally low, but there were differences related to age, gender and type of drugs. The most common reason reported for non-redemption was that the prescription was not needed, but some patients were unaware that prescriptions were issued and transmitted to the pharmacy. Accepted May 2003 Introduction 26 Non-compliance with prescribed drugs may occur at various stages of the medication-taking process. Most literature on non-compliance concerns the problem of whether patients take their drugs as intended by the prescribing doctor, often referred to as secondary noncompliance 1, 2. ‘Primary non-compliance’ has been defined as ‘any prescription issued to a patient for which no medication is received’ 3, and has received far less attention. However, knowledge of primary non-compliance is important in the audit of drug prescribing and utilisation. Primary non-compliance consists of two separate phenomena: (1) unpresented prescriptions – those that never are received in the pharmacy for dispensing and (2) unclaimed prescriptions – those that arrive at the pharmacy to be filled but are never retrieved by the patient 4. Automated transmittals of prescriptions (e-prescriptions) from physicians´ computers directly to the pharmacy computers have become common in Sweden since 1995. As all electronically transmitted prescriptions are presented to the pharmacies, unclaimed e-prescriptions represent an opportunity to measure the extent of primary non-compliance. Of studies on primary non-compliance, reported after 1986, only a few have involved more than a single pharmacy, presented ratios of unclaimed and prescribed drugs 1, 4-6 or non-redemption rates with regard to patient age and gender 1, 5, 7. To the best of our knowledge, no study has presented data on the nonredemption rate of different drugs with regard to patient age and gender. Only one study on the reasons why prescriptions remain unclaimed has involved patients from more than one pharmacy 8. The objective of the present study was to investigate (1) the extent of e-prescriptions that remain unclaimed at the pharmacies with regard to dispensed prescriptions, type of drug and patient’s age and gender, and (2) the reasons patients report for not claiming their e-prescription medicines at the pharmacies. Methods Setting In southernmost Sweden, 3 healthcare districts with a population of about 240,000 inhabitants and a high degree of e-prescriptions were selected for the study, 2 in Blekinge county and 1 in Skåne county. There are 4 hospitals, 20 community health care centres and 22 pharmacies (4 hospital and 18 community pharmacies) within the area. Automated prescription transmittals (e-prescriptions) All general practitioners at county council administered health care centres in the three health care districts use the same system for computerised patient records and transmit e-prescriptions to the pharmacies. In Blekinge, prescribing district nurses and midwives transmit e-prescriptions and in the health care district in Skåne hospital physicians also transmit this type of prescription. An e-prescription only remains in the pharmacy computer system until printout. The first fill of a new e-prescription is recorded as a dispensed e-prescription, provided the fill is done concomitantly with printout at the pharmacy. Dispensing of ‘old e-prescriptions’, i.e., the first fill of an e-prescription previously collected without concomitant dispensing and the subsequent fills of iterated e-prescriptions, are not recorded as dispensing of e-prescriptions. Interviews. An information letter on the study and the patient’s unclaimed e-prescription was sent to the patients within two weeks after identification. Patients ⱖ 15 years of age were contacted by phone within 2 weeks after the information letter had been sent. For 0 to 14-year-old patients, a parent was interviewed. At least five attempts during two different weeks, and at least two of these attempts after working time, were made to contact each patient (parent). In case of returned mail due to invalid address, no further attempts Part I. Extent of primary non-compliance Outcome measure. Primary non-compliance rates of were made. To obtain a semi-quantitative estimate of the important reasons why e-prescriptions remain une-prescriptions. claimed, the objective was to interview about 100 paDesign, inclusion and exclusion criteria. Cross-sectional tients. study of e-prescriptions on drugs to humans, transmitted during 3 months, from 1 March through 31 May Statistics. Fisher’s exact test were used to analyse dif2000, remaining unclaimed by the 1st of October ferences between observed and expected non-re2000 (after 4–7 months). Unclaimed prescriptions re- demption rates. Standardized expected rates were calceived by telephone or fax, intended for animals and culated using the observed specific rate for each genprescriptions for non-drugs (utilities, nutritional or di- der and age group (0–6, 7–14, 15–24, 25–34, 35–44, 45–54, 55–64, 65–74, 75–84 and 85+). An alphaetary products) were excluded. Based on previously collected data from 11 pharma- level of P < 0.05 was considered statistically significies, the capture was estimated to about 25 unclaimed cant. Linear regression was used to correlate non-ree-prescriptions per pharmacy and month. The investi- demption rates at different pharmacies and in different gation period was set to 3 months to obtain about age groups. Class means were used in calculations. 1500 unclaimed prescriptions. Ethics. The project was approved by the Research EthData collection. The following data were recorded: ics Committee at Lund University. pharmacy; patient: year of birth and gender; prescriber: category (specialty, clinic/office); prescription: date, prescription items, any dispensing; drug(s): Results brand name, preparation, strength, package size and anatomical therapeutic chemical (ATC) code. The in- Part I. Extent of primary non-compliance formation was collected by one of the authors (NM) at Out of 22 pharmacies, 21 participated, constituting each pharmacy in October 2000. Data were entered in 98.5% of all dispensed prescriptions and 98.6% of all a form with controlled fields (Microsoft AccessT) to e-prescriptions. During March through May 2000, minimize errors. Each prescription was double- 89,533 e-prescriptions were dispensed, constituting checked with entered data. Data on all dispensed pre- 21.7% of all dispensed prescriptions. In total, 2,171 scriptions at these pharmacies during March, April and unclaimed e-prescriptions to 1,893 patients were May 2000 were obtained from the National Prescrip- identified. The non-redeemed e-prescriptions constition Survey. Information on all dispensed e-prescrip- tuted 85.4% of all unclaimed prescriptions, and tions at each pharmacy the corresponding period was 2.37% of all e-prescriptions (dispensed and unobtained from Apoteket AB (Göran Lindén, personal claimed) (see Table 1). communication). There was a large variation of the proportion e-prescriptions/all dispensed prescriptions between the parPart II. Telephone interviews ticipating pharmacies (range 5.9; 40.0%), and the priOutcome measure. Reasons, patients report for not mary non-compliance rate per pharmacy varied claiming their e-prescriptions. 3-fold, from 1.4 to 4.2% (2.42 ± 0.72%; mean ± SD). However, there were no correlations between primary Design. Semi-structured telephone interviews with non-compliance rates whether to pharmacy size or the patients not claiming their e-prescriptions at pharma- ratio of e-prescriptions/all dispensed prescriptions. cies. The final design was decided after a pilot study on The male/female ratios for e-prescriptions (39/61%) six patients with unclaimed e-prescriptions. Four phar- and all other prescriptions (40/59%) were similar, macies, two community pharmacies in rural areas, one both overall and in all age groups. However, there was community and one hospital pharmacy in an urban a wide variation in the ratio e-prescription/all disarea were selected for the study. pensed prescriptions with regard to age, the highest proportion, 48%, for young children, 0–6 years, and Inclusion and exclusion criteria, data collection. Patients the lowest, 18%, for 55 to 64 years old (see Figure 1). with e-prescriptions on drugs, transmitted during 3 Non-redemption rates varied 6-fold with age, the weeks, from 1 through 21 September 2001, were in- peak non-redemption rates, 5.5%, were seen by adocluded if an e-prescription remained unclaimed during lescents and young adults, 15 to 24 years old, thereafthe first week of October 2001 (after 10 to 40 days). ter decreasing to the lowest ratio, 1.0%, at 65 to 74 Patients were excluded if the e-prescription(s) was years (see Figure 2). claimed before the information letter was received. Men had an overall higher primary non-compliance The unclaimed e-prescriptions were identified and the rate than women, 2.62 vs 2.19%, due to significantly data collected by one of the authors (NM) at the phar- higher non-redemption rates in each 10-year age macies in October 2001. group from 25 to 64 years. The peak non-redemption Identification of unclaimed prescriptions/ascertainment Unclaimed e-prescriptions were identified at the pharmacies both in the computer files of unclaimed e-prescriptions and in the manual archives. Each pharmacy was asked about routines for unclaimed e-prescriptions to ascertain that all relevant prescriptions were identified. 27