Vision and Value

in Health Information

Edited by
Michael Rigby
Reader in Health Planning and Management
Keele University
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2004 Michael Rigby

ISBN 9781857758634 pbk

No claim to original U.S. Government works

This book contains information obtained from authentic and highly regarded sources. While all reasonable
efforts have been made to publish reliable data and information, neither the author[s] nor the publisher can
accept any legal respon-sibility or liability for any errors or omissions that may be made. The publishers
wish to make clear that any views or opinions expressed in this book by individual editors, authors or
contributors are personal to them and do not neces-sarily reflect the views/opinions of the publishers. The
information or guidance contained in this book is intended for use by medical, scientific or health-care
professionals and is provided strictly as a supplement to the medical or other professional’s own
judgement, their knowledge of the patient’s medical history, relevant manufacturer’s instructions and the
appropriate best practice guidelines. Because of the rapid advances in medical science, any information or
advice on dosages, procedures or diagnoses should be independently verified. The reader is strongly urged
to consult the relevant national drug formulary and the drug companies’ and device or material
manufacturers’ printed instructions, and their websites, before administering or utilizing any of the
drugs, devices or materials mentioned in this book. This book does not indicate whether a particular
treatment is appropriate or suitable for a particular individual. Ultimately it is the sole responsibility of
the medical professional to make his or her own professional judgements, so as to advise and treat
patients appropriately. The authors and publishers have also attempted to trace the copyright holders of all
mate-rial reproduced in this publication and apologize to copyright holders if permission to publish in this
form has not been obtained. If any copyright material has not been acknowledged please write and let us
know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or uti-lized in any form by any electronic, mechanical, or other means, now known or hereafter
invented, including photocopy-ing, microfilming, and recording, or in any information storage or retrieval
system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access
www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC),
222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that
provides licenses and registration for a variety of users. For organizations that have been granted a
photocopy license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library.

Typeset by Aarontype Ltd, Easton, Bristol

Printed and bound by TJ International Ltd, Padstow, Cornwall

A tribute to the inspiration of Edith K˛rner
 Contents

Dedication iii
Foreword vi
Preface x
About the editor xiv
About the contributors xvi
Introduction xxi

CHAPTER 1
Edith K˛rner: visionary, NHS reformer and friend 1
Alastair Mason

CHAPTER 2
Learning from history 11
Jeanette Murphy

CHAPTER 3
Improving the United Kingdom’s health system: an adaptive model 31
to harness information and evidence
Don E Detmer

CHAPTER 4
The need for a new healthcare paradigm: patient-centred and 47
knowledge-driven
JA Muir Gray and Harry Rutter

CHAPTER 5
Information as the patient’s advocate 57
Michael Rigby

CHAPTER 6
Information for practice improvement 69
John G Williams
 Contents v

CHAPTER 7
Information for good governance 79
Ellie Scrivens

CHAPTER 8
Information for the assessment of health outcomes 91
Azim Lakhani

CHAPTER 9
Principles and purpose for child health informatics 107
Mitch Blair and Michael Rigby

CHAPTER 10
New methods of documenting health visiting practice 121
June Clark

CHAPTER 11
Using information for public bene¢t 143
Peter Tiplady

CHAPTER 12
Globalisation or localisation: common truths or local knowledge? 149
Michael Rigby

Index 159
 Foreword

It gives me great pleasure to write the foreword to this series of essays on inno-
vation in the presentation and use of information in healthcare. Throughout
my career, and not least in my current position as Chair of the Commission for
Health Improvement (CHI), I have found that recognising the importance of
information, innovation and investment in its use, is key to determining success
or failure.
As an undergraduate and postgraduate student of medicine in Wales I had
the privilege of being taught by one of the most innovative thinkers in the infor-
mation ¢eld, Archie Cochrane, who as Director of the MRC Epidemiology Unit in
Cardi¡, became a legend in his own right in pioneering a rigorous scienti¢c
approach to information and evidence. That legacy lives on, quite rightly, in
the Cochrane collaboration that was inspired by his approach and named not
so much in his honour as because the very citation of his name encapsulates
the vision and the standards. And in a di¡erent but not unrelated way, Julian
Tudor Hart was bringing a rigorous approach to record-keeping and data cap-
ture to general practice in Glyncorrwg, in a way which inspired a very indus-
trial practice population, as well as demonstrating how to address the Inverse
Care Law that he identi¢ed. So as a result, healthcare delivery in South Wales
was very much in£uenced by this recognition of information and establishment
of innovation in its application.
In my own early practice in community medicine, aspects of information were
key both in terms of providing best evidence, but ^ in a sense more impor-
tantly ^ in terms of recording local service availability and individual patient
need. And so with my progression to Chief Medical O⁄cer for Wales, it was only
natural that information should be a key resource which I valued and nour-
ished. We were able to watch what was happening not only in England but else-
where, and to interpret and adapt those initiatives to apply them to the speci¢c
challenges of the Principality. At the same time, though, we sought to put great
emphasis upon innovative and successful completion of information invest-
ment, rather than merely the initial pioneering idea. These were exciting times,
with the Strategies for Health Gain just one signi¢cant example. These were
¢rmly grounded in information ^ information as to need, information as to
performance, but above all setting measurable indices of progress counted in
terms of measured bene¢t to the population.
 Foreword vii

The technical infrastructure to obtain and use good-quality data was not
overlooked either. One particular achievement was the Digital All Wales Net-
work (DAWN), which rolled out to the whole of the NHS family throughout
Wales, including primary care, the bene¢ts of a dedicated health infrastructure.
This was later followed by developments such as the private intranet, Cymru-
Web. We also placed importance in investing in an information-handling
resource, best known as the Health Intelligence Unit, which later developed
into Health Solutions Wales.
One of the other major initiatives I had the challenge of working on, together
with Sir Kenneth Calman in England, was the total review of cancer services in
England and Wales. This was a challenging brief, and our recommendations
were equally challenging to those practising in this important ¢eld throughout
England and Wales. That we called for a radical reshaping of service provision is
now well known, and the rationale for this was entirely evidence-based. It was
upon review of information about activity and results that we were able to
reach the conclusions we did, motivate so many organisations and individuals
to undertake di⁄cult change with comparative smoothness, and as a result,
bring about lasting bene¢ts to the population and patients.
From within Wales we were not directly involved in the work of the Steering
Group on Health Services Information, and the drive of Mrs K˛rner. However,
as with many other innovations in England, we watched with a mixture of envy
and relief, but above all with the great opportunity of learning from observa-
tion. Indeed, on a ¢ne day Mrs K˛rner’s home region could be viewed across
the water from my own constituency! The challenges that the Steering Group
faced were clearly major, both in terms of the size of the task and in terms of the
natural institution of organisational inertia through the long tradition of the
pre-existing statistical systems. But what was clear was that Edith K˛rner
bought a combination of vision, inspiration and sheer determination which
motivated her Steering Group, charmed and persuaded the wider constituency,
and drove her immediate support team. From her work came a vision, and then
a demonstration, of how to harness modern management approaches to infor-
mation as a resource and apply them to the huge and technically precise
domain of health information.
Now, in my current position as Chair of the CHI, I continue to be very con-
vinced of the importance of information in healthcare management and clinical
practice. In our work in the CHI we see innovation and good practice; inevitably,
we also have to focus on the more disappointing areas of poor practice and other
problems. And if I were asked to identify any particular factors which seem to
di¡erentiate between success and failure, I have to say that nearly all the loca-
tions where we have to deal with problems can be identi¢ed as those with either
poor information systems, or scant understanding of how to use that informa-
tion in a diagnostic and material way. And similarly, with nearly every example
of good practice can be seen investment in information, which is a key factor.
viii Foreword

So it therefore gives me great pleasure to welcome and introduce this set of

essays on the innovative use and development of information in healthcare.
Firstly, because I think it rightly acknowledges and pays tribute to the work of
one of the key visionaries in this ¢eld, Edith K˛rner, who in line with many other
pioneers was not fully recognised or appreciated at the time. And secondly,
because I think it is important that we continue to innovate and develop our
information-handling abilities, and this collection comprises a number of very
di¡erent but equally important visionary approaches and applications described
by colleagues whose work is well known to and highly respected by me.
I hope that you enjoy reading these as much as I did. More importantly, I hope
that this collection inspires further practical use of information in healthcare.
The production of this volume is an important initiative by the Nu⁄eld Trust,
and I welcome it.

Dame Deirdre Hine

President Royal Society of Medicine
July 2003
 Preface

Edith K˛rner: analyst and visionary

Perhaps once in a lifetime, if one is lucky, one is privileged to encounter the
in£uence of an individual who combines vision, analysis and commitment in a
way that commands respect whilst providing inspiration. Edith K˛rner was
such a memorable person.
Edith K˛rner arrived in this country with little choice, as a pre-war refugee.
In modern parlance, she would be classed as an asylum-seeker, arriving on her
own as a homeless teenager unable to speak English. She proceeded in due
course to give to her adopted country an input and legacy that multiply
rewarded that initial asylum and hospitality.
This volume re£ects on merely one facet of her public service contribution in
Great Britain. There were earlier and later phases in very di¡erent spheres,
which makes her contribution all the more impressive. And of course there
were roles as supportive wife and caring mother ¢tted in as well, indeed ¢rst
and foremost.
However, for that small group who had the opportunity of working with
Edith K˛rner in seeking to develop systematic health information, this was a
privilege and an inspiration. And for the much wider NHS community, who
even 20 years later use her name daily as a term describing speci¢c information
concepts, she has by de¢nition become something of a legend. Few can have
their work endure in this way, even if the amount of achievement made was
far less than she wished and hoped.
This volume is not a retrospective study, though contributions start with an
analysis of her work ^ one by a contemporary and one by a retrospective dis-
coverer of a legacy deeper than is usually appreciated. Rather, the intention
is to pay tribute to Edith K˛rner’s skills and vision by seeking, however mod-
estly, to continue that visionary thinking along di¡erent new strands in a way
that, regrettably, is usually excluded in the modern task-driven and problem-
orientated public sector.
What were the magic ingredients that enabled Edith K˛rner not only to be so
successful with her initial daunting task, but to lead others to want to build
on it? Working with her was to experience a clarity of thought, and a depth of
analysis, that were powerful and enlightening at the same time. She not only
 Preface xi

had a mental master plan for her work, but she was active in small group
discussions of every detail. With her at the table the debate was always friendly,
and tolerant of exploratory ideas, yet any unfounded assumption or unjusti¢ed
conclusion was pounced on for what it was ^ unsustainable. The precise tones
of the accent with its surviving hint of east-European origins would pose
the question that, though not strictly rhetorical, by its very asking provided the
answer. The careful phraseology, the momentary pause to produce every time
the absolutely apt word or phrase, taught many of us a new level of analytic and
re£ective discussion. Yet this was always amicable, and business meetings were
like (indeed were) meetings of friends to which one always looked forward.
And to the work itself, Mrs K˛rner brought the background experience of
being the vice chair of a regional health authority, and therefore strong aware-
ness of the issues and challenges of obtaining adequate information for strategic
management. Yet, in modern parlance, she was a ‘bottom up’ as much as a ‘top
down’ person. Running throughout her reports was the express principle that it
was di⁄cult to think of any information item or analysis that was relevant at a
higher level if it was not relevant at the level below it. This was true respect for
those undertaking the operational work of the NHS, whilst also recognising the
need for modern management of an informed nature, in rather the way which
was articulated a decade later by Sir Roy Gri⁄ths.
The freshness of Edith K˛rner’s approach ^ the legacy upon which this
volume seeks to build ^ was not framed in speci¢c recommendations and
reports, important though they were at the time and in many ways still con-
tinue to have in£uence. Rather, it was the important recognition of information
as a key management tool, with its concomitant need to apply consistent objec-
tive principles and methods in the NHS to its recording, gathering and use that
was the beacon. The reports of her steering group start with an analysis of the
requirements for information, and many of the concepts she described were
articulated in ways well ahead of their time. The terms ‘health gain’ or ‘cost-
e¡ectiveness’ may not have been current at the time, but those underlying
concepts can clearly be seen in the arguments expressed. Alistair Mason, who
provides our ¢rst contribution, was the secretary to the Steering Group, and
therefore worked with Edith K˛rner closely on a daily basis. My own role at the
time was much more humble, as a mere co-optee to one working group, but
the fact that we were both treated in the same way, and both have the same
warm yet respectful recalls, is itself an example of Edith K˛rner’s inspiring open-
ness and inclusiveness in every sense of the word.
The great disappointment of Edith K˛rner’s work was that it never reached
true fruition. Very seldom does the public service, not least the NHS organ-
isation, have capacity for a visionary. Mrs K˛rner started o¡ along an important
road towards the development of e¡ective purposeful information systems;
the powers that be decided that her ¢rst, interim, reports were perfectly ade-
quate as the ¢nal answer. For them the future was for others, and so not their
xii Preface

responsibility. Edith K˛rner and her immediate colleagues were bitterly disap-
pointed; the NHS and healthcare have su¡ered seriously as a result. I hope this
volume provides a modest opportunity to give visionary thinking an appropri-
ate outlet.
However, Edith K˛rner’s work, although refreshingly clear and objective,
was neither cold nor divorced from real life. That, indeed, is part of the endear-
ing richness of the example she left. It is important not to let leading-edge think-
ing, nor top-level national responsibilities, become either remote or cold. Those
who worked with Edith K˛rner will each have a favourite personal anecdote.
One legend concerns a time when Mrs K˛rner was invited to speak to the
young members’ branch of a national professional organisation. After the for-
mal proceedings, Mrs K˛rner made it clear that she preferred to socialise with
the young members rather than the platform party. Someone di⁄dently invited
her to join the table tennis party, which she willing accepted, and play com-
menced in a gentle style with due respect for the dignity of their older guest.
Sometime later, when the visiting speaker had thoroughly trounced all the
young members, she gently asked whether they were unaware that she had
been a youth champion in the country of her birth. Like her cognitive skills,
this physical performance was little diminished with age.
A quite separate anecdote relates to the time, in the midst of the full £urry of
the work of the steering group, when, after a strenuous week, a relaxing week-
end was clearly deserved and needed. However, on subsequently enquiring
whether rest and recreation had indeed been the order of the weekend, the
reply was that she had completed some home plumbing. When those present
argued that this was not what was planned or appropriate, her comment was
archetypical of the lady, ‘The job had to be done and no plumber was available. So I
thought of the IQ of a plumber, I thought of my IQ, and wondered why I was waiting and
then going to pay a large sum of money, so I went out and bought a book and did the job
myself.’ This was not in any way an anti-tradesman statement, but rather a phi-
losophy that if something needed doing the central duty was to acquire the skills
and do the job, and not to consider any essential task demeaning.
Those two anecdotes encapsulate the humanitarian inspiration which Edith
K˛rner provided. She was able to mix and enjoy the company of anyone, and to
tackle any necessary task, but in both cases to do them with good grace, to a
high standard, and above all, with enjoyment.
Edith K˛rner may have been ahead of her time, or conversely the times may
not have been ready for her. She brought a blend of experience, common sense,
rigorous analysis and vision to an important task, and that is a powerful mix-
ture that serves as a model framework. She may not have ¢nished her task as
she saw it, or indeed as others hoped it would be undertaken, but there is no
doubt that she left a legacy ^ both of practice and of principle ^ that has
endured. These are important lessons for the future of health information ^ to
have objectivity, clarity and vision; to recognise the skills and the humanity of
 Preface xiii

those delivering care at the operational level; and at the same time to recognise
the importance of accurate and timely management information derived from
operational data. Those principles may have been overshadowed by political
expediency in aspects of recent policy, which has itself proved far less enduring,
but their appropriateness is shown by the fact that they can be recognised as
themes in each of the papers in this collection.

Michael Rigby
Reader in Health Planning and Management
Keele University
July 2003
 About the editor

Michael Rigby BA FSS

Reader
Centre for Health Planning and Management, Keele University
Michael Rigby’s initial career was in the National Health Service, commencing
as a researcher for the then Cheshire County Council Health Department. He pro-
gressed through information, planning, and computer records positions in
Cheshire Area Health Authority, culminating in appointment as Regional Ser-
vice Planning O⁄cer for Mersey Regional Health Authority. Within this period
he was a member of the Community Services Working Group of the Steering
Group on Health Services Information, chaired by Mrs K˛rner. He then moved
to the Centre for Health Planning and Management as Lecturer, with special
interests in information and related systems, healthcare reform, and non-acute
services. He teaches and leads courses on the two MBA programmes and other
programmes run by the Centre, and undertakes a range of research projects on
information issues in health and healthcare policy and practice. He has a wide
range of publications, is founding editor of the Harnessing Health Information
series with Radcli¡e Medical Press, and lead editor of Taking Health Telematics
into the 21st Century, also from Radcli¡e Medical Press with support from the
Nu⁄eld Trust.
 About the contributors

Dr Mitch Blair MBBS BSc MSc FRCPCH FRCP FRIPH

Consultant Reader in Paediatrics and Child Public Health, Imperial College
Mitch Blair trained in London as a paediatrician, then moved to Nottingham as
Senior Lecturer in Community Paediatrics from 1990 to 1998. There he devel-
oped his interest in the role of information systems in supporting child health.
He co-chaired the Royal College of Paediatrics and Child Health Interprofes-
sional Forum on Child Health Informatics. He is currently an academic clinical
paediatrician working at Northwick Park Hospital and active member of the
National Child Health Informatics Consortium, and has continued to in£uence
local and national policy on child public health through his teaching and
research. The original K˛rner reports have remained an important in£uence
throughout his career.

Professor Dame June Clark BA MPhil PhD RN RHV FRCN

Emeritus Professor of Community Nursing, University of Wales Swansea
Visiting Professor, University of Iowa, USA
Visiting Professor, University of Maribor, Slovenia
June Clark worked in the NHS for 35 years as a health visitor, Director of Com-
munity Nursing Services and Chief Nurse to a Health Authority before moving
into higher education in 1990. Her special interest is the development of a mini-
mum data set and standardised terminology to describe nursing in patient
records. She has acted as consultant to the International Council of Nurses
and to the World Health Organization, and has held many leadership positions
in nursing including President of the Royal College of Nursing 1990^94. She is
a Non-executive Director of Carmarthenshire NHS Trust.

Professor Don E Detmer MD MA FAAAS FACMI FACS

Dennis Gillings Professor of Health Management, University of Cambridge
Professor Emeritus and Professor of Medical Education, University of
Virginia, USA
Don Detmer practiced as a vascular and sports medicine surgeon for 25 years and
also developed the ¢rst Masters level management and leadership educational
 About the contributors xvii

programme for clinician executives in the USA (University of Wisconsin-

Madison 1973^84), as well as administering two academic health science
centres (Universities of Utah and Virginia 1984^95). A lifetime associate of the
National Academies, he has been active in health policy in Washington DC,
having chaired the Board of Regents of the National Library of Medicine, the
National Committee on Vital and Health Statistics, and the Board on Health
Care Services of the Institute of Medicine. He has been instrumental in the
development of the National Health Information Infrastructure in the USA.
He has directed Cambridge University Health for the past four years.

Dr JA Muir Gray CBE MD

Director of the National electronic Library for Health
Director of National Screening Programmes
Institute of Health Sciences, Oxford
Muir Gray’s main interest has for many years been getting research into practice.

Dr Azim Lakhani MA FFPHM

Director of the National Centre for Health Outcomes Development (NCHOD)
Azim Lakhani is an accredited specialist in public health medicine. He was
previously Director of Public Health for West Lambeth Health Authority, subse-
quently a Principal Medical O⁄cer and Director of the Central Health Outcomes
Unit of the Department of Health. Alongside overall current responsibility for
NCHOD’s research and development work concerning various aspects of health
outcomes assessment, Azim works directly on the development and production
of some 150 clinical and health indicators used by the National Health Service
in England (Compendium of Clinical and Health Indicators) plus dissemination of
information on health outcomes via an NHS website ^ the Clinical and Health
Outcomes Knowledge Base.

Dr Alastair Mason FRCP (London) FFPHM

Consultant Epidemiologist
National Centre for Health Outcomes Development
Alastair Mason was Secretary to the Steering Group on Health Services Infor-
mation for its duration. He was Regional Public Health Director of the South
Western RHA from 1987 to 1994. He is currently a freelance public health con-
sultant with part-time appointments with the National Centre for Health Out-
comes Development and the Royal College of Physicians Clinical E¡ectiveness
and Evaluation Unit. He edited the volume of essays, published by the King’s
Fund to honour Mrs K˛rner, Walk don’t Run.
xviii About the contributors

Jeannette Murphy BSc DipComputing

Senior Lecturer in Health Informatics
Centre for Health Informatics and Multiprofessional Education (CHIME), Royal
Free and University College Medical School, London
Jeanette Murphy’s special interest is in demystifying health informatics so that
doctors, nurses, managers and administrative sta¡ can see its relevance to their
work and to patient care; she is also interested in understanding why informa-
tion systems do and do not deliver bene¢ts. She does this as Graduate Tutor in
CHIME, with overall responsibility for matters relating to education, training
and professional development. She oversees research degree students and two
MSc programmes (Clinical Risk Management and Health Informatics). Jeanette
is also a visiting lecturer at St George’s Hospital Medical School (a Health Infor-
matics module), Keele University Postgraduate Medical School and the School
of Pharmacy (University of London). She organises IT skills training for medical
students, and special study modules in health informatics. When she was
initially trying to understand the challenges facing the NHS in relation to infor-
mation management she chanced upon the ‘K˛rner Reports’, and the breadth
and vision of this work shaped her understanding of what we now call health
informatics, and in£uenced her career.

Dr Harry Rutter MA MB BChir MSc MFPH

Specialist Registrar, South East Public Health Observatory
Fellow, Department of Public Health, Oxford University
Harry Rutter worked in a range of clinical jobs before training in public health.
He is interested in sustainability, transport and health, and the impact of tech-
nology and communications on the way in which we practice as professionals.

Professor Ellie Scrivens BA PhD

Professor of Health Policy and Director of the NHS Controls Assurance Support
Unit
Keele University, Sta¡ordshire
Ellie Scrivens is Professor of Health Policy at Keele University and Director of
the Controls Assurance Support Unit ^ a national Department of Health-funded
unit to promote risk management and improved control systems within the
NHS. She is also Vice-Chair of Shropshire and Sta¡ordshire Strategic Health
Authority, a member of the accreditation panel for the Scottish Prison Service,
is on the board of a number of NHS national organisations concerned with stan-
dards and quality of healthcare, and is consultant to the Department of Health
on the development of national standards for healthcare. She was a member of
the Community Services Working Group of the Steering Group on Health Ser-
vices. She has written extensively on quality in healthcare, and with Professor
Rudolf Klein contributed to Walk don’t Run, the ¢rst volume of essays devoted to
Edith K˛rner’s work.
 About the contributors xix

Peter Tiplady MB BS MRCGP FFPHM

Honorary Consultant in Public Health Medicine, Cumbria and Lancashire Stra-
tegic Health Authority
Formerly Director of Public Health, North Cumbria Health Authority
Peter Tiplady recently retired after 30 years in public health, but continues with
a number of professional and charitable voluntary activities. He has had a
career-long interest in information for health, in which his special interests
included information for child health and community health services. More
recently he has been concerned with the epidemiology of diabetes, issues of
foot and mouth disease, complementary healthcare and medical ethics. His
published and current research include Section 47 of the NHS Act, radioiodine
thyroid cancer in Cumbria, foot and mouth disease, control of diabetes during
the foot and mouth disease epidemic (in progress) and evaluation of healing by
light touch. He was Vice Chair of Working Group ‘D’ (community services infor-
mation) of the Steering Group on Health Services Information.

Professor John G Williams MA MSc FRCP

Professor of Health Services Research, Clinical School, University of Wales
Swansea
Consultant Gastroenterologist, Bro Morgannwg NHS Trust
John Williams has a longstanding interest in improving the availability, utility
and validity of information for patient care and service management. He is
Director of the Royal College of Physicians’ Health Informatics Unit, which
is promoting the education and culture change needed amongst physicians,
and developing evidence-based standards for medical records. He has a parti-
cular interest in the routine capture of patient-focused outcomes and in the use
of routine data for service planning and research.
 Introduction

This publication, Vision and Value in Health Information, not only celebrates the
contribution of Edith K˛rner over the years but also o¡ers a signi¢cant chal-
lenge for a proper place for health information in the modernisation of health
services in the UK.
The Nu⁄eld Trust, in its Policy Futures for UK Health1 project, was attempting
to analyse the broad environment for health in this country in the year 2015,
and the implications of that for current UK health policy. The trust indicated
some areas where the government could take action now in order to anticipate
the likely circumstances of 2015. This time frame was chosen carefully with a
view to making a constructive contribution to policy development and not
creating a piece of abstract futurology. The period under review extended
beyond the usual constraint of the electoral cycle but was short enough to
allow a realistic assessment of future developments.
The report was in the Nu⁄eld Trust tradition of stimulating change in the
culture of the health policy-making process, and of encouraging thinking and
analysis based on evidence. There were three messages to the prime minister:

. not to forget the issue of health while reforming the NHS

. how to manage public expectations and to pay for them
. the importance of thinking through fully and well in advance the implica-
tions of change for the people who work in the health service.

The study foresaw very radical new developments in, among other things,
information and communication technology, that would move conventional
professional boundaries within the workforce and change the locations of
care. All three areas highlighted that an e¡ective policy for information and
information technology (IT) was essential. No business runs on as little infor-
mation or IT expertise and capability as does the NHS, or has invested so little
in general sta¡ training or in raising awareness of the importance of data, infor-
mation and intelligence. Unless information is properly gathered, organised,
analysed and used, the NHS will continue to lack some of the most basic tools
of policy-making and may be unable to cope with the changes that lie ahead,
still less with moderate public expectations in relation to them.
Central to stimulating debate on the problems that we now face and will
face in the future, the Nu⁄eld Trust is pursuing its interest in benchmarking.
xxii Introduction

A great deal of interesting work is being done on this, for example by the Orga-
nization for Economic and Commercial Development (OECD) and the World
Health Organization (WHO). One central question will be: how can we tell
how well what we are doing in the UK works? It is essential that indicators are
developed to permit accurate assessment of where we are now, so that sensible
benchmarks can be set on where we want to be by the year 2015. This emphasis
is about perpetuating the K˛rner cause.
We are at an in£exion point in healthcare, with advances in biomedical
science, and looming developments in genetics, pharmacology, computing and
nanotechnology all acting as agents for change. But our health systems are still
unable to deliver a consistently high level of error-free healthcare. Data that
support the development of evidence should be of proven value to individuals,
the population and patients. Sound policy and IT should assist everyone con-
cerned with the healthcare system, but above all should help clinicians to
improve their operational decisions. We also need to provide consumers and
patients with accurate information that stimulates them to demand action to
improve the service they receive. Measurement and reporting lie at the heart of
the ability to identify where errors are occurring, and the need to deal with
them, as well as giving those funding the health system the ability to control
and direct their resources more accurately. Empowering the public is one of
the potentially largest drivers for improvement in our healthcare system.
There is a degree of dissatisfaction among patients with the outcomes in the
present NHS and this may be a potent means of building pressure for change.
Information initiatives are important.
The role of information and IT should also be central to improving the quality
of healthcare. It is already almost impossible for clinicians to keep abreast of
advances in their ¢elds, and even the present rate of advance in knowledge
might soon overwhelm doctors, so it is clear that it is impossible for individual
doctors to cope with the impending genetics revolution. There is a need for
information and IT systems that assist doctors and patients at the point of
decision. We need visionary information, information systems and communica-
tions technology projects almost akin to putting a man on the moon to achieve a
goal which would not only be of immense bene¢t to societies in industrialised
countries, but could also be of relevance in developing countries.
Lastly, I would like to focus on public interest in information. We now live in a
period where public disclosure of information about quality at the level of
named hospitals, doctors and primary care organisations, is becoming the
norm. In the USA this movement is being driven by a variety of stakeholders ^
government, business and the public. In the UK, the government (with some
public support) is the primary force behind the movement. At the core of this
movement is the concept that producing public information about the quality
of care actually provided will complement all those who are putting systems in
place to improve quality. The public disclosure of information should help the
 Introduction xxiii

quality of care improve rapidly in the ¢rst three decades in the 21st cen-
tury, faster than it did in the last three decades of the 20th century. The hope is
that the provision of appropriate data, information and intelligence will contri-
bute to more rapid improvement in the delivery of quality healthcare ^ no over-
use, underuse or misuse of medical interventions. This is the continuing K˛rner
challenge.

John Wyn Owen CB

Secretary
Nu⁄eld Trust
April 2003

Reference
1 Dargie C, Dawson S and Garside P (2000) Policy Futures for UK Health. 2000 report. Nuf-
¢eld Trust, London.
CHAPTER 1

Edith K˛rner: visionary,

NHS reformer and friend
Alastair Mason

The background
In 1975, Prime Minister Harold Wilson set up a Royal Commission to consider
the best use and management of the ¢nancial and sta¡ resources of the National
Health Service (NHS). It reported in July 1979 to a Conservative government
that accepted the larger part of the 117 recommendations.
As usual, implementation was dominated by structural change. The restruc-
turing of 1982 adopted the key recommendation to abolish area health author-
ities (AHAs). The new structure was to comprise 14 regional health authorities
(RHAs), 192 district health authorities (DHAs) and 90 separate family practi-
tioner committees (FPCs).
Royal Commission comments about the need to strengthen local manage-
ment were also instrumental in the genesis of the management review, led by
Roy Gri⁄ths, and the consequent implementation of the general management
function in 1984.
The royal commissioners severely criticised the state of the national infor-
mation systems, which were broadly unchanged from those existing before the
previous NHS re-organisation of 1974. In particular, they were concerned
about the lack of information available for monitoring and controlling resource
utilisation.
The regionally organised computerised hospital activity analysis statistical
system contained data about all inpatients discharged from general hospitals.
From this regional database a 10% sample was submitted to the Department of
Health to form the national hospital inpatient enquiry database. There was a
separate mental health enquiry containing information about discharges from
psychiatric hospitals and units. A variety of systems collected data about mater-
nity events. The major managerial data sources were Form SH 3, the annual
2 Vision and value in health information

facilities return from each hospital, and Form SBH 203, containing waiting list
data. These systems formed a body of statistical data for policy use by the
Department of Health, from which standard summary tables were published in
report form, inevitably with a time delay. There were major concerns about
data accuracy and timeliness, and the relevance to contemporary clinical prac-
tice, of the de¢nitions and classi¢cations of all these systems.
However, the royal commissioners also identi¢ed some promising informa-
tion research projects. As part of the ‘clinicians in management’ initiative in
the late 1970s, two former district administrators, John Yates and Iden Wick-
ings, were successfully developing new techniques for performance indicators
and clinical budgeting. In 1982, concerns about deaths after operations led to
the con¢dential enquiry into peri-operative deaths being set up. Some comput-
ing projects were beginning to show the bene¢ts that electronic data capture
and storage might bring.
It was against this background that the NHS/DHSS Steering Group on Health
Services Information carried out its work from 1980 to 1985.

The steering group

The terms of reference of the steering group were as follows:

To agree, implement and keep under review principles and procedures to

guide the future development of health service information systems; to iden-
tify and resolve health service information issues requiring a co-ordinated
approach; to review health service information systems; and to consider
proposals for changes to, or developments in, health service information
systems arising elsewhere and, if acceptable, to assess priorities for their
development and implementation.

The steering group was to be permanent (to ‘keep under review’), to have
primacy in information matters (to ‘resolve issues requiring a co-ordinated
approach’) and to be responsible for implementing its proposals. Mrs Edith
K˛rner, vice-chair of the South Western Regional Health Authority, was
appointed as chair.
At the time civil servants were very uncomfortable with the concept that an
independent, and thus di⁄cult to control, group should be given responsibility
to implement its own recommendations. Mrs K˛rner took her case direct to
Secretary of State, Patrick Jenkin, and obtained the implementation commit-
ment that she wanted. The political timing was opportune because in 1980
there was emphasis on greater decentralisation of the NHS, and recognition by
ministers that those in the periphery should have more in£uence on the centre.
 Edith Körner: visionary, NHS reformer and friend 3

Although the remit of the steering group was theoretically wide-ranging, in

practice it was seriously limited by not covering primary care or IT and compu-
ter systems. The 1982 restructuring had kept the management and delivery of
primary care services separate from the hospital and community services and
those responsible for policy in the Department of Health and Social Security
(DHSS) fought successfully to have general practice information omitted from
the scutinity of the steering group. NHS computer personnel, worried about
their perceived loss of power, managed to persuade the DHSS to set up a sepa-
rate NHS Computer Policy Committee in 1981. Despite the original commit-
ment in its remit to the primacy of the steering group, the Computer Policy
Committee was given prime responsibility for IT, thus setting the pattern for
the next 20 years, with NHS computing ¢rmly in the hands of ‘technocrats’
and not users.
The steering group focused on information for health service management.
Although extremely interested in all aspects of information, it did not directly
consider epidemiological information about the occurrence of disease, the
health needs of populations or health status. Information for clinicians to eval-
uate their care was not considered unless there were clinical groups willing to
work with them. Sophisticated data sets were developed only for accident and
emergency consultants and maternity services professionals.
Mrs K˛rner controlled the membership of her group tightly. Members were
chosen for their expertise and experience. The only people there as representa-
tives were consultants nominated by the Joint Consultants’ Committee. NHS
administrators, ¢nance o⁄cers and information experts heavily outnumbered
the few civil servants.
An innovation insisted on by Mrs K˛rner was that the secretariat should be
chosen by her and owe its loyalty to her and not to the civil service from which
it came. Indeed, in the ¢rst three months a founder member was replaced for
failing to comply with this requirement. Although essential for the smooth
working of the steering group, such allegiance did not necessarily advance the
future careers of civil servants forming the secretariat!
Much to the consternation of the civil service, Mrs K˛rner insisted that all the
papers of the steering group should be made available to anybody who wished
to see them. The thoughts and deliberations of the group were not to be kept
secret, and interested people were actively encouraged to contribute to its work.

The chair
By choosing the hitherto little-known vice-chair of the South Western RHA to
become chair of the steering group, ministers cannot have imagined the force
they were unleashing. Mrs K˛rner may not have had a reputation outside her
4 Vision and value in health information

region, but within it her ability to tackle and solve the really di⁄cult problems
without fear or favour was legendary.
I ¢nd it very di⁄cult to write objectively about this extraordinary woman.
As so many adverse things have happened to the NHS it is impossible not to look
back at the early 1980s as a golden age. Memories thus tend to be rose-coloured.
However, David King, one of Mrs K˛rner’s closest NHS friends, summed up
her qualities admirably in Walk Don’t Run, the original set of essays to honour
Mrs Körner, published in 1985:

She established her authority by brute ability, being better informed and
working harder than anyone else. Everyone accepts that she knows more
about the subject and has read more (in several languages) about it than
any two other people. Hers is not merely a detailed knowledge of a limited
technical subject, for her grasp of health services internationally and the
changes they are undergoing provide a general context into which the
invaluable auxiliary information ¢ts.

Mrs K˛rner’s commitment to the task and her capacity for hard work never
£agged, even when, in addition to everything else, she was found to be
rattling over every inch of British Rail’s track to attend or address meet-
ings. Happily, these formidable qualities are leavened with a keen sense of
humour, and when K˛rner activity was in full £ood, it was rare to meet
anyone engaged in it who did not start the conversation with the latest
anecdote or bon mot.

Powerful intellect and incisive humour are not always an endearing com-
bination unless, as in her case, they go with a genuine regard for others,
whatever their station and ability. This respect and a¡ection for the NHS
and its sta¡ were apparent to all, and people sensed that here was a refor-
mer with her heart in the right place.

The friend
Colleagues who have su¡ered over the last 15 years with chairs, bent solely on
supporting the political fancies of the day, will envy the relationships we had
within the steering group. Mrs K˛rner was a Jewess, born originally in Czecho-
slovakia, and blessed with many of the talents of this culturally rich group of
immigrants who arrived during the 1930s.
Like the archetype Jewish grandmother she worried. She worried about the
progress of the work; she worried about us in the secretariat, and she worried,
above all, from day one, about how recommendations were to be implemented.
Indeed, Mrs K˛rner was only happy when she was worrying. Many a Friday
evening I received a telephone call about a new danger facing us, allowing
 Edith Körner: visionary, NHS reformer and friend 5

me to share it with her over the weekend. This constant concern about what
was, and what might happen, underpinned an attention to detail and the rapid
resolution of potential con£icts that were essential to the successful completion
of the work
Mrs K˛rner’s command of the English language was unsurpassed. Tom Stop-
pard, a fellow countryman, sent her his plays in draft for comment. She wrote
brilliant satirical pieces for the health and social services journal, including the
cult-status column ‘Dear Charles’. Although we in the secretariat did the tech-
nical writing, Mrs K˛rner meticulously corrected the punctuation, added a ¢nal
literary gloss and chose the frequently obscure but always apt quotation asso-
ciated with each publication. The clarity of the writing was a major reason for
the ready acceptance of the recommendations by NHS sta¡.
We travelled endlessly to market our proposals and to hear the views of the
NHS face-to-face. However inept the comment or malicious the intent, she trea-
ted all questioners as intellectual equals and never put down or embarrassed
them in public. This active engagement with NHS rank and ¢le engendered a
respect and a¡ection still held today by all who were involved. There was feeling
amongst many that although they may not have understood the technical
merits of the recommendations they must be good because she was.
Mrs K˛rner took a genuine interest in our families, sharing concern about
childhood ailments and o¡ering excellent advice about educational and beha-
vioural problems. One of my children in his teens needed some time and space
to himself, and she and her husband, typically generous, entertained him for a
week with very successful results.
Indeed, the most memorable experience of that period was a personal moment
not a work one. Three years into the project we were invited to a World Health
Organization (WHO) conference in Dˇsseldorf. This was the ¢rst time she had
returned to Germany since travelling through it alone in 1938 in a sealed rail-
way carriage as a 17-year-old girl on the way to freedom in England. Though
very apprehensive she behaved with dignity and supreme professionalism. The
only clue to the inner turmoil of emotions was the doubling of an already signif-
icant nicotine habit. It was a privilege to have been with her.

The vision
The work done by the steering group was informed by the clear vision developed
by its chair. The activity focused on information for health service manage-
ment. The guiding principle governing the approach was that data should be
collected because they are essential for operational purposes. User-orientated
information yields bene¢ts to those who collect it, and thus provides an incen-
tive for accuracy and expedition. From this simple central vision there emerged
a number of logical consequences.
6 Vision and value in health information

. The central DHSS returns, which at that time totally dominated the develop-
ment of information systems, must be determined by the data required locally.
The information needs of district management should be paramount and the
data submitted to the DHSS con¢ned to a small subset of the district data set.
. Every DHA and its o⁄cers need a minimum amount of data available to
them, collected as a by-product of operational processes, to carry out their
management function. An authority not regularly using such data is handi-
capped by being inadequately informed when ful¢lling its responsibilities.
. The recommendations about data items to be included in the district mini-
mum data set should be a compromise between the desirable, the feasible
and the a¡ordable. It was felt that this pragmatic approach would lead to
proposals adopted in a reasonable timescale, with feasibility being de¢ned
as implementable within three years
. In order to make informed judgements about their own performance, mem-
bers and o⁄cers require information not only about their own district but
also comparable information about others. To enhance the validity of such
comparisons, standard de¢nitions and classi¢cations should be developed for
each requisite data item.
. Although the district minimum data set is relevant to operational managers,
most units and departments will wish to collect more data for their own
purposes. However, the variety of local arrangements militates against the
national prescription of standard data sets for every managerial entity.
. To meet the increasing need for IT within districts, considerable design
work is essential to ensure compatibility between di¡erent computer sys-
tems. Investment in computer systems was considered by the steering group
to be only justi¢ed when the bene¢ts of better information were added to the
improvements in operational processes.
. As clinical and management practices change, data collected about hospital
patients and resources have to be altered to accord with current reality. The
minimum data set is therefore not set for all time. Its relevance must be reg-
ularly reviewed and its content appropriately updated.

Rarely can a national group reforming the NHS have had such a simple sustain-
able vision; one that is as relevant to today’s health economies as it was to the
DHAs of the 1980s.

The work
The work for which the steering group became best known was the develop-
ment of the district minimum data sets. Seven working groups were set up to
review the information needed to cover all aspects of heath authority activity,
 Edith Körner: visionary, NHS reformer and friend 7

manpower and management accounting. To re£ect the vision, the work in each
area was done in two stages. Groups of NHS experts were brought together to
develop the minimum data set. No civil servants, no representatives of special
interest groups or external management consultants were involved. The talent
was all from within the NHS.
Once the district needs had been identi¢ed, a second group (with civil servants
as well as NHS sta¡) agreed the subset of the minimum data set that would
be made available to the DHSS. All the recommended data items were de¢ned
and classi¢ed. Any new data items, de¢nitions or classi¢cations were piloted
for feasibility of collection in four participating health districts.
Extensive consultation took place on each report. In addition to the usual
formal written exercise, numerous meetings were held with specialist groups
nationally and regional gatherings all over the country. An innovative feature
was the setting up of K˛rner Klubs at four regional centres which met quarterly
to be kept abreast of developments and to feed back their views and comments.
In view of subsequent events it is interesting that the only signi¢cant criti-
cism came from computer professionals who objected not to the products but
to the lack of one of their systems-based methodologies to support them. Great
care was taken to ensure that the ¢nal reports showed signs of having been
changed as a result of the consultation process.
Given this open process in which the NHS was totally involved, when pub-
lished, the reports contained no surprises and were in tune with the needs of
health authorities and their management teams. Indeed, everything was done
to make implementation feel a natural extension of the development work.
It became clear at the outset that recommendations about new data would
not of themselves ensure the e¡ective provision of information for health service
management. The steering group therefore embarked on a parallel programme
to improve the environment in which data were collected and information pro-
duced. A working group was set up to develop standards for maintaining the
con¢dentiality of patient and employee data. Numerous workshops of NHS
experts were held to develop best practice for medical records, district informa-
tion services, data standards and library services. Training materials were
developed using, for the ¢rst time in the NHS, videos and computer-based inter-
active techniques.
Despite not having formal responsibility for computing issues, the steering
group supported two important IT clinical developments. The computer-based
accident and emergency records project was based in Leeds and piloted at three
sites. The maternity project involved funding seven sites to develop software, all
using the minimum data set.
One area that the steering group decided not to address directly was the
development and speci¢cation of the information that should be presented to
health authorities. It was felt very strongly that initiatives of this kind would
depress the creativity of NHS managers. However, the group did commission
8 Vision and value in health information

Christopher Day of the Health Services Management Centre at Birmingham

University to write a book to help members of authorities to identify their infor-
mation needs.

The publications
The steering group’s publications are shown below (see opposite). There were
seven reports to the Secretary of State. Two working group reports were
published directly by the steering group. Seven occasional papers on information
issues were published by the King’s Fund. As a tribute to Mrs Körner, a
collection of essays by key participants in the work was published in 1985.

The outcome
As the minimum data set reports came out from 1982 onwards and the moment
for pressing the implementation button arrived, there was a crucial change in
ministers. Out went devolver Patrick Jenkin and in came centraliser Norman
Fowler.
Although ministers praised the work, there was no way they felt they could
entrust the crucial task of implementation to the steering group. Mrs K˛rner,
knowing that key civil servants did not share her vision and that they were very
concerned about their loss of central control and power, insisted that the new
ministers honoured the commitments of their predecessors. With an impasse
of this nature there is only one outcome. Mrs K˛rner resigned from the steer-
ing group and the computer technocrats in the DHSS became responsible for
implementation.
Without Mrs K˛rner the steering group limped on for another six months,
completing the development work before being put out of its misery in 1985.
Despite taking every conceivable precaution to ensure that the vision could be
sustained she was beaten by the political process. Within a couple of years the
NHS information culture had reverted. The needs of the DHSS were again para-
mount. The central returns, now ironically called K˛rner returns, multiplied
and assumed ever-increasing importance.

The legacy
In the introduction to Walk Don’t Run, Robert Maxwell, Director of the King’s
Fund, encapsulated the hopes of many senior NHS personnel when he wrote:
 Edith Körner: visionary, NHS reformer and friend 9

Steering group publications

Reports to the Secretary of State
. First Report to the Secretary of State: a report on the collection and use of
information about hospital clinical activity, 1982
. Collection and Use of Information about Transport Services, 1984
. Collection and Use of Information about Health Services Manpower, 1984
. Collection and Use of Information about Activity in Hospitals and the
Community, 1984
. Collection and Use of Information about Services For and in the Com-
munity, 1984
. Collection and Use of Information about Health Services Finance, 1984
. Collection and Use of Information about Maternity Services, 1985

Reports published by the steering group

. Con¢dentiality working group: The Protection and Maintenance of
Confidentiality of Patient and Employee Data, 1984
. Dental working group: The Collection and Use of Information about
Dental Services, 1985

Occasional papers published by the King’s Fund

. Converting Data into Information, 1982
. Introducing IT in the District Office, 1983
. Developing a District IT Policy, 1983
. Piloting Körner, 1984
. Making Data Credible, 1984
. Enabling Clinical Work, 1985
. Providing a District Library Service, 1985

Books published by the King’s Fund

. Walk Don’t Run, edited by Alastair Mason and Victor Morrison, 1985
. From Facts to Figures, by Christopher Day, 1985

Perhaps I may make two ¢nal comments that are not directly connected
with the task of implementing the Steering Group’s ¢ndings. The ¢rst is
a plea that the energy released by the device of a genuine partnership
between the NHS and the DHSS will be remembered. It points to the advan-
tage of challenging people at all levels in the service to take an active rather
than a passive role in problem resolution. It also underlines that NHS
people and DHSS people can be far more e¡ective working closely together
rather than at arm’s length.
10 Vision and value in health information

The second comment concerns the need to see the K˛rner reports as an
important stage in the continuing evolution of health services information,
not as an end point. It would show a sad unawareness of Mrs K˛rner’s
quickness and breadth of mind to allow NHS information systems to set in
concrete, rather than continue to develop.

Sadly, the challenge was not forthcoming again and the concrete set. If any
contemporary NHS manager thinks of K˛rner it is in the context of the ever
more burdensome central returns and the use of the ¢nished consultant epi-
sode as the currency of the internal market. Both perceptions are nothing to do
with the work of the steering group but re£ect the way that civil servants have
used the K˛rner name to maintain their centrally determined approach to
national information systems.
Mrs K˛rner had a powerful but simple vision that was ably communicated to,
and accepted by, the NHS but was not palatable to the DHSS. A series of techni-
cally competent products about data content and the information environment
spelt out the necessary detail to ensure the vision could be implemented.
All this was for naught when, in one of the periodic political swings from
decentralisation to centralisation, ministers and their advisers decided that she
could not be trusted to implement her proposals in a way that would be accep-
table to the DHSS. Instead, they handed over this crucial task to her greatest
critics. The primacy of the user was replaced by dominance of the computer
technocrat, and devolved management again dominated by central demands.
Unfortunately, in that process the name of the visionary driving force has
been tarnished through association with an approach which breached her own
clear principles.
NHS information services, and consequently all attempts at NHS reform,
have su¡ered ever since. Moreover, the enduring lessons on e¡ective manage-
ment of innovation in information systems have continued to go unheeded.
CHAPTER 2

Learning from history

Jeannette Muphy

Introduction
It is now more than 20 years since the Department of Health and Social Security
(DHSS) set up the NHS/DHSS Steering Group on Health Services Informa-
tion, chaired by Edith K˛rner. This initiative was the ¢rst comprehensive and
detailed review of the statistics available for health service management since
the health service was founded in 1948.
Is there any justi¢cation now for revisiting and reassessing the achievements
of Mrs K˛rner and her colleagues? A quick glance through the indexes of several
contemporary textbooks of health informatics suggests that the current genera-
tion of health information students and practitioners is very unlikely to be
exposed to the goals and methods of the steering group. At the same time, an
older generation of clinicians and managers may feel that K˛rner’s work is quite
irrelevant to the challenges they face in the wake of Information for Health.1 For
some, the steering group may seem an anachronism, a throwback to the days
before the internet, clinical codes, electronic patient records and trust intranets.
If we subscribe to the view that the solution to current information problems
in the NHS lies with advances in communication and information technologies
then Edith K˛rner seems to have little relevance to the challenges facing the
health service today.
The goal of this chapter is to challenge this perception by demonstrating that
Mrs K˛rner’s insight and initiative still have relevance for the NHS and to all
those involved with the modernisation agenda. The reason for her continuing
relevance lies in the fact that she identi¢ed a set of methodological and episte-
mological problems which are as relevant today as they were 20 years ago.
More to the point, some of her recommendations that were not implemented
are worth revisiting.

K˛rner’s vision
I remember how astonished I was when I ¢rst encountered the six reports pro-
duced by the steering group chaired by Edith K˛rner from 1980 to 1984. It was
12 Vision and value in health information

in the early 1990s ( just before the launch of the ¢rst national health informa-
tion strategy) and I was designing a health informatics course for health science
students. When I started to read about Mrs K˛rner’s work, the ¢rst thing that
struck me was the scope of her project and how much she and her team had
achieved in such a brief time. In the space of four years, the steering group
looked at the collection and use of information about:

. hospital facilities and diagnostic services

. patient transport services
. health services manpower
. paramedical services and miscellaneous
. community services
. health services ¢nance.

I was intrigued to know more about the chair ^ who was she; how did she come
to take on the task of reforming the way in which information was captured and
used; and what methods did she use to carry out this herculean task? But above
all, I wanted to know what motivated her to devote ¢ve years of her life to this
project. What was her vision? The ¢rst surprise when I started to investigate
was the fact that she did not come from either a ‘techie’ or an information
science background. Nor was she a clinician. Uniquely, Mrs K˛rner brought to
the job her managerial experience, a capacity for hard work and a commitment
to an open, participatory approach to the task of rethinking the information
needs of the NHS. As I read about her and the work of the steering group, I was
struck by the clarity of their vision and the revolutionary implications of what
they were seeking to achieve.

What was Edith K˛rner’s agenda?

One quote I came across seemed to encapsulate the K˛rner project:

the majority have preferred to denigrate the statistics rather than improve
them and face the music2

Edith K˛rner was out of step with the majority; she was committed to ¢nding a
way to improve the way that data were collected and statistics generated. But
the improvements she was seeking extended far beyond better data quality and
information systems. On the surface, it may seem that the focus of the steering
group was on number-crunching: statistics, data de¢nitions and minimum data
sets. But in fact, from reading the reports and other documents produced by the
steering group and the working groups, it is clear that the protocols for data
 Learning from history 13

collection and the statistics were just a means to an end. The real goal was
to improve the quality of healthcare. Speaking at a conference in 1983, Mrs
K˛rner said that she saw the working party as the means of reaching the true
objectives of the NHS, namely better health for more people.3
My reading of Edith K˛rner’s approach to reforming the system of data collec-
tion suggests that she was aware of trends in the commercial sector where
chief executives were waking up to the value of information and the need to
have a corporate approach to information. Under her chairmanship, the steer-
ing group adopted this concept:

Edith K˛rner’s review of NHS information marked a signi¢cant turning

point in central returns as well as other areas of information around non-
aggregated data. For the ¢rst time in an overt way the starting point was to
talk about the business of the NHS and then construct the data needed
to support the business functions. 4

The steering group saw its project as being at the very heart of the changes that
were taking place in the health service. It could see that political and economic
pressures would bring new demands for accountability, and at the same time
the introduction of new models of management was imminent. What needed
to be done was ‘to align the collection, processing and dissemination of informa-
tion in the NHS to clinical and economic realities of the present and the future’.5
In this emerging climate, managers and clinicians needed better-quality data
to allow them to determine:

. the population they serve

. whether the population receives the services it needs
. how those services are provided
. whether services are as good as one would like them to be.6

What strikes me as signi¢cant about Mrs K˛rner’s vision of the role of health
information is that it was grounded in a shrewd reading of where the world
was moving and how this would impinge on public services in general and
the health services in particular. But her vision was wedded to an unshake-
able commitment to the core values of the NHS. Although the K˛rner reports are
often characterised as being pragmatic, this does not mean they were devoid
of ideals.
The challenge facing Mrs K˛rner was to convince a wide range of stake-
holders that changes were needed, and to give clear guidance on how things
needed to change. To appreciate the scale of this challenge we need to consider
what the world was like in 1980 when the Secretary of State for Social Services
set up the NHS/DHSS Steering Group on Health Services Information.
14 Vision and value in health information

The state of health information systems in 1980

. NHS information systems had evolved in a haphazard manner over 30 years.
New information systems were created in an ad hoc manner in response to
perceived information requirements.
. There was confusion over the ownership and purpose of existing information
systems.
. Di¡erent people collected rather similar information for di¡erent masters.
. Most systems of data collection were centrally imposed.
. Consequently, there were many separate systems built to provide data for the
DHSS that were not used locally
. Sta¡ on the ground were fearful of interpreting the data they collected.
. No single person or department had responsibility for making sense of data
£ows.
. An absence of technology meant that manual systems were used for enter-
ing data. Data entry was slow, time-consuming, labour-intensive and prone
to errors.
. In regional and national centres, there were computers but these were large
cumbersome, mainframe machines.7

Immediate drivers for change

In the decade leading up to K˛rner, critics of the way in which health informa-
tion was gathered and used became more vocal. They pointed out that in the
absence of a national information system which makes it possible to compare
performance, we have not so much an NHS as a collection of local health ser-
vices. The decision to commission a comprehensive review of health informa-
tion services in 1980 was a response to four separate reports in the 1970s.
In 1972, the report on management arrangements noted that ‘existing infor-
mation is sometimes unreliable, of doubtful relevance and out of date, and there
are gaps in what is available’.8
In 1976, the three chairman’s reports noted that ‘much information [was]
demanded quite unnecessarily and . . . that the Department and Service should
jointly embark upon (a) study based upon the simple question: ‘‘Is what you are
collecting of value; and who actually uses it?’’ ’.9
Also in 1976, the resource allocation working party identi¢ed ‘a clear and
pressing need for improvement in the data routinely collected’.10
Lastly, in 1979 the Merrison Commission provided a very damning appraisal
of NHS information: ‘The information available to assist decision makers in the
NHS leaves much to be desired. Relevant information may not be available at
all, or in the wrong form. Information that is produced is often too late to assist
decisions or may be of dubious accuracy’.11
 Learning from history 15

To summarise, the charges were: that there was a lack of clarity about what
information was needed; that data quality was suspect; that there were gaps in
provision; and that there were problems with accessibility, timeliness and pre-
sentation. Clearly, there was urgent need for reform.
The terms of reference for the steering group were very broad:
. to agree, implement and keep under review principles and procedures to
guide future development of health service information systems
. to identify and resolve health service information issues requiring a co-
ordinated approach
. to review existing health service information systems
. to consider proposals for changes to, or development in, health service infor-
mation systems arising elsewhere and, if acceptable, to assess priorities for
their development and implementation.
To appreciate what Edith K˛rner achieved, given this broad brief and the tight
timescale for delivering the reports, it is useful to look at the process by which the
steering group collected its evidence, as well as some of its recommendations.

Creating the evidence base for the

proposed reforms
Although the members of the steering group may have had their private views
as to the best way of overhauling NHS information services, the approach they

Steering group working methods

1 Find out about the information requirements of a district health author-

ity (DHA) and its management team.
2 Investigate any additional needs ^ of regional health authorities
(RHAs) and central government departments.
3 Develop de¢nitions and classi¢cations for the data items that are to be
collected to satisfy identi¢ed requirements.
4 Field-test the recommendations in at least four health districts.
5 Consult about the recommendations, both formally (for example, cir-
culation of working papers and interim versions of recommendations)
and informally (through a series of regional seminars).
6 Finalise recommendations in the light of ¢eld-testing and consulta-
tions; point out the likely resource consequences and a feasible time-
table for implementation.
16 Vision and value in health information

adopted to their work was very pragmatic. Reading the reports and the sup-
plementary material, it is very apparent that members of this group were not
theoreticians or academics, but people who were very connected to the realities
of the NHS. Furthermore, they were aware of the need to test their assumptions,
to ¢nd out if their proposed reforms were feasible (what we would nowadays
think of as ‘rapid prototyping’).
Their working method was to build a model (such as their interim reports)
and then test in the reality of the workplace whether the ideas were feasible.
Four NHS health districts volunteered to act as test-beds for this work. From
the report of their experiences,12 the working methods of the steering group
can be summarised (see box).
Whatever one might think of what the steering group achieved, it is very hard to
fault its method. If you are trying to make sense of chaos, and to engage all the
stakeholders, the K˛rner approach seems an eminently sensible way forward.

What did Mrs K˛rner achieve? How much of

her vision survives?
The following are the nine elements that make up Edith K˛rner’s vision of infor-
mation management in the NHS:

1 Data would be captured as close to the patient as possible.

2 Clinicians would be more involved in data capture and have a greater sense
of ownership of the data.
3 Only relevant, useful data would be recorded.
4 Data would be accurate and timely.
5 It would be easy to aggregate data and make comparisons.
6 Both managers and clinicians would trust the data.
7 Patients could be con¢dent that their personal information was secure and
protected.
8 All sta¡ would have ongoing education and training to ensure they were
able to both collect and use data.
9 The minimum data set would be a starting point for districts, which would
add additional data to meet their needs.

Has this vision stood the test of time? I think that Mrs K˛rner’s vision has proved
to be remarkably robust. Apart from queries about the scope of the minimum
data set, most of us today could still sign up to her conception of what a
well-organised information system would be like. The principles she enunciated
20 years ago mirror the rules and exhortations found in most contempor-
ary books on the design of information systems. Her vision encompasses good
 Learning from history 17

practice with respect to data quality and information capture, and also demon-
strates a ¢rm grasp of socio-organisational issues. The one element which you
might say is absent from this sketch of Mrs K˛rner’s vision is a reference to tech-
nology. This omission is quite deliberate because my reading of her suggests
that she ¢rmly believed that technology could not deliver bene¢ts unless there
is an integrated concept of the objectives of the organisation. So technology is a
way of delivering the vision but it is not part of the vision.
To achieve the vision, the steering group argued that the following actions
needed to be taken:

. Create an information environment in which all who work in the NHS care
about data and have the education, training and support needed to work
with data.
. Establish the credibility of data ^ ensure it is trusted, easy to access, easy to
understand, and, most importantly, easy to turn into information.
. Foster a culture in the NHS in which clinicians and managers use informa-
tion to inform decision, to plan for the future, to monitor performance and
outcomes.

In a sense these are all actions that had to be part and parcel of the implementa-
tion phase. The thinking behind all this is that if these conditions are ful¢lled,
patients will bene¢t. Edith K˛rner’s vision anticipated a rational NHS, where
decisions were based on information, intelligence and evidence.

On what basis should we judge the

K˛rner reports?
When trying to ¢ll out a scorecard on the K˛rner report we need to decide by
whose standards we should appraise the work. The analysis that follows was
derived from two sets of criteria: ¢rst, a checklist of logical measures of success;
second, the objectives Mrs K˛rner set for herself and the steering group. The
verdict of her critics will be considered towards the end of this chapter.

. Did the steering group ¢nish the task it set out to do in the designated time? Here,
the answer is ‘yes’. The steering group wound up its work in 1984 with pub-
lication of its sixth report.
. Did the steering group use an explicit method and did it consult widely? Full
marks here. Its method is clearly spelled out and the consultative process was
extensive.
. Were the recommendations of the steering group accepted? Yes ^ in April 1984
the government decided to back the national implementation of the K˛rner
18 Vision and value in health information

recommendations and issued HC84(10), which set out the implementation

timetable.
. Were the recommendations of the steering group implemented? Here the answer
is both ‘yes’ and ‘no’. Although the reports were accepted, and a deadline
for implementation was set (1987), not all the recommendations were
implemented.

Writing from the perspective of an IT professional, in 1986 Windsor13 docu-

mented the missed deadlines and the fact that implementation was only partial.
He cites three reasons for this:

. unreal expectations about IT

. an underestimation of the work involved
. insu⁄cient sta¡ training and involvement.

Leaving Windsor’s caveats to one side, by and large on the basis of the four
questions asked earlier, the K˛rner project should be judged to have been a suc-
cess. But, as the steering group clearly recognised, this would be misleading:

the success or failure of our work will be judged not by the quality of our
reports or by the successful implementation of our recommendations about
data and systems. The key criterion by which our performance must be assessed
is the extent to which the information derived from the new data sets is used
to make decisions about the allocation, planning and review of resources [author’s
italics].14

In other words, Edith K˛rner set very stringent criteria by which to judge the
success of her work. Did the K˛rner reports break the vicious cycle of underuse
of information and poor quality of data collected? Here I think the answer is
‘no’. To back up this assertion, there is evidence to suggest that in the post-
K˛rner environment, the availability and use of health information did not
show marked improvement.
In 1988, the Social Services Committee had this to say about the quality and
usefulness of health information:

The last major weakness of the National Health Service is that it is not pos-
sible to tell whether or not it works. There are no outcome measures to
speak of other than that of crude numbers of patients treated. There is
little monitoring on behalf of the public. As a result, the correct level of
funding for the NHS cannot be determined.15

Commenting about NHS information in 1992 (¢ve years after the K˛rner
reports were meant to have been implemented), Cyril Chantler noted that
 Learning from history 19

‘To outside observers the lack of relevant information for management in the
NHS is surprising.’16
Barbara Young, writing as President of the Institute of Health Services
Management, in the Foreword to Information for Action, also suggested that,
despite the K˛rner reforms, managers remained disengaged from the informa-
tion agenda:

As a result of the K˛rner initiative, and the widespread introduction of

information technology, Health Service managers now have access to
more information than ever before. And yet many managers believe that
they are not yet making the most of the information that is available.17

Further evidence that Mrs K˛rner failed to bring about changes in information
practices can be inferred from the list of questions (see box) the British Medical
Association (BMA), in the early 1990s, said needed to be asked about informa-
tion. These questions show considerable overlap with those she posed, suggest-
ing again that the reports had not had the desired impact.

Questions the BMA thought needed to be answered about health

information

. How much information is needed?

. What information is required?
. How may it be collected most cost-e¡ectively?
. How should it be made available and to whom?
. Where do priorities lie in funding health, administrative and manage-
ment information systems?

Explaining why the NHS failed to take on board

Edith K˛rner’s vision
In trying to disentangle the various threads of what was been attempted, what
was achieved and what did not succeeded, it is useful to revisit Mrs K˛rner’s
thinking at the time of the publication of the First Report. What we ¢nd is that
from the outset Edith K˛rner was aware that even if the recommendations of the
steering group about what data to be collected routinely were implemented, this
would not ensure the e¡ective provision of information for health service man-
agement (First Report, p 9).14 Quite rightly Mrs K˛rner maintained that the
success or failure of her work rested on bringing about cultural change in
the NHS so that managers were willing to use information:
20 Vision and value in health information

there is still, deep in the NHS management culture, a strong resistance to

use even the simplest qualitative techniques.18

Managers give numerous excuses to justify their failure to be informed, one of the
most common being that data cannot be trusted. Although Mrs K˛rner recog-
nised that there was some truth to this claim, she felt it also provided an excuse
for inaction. To change this situation, a whole range of measures needed to be
put in place to develop an information culture, for example education and train-
ing, skills audit, development of data standards, development of best admin-
istrative practice for safeguarding patient data and promotion of the use of IT:

We are also concerned with assisting in the creation of an environment

which will allow and encourage the e⁄cient collection, collation, proces-
sing and transmission of data.18

Looking back now more than 20 years since the steering group began its work,
we must admit that the necessary cultural and organisational changes have
still not occurred. But do we place this failure at Edith K˛rner’s doorstep?
I think it would be most unfair to blame the steering group for having failed to
promote better use of information in the NHS. Nor does the fact that Edith K˛rner
did not succeed in achieving what she had set out to do mean the vision was
£awed. To understand the fact that we are still struggling to cultivate a culture
which values and uses information we need to look at what happened between
the time the reports were commissioned and the time they were delivered. With
hindsight, we can identify six interconnected developments which undermined
Mrs K˛rner’s agenda.
First, there were major political and economic changes; the whole landscape
of the public sector changed signi¢cantly from the time Mrs K˛rner began her
work. The phrase that is sometimes used to characterise this change is ‘the end
of the era of optimism’. The medical sociologist, Mary Ann Elston, describes the
emergence of a new climate in which questioning the e⁄ciency and e¡ectiveness
of medicine’s use of resources became a more legitimate activity for politicians.
‘The swift introduction, following the Gri⁄ths Report in 1983, of general man-
agers charged with responsibility for the e⁄cient use of resources,’ Elston
suggests, ‘apparently cut a swathe across established lines of professional
responsibility and clinical freedom.’19 ‘By 1985, it was possible to identify a
series of (government) moves which arguably at least are beginning to amount
to a confrontation with the medical profession.’19
The reaction of the professions is encapsulated in Trevor Clay’s pronounce-
ment, ‘The Gri⁄ths Inquiry . . . signalled the demise of professional power in
the NHS. The doctors were deemed important only in so far as they could be
nudged into management’ (as quoted by Elston pp. 68^69).19
 Learning from history 21

Second, in response to this demise of consensus politics, clinicians were less

receptive to some of the thinking behind Edith K˛rner. Feeling under pressure
as a result of the changes taking place in the NHS, they were mistrustful of an
initiative which required them to be more open and transparent about their
work and their use of resources. IT, along with information systems, were
coming to be seen as a tool for being able to control doctors by providing data
on cost and outcome of medical procedures.
Third, Mrs K˛rner was adamant that her proposals for improving information
quality rested on assumptions about the competence and political will of author-
ities and their o⁄cers.20 By the late 1980s, health service managers were also
experiencing new pressures and were inclined to be wary about something
as potentially subversive as well-presented information that allows valid com-
parisons. Managers, Yates argues, are frightened to explore some of the issues
raised by statistical data. Information can be used both to solve problems and to
identify problems.21 Without the backing and goodwill of doctors and man-
agers, the success of the K˛rner project was in jeopardy from the outset.
Fourth, the changes recommended by the K˛rner steering group were far more
costly to implement than had been imagined. To capture the minimum data sets
required investment in systems and training. To set up and manage information
services new monies needed to be identi¢ed. And then, of course, there was
the looming cost of computerisation. Many observers have pointed out that the
necessary investment in both people and technology was not forthcoming.
Fifth, although computers were supposed to be part of the solution, in some
ways they were part of the problem. The arrival of more computers, with the
launch of ¢rst national Information Strategy in 1992, led to a shift in the dis-
course from information to technology. Willcocks and Marks maintain that
given the lack of sta¡ resources which made it di⁄cult to meet the implementa-
tion deadlines, computerisation came to be seen as a prerequisite for achieving
the K˛rner implementation dates.

the NHS has been pushed from information into Information Technology
to the extent the two have become synonymous, especially in general man-
agement’s eyes.22

Edith K˛rner never thought the computer would be the panacea to the in-
formation problems of the NHS information. In her preface to the workshop
proceedings, Developing a District IT Policy, she put her ¢nger on the key
problem ^ technology is seen as the holy grail, but for technology to deliver ben-
e¢ts, managers need to be understand what information systems can and
cannot do. She perceived that we are in a Heraclidean landscape, with the river
of technology changing from moment to moment. To survive in this situation,
we need to be nimble and adaptable, ‘only a process in which teaching learning
22 Vision and value in health information

and doing interact constantly with each other, will allow us to acquire and
maintain the skills which we shall need in the years to come’.23
Mrs K˛rner was also percipient in her observation that the problems of apply-
ing IT solutions in a healthcare environment are not just technical, ‘they
extend to structural concerns of centralisation and decentralisation and to the
respective roles of technical experts vis-a'-vis lay computer users’. To rephrase
this in modern jargon, Edith K˛rner was aware that when developing a coher-
ent information policy and an IT system, some process re-engineering may well
need to take place.
Sixth, the process of culture change was much more di⁄cult to achieve than
anticipated. Despite K˛rner’s repeated messages about the need to invest in
people, a coherent education, training and professional development strategy
was not set in place. Even today we are still wrestling with the problem of how
best to prepare clinicians, managers and administrative and clerical sta¡ to
value and use information. To paraphrase the American sociologist Daniel Bell,
we all live in an information society; we are all information workers, but few of us
are prepared to play this role.
In the ¢nal chapter of Walk Don’t Run Yates re£ects on the costs of using infor-
mation.21 His appraisal goes beyond the obvious ¢nancial costs. One of the costs
Yates identi¢es is a willingness to change working practices on the basis of
information. This, in turn, means being able to take on the vested interests
that prefer to maintain the status quo. Yates hints that Edith herself paid the
price of ¢ghting to establish an ‘evidence-based health care system’. He suggests
that Mrs K˛rner was not surprised by some of the reactions to her work, ‘she
recognised from the outset that spring cleaning the NHS information system
would be a mammoth task, and probably completely thankless.’
Throughout my research into Edith K˛rner’s work, and her legacy, I have con-
tinued to puzzle over what sort of person she was. My abiding impression is that
she was often misunderstood and criticised for what she did not do. Her critics
seemed reluctant to give her credit for what she had attempted. I empathise
with her long struggle to combat the view that a commitment to rigorous
methods is somehow incompatible with humanistic values. As a teacher, I have
very vivid memories of my struggle to convince nurses and social workers of the
need to be numerate and computer-literate. They justi¢ed their resistance on the
grounds they were ‘people-focused’ and they did not see why I wanted them to
deal with numbers and computers. When addressing the Annual Study Confer-
ence of Health Visitors in 1983, Mrs K˛rner was taken to task by her audience
for thinking it was possible to quantify their work. Health visitors expressed
doubts as to whether the breadth and depth of what they do on a home visit
could be portrayed by a statistic. Her reply was both poignant and revealing,
‘Statistics’, she said, ‘are people with the tears wiped o¡.’3
Edith K˛rner was well aware that her work was far from neutral. A study of
information-gathering in the NHS was potentially subversive for two reasons.
 Learning from history 23

First, because it exposed poor organisational design and a lack of rigorous think-
ing about the way clinical work was scheduled and delivered. Second, the crea-
tion of credible data would make it more di⁄cult for clinicians and managers to
conceal poor practice. Her awareness of the political implications of her work is
nicely demonstrated in the choice of quotes used at the start of the six pamphlets
illuminating di¡erent phases of the work of the steering group. The quotations
range from Machiavelli: The Prince, to John Galbraith: The Age of Uncertainty and
John Stewart Mill: On Liberty.
There is little doubt that even today, after two national information strategies,
there is still a gap between data collection and the intelligent use of information.
Does this means that Edith K˛rner failed? I feel this judgement would be too
harsh. The fact that she did not create an environment where all participants
were able to understand their role in the information agenda can be interpreted
in various ways. As suggested previously, the challenges of organisational
development and providing adequate education and training for all who collect
and use information were greater than anticipated. To achieve this demanded a
level of investment and sustained leadership that was not forthcoming. Further-
more, although there were various drivers for change, the really big seismic
forces had not yet hit the NHS in the mid-1980s. Audit, risk management and
clinical governance were still on the distant horizon.

Edith K˛rner: a beginning not an end

Many of Mrs K˛rner’s critics have accused her of bequeathing the NHS an in£ex-
ible legacy, the K˛rner returns. Another frequently heard charge is that her
minimum data sets turned into normative data sets that sti£ed innovation and
development. The implication of these comments is that she had a very narrow
concept of the information requirements of the NHS. But if we listen to her own
words, it is very clear that Mrs K˛rner saw her work as a beginning, not an end.
Writing for the Health Service Journal in 1987, she made clear that information
systems need to be kept under review and changed as the world changes:

The health service ¢nds itself today at a dangerous juncture in information

development. Most of the data speci¢ed as the national minimum are, or
soon will be collected.24

At the stage she was writing (October 1987), the steering group reports had
been accepted and were in the process of being implemented. But far from feel-
ing that the battle had been won, Mrs K˛rner was all too aware of the fragility of
what had been achieved ‘because too many people mistake the milestone for the
¢nishing post’.
24 Vision and value in health information

The danger stemmed from the fact that people might not appreciate that what
was needed was not a once and forever reform of information systems but a
state of permanent revolution. In her words, ‘It is essential that everyone con-
cerned be made to realise and accept that we are moving towards not an event
but a process’. Information management is not ‘a task that can be put to bed
when completed, or a routine which, once established, can be left to run by its
own momentum. It is a demanding and iterative task’.24 (Those involved in
implementing Information for Health1 and Building the Information Core25 might
consider having these words emblazoned on their doors.)
Edith K˛rner was particularly concerned to warn the NHS that in relation to
information there were no simple cures, and that it would be a terrible mistake
for her work to be rei¢ed ^ for the minimum data set to come to be seen as
immutable and for thinking to be frozen as a result of her reports. As far as
Mrs Ko«rner was concerned the NHS information edi¢ce was not complete and
she was utterly opposed to the prospect that information strategies might be
delegated to experts.
K˛rner saw information as a way of mapping the real world. As the world
changes, so the information needed to describe it changes. Edith echoed the
Greek philosopher Heraclides when she said, ‘Thus, while the information task
is iterative, it is not like painting the Forth Bridge: both the bridge and the
method of painting change even as we are painting it.’24
In 1987 Mrs K˛rner recognised that already the data items proposed by the
steering group were ‘feeling the tooth of time’. ‘In the ¢ve years which have
elapsed since publication of the ¢rst report, the NHS world has changed su⁄-
ciently for some of our data de¢nitions and data sets to need revision.’24

Appraisals of Edith K˛rner: diverse views

of the legacy
Appraisals of Edith K˛rner’s work fall into three distinct categories: those that
see her as helping to liberate health information services; those that deplore the
K˛rner legacy; and those that view her as a mixed blessing.
Let us start by hearing from those who applaud the K˛rner legacy:

On re£ection, the K˛rner statistical revolution probably remains the single

most important event in the ¢rst 50 years of NHS information and will
in£uence thinking at the start of the new millennium. It raised the pro¢le
of information but perhaps more importantly, for the ¢rst time took a cor-
porate rather than piecemeal look at the needs of the service, ending the
separation of data for local and central requirements.4
 Learning from history 25

The health service after K˛rner . . . is at least a more rational system. The
management and use of information within the system is better under-
stood, de¢ned and standardised, which must lead, particularly with the
national introduction of health care computers, to better patient care.26

The Steering Group . . . provided the basic building blocks of information for
managing the health services.27

Mrs K˛rner’s great contribution to the development of IT was to discern

as a prime objective the production of credible information for managers
and clinicians; and it is this objective that determines the technology
required.28

The K˛rner reports may also be seen as a natural ¢rst step in preparing for
future managerial and organisational changes in the NHS. . . . both the
government and K˛rner had such policy directions in mind when develop-
ing NHS information requirements in the early 1980s.22

Many of Edith K˛rner’s critics focus on the unintended consequences of the

steering group reports and recommendations, and the problems not tackled:

K˛rner’s system is still with us, and forms the basis of the national perfor-
mance indicators. These have been widely criticised ^ perhaps condemned
is a better word ^ principally in the basis that they re£ect input and process
measures, and largely omit outputs and outcomes. Moreover, though in-
tended for local use they re£ected a centralised view of the world, and so
were of limited use to those who collected them. The practical consequence
for local sites was that most of their data collection e¡orts went on data of
little local relevance and most felt they could not commit substantial extra
resources to the collection of other, more relevant data.29

With hindsight the work by Edith K˛rner and her colleagues can be seen
to have underpinned subsequent moves to the general management con-
cept and then the internal market. It is unfortunate that much of their
initial logic and work was seen by the DHSS as making a stick to beat
the NHS with: the resultant stick was called the ‘Department of Health Cen-
tral Returns’.30

Knox and colleagues spelled out the de¢ciencies of the K˛rner reports from the
perspective of community medicine. These authors took issue with the terms of
reference to which the steering group worked:

The data sets delineated by the Steering Group were directed more towards
managing the institutions, the ¢nances and the resources of the NHS and less
26 Vision and value in health information

toward evaluating, monitoring, enhancing and managing the function and

e¡ectiveness of health- and health-care programmes.31

Some commentators, whilst critical of some of the consequences (largely unin-

tended) of the K˛rner reports, do give her credit for her vision. Gowing, for
example, sees both the negative and positive side to the K˛rner legacy:

These aims were admirable and, if achieved, were to support the develop-
ment of general management in the NHS. However, the real outcome was
that ‘implementing K˛rner’ became an end point in itself. Whilst K˛rner
returns may have been helpful to the Department of Health, their useful-
ness within health authorities, in hospitals and for doctors has been extre-
mely questionable.32

K˛rner did, however, ensure that information and IT were put on the man-
agement agenda . . . The need for systems in turn led to resources being
assigned to this area and particularly to the establishment of information
and IT departments.32

Similarly, Windsor, at the start of his critique of K˛rner, pays tribute to her
breadth of vision:

There is no doubt that K˛rner is the most important event ever to hit man-
agement information in the NHS. Nor is there any doubt that it is, in scale,
breadth and quality, a remarkable achievement. . . . K˛rner broke with the
pattern of nearly forty years of apathy about information systems in
the NHS. For the ¢rst time the corporate and not the piecemeal needs of the
service have been structured: one of the reports’ main achievements must
be the end of the nonsense of separate data for local and central require-
ments. . . . Its breath of coverage is unrivalled.13

What is also interesting about Windsor’s re£ections on Mrs K˛rner is that he

feels that, far from being disloyal in seeking to identify the £aws in the K˛rner
recommendations, constructive criticism would help to ensure the success of
the enterprise.

Un¢nished business
It would be beyond the scope of the current chapter to respond to each of these
criticisms. My general view is that many of the critics focus on what Edith
 Learning from history 27

K˛rner did not do and ignore the cogent reasons she gives for limiting her inves-
tigation to the needs of NHS managers and her rationale for not addressing
epidemiological concerns:

The Steering Group’s main concern is with information for health services
management. Thus we have not tackled speci¢cally the information
needed by health professionals to evaluate the results of their care; nor
that needed by individual professional bodies to review the resources to
and the professional work of their members.14

But, as the preamble went on to say, clinicians would ¢nd much in the report to
assist them when reviewing their performance and the amount of resources
devoted to their activities. As for what we now call ‘health needs assessment’,
the steering group had this to say:

In our work we have not directly considered information about the occur-
rence of disease or about the health needs of populations except in so far as
these can be inferred from data about hospital episodes and certain com-
munity health programmes; nor have we made recommendations about
data describing health status or the clinical and social outcomes of the use
of health services.14

Mrs K˛rner’s justi¢cation for excluding epidemiological data was that it lay out-
side the scope of her work. And, regarding health status and social outcomes,
ever the realist, K˛rner observed that:

Methods of measuring these important variables are not su⁄ciently devel-

oped to allow their introduction as a routine in health service information
systems. When further research has identi¢ed those indicators which are
not only sought by health service managers but are also capable of routine
collection in all health districts consideration will be given to their inclu-
sion in the minimum data set.14

Final re£ections
In any consideration of a vision for health information, the last word must be
Edith K˛rner’s. In 1984, after all the recommendations of the steering group
had been accepted by the Secretary of State, Mrs K˛rner wrote a article for
the British Medical Journal in which she issued a wake-up call to clinicians. She
warned doctors of the big battles that were raging over resources and made it
clear that they could not expect to be shielded from these. Whilst in the past
such debates took place behind closed doors between the Cabinet and the
28 Vision and value in health information

DHSS, in the years ahead this struggle for funds, she predicted, would become a
familiar problem to all who work in the NHS. Edith K˛rner’s advice was blunt:
to survive you will need evidence. Guesses, unsupported opinions and anecdo-
tal evidence will not prove e¡ective weapons in the battle for resources. This
message needs to be placed alongside Mrs K˛rner’s earlier exhortation to doc-
tors to improve the quality of data on diagnoses and operative procedures. She
pointed out that as this data is recorded by doctors in their clinical notes, it is
they who are solely responsible for the quality of this data. She was well aware
of the implications of this advice and why it might be ignored. ‘The procedures
to maintain high levels of accuracy and completeness of these data items may
impose an unwelcome discipline on doctors who record them.’18
In 1984 Edith K˛rner stood down as the chair of the steering group. She
re£ected on what she had achieved and the challenges that lay ahead. This
was her parting message to the health service:

The Steering Group has led the NHS to the water and cleansed it of sedi-
ment and impurities. Given the country’s economic and political condition,
the service will refuse to drink at its peril.5

References
1 NHS Executive (1998) Information for Health. NHS Executive, Leeds.

2 King D (1985) Mrs K˛rner and her Steering Group. In: A Mason and V Morrison (eds)
Walk Don’t Run: a collection of essays on information issues published to honour Mrs Edith
K˛rner CBE. King Edward’s Hospital Fund for London, London.

3 K˛rner E (1983) Statistics are people with the tears wiped o¡. Health Visitor. 56: 441^2.
4 Kempner V (2000) Statistical returns: past, present and future. In: D Leadbeter (ed)
Harnessing O⁄cial Statistics. Radcli¡e Medical Press, Oxford.
5 K˛rner E (1984) Improving information for the NHS. BMJ. 289: 1635^6.

6 Day C (1985) From Figures to Facts. King’s Fund (on behalf of NHS/DHSS Health Services
Information Steering Group), London.
7 Mason A and Morrison V (eds) (1985). Walk Don’t Run: a collection of essays on informa-
tion issues published to honour Mrs Edith K˛rner CBE. King Edward’s Hospital Fund for
London, London.
8 Department of Health and Social Security (1972). Management Arrangements for the
Reorganised National Health Service (‘The Grey Book’). HMSO, London.
9 Department of Health and Social Security (1976). The Regional Chairmen’s Enquiry
into the Working of the DHSS in Relation to Regional Health Authorities. Report (the three
chairmen’s report). Department of Health and Social Security, London.
 Learning from history 29

10 Resource Allocation Working Party (1976) Sharing Resources for Health in England
(RAWP Report). HMSO, London.
11 Royal Commission on the National Health Service (Chairman Sir Alec Merrison) (1979)
Report. Cmnd 7615. HMSO, London.
12 NHS/DHSS Health Services Information Steering Group (1983) Piloting K˛rner: the views
of senior administrators from the four districts who piloted the interim reports of working
groups A to E from 1981 to 1983. King’s Fund, London.
13 Windsor P (1986) Introducing K˛rner: a critical guide to the work and recommendations of the
Steering Group on Health Services Information. British Journal of Healthcare Computing
Publications, Weybridge.
14 Steering Group on Health Services Information (1982) First Report to the Secretary of
State: a report on the collection and use of information about hospital clinical activity in the
National Health Service. HMSO, London.
15 Coulter A (1991) Evaluating the outcomes of health care. In: J Gabe, M Calnan and
M Bury (eds) The Sociology of the Health Service. Routledge, London.
16 Chantler C (1992) Management and information, in the future of health care: articles
published in the British Medical Journal, BMJ, London. 304: 632^5.
17 Lattimer B and Mason A (no date) Information for Action. Institute of Health Services
Management, London.
18 NHS/DHSS Health Services Information Steering Group (1984) Making Data Credible.
King’s Fund, London.
19 Elston MA (1991) The politics of professional power. In: J Gabe, M Calnan and M Bury
(eds) The Sociology of the Health Service. Routledge, London.
20 NHS/DHSS Health Services Information Steering Group (1982) Converting Data into
Information. King’s Fund, London.
21 Yates J (1985) Using Information. In: A Mason and V Morrison (eds) Walk Don’t Run:
a collection of essays on information issues published to honour Mrs Edith K˛rner CBE. King
Edward’s Hospital Fund for London, London.
22 Willcock LP and Mark AL (1988) Information for Management? A review of progress on
information technology and general management in the UK National Health Service. Working
Paper No 92, City University Business School, London.
23 NHS/DHSS Health Services Information Steering Group (1983) Developing a District IT
Policy. King’s Fund, London.
24 K˛rner E (1987) Too important for the experts. Health Service Journal. 29 Oct: 1258^60.
25 Department of Health. Building the Information Core . Department of Health, London.
26 Strickland-Hodge B, Allan B and Livesey B (1988) Information Technology and Health
Care. Gower, Aldershot.
27 Scrivens E (1985) Policy, Power and Information Technology in the National Health Service.
(Bath Social Policy Paper No. 3) University of Bath, Bath.
30 Vision and value in health information

28 Peel V (1995) Information management and technology strategy for healthcare organi-
sations. In: R Shea¡ and V Peel (eds) Managing Health Service Information Systems:
an introduction. Open University Press, Buckingham.
29 Keen J (1994) Information policy in the National Health Service. In: J Keen (ed.) Informa-
tion Management in Health Services. Open University Press, Buckingham.

30 Nicholson L and Peel V (1995) Manpower development for NHS information systems.
In: R Shea¡ and V Peel (eds) Managing Health Service Information Systems: an introduction.
Open University Press, Buckingham.

31 Knox EG (1987) Health-Care Information. Nu⁄eld Provincial Hospitals Trust, London.

32 Gowing W (1994) Operational systems. In: J Keen (ed.) (1994) Information Management
in Health Services. Open University Press, Buckingham.
CHAPTER 3

Improving the United

Kingdom’s health system:
an adaptive model to harness
information and evidence
Don E Detmer

Introduction
These are stressful times for the healthcare sectors of all nations, and the United
Kingdom is no exception.1^6 Health system shortcomings are widely publicised,
funding remains tight, public expectations for improved performance are grow-
ing, and many people want more direct involvement in their own care. New
knowledge and technologies create an abundance of opportunities to improve
health, but inadequate di¡usion mechanisms or limited resources often hinder
their impact. Healthcare in the near future must continue to deliver acute,
episodic care to individuals (the old business) and simultaneously provide
much better chronic illness management for both individuals and populations
(a newer business). The challenge for all nations is to redesign their complex
care systems.7,8 Neither a straight top-down nor a bottom-up approach will
work; rather, change must occur throughout systems over time.
Despite a strong commitment from health professionals to improve health,
their e¡ectiveness is determined largely by the system in which they operate.
A healthcare system is only as good as the skills and ideas of its people, the pro-
cesses and infrastructure they use to accomplish their work, and the culture
that supports their e¡orts. Today each of these critical components is distressed
and only the adoption of a clear longer-range strategy will truly integrate e¡orts
and lead to a modernised service. Considerable e¡ort is already being devoted to
identifying the kinds of changes needed and how best to implement them. It is
32 Vision and value in health information

timely and appropriate to identify the critical leverage points for transformation
of the health system.
The times call for vision and leadership ^ a vision simple enough to carry in
one’s head and leadership that both motivates reluctant souls to respond to
relevant messages and allows the experimentation needed for changes to occur.
A strong vision that is e¡ectively communicated by leaders and eventually
embraced by the broader health community is not, however, su⁄cient to reach
the objective of a transformed health system for the 21st century. Leaders must
assure that organisational systems are both £exible and su⁄ciently aligned with
stated goals. They must support a culture whose values drive them toward
needed actions. Further, the vision that will shape the evolution of the health
system must be constantly re¢ned to accommodate the ever-growing base of
knowledge on what constitutes e¡ective health services. There needs to be a
permanent means of ensuring that the vision stays ahead of the system and that
the system’s development is generally consistent with the vision and available
evidence. And, a proper knowledge and information technology infrastructure
(what Canadians call an ‘infostructure’) must underpin it all.9
The four recommendations presented here create an integrated regimen for
improving performance of the United Kingdom’s health systems. The ¢rst recom-
mendation calls for a national academy of health to provide continuous visible-
thought leadership for the health system that can withstand and channel the
winds of open political discourse and media hype. The second recommendation
calls for a set of principles to guide reform e¡orts. The third recommenda-
tion develops informed leadership through a clinical scholars’ programme and
a corresponding redesign of the work setting to provide prepared clinician-
executives with the time, authority and resources needed to perform their lea-
dership roles. The fourth recommendation establishes an information and
knowledge platform as essential to improving the organisation and delivery of
care. These recommendations will complement the recent set of policy initia-
tives by the Department of Health and the other home countries’ health depart-
ments, and enhance performance of whatever system emerges from the current
reform debate in the United Kingdom.

Recommendation 1: establish a national,

independent body to guide health
policy development
This politicization of technical debates about health policy greatly impedes ^
one might say, prevents ^ an informed, calm, public debate about the future of the
health care system.
RGA Feachem
 Improving the United Kingdom’s health system 33

One of the major impediments to true transformation for the NHS is the degree to
which the NHS is exposed to political winds and expected to respond to the latest
public concern. The UK lacks a representative group of health experts to bring
su⁄ciently broad and seasoned input to the table to contemplate and create new
and more acceptably framed programmes for change that are not linked to any
particular political party. The current situation needs to be replaced by a steady,
evidence-based course of action that builds on past experiences and policies
so that there is a long-term sense of policy direction and better continuity for
policy that in£uences individuals working within the NHS.
A Royal (National) Academy of Health (R/NAH) would provide needed
authority, independence and credibility for UK health policy and NHS strategy
development. The R/NAH would be a membership organisation modelled on the
Institute of Medicine (IOM) in the USA (as described below). It would be broadly
representative of the health professions and include distinguished individuals
of related disciplines who are willing to volunteer their relevant expertise to
critical health issues facing the nation, much like the Foresight Programme or
the Academy of Medical Sciences, but with all relevant disciplines represented.
The new organisation would provide guidance to the governments on critical
health policy questions while remaining independent of immediate political
demands. The primary role of the R/NAH would be to provide unequivocal evi-
dence derived from analysis, synthesis and constructive deliberation. Reliance
on scienti¢c rigor and consensus development would provide crucial ballast
against shifting political winds and a bu¡er against an incessant media that is
interested in ‘human interest’ for hype and sales as much as constructive social
change. In addition to producing useful, evidence-based reports, the R/NAH
would also provide a mechanism for creating more mature informed leaders.
The R/NAH would respond to speci¢c policy questions (for example, what are
the implications of wait times on health status; when are wait times acceptable
and when are they detrimental; is this policy generally functional or dysfunc-
tional?), would help to de¢ne the health agenda for the UK (such as, which
of the myriad white papers and reports should form the basis for action; what
are the most important tasks for health system reform; what should the trans-
formation platform look like?) and articulate policy in terms of action steps for a
variety of stakeholders. Further, the R/NAH would play an important role in
educating the public on the complexity of health policy issues.
The IOM was established in 1970 as a non-pro¢t organisation that is a
branch of the National Academy of Sciences. Its mission is to ‘advance and dis-
seminate scienti¢c knowledge to improve human health’.10 The IOM strives to
‘obtain the most authoritative, objective, and scienti¢cally balanced answers
to di⁄cult questions of national importance’. It provides information and advice
concerning health and science policy to government, the corporate sector, the
professions and the public through its studies and reports. The federal govern-
ment requests and funds the majority of IOM studies but committees of volunteer
34 Vision and value in health information

scientists selected totally by the IOM processes conduct the actual studies. The
IOM carefully composes these committees to assure the necessary expertise,
and to mitigate complicating bias or con£ict of interest. The committee reports
undergo extensive review and evaluation for scienti¢c merit and rigorous evi-
dence by a group of relevant experts who remain anonymous until the study is
published.
The IOM also organises round tables, workshops and symposia to provide an
opportunity for public and private sector experts to ‘openly discuss contentious
issues in an environment that promotes evidence-based dialogue’. The IOM also
manages the Robert Wood Johnson Health Policy Fellowship Program,
designed to develop the capacity of outstanding mid-career health professionals
in academic and community-based settings to assume leadership roles in health
policy and management. (In time, such a programme would be a useful adjunct
to the clinical scholars’ initiatives discussed below.)
In addition, the IOM is an honori¢c membership organisation. Members are
elected on the basis of their professional achievements and serve without com-
pensation in studies and other activities on health policy issues. One-quarter
of members must be selected from professions other than those primarily con-
cerned with medicine and health; members are represented from natural, social
and behavioural sciences, law, administration and engineering. This mixture
is crucial since it creates a rich pool of experts to allow panels to be created on
a wide variety of studies. Foreign members add greatly to the quality of delib-
erations. Non-members are also invited to participate in study committees to
ensure balanced representation of views and the highest calibre of technical
expertise. The IOM works through a set of boards composed of members and
sta¡ed by the IOM. The board structure assures that work is done across a num-
ber of relevant topic domains that include children, youth and families, food and
nutrition, global health, healthcare services, health promotion and disease pre-
vention, health sciences policy, cancer policy, neuroscience and behavioural
health, and health policy programmes and fellowships.
Over its 30-year history, IOM studies have addressed a wide variety of issues
ranging from disparities in healthcare among races and ethnic groups to bio-
logical threats and terrorism, from improving palliative care for cancer to family
violence.11^14 The IOM special initiative on healthcare quality resulted in two
pivotal reports on patient safety and healthcare quality that provided a call to
action and a framework for achieving threshold improvements in safety and
quality.2,3 These reports, To Err is Human and Crossing the Quality Chasm, have
received some attention in the UK. The quality initiative is ongoing and now
has a major IT component as well to address speci¢c policy issues raised in
those reports, such as identi¢cation of priority areas for quality improvement,
outlining a national healthcare quality report, guiding development of patient
safety data standards and creating a robust national information infrastructure.
These reports and national events led the US Secretary of Health and Human
 Improving the United Kingdom’s health system 35

Services to request a study to o¡er the government advice on next steps. This
report, Fostering Rapid Advances in Health Care: learning from system demonstra-
tions, outlines a set of major statewide initiatives to create the seeds for the
future US healthcare system.15 It has received a strongly positive response
from the media in the USA.
In the area of IT and information policy alone, the IOM and the National
Research Council of the National Academy of Sciences have produced a series
of reports that have brought important health information technology issues to
the attention of the federal government. Indeed, they helped put these issues on
the national agenda. These reports include:

. The Computer-based Patient Record: an essential technology for health care16,17

. Health Data in the Information Age: use disclosure and privacy18
. Telemedicine: a guide for assessing telecommunications in health care19
. For the Record: protecting electronic health information20
. Trust in Cyberspace21
. Networking Health: prescriptions for the internet 22

Two important recent reports from the US government:

. Transforming Health Care Through Information Technology (President’s Infor-

mation Technology Advisory Committee, 2001)23
. Information for Health: a strategy for building the national health information
infrastructure (National Committee on Vital and Health Statistics, 2001)24

drew heavily from the conclusions and recommendations of the IOM and NRC
reports and the individuals who had been part of the studies.

Recommendation 2: adopt a ¢nite set of simple

rules to guide major change
The maxims of complex adaptive systems (CAS) o¡er new insights on how to
achieve profound change within a dynamic environment and are replacing
the command and control models for organisation and change management.25
The science of CAS recognises that many systems, including healthcare, are not
fully predictable. Not only can the elements of CAS change themselves, but
creativity is also a natural part of any complex system. As a result, it is di⁄cult
to predict how a CAS will behave. Yet despite these considerations, a CAS can be
orderly without central control. A few simple rules applied locally can yield
36 Vision and value in health information

complex outcomes (for example, maintenance of a huge colony of insects or

the Internet).
A fundamental lesson from CAS for the reformulation of the health system
is to provide a ‘good enough’ vision along with a set of simple rules, and then
to allow generous latitude for innovation. Without any rules, complex systems
will experience chaos rather than progress because there are too many possible
paths to follow. With a set of rules or values to guide thinking and action, multi-
ple small changes throughout complex systems can result in major improve-
ments over time through a ‘biological’ kind of evolution. Immediate change is
not the goal. Rather, one seeks to use simple rules and minimum speci¢cations
so that gradually changes evolve from the natural creativity of actors making
improvements throughout the system. In time less adaptive processes are then
dropped in an evolutionary fashion. When one is uncertain about what to do in
either building a component of a new system or approaching a patient, one can
examine the potential choices for action against the basic rules.
There is a distinct need for clarity. Participants in the system need to be able
to discern the rules easily. The blizzard of new policy initiatives and visions, and
targets and ‘reform’ directives emanating from Whitehall over the past few
years would be su⁄cient to confuse any group of leaders as well as potential
followers. Further, too many rules limit £exibility and innovation, and thwart
constructive incremental change.
What then constitutes the ‘good enough’ vision and what kind of rules are
needed to guide the redesign of the health system? The IOM has provided these
pivotal elements and more for the US healthcare sector in its report, Crossing
the Quality Chasm: a new health system for the 21st century.3 The IOM vision com-
prises two components ^ a statement of purpose and a set of six aims for the
system.
Purpose: The purpose of the healthcare system is ‘to continually reduce the
burden of illness, injury, and disability, and to improve the health and
functioning of the people’ of the nation (p. 6).

Aims: ‘. . . [C]are must be delivered by systems that are carefully and con-
sciously designed to provide care that is safe, e¡ective, patient-centered,
timely, e⁄cient, and equitable’ (p. 7).3

Health professionals and decision-makers can easily remember the purpose and
aims since they are clear and basic and they share tacit validity. Despite their
simplicity, they are comprehensive and relevant for all health systems.
Certainly, one wishes to have care that is safe. The precept of doing no harm
while seeking to help the patient is an ancient one. An ethical system of care
should not waste resources on care whose bene¢ts are unproven, hence e¡ec-
tiveness is important. Further, if one pursues e¡ectiveness wherever it may
 Improving the United Kingdom’s health system 37

lead, signi¢cant changes in care systems may well develop over time. Care,
de¢ned as health services, may possibly change one day into more e¡ective
community education policies. Patient-centred care is subtle and can be a chal-
lenge to focus upon since one can rather easily say that virtually everything is
there to some extent to focus on the patient. What is intended is to have, to the
maximal extent possible, patients at the centre of their care with respect to deci-
sions, responsibility, and style and manner of delivery. This has not been a
design feature of many of our care processes. This becomes even more impor-
tant as chronic illness becomes a more dominant component of illness patterns
(World Health Organization, 2001).26
Care should be timely. Some problems can be safely put aside, whereas good
care in other instances must be prompt care to be any care at all. As the old
adage goes, ‘a stitch in time saves nine’. A care system must have a way to
determine what can wait and what cannot wait and triage patients accord-
ingly. E⁄cient care is important since there are always insu⁄cient resources
and wastefulness is essentially care that is not given to someone else who
might bene¢t from it. Lastly, care should be equitable. Obviously, there may be
times when these criteria come into con£ict but much of the time they do not.
To achieve the six aims, healthcare organisations need to redesign their sys-
tems. According to the IOM, healthcare process redesign should be based as well
on 10 principles (see box). Health professionals need to be supported in their
e¡orts to implement these redesign rules. The health system must o¡er appro-
priate training opportunities and incentives that encourage behaviour consis-
tent with the rules. Information systems will play a critical role in putting these
rules into routine practice (see rule 4 below).

Redesign rules for healthcare processes3

1 Care is based on continuous healing relationships

2 Care is customised according to patients’ needs and values
3 The patient is the source of control
4 Knowledge is shared and information £ows freely
5 Decision-making is evidence-based
6 Safety is a system priority
7 Transparency is necessary for patients and their families to make
informed decisions
8 Needs are anticipated
9 Waste is continuously reduced
10 Co-operation among clinicians is a priority
38 Vision and value in health information

Recommendation 3: strengthen leadership

throughout the health system through
educational opportunities and restructured
roles within their work settings
E¡ective leaders move their organisations or communities from vision to reality,
idea to action and the abstract to the concrete.27 The transformation process
entails four basic steps. First, leaders must be able to convince their constituents
of the need for change and motivate them to act by describing their vision and
delineating goals, often in the form of stories. Second, leaders must enable
and encourage the organisation to succeed by building the platform (skills,
tools and incentives) that will enable and encourage change. Third, the mem-
bers of the organisation must act. Fourth, the organisation, including its leaders,
must learn from successes and mistakes to improve its performance.
The nature of leadership has changed in recent years to accommodate
changes in organisational structure, culture and practices. E¡ective leaders are
more likely to be found throughout organisations and society, to share informa-
tion and power, and to possess both strong technical expertise and ‘emotional
intelligence’.28 The leadership needed to initiate and support dramatic change
will not appear deus ex machina. It must be cultivated by organisations, and
future leaders need opportunities to re¢ne their skills.
Further, there is a need for both organisations and individuals to understand
the distinction between being a manager and being a leader. Managers are
important for smooth operations while leaders determine future directions
for their organisation or unit. Leaders set the tone for their organisation through
their communications, behaviours and the goals they set for the organisation,
its personnel and themselves. It is particularly important that leaders under-
stand the various constituencies that are part of their organisation so that they
can translate their message to suit di¡erent audiences.
Leaders assume that their managers and sta¡ are pro¢cient in three critical
information skills ^ ¢nding and using evidence to guide clinical and adminis-
trative decisions, using data to generate new knowledge and evidence, and
redesigning clinical and administrative processes. Success in the new health
system will depend on a leader’s ability to foster collaboration and cross-
fertilisation among di¡erent disciplines and across organisational lines. Leaders
must be externally focused and play a critical role in forging organisational
relationships that will enhance its ability to succeed. Lastly, leaders must be
willing to make decisions, take risks, act on their decisions and learn from both
their successes and failures.
The NHS has a serious problem in the knowledge and skill base of its most cru-
cial group of change agents. Its highly respected clinical leaders are competent
 Improving the United Kingdom’s health system 39

in their clinical disciplines but those who are tagged to become leaders or who
have the instincts and capabilities for being ¢rst-rate medical or nursing direc-
tors or leads on a variety of new programmes move into these posts with a mini-
mum of appropriate academic preparation. It is no surprise that many people in
such positions do not give the job adequate attention and think it is less impor-
tant than the clinical backlogs they also face. Yet it is the innovative thinking of
these individuals that is needed to lead the system redesign that will help to
reduce these backlogs.
Developing leadership skills is not a quick process and two-day workshops are
not adequate to prepare rising leaders of the NHS. The Leadership Centre within
the NHS Modernisation Agency is an important step towards addressing cur-
rent shortcomings.29 Its o¡erings need to be strengthened, however, through
the creation of a clinical scholars’ degree programme that is aimed at fostering
a broad perspective and strengthening analytical skills of its participants. Only
one or two ¢rst rank programmes would be required to meet the needs for the
entire UK and Europe. Moreover, the resources needed to do the job properly are
such that only one site in the UK will be likely to generate su⁄cient funding to
do the job right.
The model programme would include learners pursuing master’s degrees,
some studying at the doctoral level, and some studying simply to gain skills
and knowledge in one particular area. Those learners enrolled for the master’s
level courses could take them either as modular courses to meet the degree
requirements or simply as continuing education courses. However, by being
part of an organised curriculum, their courses would comprise a rigorous, com-
prehensive set of knowledge and skills needed to excel at the role of clinical
leader. Approximately one-half of the curriculum would be from the core curri-
culum for a Master of Business Administration (MBA) degree; one-¢fth would be
on healthcare systems and clinical informatics; one-¢fth would be focused on
quality and safety; one-¢fth would contain elements from a Doctor of Public
Health curriculum; and the balance would be electives particularly suited to
helping learners to improve their current work environment and the problems
it faces. Academic faculty and an appropriate worksite ‘faculty member’ would
collaborate in shaping simultaneously a scholarly and practical e¡ort to address
a current problem facing the organisation and then solve it. This will assure that
the programme actually delivers relevant knowledge and theory while also
developing the learners’ practice skills. The learning goal is not knowledge
per se but useful knowledge, including the skills to create and manage change.
A high priority of the programme is a multidisciplinary character, with both
the teachers and learners being from various backgrounds and disciplines.
Much of the current discontinuity in the present system exists because partici-
pants do not see themselves as members of teams where the role of each is
known and is highly valued. Those who leave the programme will have chan-
ged their perspectives and sense of values. They will be more prepared to face
40 Vision and value in health information

tough issues and help those with more focused responsibilities and skills to
create e¡ective solutions combining both human resources development and
relevant communication and information systems. The practicum ^ the practi-
cal component ^ will also tend to keep learners committed to their current
practice setting while it builds a tradition of supporting the growth and devel-
opment of talented younger people within a variety of settings (for example,
primary care, community resources, hospitals). It scarcely serves the system to
have people constantly shifting from one locale to another with the lost produc-
tivity such moves entail.
Those few individuals who pursue doctoral-level training will become the
generation of health services researchers and leaders who help to ensure that
top-quality research is pursued and that top-level health policy meets the qual-
ity standards needed for progress in healthcare systems. These people will serve
as mentors and colleagues to rising generations of health professionals. Their
training will have made them international in their outlook so that the advances
from elsewhere are fed into the system.
In addition to a robust educational programme to develop clinical scholarship
and leadership, clinical leadership roles within the NHS must be redesigned to
allow those with the knowledge and skills to reinvent their environments and
its systems. Achieving dramatic progress with respect to IT in the UK will involve
the skills of strategic assessment and implementation as much as knowledge of
information hardware and software. The chief information o⁄cer needs to be
seated at the board level but only if he or she has the knowledge and skills
needed to o¡er strategic and tactical direction.
Consider the model of how our cells work. There is important information
sent from the nucleus in the form of messenger RNA to ribosomes. The ribosome
is one-half nucleus and one-half cytoplasmic in its orientation. Having an inti-
mate knowledge of both the centre and the periphery, it can create proteins
(that is, policies and procedures) that are speci¢c to cellular needs so that the
cell can adapt to changing circumstances and still remain healthy. After a set
of clinical scholars begin to populate the work settings as capable ‘ribosomes’,
they must be given the materials (substrates) they need to revise, reform and
create more appropriate structures, systems and modes of communication so
that meaningful acceptable change can be accomplished without intense angst
or strain. Indeed, much of this will eventually become ‘second nature’ since it
basically is an organic approach used by complex adaptive systems as a way
of sense-making.
The poor opinion of management and leadership within the ranks of many
health professionals would suggest that an explicit programme be developed to
raise the visibility and acceptability of these crucial functions. This can involve
little more than acknowledging to the inside and regional community when the
e¡orts of a team and its leader have led to clear improvements in care. Publicly
 Improving the United Kingdom’s health system 41

acknowledging genuinely good deeds is a time-honoured way of giving a sense

of acceptability to traditionally less rewarded activities.

Recommendation 4: build the information

and knowledge platform that will enable
practitioners to translate the envisioned
health system into practice
What practices and behaviours can we expect to see in the new health system?
All decisions within the health system ^ whether a clinical decision for an
individual patient or a decision to fund a particular treatment for the entire
system ^ will be increasingly based on evidence rather than opinion. Quality
improvement will be a routine part of healthcare practice rather than an
adjunct activity. Every health service organisation will become a laboratory in
a sense and strive to learn from its successful and unsuccessful new pilot projects
as well as newly examined routines. Integration of services and collaboration
among health professionals, a range of organisations and with patients and
patient groups will be the norm. Citizens will assume greater responsibility for
and play a major collaborative role in managing their own health challenges.
Healthcare professionals and organisations will view the determinants of health
broadly and pay greater attention to the health of populations.30
Robust information and knowledge management capabilities will support vir-
tually every interaction in the health system and will be pivotal to successful
transformation of the health system. IT and knowledge management are critical
enablers of the system redesign rules developed by the IOM (see box on p. 37).
Citizens and patients can use information technology to develop both personal
and intergenerational health histories, identify risk factors and communicate
with health professionals. Facile access to reliable knowledge resources will
help citizens to determine when they need to contact a health professional and
aid them as they participate in decisions about treatment. Increasingly, technol-
ogy is making it possible for patients with chronic disease (for example, diabetes)
to monitor their health status at home while still maintaining contact with
health professionals.31
IT is the means by which health organisations and professionals can assure
that they are providing quality care to individual patients. Computer-based
patient records make it possible to build complete health histories across time
and across settings of care. Decision support systems bring the latest evidence
to clinicians at the point and time of care. Automation of routine tasks (for exam-
ple, physician order entry) reduces the potential for human error. For example,
42 Vision and value in health information

a computerised physician order entry system with drug allergy and drug^drug
interaction warnings reduced non-missed dose medication errors by 80%.32
Electronic mail o¡ers a convenient means of maintaining contact with patients
and strengthening the relationship between patients and professionals.
IT is also the means by which health organisations and professionals can
improve the quality of health services for the population as a whole by strength-
ening the body of evidence on which health decisions are based. A system
worth its salt will actively seek evidence of clinical impact in a manner that
allows conclusions to be drawn based upon experience. In a managed informa-
tion environment, when problems are identi¢ed that meet criteria for care, and
there is sound evidence on how to obtain excellent results, one follows decision
support rules so that performance does not slip and slide along in an uncertain
manner. Clinicians can choose to override the decision support system when
selecting among care options. However, the data on all patients is tracked
over time so health professionals and system managers can review both the
quality of the protocol and the impact of their decisions to override it. They
can either revise the protocol to become more robust or revise their thinking
and behaviour.
The decision support system is ‘adapted’ over time based upon the relevant
evidence, both distant and local, to assure better processes of care and better
outcomes. The emerging term for such decision support systems for clinical
environments is ‘evidence adaptive clinical decision-support systems’.33 The
hallmark of these systems will be their ability to provide the most current litera-
ture-based evidence and local practice-based evidence to clinicians. Getting
‘just-in-time’ knowledge that is needed at the ‘point-of-decision’ to make sound
decisions is a role for the ‘infostructure’. The architecture for the information
infrastructure must be able to draw upon personal health records, patient
health records and population health records as needed.24 Only with proper
design will the system allow such data to be assembled and studied so that
leaders at di¡erent levels of the system can ask relevant questions and receive
timely answers.
The cycle described above presumes the existence of robust and formal knowl-
edge management capabilities within organisations and across the health sector
as a whole. Information and knowledge management system developers and
users need to understand that information systems are in reality echoes of face-
to-face conversations or other relevant communications. The goal is to both
enhance communications in the ¢rst instance and also bring to later decisions
the value that was derived from earlier investigations and conversations.
Further, the bene¢ts of IT can only accrue if the information that is being trans-
mitted or used for analysis is of high quality, if the people using the information
are skilled in its application, and organisational processes are modi¢ed to take
advantage of improved access to information and incorporate lessons learned
from analysis of the information.
 Improving the United Kingdom’s health system 43

Developing a robust ‘infostructure’ in and of itself represents a formidable,

but important and achievable challenge for each nation seeking to improve
health system performance. In the UK (as in other countries), there is a crying
need for greater central leadership to be taken with respect to system architec-
ture, standards development, and movement to the internet. The appointment
of the Director General of NHS Information Technology is a strongly positive
move in this direction. At late-2002, the UK is literally desperate for such help,
and crucial progress made at the grassroots over the past decade is likely to
become more and more local in its outlook and functionalities if stronger direc-
tion is not o¡ered from the centre. Patients, health professionals and system
managers must all acquire and continue to re¢ne their skills to use emerging
information systems e¡ectively.

Conclusion
Well done is better than well said.
Benjamin Franklin

The UK and the USA are both grappling with how to overcome the short-
comings of their current healthcare systems. They are striving to improve their
approaches to rationing while dramatically improving the quality of their health
services and, in so doing, ultimately improve the health status of their popula-
tions. The complexity of the endeavour should not be underestimated. Clear
vision, su⁄cient resources and countless hours of e¡ort must be combined with
patience and creativity to transform the respective health systems. The impor-
tance of these e¡orts cannot be overstated, improving the health of individuals
and populations always has been and always will be one of the noblest tasks
for professionals and one of the most important functions of government.
Responsibility for building the 21st century health system does not lie with
‘them’. It falls to each citizen, each patient, each health professional, each
health system executive, each policy analyst and each legislator to do a bit more
than their fair share. Citizens and patients have a responsibility to pursue healthy
lifestyles and to participate actively in decisions about their healthcare. Health
professionals have a responsibility to collaborate with their peers and with
patients, to grow as professionals, and to embrace the forthcoming changes.
Policy-makers and legislators have the responsibility for providing needed tools
and adequate resources. System executives have the responsibility for helping
their sta¡ see the possibilities of a 21st century health system and encouraging
experimentation that will yield new practices.
To achieve the greatest bene¢t from available resources, we must all be evi-
dence-based in our decision-making and align our actions with the rules of the
new health system. IT is the means by which evidence can be systematically
44 Vision and value in health information

developed, disseminated and applied throughout the system. Sound informa-

tion systems combined with practices based on evidence will provide a strong
foundation for the future health system and for healthier and happier lives.

References
1 Feachem RGA (2000) The future of the NHS: confronting the big questions. Health
A¡airs 19: 128^9.
2 Institute of Medicine (2000) To Err is Human: building a safer health system. National
Academy Press, Washington DC.
3 Institute of Medicine (2001) Crossing the Quality Chasm: a new health system for the 21st
century. National Academy Press, Washington DC.
4 Klein R (1999) Why Britain is reorganizing its National Health Service ^ yet again.
Health A¡airs 17: 111^25.
5 Smith PC (2002) Performance management in British health care: will it deliver?. Health
A¡airs 21: 103^15.
6 Smith R (2002) Oh, NHS, thou art sick. BMJ 324: 127^8.
7 King’s Fund (2002) The Future of the NHS. A Framework for Debate. King’s Fund, London.
8 Wanless D (2002) Securing Our Future Health: taking a long-term view. April. HM Treas-
ury, London.
9 O⁄ce of Health and the Information Highway (2002) Canadian Health Infostructure
(CHI): introduction. (www.hc-sc.gc.ca/ohih-bsi/chi ics/index e.html)
10 Institute of Medicine (2001) Informing the Future: critical issues in health. (www.iom.edu)
11 Institute of Medicine (2001) Improving Palliative Care for Cancer. National Academy
Press, Washington DC.
12 Institute of Medicine (2002) Unequal Treatment: confronting racial and ethnic disparities in
health care. National Academy Press, Washington DC.
13 Institute of Medicine (2002) Biological Threats and Terrorism: assessing the science and
response capabilities. National Academy Press, Washington DC.
14 Institute of Medicine (2002) Confronting Chronic Neglect: the education and training of
health professionals on family violence. National Academy Press, Washington DC.
15 Institute of Medicine (2002) Fostering Rapid Advances in Health Care: learning from system
demonstrations. National Academy Press, Washington DC.
16 Institute of Medicine (1991) The Computer-based Patient Record: an essential technology for
health care. RS Dick and EB Steen (eds). National Academy Press, Washington DC.
17 Institute of Medicine (1997) The Computer-based Patient Record: an essential technology for
health care. (rev. edn.) RS Dick, EB Steen and DE Detmer (eds). National Academy Press,
Washington DC.
 Improving the United Kingdom’s health system 45

18 Institute of Medicine (1994) Health Data in the Information Age: use, disclosure, and
privacy. MS Donaldson and KN Lohr (eds). National Academy Press, Washington DC.
19 Institute of Medicine (1996) Telemedicine: a guide to assessing telecommunications in health
care. MJ Field (ed). National Academy Press, Washington DC.
20 National Research Council (NRC) (1997) For the Record: protecting electronic health infor-
mation. National Academy Press, Washington DC.
21 National Research Council (1999) Trust in Cyberspace. FB Schneider (ed). National
Academy Press, Washington DC.
22 National Research Council (2000) Networking Health: prescriptions for the internet.
National Academy Press, Washington DC.
23 President’s Information Technology Advisory Committee (PITAC) (2001) Transforming
Health Care through Information Technology. (www.ccic.gov/pubs/pitac.index.html)
24 National Committee on Vital and Health Statistics (2001) Information for Health: a strat-
egy for building the national health information infrastructure. United States Department of
Health and Human Services, Washington DC.
25 Plesk P (2001) Redesigning health care with insights from the science of complex adap-
tive systems. In: Institute of Medicine Crossing the Quality Chasm: a new health system for
the 21st century. National Academy Press, Washington DC.
26 World Health Organization (2001) Informal Meeting on Innovative Care for Chronic Condi-
tion. (ICCC) 30^31 May. Geneva, World Health Organization.
27 Couto RA (2002) To Give Their Gifts: community, health, and democracy. Vanderbilt Uni-
versity Press, Nashville, Tennessee.
28 Goleman D (1998) Working with Emotional Intelligence. Bantam Books, New York.
29 NHS Modernisation Agency (2002) Leadership Development. (www.modernnhs.uk)
30 Blue Ridge Academic Health Group (1998) Promoting Value and Expanded Coverage: good
health is good business. Cap Gemini Ernst & Young, Washington DC.
31 Shea S, Starren J, Weinstock RW et al. (2002) Columbia University’s Informatics for Dia-
betes and Telemedicine (IDEATel) Project. Journal of the American Medical Informatics
Association 9: 49^55.
32 Bates DW, Teich JM, Lee J et al. (1999) The impact of computerized physician order entry
on medication error prevention. Journal of the American Medical Informatics Association
6: 313^21.
33 Sim I, Gorman P, Greenes RA et al. (2001) Clinical decision support systems for the prac-
tice of evidence-based medicine. Journal of the American Medical Informatics Association
8: 299^308.
CHAPTER 4

The need for a new healthcare

paradigm: patient-centred
and knowledge-driven
JA Muir Gray and Harry Rutter

Introduction
Never has healthcare been so e¡ective, or so criticised. About three of the ¢ve
extra years of life that people have gained in the last half-century are due to the
e¡ects of progress in healthcare.1 The developments, many of which we now
take for granted, have been astonishing, ranging from renal transplantation to
the treatment of migraine. However, those who provide and pay for healthcare
face continuing problems as the need and demand for healthcare increases
inexorably owing to population ageing, technological development and rising
expectations. In almost every country the rate of increase of need and demand
is greater than the rate at which additional resources can be made available by
the state.
For these reasons healthcare systems always fail to meet both need and
demand, leading to increased pressure for more resources. This is resisted in
many countries because of the high proportion of the wealth of societies already
invested in healthcare. The gap between public expectations and the reality of
service delivery also leads to demands for ever higher standards of management
of those resources that are available, and managerial ingenuity has been in-
vested in the development of better systems for managing human resources, for
funding healthcare, for motivating clinicians, and in new methods of organis-
ing and delivering healthcare.
One reason for our continuing problems is that we are not moving quickly
enough to the new paradigm described in The NHS Plan and related documents,
such as An Organisation with a Memory2 and The Expert Patient,3 described by the
Chief Executive of the NHS at the South East Region conference on Information
48 Vision and value in health information

for Health in June 2001. The di¡erence in emphasis between the old and the new
paradigms is set out in Table 4.1.

Table 4.1 The old and the new knowledge paradigms

20th century paradigm 21st century paradigm

Professional control and responsibility Shared control and responsibility

Single professions Teams
Expertise Accountability
Institutions Networks
Services Systems
Paper Digital
Finance as key constraint Knowledge and people as key constraint
Quantity Quality and safety
Averages Inequalities
Strategy Delivery
Science Humanity
Centred on professionals Centred on patients

The response of the Department of Health to the Kennedy Report4 on pae-

diatric heart surgery, Learning from Bristol, emphasises the need to move to the
new paradigm.
Many changes are needed to bring about the new paradigm. One of the most
important of these is the need to realise the potential value of information and
knowledge for patients, clinicians, managers and commissioners. The new para-
digm has the following characteristics:

. it assumes the patient is competent and responsible

. it creates an electronic patient record as the hub of the system
. it integrates every type of document to provide a single source
. it supports clinical decision-making by delivering knowledge when and
where and how it is needed
. it uses knowledge to drive quality improvement
. it allows changes in policy to be immediately communicated and implemented
. it facilitates systems and networks
. it o¡ers research opportunities to clinicians and patients.

Coping with the explosion of knowledge

More than two million research reports are published every year. Vast amounts
of data are collected about patients. Huge numbers of people work in health-
care, each of whom has expertise and experience that is rarely made explicit;
 The need for a new healthcare paradigm 49

it is thus tacit knowledge. Patients also have a large amount of tacit knowledge
that can be made explicit and useful. However, the potential usefulness of
this knowledge is not fully realised because it is not managed well enough.
Healthcare organisations manage money, human resources and their buildings
with care and responsibility, but knowledge is not managed to this standard.
Librarians manage the knowledge that is sent to libraries, but much important
knowledge is not sent to libraries ^ knowledge about health service use and
outcomes, for example. Furthermore, few libraries are properly used in support-
ing board decisions because board members often do not see the need to use the
knowledge and skills contained in libraries for management decision-making.
The internet changes all aspects of knowledge management from research to
publishing and from distribution to implementation, but few health services
are realising the potential that the new paradigm o¡ers.

Knowledge management problems

Although healthcare depends on capital assets, such as buildings, it is an excel-
lent example of a knowledge economy, namely one that is based on knowledge
as an asset and in which the main products ^ better clinical decision-making
and better treatment ^ depend on the e¡ective management of knowledge.
Furthermore, healthcare has it own huge research infrastructure producing
knowledge speci¢cally for the healthcare industry.
However, as a knowledge economy healthcare is cavalier about the way in
which it manages knowledge, and a number of weaknesses in its knowledge
management are easily observable, notably:

. poor co-ordination of knowledge sources

. poor control of knowledge £ows into organisations
. poor control of knowledge £ows within organisations
. poor document management
. wide variation in the knowledge available to key professionals
. failure to get new knowledge quickly and consistently into practice
. haphazard and low quality control over the knowledge produced by health-
care organisations, for example knowledge for patients.

The magnitude of these problems is immense; for example, half of all medication
errors are knowledge-based, and poor document management is overloading
practitioners with information: almost 23 kg of paper guidelines had been sent
to all 30 000 general practitioners in the eight years to 1998.5
All of these problems can be tackled by more e¡ective knowledge management.
50 Vision and value in health information

Aims and objectives of knowledge

management
The aim of knowledge management is to help individuals and organisations to
achieve their objectives more e⁄ciently through the generation, organisation,
mobilisation and utilisation of knowledge. The speci¢c objectives of knowledge
management are to:
. identify the knowledge needs of each professional, patient and organisation
. develop the necessary culture, systems and structure within organisations
. ensure that the knowledge required is produced
. provide answers to questions
. convert tacit knowledge into explicit forms
. deliver knowledge when and where and how it is needed
. develop the knowledge management skills of individuals
. manage documents e¡ectively and e⁄ciently
. create opportunities for learning.

Improving the three dimensions of

knowledge management
To improve knowledge management three inter-linked initiatives are required
(Figure 4.1).
Personal
development

Organisational Document
development management

Figure 4.1 Three inter-linked initiatives to improve knowledge management.

Personal development
Although healthcare is based on knowledge and uses large amounts of resources,
it is an activity delivered by people for people, and personal development is of
central importance in improving knowledge management.
The level of performance of individuals within an organisation is a function of
three variables: it is directly related to their competence and motivation, and
 The need for a new healthcare paradigm 51

inversely related to the barriers that they have to overcome. A personal devel-
opment strategy therefore has to address all three of these issues (Figure 4.2).

Competence

Motivation Barriers

Figure 4.2 Three issues to be addressed in a personal development strategy.

Professionals and patients are highly motivated to manage knowledge to best

e¡ect but they face signi¢cant barriers in doing so, such as di⁄culties getting
hold of knowledge. However, even with these barriers removed individuals
may not have the skills that are required to:

. structure questions to make best use of knowledge

. ¢nd the relevant knowledge
. appraise the quality of the knowledge found
. tailor the knowledge to take into account the needs and values of the indivi-
dual patient or population
. communicate and implement the knowledge.

Fortunately, there are now resources available to help professionals acquire or

improve these skills, such as the Critical Appraisal Skills Programme (CASP)
based in Oxford.

Organisational development
Three aspects of an organisation impinge upon the ability of organisations to
manage knowledge culture, systems and structure (Figure 4.3).

Culture

Systems Structure

Figure 4.3 Three aspects of organisations that impinge upon their ability to manage
knowledge.
52 Vision and value in health information

Organisational change has hitherto focused primarily on structural reorganisa-

tion. But, structural reorganisation will never produce a perfect organisation
for an activity as complex as healthcare. Transformation through cultural
change and new knowledge management systems provides better opportunities
for tackling the ubiquitous seven healthcare problems (see box) and maximising
the value from the resources invested in healthcare. Some structural change is
sometimes required, for example someone at board level needs to be identi¢ed as
the Chief Knowledge O⁄cer, analogous to a Chief Finance O⁄cer. But the main
changes that need to take place are in culture and systems.

The ubiquitous seven healthcare problems

. Clinical errors
. Unknowing variations in policy and practice
. Failure to get knowledge into practice
. Overenthusiastic adoption of new technology of low value
. Poor clinical quality
. Poor patient experience
. Waste

Document management
The term ‘knowledge’ is sometimes interpreted as meaning evidence derived
from research, but this is only one meaning and it is too narrow. A second

Table 4.2 Document types and the knowledge they contain

Document type Example of embedded knowledge

Policy General principles to guide prioritisation and resource allocation

Knowledge Information derived from data, experience or research, tailored for
decision-making
Quality How services should be provided to achieve explicit standards
Patients Data about individual patients
Administration Information used to organise services
Returns Data provided for performance monitoring
Education Resources to increase knowledge and skills
Research Data collection questions
 The need for a new healthcare paradigm 53

approach is to distinguish tacit knowledge, which is within the head of the

person who has developed or acquired it, from explicit knowledge that has
been written down and published. Fortunately, tacit knowledge can usually be
converted into explicit knowledge without too much di⁄culty. Thus, people
other than those who have met and spoken to the holder of the tacit knowledge
can also use the knowledge that has been acquired from experience. Perhaps
the simplest way to de¢ne the di¡erent types of knowledge is to classify knowl-
edge by the objectives for which it has been prepared. Table 4.2 lists the di¡erent
types of document used in healthcare.
To improve document management it is necessary to clarify which types of
content will be required for particular decisions, professionals or organisations,
and then, using web-based systems, to organise those documents in a way
which maximises their:

. accessibility ^ the ease with which a document can be found, moved or stored
. usability ^ the ease with which it can be used for the purpose intended
. functionality ^ the ease with which the documents and their systems inter-
act with all other information systems, such as the electronic patient record.

Improving knowledge management

To improve knowledge management a two-stage process is necessary in which
each of these three dimensions ^ professional development, organisational
development and document management ^ is analysed and then changed to
improve e¡ectiveness and e⁄ciency.
The adoption of these new healthcare paradigms o¡ers a di¡erent way of orga-
nising healthcare compared with the patient-centred traditional model that is
based on discrete organisations such as hospitals or primary care. General
moves to strengthen the infrastructure, such as increased access to broadband,
are also essential, but the development and implementation of new medical
knowledge paradigms o¡ers a number of bene¢ts, notably:

. the opportunity to interest and motivate many clinicians who are interested
in reducing clinical errors, but who would be uninterested in, or cynical
about, concepts such as knowledge management
. improvements in quality management
. support for national programmes of disease control by providing a knowl-
edge framework on which better control of epilepsy or asthma could be built.

Change can be facilitated and promoted through the use of concepts, but it
is essential to complement these with structural and systems measures to
54 Vision and value in health information

transform healthcare organisations from those in which knowledge plays only

a minor and peripheral part in the life of the organisation to those in which
knowledge is managed as carefully as money or human resources.
The knowledge system built for the Rapid Access Chest Pain Clinic web folder6
is an example of a knowledge system that incorporates many of the features
listed above and which will incorporate all the characteristics listed earlier.

Building knowledge into quality improvement

Knowledge is derived not only from research and experience. Routinely col-
lected data produce knowledge; the Public Health Observatory is an excellent
example of good knowledge management. We need to link all types of knowledge
to the most important task of the NHS ^ looking after patients and populations,
and continually improving quality and safety to improve health (Figure 4.4).
The new paradigm for healthcare could be created by, and would be driven
by, well-managed knowledge.

Knowledge
about disease
and its control

Prepare

Knowledge
Knowledge
about the
about the NHS Change Do
individual
as a whole
patient

Reflect

Knowledge
about all the
processes and
outcomes of all the
patients treated
by a service

Figure 4.4 Knowledge tasks in the NHS.

 The need for a new healthcare paradigm 55

References
1 Bunker J (2001) Medicine Matters after All. The Nu⁄eld Trust, London.
2 Department of Health (2000) An Organisation with a Memory. The Stationery O⁄ce,
London.
3 Department of Health (2001) The Expert Patient: a new approach to chronic disease manage-
ment for the 21st Century. The Stationery O⁄ce, London.
4 Department of Health (2002) Learning from Bristol: the Department of Health’s response to
the report of the public inquiry into children’s heart surgery at the Bristol Royal In¢rmary
1984^1995. The Stationery O⁄ce, London.
5 Hibble A, Kanka D, Pencheon D et al. (1998) Guidelines in general practice: the new
tower of Babel? BMJ 317: 862^3.
6 National Electronic Library for Health Rapid Access Chest Pain Clinic project. (www.
nelh.nhs.uk/heart/racpcs/dataset/racpc intro.htm )
CHAPTER 5

Information as the
patient’s advocate
Michael Rigby

Introduction
Traditionally, the medical record has been an essential information store captur-
ing the patient’s history of diagnosis and treatment with a healthcare provider.
By reference to this record subsequent clinicians can scan the patient’s previous
health experiences, but more importantly can assess the possible causes of new
morbidity, and previous patterns of response to treatments, as the record speaks
for the patient in a way more technically eloquent than patients themselves
could achieve. Medical and nursing practice has considered the record as some-
thing which belonged to the organisation and was there for the bene¢t of
clinicians, whereas modern thinking is that it should be considered far more as
being a representation of the patient’s interest.
Now, modern technology and the opportunity to interlink records give this
advocacy function much greater power and wider recognition. There are also
sub-groups of patients who are not able to represent themselves easily in health-
care encounters ^ these include: infants; patients who are comatose confused
or distressed; and under some circumstances, those with a mental illness or
intellectual disability. Thus as organisational concepts such as care planning
become more focal to the healthcare process, so the patient’s case for bene¢ting
from a share of available healthcare resources becomes dependent upon the
strength of the representation of their need as encapsulated in the record. This
is additional to the traditional indirect but important function of epidemiology
and public health in taking a statistical representation of a de¢ned group of
people and promoting their interests through public health measures, including
identi¢cation of resource needs.
Whilst some aspects of this representation approach are rooted in history, the
work of the Steering Group on Health Services Information, led by Edith K˛rner,
58 Vision and value in health information

represented a major step forward. In particular, the minimum data set (speci¢ed
according to local functional need) is a concept that ensures comparability of
representation. Some subsequent initiatives have sought to facilitate this repre-
sentational role. However, as with so many initiatives that Mrs K˛rner’s work
instigated, the follow-through has been poorly focused and has not captured
the imagination or striven for the goals that are desirable.

Development of record-keeping
Over the years record-keeping has been seen as a fundamental and essential part
of clinical practice ^ for doctors, for midwives, for nurses and for other profes-
sionals. As a result, in recognition of the vital importance of the contribution of
an accurate historic record to the care of the individual, professional bodies have
established principles and professional standards for this task. For the individual
clinician failure to study the record before treatment is perceived as reckless risk-
taking (except in emergencies when the record is not available and life is at
immediate risk), and concomitantly the keeping of accurate records is seen as
an essential professional requirement, failure to comply with which is a disci-
plinary matter. In the more complex world of hospitals, with a large range of
diagnostic and other information, and tremendous demands upon the records
library function, a distinct medical records profession has become established,
together with principles and techniques.
However, these principles and practices have emerged from centuries of
keeping of paper records, largely completed by hand or more recently machine
printout, and perforce restricted to the individual healthcare organisation. The
record for individual patients could be held within the controlled environment of
the treatment location, and thereby also linked to the particular health prob-
lems being treated there.
Given these physical constructs, communication between di¡erent treatment
locations was by means of letter. Although the message would be about an indi-
vidual patient ^ a request for a diagnostic procedure, a request for treatment for
a purported condition or a discharge noti¢cation with summation of the action
taken and recommendations for onward care ^ the views expressed were clini-
cal interpretations. They were messages from one health professional or team
to another, expressing views about the patient, and asking another clinician to
undertake speci¢c action. Although these messages concern a patient, the views
are those of health professionals, and the speci¢c wishes of the patient may or
may not be strongly articulated within them.
This is a ‘message in a cleft stick’ approach to clinical communications, and
the health records for patients begin to acquire a collection of the messages
received. A particular provider’s record for a patient thus extends to being a
representation of the diagnosis and treatment of that patient by that health
 Information as the patient’s advocate 59

organisation, supplemented by a series of distilled views through this collection

of messages from elsewhere.

Electronic opportunity, electronic challenge

This organisational messaging approach to record-keeping was the best that
could be achieved in a paper-based era, but that started to change in the last quar-
ter of the twentieth century. The advent of modern telecommunications led
to widespread parts of industry and commerce dispensing with paper and relying
upon electronic media for both transmission and storage of information. In many
cases this passed through interim phases, such as use of facsimile machines to
transmit a letter instantaneously (though with the input and output still being
paper-based), but it soon migrated to full, electronic inter-connectivity. In order
to bring some semblance of order to this world, standard messaging protocols
became a key issue, and health organisations both within individual country
health systems and internationally became involved in developing standard
clinical messages. Pathology and other areas of strongly numerate clinical
data, which were amenable to structuring yet where accuracy of detail was
essential, led in this ¢eld. Less straightforward in the ¢rst instance were the stan-
dards to convey word-based or narrative-based information.
However, the concept of greater inter-connectivity frightened healthcare
practice almost as much as it enthused industry and commerce. For instance,
retail banking underwent a revolution as it grasped not just electronic recording
but consumer-focused electronic interactivity. The era of customers going to a
bank during set working hours, to ¢ll in a form in order to get some of their own
money put into their hand, was swept away with the advent of automatic teller
machines (ATMs) available 24 hours per day and in locations convenient to
consumers, who no longer needed to conform with the provider organisation’s
working practices in order to obtain their own money. This progressed to creat-
ing a wider range of banking services and stimulated a consumer-led refocusing
of ¢nancial services. Banks had to change radically, and their structure and
sta⁄ng adjusted to support consumer needs. Less conspicuously but equally
signi¢cant, manufacturing organisations interlinked their operational systems,
so that supplies of materials and components were ordered electronically as
needed, and the whole transportation industry transformed to logistic support
and just-in-time (JIT) deliveries.1
The health sector response to these opportunities was pedestrian. Some diag-
nostic departments moved rapidly to computerisation and electronic media
because these signi¢cantly enhanced their patterns of working. Slowly there
were moves to forms of electronic record-keeping, but these were initially very
much focused upon patient registration, patient administration and business
60 Vision and value in health information

transactions. The opportunity to re-engineer the organisation and the processes

was but a gentle breeze blowing against the historic temple of organisational
structure and its long-established ritual procedures.

Of bright lights and mirrors,

and deeper shadows
However, as the use of electronic data processing and recording crept slowly
into healthcare, so some pioneers saw the light which beckoned towards
greater bene¢ts than merely the electronic automation of messaging. The
work which captured the imagination was that of the US Academy of Sciences
Institute of Medicine in 1991, which reached the public as a seminal volume.2
This highlighted the bene¢ts to the patient of information about them being
held in any treatment location and being available in real time to clinicians at
a later date in a totally di¡erent treatment location and regardless of the health
problem being addressed.
Although the work rapidly ¢red the imagination, its limitations were not
widely noticed. There was the lack of any costing, and no modelling of the
volume of transactions and thus the electronic demands that would be made,
but more signi¢cant was the fact that the means of representation of patients
between healthcare organisations and locations was to be totally changed ^
letters and other inter-personal communication were out, and direct access to
the record was in. So instead of the remote clinician initiating a person-to-
person communication between health professionals that summed up the
previous history or health issue, the record written by a previous clinician
itself now spoke directly for the patient ^ it was now the patient’s advocate.
It spoke of the patient’s wider healthcare history, with their consent (if operated
appropriately), but without their needing to be present to tell the story. This was
a new and unremarked attitude, whereby the record itself was assumed
to be able to represent patients’ full interests without clinician intervention or
interpretation.
The Institute of Medicine report highlighted, appropriately, the bene¢ts of
much wider and immediate access to essential information. But inevitably it
assumed that the information was largely scienti¢c and biophysical. Under-
neath, the vision was built largely on the unexplored assumption that health
professionals thought along similar lines, recorded by use of the same notations
and used language in similar ways. And it gave little thought to how records in
di¡erent settings interpreted patients’ views, feelings, preferences or their life
circumstances speci¢c to their (often not fully described) personal and social
contexts. It assumed a common inter-organisational language of care that had
barely begun to be developed.3
 Information as the patient’s advocate 61

In a healthcare world in which 10% of morbidity relates to mental illness

(in its broadest de¢nition), where family and social support is seen as a nor-
mal requirement but is intensely personalised, and where a high proportion of
adults have some form of caring or supporting responsibility, this simple
approach to indicating health activity with electronically held records is bound
to be regressive towards the medicalisation of healthcare. Representation of
patients becomes unduly focused on their biological history and previous health-
care interventions. Representation of patients’ personalities and daily circum-
stances, the impact on daily life of functional limitations of health problems,
and their aspirations and priorities for treatment outcomes, which hitherto
came through the clinician-to-clinician message (where even then it was likely
to be under-represented) rather than from the formal records, were now at risk
of further downgrading.

Recognising the wider role of the record

One of the key issues in moving to a modern information environment in
healthcare is to recognise that the traditional format of the medical record ^
a sequential chronology or history with some additional noting of external
material ^ has been a necessity of limited technology, not a framework of ¢rst
choice. One of the earlier proponents of a more structured approach, even in a
paper-based era, came from Laurence Weed in America, who introduced the
concept of the Problem Orientated Medical Record (POMR).4,5 This sought the
drawing together of patients’ complaints into a structured series of problems
rather than clinical phenomena, so that a more integrated or holistic approach
could be taken. Concomitant with this was the use of the systematic ‘SOAP’
approach of seeking Subjective data then Objective data, Assessing the problem
and then Planning management of the case.6
This, in turn, led to a focus on the planning of care. Broadly speaking, care
planning was developed within nursing care to recognise that a range of activ-
ities need to be undertaken regularly, at pre-planned times, but above all for
expressed purposes and with intended outcomes, and that the nursing resource
should be available to meet these needs.7 This manifests in one form in inpatient
care, but the same principles apply to domiciliary care for either physical or
mental conditions, where it is important that procedures are undertaken at the
right time and place. In more integrated form and by various other therapy-
speci¢c titles, such as care protocols and care pathways, care planning is now a
recognised approach. Indeed, it is seen as important that patients are fully
aware of their care plan.
However, this is only the ¢rst move towards advocacy for patients, through
ensuring that future resource requirements are recorded and thus patients’
62 Vision and value in health information

interests protected. This might be seen as resource requirement representation,

rather than full advocacy. And even then, apart from speci¢c intensive settings
such as theatre rostering and acute hospital dependency-based nurse roster-
ing, there are few examples of patients’ care plans directly fuelling a health
resource management function in a way that modern commercial functions
would see as essential good management and service, based on customer
orders or daily sales.1
The latent need ^ the true record advocacy that remains to be understood or
achieved ^ is the further development of the record so that it truly acts on behalf
of patients as their advocate under a range of di¡erent circumstances. This is
particularly important bearing in mind that patients are seldom present when
decisions are made that a¡ect their treatment ^ decisions such as scheduling
of admissions, deployment of resources or in the longer term the development of
health services. And, in a range of not uncommon situations, although patients
are present they are not able to represent their own interests fully at that time.

Advocacy roles of the record

Looking forward to the opportunities that ought to be created with the introduc-
tion of electronic records, and the virtual electronic record created by wider
networks, there is now a new opportunity for records to act positively as
patients’ advocates, representing their needs in a way that cannot be overlooked.
The following types of advocacy through the patient record can be considered.

Advocacy for treatment

A patient presenting and not fully able to explain themselves through reasons
of illness or confusion, as well as patients being considered at a case confer-
ence, should be capable of representation from a record which presents not just
their clinical symptoms but that also represents them as human beings in a
social context, with personal needs and views. Thus their preferences, their
responsibilities and the support available to them are equally important. Some
proponents of networked electronic records in mental health have advocated
these solutions and the results and bene¢ts.8,9 In a human values-based service,
recognition of patients’ preferences and beliefs should also be important, and
should move from the current largely negative approach of not undertaking
treatments to which patients will object, to a more positive one of recognising
how patients would prefer to be treated.10
 Information as the patient’s advocate 63

Resource advocacy
Healthcare resources are always overstretched, and even in the best-planned
circles illness of sta¡ or other adverse events will always disturb the pre-planned
schedule and require immediate decisions. Care may be delayed, rescheduled,
reduced or cancelled. In these circumstances patients are only represented by
their records or their scheduled requirements.

Delivery pattern advocacy

Whilst many aspects of clinical priority are self-evident, others are lost by a
simple listing of clinical procedures ^ for instance, rescheduling an activity to
a date when a patient has a clashing commitment is likely to waste resources.
An integrated or inter-linked record with details of all planned care can
avoid this.

Protection advocacy
Possibly the least-appreciated of all, the record can act to protect patients.
A strong yet under-realised example is in child protection, where abused chil-
dren are usually presented to di¡erent health facilities in di¡erent locations so
as to disguise the pattern of induced illness or injury. Subsequent studies have
shown all too frequently that all the relevant information is available, but is
locked into separate unlinked record systems.11 A means of pooling, within an
ethical framework, of this otherwise comparatively insigni¢cant information in
order to build up a big picture would have a protective e¡ect. By presenting the
pattern of previous presentations and problems to a health or social care profes-
sional, the record would speak on behalf of the child.

Prevention advocacy
This is more widely understood, and is based on the emergence of public health
and epidemiological principles. Illness in individuals will be treated on an
individual basis, but if summaries of that information are pooled then epidemio-
logical pictures build up. Action can then be taken to address the course of a
problem and to prevent further recurrence. The bene¢ciary of this form of popu-
lation advocacy is not the group of patients who have already su¡ered, but their
peers in the social group who are protected from also joining them. The social
group itself may not have any visible cohesion. The traditional example is the
64 Vision and value in health information

identi¢cation of the Broad Street pump as a source of cholera, but in modern

times possibly the least-cohesive group was people who had walked down a
particular London street, who thereby contracted Legionnaire’s disease. An ex-
tensive public health investigation identi¢ed the source as being a building’s air
conditioning system, and its elimination protected the general population.

Requirements for record advocacy

Realisation of the advocacy referral of the electronic record requires three
things. First, electronic records need to be available in a locality, and with su⁄-
cient coverage of use to speak for all those in a particular category of need, or
service bene¢ciaries. Although the core concept may have arisen from the
concept of the computerised patient record,2 it is wider, community-based net-
working that has made the foundations available through comprehensive cov-
erage.12 This broad concept is now becoming more widely enshrined in national
policy, as with the initial NHS policy document, Information for Health.13
Second, as already indicated, there is considerable risk that on their own
electronic records become increasingly biophysical in detail, and reduce the
representation of patients’ personal viewpoints and needs. Thus it is very impor-
tant that electronic records contain adequate information about needs, proper-
ties and preferences, so that they truly represent patients as individuals. In order
to do this appropriate elements are needed in the record structure,9 but recorded
in a sensitive yet objective way.3
Third, such systems need to represent the real world in terms of availability of
appropriate services, practical operational arrangements to manage their
deployment and general usage by the full range of healthcare professionals
and ¢rst-line managers.14,15 This needs a number of concepts to be developed
and applied to healthcare that may be commonplace in other commercial ser-
vices, but not so in this more complex world. These include the management
of resource slots and better planning of treatment patterns, and avoidance of
bottlenecks and overcommitment. Concerning topics such as the need to pro-
tect vulnerable children, the most e¡ective balance between con¢dentiality
and sharing key indicators of risk has yet to be achieved within the networked
record system.

Conclusion
As with so many developments in healthcare, there is a risk that the develop-
ment of electronic patient record systems will be driven by issues of provider
organisation e⁄ciency, and modelled largely on the apparently more expensive
 Information as the patient’s advocate 65

areas of acute care. However, as is also often the case, this will fail to look at the
yield of human bene¢ts that could and should be available by a more re£ec-
tive and consumer-orientated approach. Streamlining post-operative care, for
example, should be bene¢cial to patients and organisations. However, avoiding
missed appointments in mental health or long-term support of the frail and
elderly, or rapid adjustment of planned support because of a change in the
patient’s circumstances, will reduce the risk of adverse incidents whose human
and economic costs can be even more severe. Quantifying the bene¢ts of redu-
cing the number of children at risk of inadequate care may be di⁄cult, but there
are clearly gains on all fronts if these problems are reduced. But even beyond
these more emotional bene¢ts, there are clearly signi¢cant gains to patients
and services if the ongoing delivery of care and provision of support is more
¢nely tuned to human facets, and if necessary short-term changes to deploy-
ment caused by sta¡ sickness or other factors can minimise disadvantages
to patients.
Moreover, the issues are not restricted merely to the immediate ones. The
NHS is concerned about comparatively poor patient compliance both with
delivery mechanisms, such as keeping to appointment times, and with compli-
ance with treatment regimes. It is often not appreciated that this is exacerbated
by the apparently impersonal and somewhat mechanistic structure of the ser-
vice delivery system. If healthcare delivery based on modern record systems is
seen to be much more people-speci¢c and sensitive, it is more likely to increase
respect for the service and thus compliance with it.
In addition, management should increasingly be evidence-based, and the
necessary evidence derives in a large part from local information.16 Wherever
possible, management information should be drawn from operational systems,
to minimise cost whilst valuing accuracy and timeliness. Electronic patient
records are an ideal source of such data. But if operational and strategic deci-
sions are going to be made based on the information in patients’ records, it is
also essential to ensure that they act e¡ectively as patients’ advocates.
In order to achieve this the focus of the record needs to move from solely that
of a treatment history and a treatment booking system to one that, in all
respects, speaks for and on behalf of patients in a much more personal way as
an advocate. In a signi¢cant way Mrs K˛rner’s work laid a simple foundation,
by indicating the importance of the minimum data set to ensure that adequate
information was recorded and shared for each patient. This was a very simple,
and in the more modern technological context simplistic, approach. Never-
theless, when building upon the concept of a core set of essential data represent-
ing patients, and building into that ideas of representing patients’ interests in a
wider humanitarian sense, can be seen the foundations of a new paradigm of
representation of patient interest in the record. Now that there is a policy drive
to further speed up the implementation of electronic patient records, there is
the danger that history will repeat itself and the NHS will opt for the simplest
66 Vision and value in health information

solution simply because it will be the quickest. But as we learnt from that era of
short-term expediency, the core purpose will be signi¢cantly undermined. The
challenge in the modern implementation of electronic information and records
systems in health must be to build in the wider interests of patients, using appro-
priate representations through modern information concepts, so that the new
record structures truly represent patients as people, and promote their interests
and advocate their positions in articulate, sensitive and caring ways.

References
1 Roberts R, Rigby M and Birch K (2000) Telematics in healthcare: new paradigm, new
issues. In: M Rigby, R Roberts and M Thick (eds) Taking Health Telematics into the 21st
Century. Radcli¡e Medical Press, Oxford.
2 Institute of Medicine (1991) The Computer-Based Patient Record ^ an essential technology
for health care. RS Dick and EB Steen (eds) National Academy Press, Washington DC.
3 Swayne J (1993) A common language of healthcare? Journal of Interprofessional Care
7: 29^35.
4 Weed L (1968) Medical records that guide and teach. NEJM 278: 593^600, 652^7.
5 Weed LL (1969) Medical Records, Medical Education, and Patient Care: the problem orien-
tated record as a basic tool. Case Western Reserve University, Cleveland OH.
6 Savage P (2001) Problem Orientated Medical Records. BMJ 322: 275^6.
7 Binnie A et al. (1984) A Systematic Approach to Nursing Care ^ An Introduction. Open Uni-
versity Press, Milton Keynes.
8 Robins SC and Rigby MJ (1995) Electronic health records as a key to objective health
care needs assessment beyond the hospital boundary. In: RA Greenes, HE Peterson and
DJ Protti. Medinfo ’95 ^ Proceedings of the Eighth World Congress on Medical Informatics,
Vancouver, 23^27 July 1995. Healthcare Computing & Communications Canada Inc.,
Alberta.
9 Rigby MJ and Robins SC (1996) Building healthcare delivery systems, management and
information around the human facets. In: J Brender, JP Christensen and J-P Scherrer
et al. (eds) Medical Informatics Europe ’96 ^ human facets in information technologies.
IOS Press, Amsterdam.
10 Rigby M (1994) Patient-focused hospitals or person-focused healthcare? Health Services
Review, X: 8^10.
11 Noyes P (1991) Child Abuse ^ a study of inquiry reports 1980^1989. HMSO, London.
12 Rigby MJ and Nolder D (1994) Lessons from a child health system on opportunities and
threats to quality from networked record systems. In: F Roger-France, J Noothoven van
Goor and K Staehr-Johansen (eds) Case-based Telematic Systems Towards Equity in Health
Care (studies in Health Technology and Informatics Vol. 14). IOS Press, Amsterdam.
 Information as the patient’s advocate 67

13 NHS Executive (1998) Information for Health. Department of Health, London.

14 Rigby M and Robins S (1995) Practical success of an electronic patient record system in
community care: a manifestation of the vision and discussion of the issues. In: J Runnen-
berg J (ed) The Future of Patient Records ^ care for records for care (Proceedings of the 8th
European Health Records Conference, 21^24 May 1995, Maastricht). Dutch Associa-
tion for Medical Records Administration, Hoorn.
15 Robins SC and Rigby MJ (1995) Opportunities for community services provided by inte-
grated electronic patient records and critical success factors for achieving this in practice.
In: J van der Lei and and WPA Beckers. AMICE 95 Proceedings ^ Strategic Alliances between
Patient Documentation and Medical Informatics. VMBI (Dutch Society for Medical Infor-
matics), Amsterdam.
16 Bullas S and Ariotti D (2002) Information for Managing Healthcare Resources (Harnessing
Health Information Series No. 6). Radcli¡e Medical Press, Oxford.
CHAPTER 6

Information for practice

improvement
John G Williams

Introduction
In an ideal world, clinicians delivering care to the highest possible standards will
be fully informed about the patients who are the subjects of their care. They
will have immediate access to current evidence-based wisdom as to what is the
best route to diagnosis and cure or alleviation of the condition presented. They
will also know details of all the resources, facilities and expertise available to
support them in achieving this, and how to access these quickly and appropri-
ately. Clinicians will have knowledge and insight into the extent and limitations
of their own expertise, and will be able to monitor their own performance regu-
larly. Lastly, they will be able to plan and implement sensible changes to their
own and their teams’ practice, based on knowledge of demand, and the success
and weaknesses of the service.
Patients are the prime source of information about themselves, but recall
and accuracy deteriorate rapidly with time, and probably the only truly reliable
information from the rational patient will describe current problems and com-
plaints. The primary purpose of the patient record is to document details of
history and care that will be of use in the immediate or long-term future.
Present record systems fail to meet this primary purpose,1 for many reasons.
Most records are paper-based, and are thus often not available, particularly if
needed in more than one place at a time. Paper records grow larger, and they
deteriorate. Each provider organisation and professional group has its own
system. Surprisingly, in hospitals there is no statutory requirement governing
the structure or content of medical records, and diversity and lack of structure
or organisation are commonplace. The current concept of the patient record, as
a provider, speciality- and episode-based journal, is also £awed, preventing the
easy generation of a clear longitudinal picture of patients’ illness or health.
70 Vision and value in health information

The ideal world envisages a simple means of capturing and storing data about
patients, yielding information that is immediately available when needed, and
presentable in a variety of forms, appropriate to the context of care. It is believed
that electronic capture of data in both structured and free text form will enable
this,2 but the technical, cultural and organisational problems of achieving this
are formidable. From the technical point of view, a generic approach to elec-
tronic clinical system development has long been advocated,3 but progress is
inhibited by short-term, speciality-speci¢c requirements. It is being further
slowed by the need to monitor the implementation of condition-focused national
service frameworks that are generating their own, condition-speci¢c data sets
and data collection processes.4,5

The clinician’s perspective

From the clinician’s perspective, one of the most di⁄cult changes in the move
towards electronic records will be the development of the cognitive agility and
skills needed to organise and document clinical facts to a level of detail and
accuracy that has hitherto been uncommon.6 It will mean adherence to a
common language of health,7 agreed and understood by all professionals, and
including common headings for the organisation of the record.8 There will also
need to be an integration of professional perspectives, and a process for extrac-
tion of summary facts of importance for future care. Compartmentalisation by
sectors of care will also have to disappear, so that an overall picture of the
health and social care history and needs of the patient is obtained. Needless to
say this raises important issues of con¢dentiality and security of information,
which are further inhibiting progress.9 It is to be hoped that theoretical con-
straints will be tempered with reality in the future.10
In an era of long overdue and increasing patient empowerment,11 a change in
culture that gives greater recognition to patients’ information needs, and their
wishes and preferences with regard to treatment, needs to be supported by
an appropriate technical and information infrastructure. This must ensure
that patients’ views and preferences are made known to those who look after
them, especially during times when patients may not be capable of expres-
sing them. Similarly, the impact of care needs to be addressed more rigorously,
by use of fully validated patient-focused outcome measures that are routinely
collected and monitored during their care.12
Clinicians need information to ensure that they are up to date, delivering clin-
ical management that is proven by research to be e¡ective, or if this evidence is
not available, equates with the consensus view of experts in the area.13 Clini-
cians cannot possibly keep abreast of all the literature relevant to their clinical
practice without help, and an authoritative synthesis of critically appraised
 Information for practice improvement 71

evidence on best practice must be made available at the point of care. Thus, col-
lation of the knowledge base is required, with its delivery to clinicians where
and when needed. The National electronic Library for Health14 is a visionary
initiative that aims to pull together reliable knowledge in a form that is easily
accessible to both clinicians and patients. Ensuring that it is accessible at the
point of care: in the surgery, on the wards, in outpatients, even in the home,
remains a challenge for those who are developing the communications infra-
structure for the NHS and it is hoped that progress will now be rapid. Achieving
this vision demands more than the development of a sophisticated technical
infrastructure. Sifting, appraising and reviewing the vast knowledge base of
healthcare has been catalysed by the Cochrane collaboration15 and a cultural
shift towards evidence-based medicine,16 but its presentation to busy clinicians
who have inadequate time to talk to patients, let alone keep up to date, remains
a problem. The solution is believed to lie in producing trustworthy advice in the
form of evidence-based guidelines,17 but there remains a need to ensure their
uptake by those who deliver care. This depends on clinicians being involved
in the production of local guidelines and requires active local dissemination,
education and support,18 whether guidelines are made available in paper or
electronic form.19 Thus, organisational development, culture change, educa-
tion and training all need to move forward, hand in hand with technology.

Evidence-based medicine
Evidence-based medicine has been de¢ned as a requirement for individual
clinicians to ‘engage in life-long, self-directed learning, so that they remain
continually up-to-date with research evidence and o¡er their patients the
best available practice according to that evidence’,16 but challenged by those
who champion more medicine-based evidence, which re£ects more appropri-
ately the diversity and challenges of care at the ‘coal face’.20 This concept of
evidence-based medicine also fails to cope with the complexity of some interven-
tions, where accumulation of evidence is di⁄cult and where managing change
requires discussion of options and consensus-building among various stake-
holders.21 These di⁄culties are highlighted by the complexities of national
service frameworks for heterogenous diseases, such as cancer,22 and situations,
such as the management of elderly people.23 The explosion of information avail-
able in electronic form adds further concerns about its validity, and measures of
quality are needed.
Information about resources, expertise and access to healthcare is surpris-
ingly sparse at both local and national levels. This information is needed by
managers, clinicians and patients, and it must be accurate, updated regularly
and available at the point of care. The scarcity of such information implies that
72 Vision and value in health information

the development of organisational record systems seems not to have reached

the consciousness of the health service as yet.
The primary purpose of the patient record is to inform and support the man-
agement of the individual patient. Secondary purposes include provision of
information about activity and quality, by extraction from the record data that
can be aggregated. Local uses include research, audit and service planning.
National uses currently include: identi¢cation of activity and resource use to
inform decisions on the level and distribution of funds; the generation of orga-
nisational and clinical performance tables; and monitoring trends in diseases
and healthcare provision.24 The inadequacy of central returns in meeting
these purposes has long been known.
Edith K˛rner described how ‘inaccuracy, lack of timeliness and certain inher-
ent defects form the core of criticism levelled against NHS information’, and this
was taken as the main premise for her in-depth study of health services informa-
tion in 1982.25 The intention of her recommendations was the promotion of
better use of statistical information and healthcare intelligence by those who
manage healthcare resources.
Although Mrs K˛rner left no stone unturned in her e¡orts to rationalise infor-
mation services, she worked in an era when clinicians were thought to have no
need for information derived from aggregate data about their practice, and thus
had no perceived role in either its collection or validation. Since her report, the
need for information by clinicians has changed, but the validity of hospital epi-
sode statistics is deplorable26 and culture has remained largely static. Thus,
regrettably, clinicians continue to play very little part in the collection or vali-
dation of central returns, yet more than ever they need timely, accurate and
understandable information about their practice. Already, central returns are
being used to inspect clinical performance, and it is proposed that this will
develop further with the introduction of appraisal and revalidation.27 Thus
the vision is of timely and accurate information that is of value both to clini-
cians, who deliver the service, and to those who manage it, derived from data
held in records used in day-to-day clinical practice. For such data to be compar-
able across the NHS, generic systems3 and common clinical7,8 and technical
standards will be essential.

A vision for the future

Clinicians of the future will receive regular reports of their activities, informing
them of the numbers of patients they have treated, the presenting complaints,
the diagnoses and the actions they have taken. These reports will also tell
clinicians of the e¡ects of their actions on patients, both in terms of change in
test results, disease activity scores, adverse events and mortality, and also from
the patients’ perspective in terms of symptom change and quality of life scores.
 Information for practice improvement 73

This patient focus implies greater patient involvement in data collection,28 and
greater access by patients to computer terminals, both in waiting areas and on
wards, in primary and secondary care, where they could review data about
their illness and its treatment, and could complete patient-focused question-
naires that would enable monitoring of outcome.
Well-informed clinicians will rapidly be able to detect areas of practice that
need attention. They will be able to map referral rates to activity, so that they
can predict a rising waiting list and deal with it by recon¢guring their services.
They will monitor complications and adverse events, to pick up trends at an
early stage. They will be able to see whether their practice is deviating from
the wisdom held in guidelines, and if they are justi¢ed in doing so by the details
of the individual case. Clinicians will be able to identify their own educational
needs, and monitor the experience and results of those they train. They will be
able to search for new knowledge by analysing the details of their own practice,
and could participate in rigorously designed randomised controlled trials by
providing routinely collected data without the need for additional purpose-
designed data collection.29 They will be able to use information about their
practice to plan new services and justify investment in them.30 Clinicians will
also be able to involve individual patients in decisions about their care, enabling
them to base decisions on detailed knowledge of the clinician’s experience, and
likely outcomes or side e¡ects of treatment.11
This vision for better information, to underpin the management of both
patients and the service, applies equally to primary, secondary and community
care. In primary care disparate organisations and accountability of teams pre-
sent their own challenges,31 and the need to blur the interface between these
sectors is already clear.32
How close are we to achieving this vision? Individual patient records remain a
shambles in secondary and community care: largely paper-based, disorganised
and useless as a ready source of secondary data.1,26 Progress has been made
towards computerised records in primary care, but these remain largely unipro-
fessional, unstandardised and of limited value in the generation of secondary
information. In hospitals, automation of the processes that surround clinical
records has already achieved great bene¢ts to clinicians in many areas by
enabling on-line ordering of tests and viewing of results, electronic prescribing,
automatic communication with other professionals and scheduling of proce-
dures in discussion with patients. Capturing structured clinical data, in a form
that is easily analysed and will yield information that is comparable across the
NHS, remains more elusive because it poses far greater challenges that require
the de¢nition of standards, education, training and culture change. The health
service has not yet moved signi¢cantly from a professional culture that is domi-
nated by clinical tunnel vision, encouraging professionals to see only the needs
of the individual patients in front of them, and unable to set this in the wider
context and con£icts of the NHS. The service is also so under-resourced that
74 Vision and value in health information

clinicians are not given the time to make the cognitive e¡ort that will enable
them to be more precise in their recording of diagnoses and procedures, as well
as the discrete capture of data on other clinical parameters, such as symptoms,
signs, problems and complications.
So much for where we want to be, where we are now and the obstructions to
getting there. How can the NHS move forward?

Moving the NHS forward

First, clinicians need to be involved ^ in all aspects of change. Information
strategies for the NHS have repeatedly identi¢ed the need for operational
systems, based on professional^patient interaction, but have each time failed to
understand what this meant in terms of system design and implementation, and
clinical involvement in information management. Regrettably, Edith K˛rner
missed the opportunity to point out the futility of collecting clinical data in
central returns without any involvement of clinicians in the process or in data
validation, although she did make clear her view that her data sets were a reluc-
tant interim measure. She was insistent that further developmental work in
information culture and application was needed, but this was not progressed
by central government, much to her disillusionment33 (see Chapter 1). The con-
tinuing problems with hospital episode statistics are probably a direct result of
this, and should be an early target for change. Giving clinicians ownership and
responsibility for paper records and secondary data extraction would start the
culture change that is needed if electronic solutions are to work in the future.33
It will require resources, but these are already deployed in a discredited network
to generate invalid central returns that are so mistrusted that new networks
are being created to support national service frameworks for cancer, heart
disease, stroke and care of elderly people. It will also require education and train-
ing, building on the progress made in the late 1990s, to identify the learning
needs in this area.34 Giving clinicians regular feedback would ensure that their
involvement was productive and would engage them further in the process.
To ensure compatibility and comparability across professions, sectors, organi-
sations and sites, standards for data and systems are needed.
Much has already been achieved in the development of clinical standards for
medical records. The clinical terms project involved more than 2000 clinicians
from all disciplines in developing a thesaurus of terms to describe all aspects of
health and social care.35 The delay in translating this clinical thesaurus into a
useable coding infrastructure is frustrating, but it is hoped that SNOMED-CT,36
a collaboration between the NHS Information Authority and the American
College of Pathologists, will eventually realise this large piece of the vision
jigsaw. It is also hoped that the aborted work of the headings project8 to de¢ne
 Information for practice improvement 75

and evaluate common headings acceptable to all professionals in all healthcare

domains, will be the foundation for an evidence-based record structure, such as
that proposed by the Royal College of Physicians.37 The episodes of care pro-
ject32 described lucidly the severe limitations of the current concepts of discrete
consultant- and sector-based episodes, and the far-reaching implications of this
must be heeded.
Failure to identify the required functions of the patient record in an electronic
environment has long been an inhibition to progress, and has allowed the vision
embodied in national strategies to be submerged by short-term managerial
needs in terms of resources and contracting. The Academy of Royal Colleges
sought to redress this situation by preparing a speci¢cation of requirements for
clinical systems in secondary care.38 This process has utilised wide medical
consultation and should catalyse debate with other professionals so that multi-
professional clinical needs are truly met in the next generation of systems devel-
opment. The needs of primary care are being addressed through the Primary
Care Information Modernisation Programme,39 but community care remains
the poor relation.40 There is also outstanding work in identifying the minimum
data that should be collected routinely to enable monitoring of the quality and
quantity of NHS activity without additional data collection processes.
The National Programme for IT in England41 and Informing Health Care in
Wales42 both envisage major progress over the next few years. It is likely that
this will be patchy, but if the design of the quilt is clear the patches can ¢t
together without gaps, and if the seamstresses are skilled, the result will be
worth waiting for. Above all, it needs partnership ^ between politicians, who
can listen to those who deliver the service; between managers, who acknowl-
edge the primacy of individual patient care in determining the information
agenda; and between clinicians, who are prepared not only to change their
ways of working but to take on new responsibilities, and to work towards a
goal that may not give them immediate bene¢ts.

References
1 Audit Commission (1996) Setting the Record Straight. HMSO, London.
2 NHS Executive (1998) Information for Health. An information strategy for the modern NHS
1998^2005. NHS Executive, Leeds.
3 Williams JG, Morgan JM, Howlett PJ et al. (1993) Let there be light. British Journal of
Health Care Computing 10: 30^2.
4 Cancer Dataset Project (2000) Consultation document for cancer dataset version 1.0.
NHS Information Authority, Birmingham.
5 National Service Framework for Ischaemic Heart Disease. (www.doh.gov.uk/nsf/coronary.
htm)
76 Vision and value in health information

6 Williams JG and Morgan JM (1995) The clinician^information interface. In: RA Greenes,

HE Peterson and DJ Protti (eds) Medinfo ’95 Proceedings of the Eighth World Congress on
Medical Informatics. International Medical Informatics Association.
7 Swayne J (1993) A common language of care? Journal of Interprofessional Care 7: 29^31.
8 NHS Information Authority (2000) Headings for Communicating Clinical Information. NHS
Information Authority, Birmingham.
9 France E (1997) The impact of the law on patient con¢dentiality: likely consequences.
In: The British Computer Society Health Informatics Specialist Group. Current perspectives in
health care computing part II. BJHC Ltd, London.
10 Roberts R, Thomas J, Rigby MJ et al. (1997) Practical protection of con¢dentiality in
acute health care In: R Anderson (ed) Personal Medical Information: security, engineering
and ethics, Proceedings of Personal Information Workshop. Cambridge, June 1996. Springer-
Verlag, Berlin. pp. 67^78.
11 Elwyn G (2001) Shared Decision Making: patient involvement in clinical practice. PhD Thesis.
University of Nijmegen, Nijmegen.
12 Guyatt GH, Feeny DH and Patrick DL (1993) Measuring health-related quality of life.
Annals of Internal Medicine 118: 622^9.
13 Deighan M and Hitch S (eds) (1995) Clinical E¡ectiveness from Guidelines to Cost-e¡ective
Practice. Department of Health, London.
14 National electronic Library for Health. (www.nelh.nhs.uk)
15 The Cochrane Library. (www.cochrane.co.uk)
16 Sackett DL and Rosenberg WMC (1995) The need for evidence-based medicine. Journal of
the Royal Society of Medicine 88: 620^4.
17 Centre for Reviews and Dissemination and Nu⁄eld Institute for Health (1994) Imple-
menting clinical practice guidelines: can guidelines be used to improve clinical practice?
E¡ective Health Care 1^12.
18 Oxman AD, Thomson MA, Davis DA et al. (1995) No magic bullets: a systematic review
of 102 trials of interventions to improve professional practice. Canadian Medical Associa-
tion Journal 153: 1423^31.
19 School of Postgraduate Studies in Medical and Health Care (2001) An Assessment of an
On-line Clinical Information Service at Morriston Hospital (MOCIS). Final report to the Nuf-
¢eld Trust. University of Wales, Swansea.
20 Knottnerus JA and Dinant GJ (1997) Medicine based evidence: a prerequisite for evi-
dence-based medicine. BMJ 315: 1109^10.
21 School of Postgraduate Studies (2001) A Practical Guide to Achieving Clinical E¡ectiveness.
University of Wales, Swansea. (www.rcplondon.ac.uk/college/hiu/educationmaterial.
htm)
22 Calman K and Hine D (1995) A Policy Framework for Commissioning Cancer Services.
Department of Health, London.
23 Department of Health. National Service Framework for Older People. (www.doh.gov.uk/
nsf/olderpeople.htm)
 Information for practice improvement 77

24 NHS Executive (1998) Central Data Collections from the NHS. Health Service Circular
1998/054. NHS Executive, Leeds.
25 K˛rner E (1987) First Report to the Secretary of State of the Steering Group on Health Services
Information. HMSO, London.
26 Williams JG and Mann RY (2002) Hospital episode statistics: time for clinicians to get
involved? Clinical Medicine 2: 34^7.
27 Maynard A and Bloor K (2001) Reforming the contract of UK consultants. BMJ
322: 541^3.
28 National Assembly for Wales. Sharing Clinical Information in the Primary Care Team (SCI-
PiCT). Cardi¡, National Assembly for Wales, Cardi¡. (Unpublished).
29 Williams JG, Cheung WY, Cohen D et al. (2003) Can randomised trials rely on existing
electronic data? A feasibility study to explore the value of routine data in health technol-
ogy assessment. Health Technology Assessment 7(26).
30 Williams JG (1999) The use of clinical information to help develop new services in a dis-
trict general hospital. International Journal of Medical Informatics 56: 151^9.
31 Rigby M, Roberts R, Williams J et al. (1998) Integrated record keeping as an essential
aspect of a primary care led service. BMJ 317: 579^82.
32 Rigby MJ and Williams JG (2000) De¢nition of Episodes Project. Collaborative Exercise: ¢nal
report. NHS Information Authority, Birmingham.
33 Williams JG and Severs MP (1998) Physicians in the information age: are we keeping
pace? Journal of the Royal College of Physicians London 32: 193^4.
34 Severs MP and Pearson C (1999) Learning to Manage Health Information. A theme for clin-
ical education. NHSE, Bristol.
35 Williams JG, Ford DV and Yapp TR (1993) Capturing clinical activity: coming to terms
with information. Gut 34: 1651^2.
36 NHS Information Authority and American College of Pathologists (2000) SNOMED
Clinical Terms. A Global Leader in Healthcare Terminology. NHS Information Authority,
Birmingham.
37 www.rcplondon.ac.uk/college/hiu/recordsstandards
38 Academy of Medical Royal Colleges Information Group. Speci¢cation of Core Require-
ments for Clinical Information Systems in Support of Secondary Care. (www.aomrc.org.uk/
ACIG.htm)
39 Primary Care Information Modernisation Programme. (www.doh.gov.uk/ipu/whatnew/
itevent/tables/infoauditinprimary care.htm)
40 Audit Commission (1997) Comparing Notes: a study of information management in commu-
nity trusts. Audit Commission, London.
41 Department of Health. National Programme for IT in the NHS. http://www.doh.gov.uk/
ipu/programme/index.htm
42 Informing Healthcare: transforming healthcare using information and IT (2003) NHS
Wales, Welsh Assembly Government.
CHAPTER 7

Information for good

governance
Ellie Scrivens

Introduction
NHS organisations produce vast amounts of information for many purposes
ranging from central monitoring requirements through to that used for moni-
toring individual activities. Much of this information is produced for others and
little ¢nds its way into the hands of boards for their deliberations. Indeed, boards
often struggle to know which data sources to use in order to judge whether their
organisations are well managed. However, if they are to meet the require-
ments now placed upon them by HM Treasury to manage key risks facing their
organisations, they need to develop a structured approach to select e¡ective
information for assessing whether they have appropriate risk management
and controls in place.1,2
Edith K˛rner began the move to investing in information-based manage-
ment systems that would provide boards with the information to assess the
e¡ectiveness of their management activities. The original data sets were pro-
ducts of the understanding of consensus management at that time. With the
increasing complexity of the social, political and policy environments in which
all organisations, public and private sector, ¢nd themselves today, there has
been universal recognition of the need to ensure that organisations manage
change and risk in a structured and systematic manner. The controls assurance
agenda recognises the new information needs of boards in an increasingly com-
plex world.
In 2002, the Department of Health (DoH) produced a mandatory set of report-
ing requirements for the general management of all NHS organisations. Entitled
The Governance Standard, the reporting requirements outlined a comprehensive
business planning approach for use by boards in the planning and manage-
ment of health services.3 Compliance with The Governance Standard (see box)
80 Vision and value in health information

would be audited by the internal audit service and substantiated by the external
audit service.

The Governance Standard

. There are clear accountability arrangements in place throughout the

organisation.
. The board identi¢es the needs of stakeholders on an ongoing basis and
determines a set of key objectives and outcomes for meeting these needs,
including how it meets its duty of quality.
. The board ensures that there are proper processes in place to meet the
organisation’s objectives and secure delivery of outcomes.
. The organisation is capable of meeting its objectives and delivering appro-
priate outcomes.
. The organisation learns and improves its performance through contin-
uous monitoring and review of the systems and processes in place for
meeting its objectives and delivering appropriate outcomes.
. The board ensures that there are proper and independent assurances
given on the soundness and e¡ectiveness of the systems and processes
in place for meeting its objective and delivering appropriate outcomes.
. The board can demonstrate that it has done its reasonable best to achieve
its objectives and outcomes, including maintenance of a sound and e¡ec-
tive system of internal control.

Although NHS organisations had been exhorted to apply the principles of good
business planning to their activities for many decades, this was the ¢rst time
that these principles were a mandatory and full requirement on NHS boards.
The arguments behind the genesis of this standard drew on the logic of good
governance being promoted simultaneously across the private and public sec-
tors. The logic was derived from principles of good governance, that is, the way
in which boards should organise their work to demonstrate accountability in
the use of, in the case of the private company, shareholder funds, and in the
public sector, taxpayers’ money.
The Governance Standard was the key stone of a series of standards promoting
risk management in the NHS as part of a programme known as ‘controls assur-
ance’.3 This had begun in 1994 with the introduction of a number of pilot
studies to improve risk management in healthcare organisations. At the same
time, a set of standards describing the ¢nancial requirements placed on
NHS organisations was produced. Eighteen organisational standards were
added in 1998, each dealing with a speci¢c aspect of responsibility for the safety
 Information for good governance 81

of patients and sta¡, where failure to ensure that the requirements of the
standards were met could result in harm being caused to patients or to sta¡, or
sanctions being applied to the NHS organisation for failing to act in a responsi-
ble manner towards its patients or sta¡ through not complying with legislation
or other regulations to protect them. In addition, there was a risk management
system standard that provided the framework for the management of risks
across the whole organisation, and later, a ¢nancial management standard
and the governance standard. The ¢nance standard pulled together the original
standards, and covered areas relating to ¢nancial management, which if
ignored could put the organisation at risk. The governance standard provided
the strategic overview for the management of business risks across the whole
organisation.

Genesis of The Governance Standard

Concerns about governance gained momentum in the 1990s, after notable
exposures of questionable business practices such as those associated with
Robert Maxwell, Barings Bank and others. The conduct of business by private-
sector companies came under public scrutiny and, at a time when governments
were encouraging the participation of members of the public in share owner-
ship, there was a realisation that companies had to be prepared to demonstrate
that they were acting in the best interests of their shareholders and not in
their own self-interest. In response, many national accounting bodies, either
independently or with their governments, devised requirements for board behav-
iour. The sentiments of the reports in which these requirements were stated
shared a number of similar principles, but the most striking was a requirement
that boards should place priority on managing the key risks which threatened
their business.
The UK version of this report, colloquially referred to as ‘the Turnbull report’
after its chairman, Nigel Turnbull, required businesses to assure their share-
holders that e¡ective internal control systems were in place within their
organisations.4 The recommendations from this report were incorporated into
the Combined Code of the Stock Exchange in 2000 and set the requirements for
UK listed public companies, laid down bv the Financial Services Authority.
Good governance, it was held, requires e¡ective control systems.
So what is an e¡ective control system? Control is the processes in place that
an organisation uses to achieve its objectives. In general there are considered to
be two sorts of controls within an organisation. ‘Hard controls’ are the rules
and regulations laid down within the organisation to prescribe behaviours.
‘Soft controls’ are the culture of the organisation, which recognises that there
82 Vision and value in health information

are generally accepted behaviours in place to ensure that the people working
within the organisation are clear about the risks that the organisation ^ and
therefore its board ^ is prepared to tolerate and deal with itself. Risks are there-
fore the currency of control. Uncontrolled risks help to identify failings in
control systems. Managed risks help to identify how well the controls are man-
aged within an organisation. Boards are therefore encouraged to study the
management and the treatment of risks by their organisations to ensure that
the control systems are in place and functioning e¡ectively.
Risk management, therefore, is inextricably linked to the requirement to
develop sound processes of internal control, thereby enabling organisations
to sign the assurance statement on internal control:

Risk management grew out of internal control good practice, and the
systems it promotes are designed to give assurance to the Board and key
stakeholders that management processes are e¡ective and coherent, and
that levels of responsibility and accountability are clearly de¢ned.5

The Turnbull report, with its emphasis on internal control, required private
companies to embed risk management throughout their organisations. The UK
Treasury adopted this approach for its government departments and in 2000
required that all government departments develop a framework for risk manage-
ment, and also, sign what was referred to as a ‘Statement of Internal Control’.6
Happily, this development coincided with another government policy to pro-
mote modernisation and innovation in the public sector. A series of disastrous
central government policy implementations led to recognition of the need to
ensure that central government departments were capable of undertaking
strategic planning by use of risk management. The unsuccessful launch of the
Millennium Dome; the failure of the Passport O⁄ce to achieve its targets for
turnaround in issuing new passports; and others led to the realisation that risk
management was vitally important to the development of new public services.7

Too often Central Government does not assess the risk and walks o¡ the end
of a cli¡. It does not consider where it is going and when it marches forward
it does not manage the risk. Too often, there are no pilot projects, no train-
ing and no contingency plans.8

However, a review by the Cabinet O⁄ce revealed that departments had focused
more on better-known risks, such as safety hazards and risks associated
with scienti¢c uncertainty, than on the strategic risks of running their core
businesses.9 A National Audit O⁄ce survey revealed similar ¢ndings, in that
government departments focused on minimising ¢nancial loss or preventing
impropriety rather than on the achievement of broad policy objectives.7 The
 Information for good governance 83

result was a concerted move by the Cabinet O⁄ce to put pressure on govern-
ment departments to focus more on the achievement of their objectives by
examining the risks to achieving the objectives. In short, the emphasis was
moving from simply looking at the management of risks to safety, and moving
towards the management of risks to the achievement of the wider policy objec-
tives set for government departments.
A review of risk management by the Public Accounts Committee arrived at a
similar conclusion: ‘risk management is about ensuring the achievement of
outputs and outcomes, and having reliable contingency arrangements to deal
with the unexpected which might put service delivery at risk’.9 But in addition,
it was emphasised that risk-taking was a positive and forward-looking process
that should be encouraged. Risk management is about’ ‘changing behaviour
and having processes in place which support risk-taking and innovation’.
In the Department of Health, the various messages on broadening the scope of
risk management found resonance in the development of the controls assurance
project. Begun in 1994, with the advent of improved scrutiny of decision-
making through the introduction of audit committees, the controls assurance
agenda began to grow by recognising the need for improved patient safety.10
This was supported by developing of standards to improve the healthcare
environment through listing of hard controls covering areas such as ¢re, build-
ings, plant and equipment, medical devices, medicines management, etc., and
through the development of a standard to cover complaints, incidents or adverse
events reporting, etc. The Department of Health took this thinking further by
introducing the National Patient Safety Agency in 2001, which was to promote
the reporting of incidents and near-misses that could be of use in protecting
patients and sta¡.11 Risks to health and safety, in their broadest sense, were
being addressed. During the same period the risks associated with soft controls
also began to develop, in particular those concerning the work of professionals
employed within the NHS. Clinical governance, which had as its emphasis the
cultural and behavioural aspects of healthcare delivery, emerged to promote
greater awareness by clinical sta¡ of risks of clinical activity ^ to patients and
to themselves. The third stage of development arrived with the implementation
of Turnbull in the requirements for boards to pull non-clinical and clinical con-
trols, hard and soft together, into a single uni¢ed system of internal control that
would form the bedrock of governance in the NHS.
The uni¢ed system of internal control comprises a number of separate and
distinct control systems that thread through an organisation. These all contri-
bute to a single management system, leading to the board, which encompasses
all risk-based activities in the NHS ^ ranging from clinical skills through to
the management of estates ^ all ultimately focusing on the overall goals of the
system of healthcare set through the national priorities for healthcare issued by
central government.
84 Vision and value in health information

A call to accountability
Central government requires increased accountability, but there is little agree-
ment as to how this is to be achieved. Under the most recent government
initiative for the NHS, enshrined in the document, Shifting the Balance of Power
within the NHS: securing delivery,12 there is a commitment to allow NHS organi-
sations to take greater responsibility for controlling their functioning. In this
model the role of the centre becomes one of monitoring performance and of
ensuring that objectives are achieved. The model has been speci¢ed as central
standards and policies to be set by the relevant government department and the
Department of Health, and the monitoring to be conducted by a new inspection
and audit agency.13 Assistance with implementation will be in the hands of the
Modernisation Agency and various other agencies.
However, one problem to be tackled within the new accountability frame-
work is the degree of speci¢cation built into the design of both standards and
monitoring procedures. It is possible to write standards that are so speci¢c,
they dictate precisely how organisations should be structured and should oper-
ate. This constrains completely the freedom of local management to determine
the best use of services and resources. On the other hand, too few speci¢c stan-
dards allows local service development to become unequal, and inequity of
service provision £ourishes. The appropriate level of control is a very ¢ne bal-
ance that needs to be struck by central government implementation at local
levels. The search is for what has been described as intelligent accountability:
Intelligent accountability requires more attention to good governance and
fewer fantasies about total control. Good governance is possible only if
institutions are allowed some margin for self-governance of a form appro-
priate to their particular tasks, within a framework of ¢nancial and other
reporting.14
The challenge, therefore, is to use the management of risks, as enshrined in
good governance, to ensure that organisations can deliver the services that
the public want, in the way the public wants to receive them ^ that is, safely
and securely. Accountability, therefore, is about providing services that the
general public can trust, wholly and without thinking. And trust requires that
the public, both as taxpayers and consumers of services, is not exposed to risk.
The range of risks to be addressed is of course very wide ^ most obviously
risks associated with ¢nancial decisions but including risks to reputation,
environmental risks, risks to health and safety, anything in fact which can
adversely impact on business continuity and reduce shareholder value.15
By espousing Turnbull, and previous reports on corporate governance includ-
ing Cadbury, Greenbury and Hampel,3,16 ^18 the government has accepted the
 Information for good governance 85

notion of ‘holistic business risk management’ and is commending this as a set of

sound principles for the corporate governance of risk to deliver the account-
ability needed within the process of modernising public services.

Information requirements
O’Neill14 has stated that:

Serious and e¡ective accountability, I believe, needs to concentrate on good

governance, on obligations to tell the truth and needs to seek intelligent
accountability. I think it has to fantasise much less about herculean micro-
management by means of performance indicators or total transparency.
If we want a culture of public service, professionals and public servants
must in the end be free to serve the public rather than their paymasters.

Intelligent accountability requires that organisations can manage in such a

way as to promote trust between members of the public and the management
that provides services to them through the appropriate management of risks.
However, although there is a natural desire to live in a ‘zero-risk’ society, it is
not possible to make every activity risk-free. Therefore, society needs to ¢nd a
mechanism for balancing progressive action with appropriate or acceptable
levels of risk. In some cases, individuals must be allowed to determine the
extent of risk to which they are exposed. In other cases, individuals are taking
risks that can be prevented with relative ease. In the former case, individuals
who require invasive clinical interventions must themselves be responsible for
assessing risks. Hence the need for appropriate patient information and consent
procedures in clinical care. In the latter case, wearing a seat belt is a decision
imposed by wider society on those who choose to travel in cars. Society has to
determine the extent to which interfering in the lives individuals’ exposure
to risk is acceptable. Equally, organisations have to decide how much risk they
expose individuals to in the course of their actions, and governments have to
judge where and when controls should be put on organisations to constrain
their risk decisions. An organisation has to determine the extent to which it
exposes its sta¡ and its customers to certain risks, for example the possibility
of legionella micro-organisms in hospital water tanks. The organisation has to
determine the appropriate level of monitoring and maintenance of its water
system. The government, however, chooses to regulate the amount of air pollu-
tion produced. There is a necessary balance between what is called ‘acceptable
risk’ and the level of control placed upon the actions of all levels of society, from
organisations to work units, individual sta¡, individual consumers. In some
cases it is easy to legislate for control, in others it is very di⁄cult, expensive
86 Vision and value in health information

and time-consuming to control the actions of every individual. In some cases,

the bene¢t of freedom of choice is felt to outweigh the risks associated with indi-
vidual actions. In others, economic bene¢ts outweigh the costs of social control.
Tobacco smoking is a very risky behaviour, but society has not chosen to
outlaw the production and sale of tobacco. Striking a balance between allowing
risks to be taken and controlling to protect against risks is a sensitive and politi-
cally complex issue.
Governance is about translating these complex choices to an organisational
level. All organisations have to be aware of regulations and controls that are
placed upon them. All organisations must be aware of the risks that are being
run for their sta¡, their consumers and their local communities by the actions
they undertake. All organisations must accept that they must decide, at every
level, where they are creating risks that might cause harm or fail to achieve the
purpose for which they were created. Every board has to take responsibility for
the risks which are inherent in the business that it runs.

Information for governance

So what is needed for a board to ful¢l its duties and obligations concerning intel-
ligent accountability and good governance? The board has to understand the
risk domains for which it is responsible. It has to understand the control systems
it has in place to manage the risks created by the very fact the organisation
exists and actions are conducted in its name. The board has to convey to all
sta¡ that they are responsible for ensuring that risks are not taken unnecessa-
rily in any part of the organisation. And the board has to assure itself that all
risks are managed in a way that will be acceptable to its stakeholders, its
public and its consumers. A model of internal control has been devised to pro-
vide a framework for NHS organisations to consider the risks that face them
(Figure 7.1).
For an NHS organisation, this means ensuring that all the regulations and
laws controlling various aspects of its activities are complied with. The complex-
ity of healthcare organisations means that these are many, covering a wide
range of domains from clinical practice to NHS estates. It also means that the
wide range of activities that are left open to the discretion of individual sta¡,
senior and junior, and those that are left open to individual patients and carers,
are also managed in such a way as to ensure risks are controlled appropriately.
Dirigible policies that attempt to control all the actions of every individual
are not possible and boards have to rely on the soft controls of professional
behaviour and sta¡ commitment to ensure that all actions taken within their
provenance are acceptable.
 Information for good governance 87

Internal control Objectives

Monitor, review, learn and improve

Accountability
Communication

Consultation
Processes
Culture

Culture
Audit
Capability

Outcomes

Stakeholders

Figure 7.1 A model of internal control.

Monitoring systems are needed to demonstrate that sta¡ are able to review the
risks they and their patients face continually, and can account intelligently for
the actions that they take when pursuing their professional objectives. Perfor-
mance measures are needed to inform boards that the right balance is being
reached between risk and control. Measures derived from internal management
systems, such as complaints and incident reporting, highlight failures within
the organisational processes that can suggest where unacceptable risks are
being taken. Key performance indicators that focus on the care delivered to
patients can show where there are unacceptable risks in the whole service
delivery system.19
Boards have to determine the nature and the culture of the system of internal
control they wish to see within their organisations. They have to be responsible,
that is, accountable for all failures of systems that are designed to ensure that
services are adequately and safely delivered to patients and equally that sta¡
are safe and able to work e¡ectively.
Controls assurance has created organisational standards to help NHS organi-
sations deal with some of the obvious risks they face when managing their
organisational environments. Other control systems, such as clinical govern-
ance, information governance and research governance, are being developed to
help to manage speci¢c areas known to present high risks to NHS organisations.
88 Vision and value in health information

The remainder of the organisation must determine itself through use of behav-
ioural processes. The most commonly used is known as ‘control self-assessment’,
which uses a total quality management-style approach to team assessment of
risks in the working environment. Teams are encouraged to review the risks
they face in their day-to-day work and to put in place actions and procedures
that will help them to manage risks.20 Risks that need long-term monitoring,
or that cannot be solved by the team, are reported to a risk register, which acts as
an inventory of risks across the organisation. The risk register can be reviewed
by senior management and either acted upon or passed up to higher levels of the
organisation for consideration. At each level, signi¢cant risks are passed up
the organisation and, if serious enough, reach the board for its consideration.
It is common for risk registers also to be used to quantify risk. Each risk is
given a probability of occurring (likelihood ) and a severity rating (how serious
the outcome would be for the organisation, with death usually being the most
serious). In this way, numerical values can be used to determine the degree of
signi¢cance attached to each risk by the people who identi¢ed the risk. Boards
are alerted to actions that are of concern within the organisation.
This information has to be put alongside the key performance indicators used
by boards to monitor their overall performance. Boards have to learn how to
judge the meaning attached to information that is presented, and to decide
how to take approriate actions. Board members, who will ultimately be held
responsible for the quality of the performance of the organisation, have to exer-
cise their judgement.
Central government, in the form of the Department of Health, can dictate key
performance indicators and can dictate the structure of risk management sys-
tems. But, in a system with devolved accountability, each and every board has
to take responsibility for the actions taken within its organisation. There are no
short-cuts to the information required for intelligent accountability. Boards
have to provide environments in which sta¡ can operate intelligently, and can
communicate freely to ensure that risks are managed. Boards have to have a
clear view of their overall direction and their objectives. And boards also have
to develop ways of selecting and judging information intelligently to promote a
relationship of trust with the general public and other stakeholders.
Edith K˛rner recognised the need for management information systems to
re£ect the environments in which healthcare organisations exist, and relate
to appropriately, if they are to provide modern healthcare. The NHS modernisa-
tion agenda requires an approach to management that is able to deal with a
more demanding public, aware of risks to health and with aspirations for the
public services that exist to serve them. The governance agenda has advanced
to meet these new demands, and in so doing, itself demands a new approach to
the use of information to improve the management of healthcare services.
For further information on controls assurance and corporate governance,
visit the controls assurance website (casu.org.uk).
 Information for good governance 89

Acknowledgement
The NHS Controls Assurance Support Unit is funded by the Department of
Health. The views expressed in this chapter are those of the author and not
of the Department of Health.

References
1 HM Treasury (2002) Audit and Accountability in Central Government. The Government’s
reponse to Lord Sharman’s report. HM Treasury, London.
2 Lord Sharman (2002) Holding to Account. HM Treasury, London.

3 Department of Health (2002) The Governance Standard. Department of Health, London.

4 The Internal Control Working Party of the Institute of Chartered Accountants in Eng-
land and Wales (1999) Internal Control: guidance for directors on the combined code: the
Turnbull report. Institute of Chartered Accountants, London.
5 Minister of Supply and Services Canada (1994) Quest for Quality in Canadian Health Care.
Health Canada, Canada.

6 HM Treasury (2000) Corporate Governance:statement oninternal control. DAO(GEN)13/ 0 00.

7 National Audit O⁄ce (2000) Modernising Government ^ how the NAO are responding.
National Audit O⁄ce, London.

8 Sir John Bourne (2002) Reinforcing Positive Approaches to Risk Management in Govern-
ment. Institute of Risk Management, London.
9 Public Accounts Committee (2001) Managing Risk in Government Departments ^ ¢rst
report. House of Commons, London.

10 Public Audit Forum (2001) Propriety and Audit in the Public Sector. 1. Public Audit
Forum, London.
11 Department of Health (2000) An Organisation with a Memory: report of an expert group on
learning from adverse events. The Stationery O⁄ce, London.
12 Department of Health (2001) Shifting the Balance of Power within the NHS: securing deliv-
ery. Department of Health, London.

13 Secretary of State (2002) Delivering The NHS Plan. Cm 5503. HMSO, London.
14 O’Neill O (2002) A Question of Trust, BBC Reith Lecture, London.
15 Bibbings R (2001) If it can happen . . . . OSH World. http://www.sheilapantry.com/
oshworld/focus/2001/200108.html
16 Cadbury A (1992) The London Stock Exchange. The ¢nancial aspects of corporate governance.
Business Science Press, London.
90 Vision and value in health information

17 Greenbury Committee (1995) Directors’ Remuneration. Report of a Study Group chaired

by Sir Richard Greenbury, London.
18 Hampel Committee (1998) Committee on Corporate Governance: ¢nal report. The Stock
Exchange, London.
19 Wilkes J (2001) How are we doing? Gasette 73: 1^4.
20 Department of Health (2001) Controls Assurance Project ^ guidance ¢rst principles.
Department of Health, London.
CHAPTER 8

Information for the assessment

of health outcomes
Azim Lakhani

Introduction
The Ko«rner reports represent an important stage in the continuing evolution of
health services information in the UK. This essay is based on the experience over
a decade of the Department of Health’s former Central Health Outcomes Unit,
and the current outsourced National Centre for Health Outcomes Development,
in the use of health services data to assess outcomes ^ experience in which the
vision of Edith Ko«rner has had a signi¢cant role.
The National Centre for Health Outcomes Development is a key source of
information on assessment of levels of health and outcomes of health interven-
tions, at individual, health authority, NHS hospital trust and local authority
levels for the UK National Health Service (NHS) and the government. It is based
jointly at the London School of Hygiene and Tropical Medicine, University of
London and the Institute of Health Sciences, University of Oxford. In 1993, the
Department of Health in England set up the Centre for Health Outcomes Unit to
co-ordinate a programme of work to develop methods and systems necessary to
assess health outcomes. Following the evolution of a substantial national pro-
gramme, the unit was contracted out in 1998 and became what is now the
National Centre for Health Outcomes Development.
The following is a re£ection on three aspects of health outcomes assessment
in the NHS:

. What can we do today in the context of what needs to be done?

. What contribution have the Ko«rner reports made to this?
. How can the lessons be applied to ¢ll in gaps?
92 Vision and value in health information

An approach to the assessment of

health outcomes
Health outcome has been de¢ned variously as either a health state at a point in
time, or a change over a period, or a result.1 The National Centre for Health Out-
comes Development uses the term ‘health’ broadly to include health-related
status and risk factors a¡ecting health. From the perspective of outcome as a
change, there may either be change or no change when change is either expec-
ted or not expected. Examples include: no improvement following treatment
with antibiotics for an infection when improvement is expected; or activities
of daily living of a disabled person not deteriorating. The National Centre
for Health Outcomes Development has previously de¢ned health outcomes as
changes in health, health-related status or risk factors a¡ecting health, or lack
of change when change is expected.2
We should recognise that change may occur either by design or through a
natural process. The former is more meaningful for performance assessment
and health outcome may thus be re¢ned further and presented as a ‘result’.
The Centre for Health Outcomes Unit de¢ned health outcome as ‘attributable
e¡ect of an intervention or lack of intervention on a previous health state’.1
Here, the relationship with the cause of the health state is important. A health
state may be the result of an intervention (for example, surgery) or may be due
to a known e¡ective intervention not being applied (for example, drug treat-
ment for high blood pressure).

Practical tools for complex concepts ^

an approach
The home page of the Department of Health’s website3 states simply ‘Our aim is
to improve health and well-being of people in England’. Assessing success in the
context of this aim is, however, no simple matter.2 Health outcome is often a
cumulative result of a variety of in£uences or services, either provided or not
provided, by a variety of organisations over time. The Department of Health is
one of many players. In addition to a direct role using its own resources, the
Department of Health also has a role in acting as an advocate for health and as
a partner with other organisations concerned with health. Any serious attempt
at monitoring the achievement of stated health goals should re£ect this more
complicated reality.
Figure 8.1 illustrates a way of looking at health outcomes as results, drawing
a variety of linked health states and services together to provide a summary
health outcome overview. It shows further categorisation of ‘health improve-
ment’ to re£ect more speci¢c aspects of outcome, that is, success in:
 Information for the assessment of health outcomes 93

. reducing level of risk to health

. reducing level of disease and impairment
. reducing adverse consequences of intervention
. reducing adverse consequences of lack of intervention
. restoring function and improving quality of life
. reducing premature deaths.

Health outcome objectives Potential interventions

Proactive interventions
to avoid risk
Remove/reduce exposure to Examples:
amenable determinants [Health promotion – better
of disease/ill health diet, exercise]
Examples:
Adults who are obese
[Blood cholesterol]
Timely detection of risk and
reactive interventions
to remove/reduce risk
Reduce incidence and prevalence Example:
of disease/ill health [Smoking cessation programme]
Examples:
High blood pressure
[Diabetes]
Timely interventions
to detect and treat disease
Reduce potentially avoidable
Examples:
adverse consequences – Treated and controlled high blood
severity, secondary disease and pressure
complications of treatment [Surgery rates in the context of need]
Examples:
Amputations among people with
diabetes
Incidence of stroke
[Complications following surgery] Late interventions
to minimise consequences
Examples:
Optimise function and quality of
[Emergency treatment after stroke]
life
Examples:
Timely discharge from hospital
[Quality of life of people with asthma]

Reduce premature death

Examples:
Deaths from potentially avoidable causes Note: Comparative data at national level
Deaths following treatment in hospital are not currently available for examples
Survival after diagnosis of cancer in [ ].

Figure 8.1 A way of looking at health outcomes as results.

94 Vision and value in health information

Similarly, the action needed to achieve such success is divided into more precise
categories:

. proactive interventions to avoid risk to health

. timely detection of risk and removal/reduction of existing risk to health
. timely interventions to detect and treat disease/ill health
. late interventions to minimise consequences of disease/ill health.

This way of presentation helps to clarify and highlight what we are trying to
achieve, covering positive aspects, such as improved quality of life, and negative
aspects, such as avoidable disease and ill health. It re£ects aspects such as clin-
ical signs and disease as well as those of more immediate concern to patients,
such as handicap and impact on quality of life. It shows how lack of action
may lead to a chain of potentially avoidable consequences, from risk to disease,
complications of disease, poor quality of life and premature death, and shows
the kinds of action that may alter these consequences for the better. It should
be noted that presentation of data in the context of such a ‘health outcomes
overview’ describes mismatching outcomes at a point in time. The data on risk
show risk to future health. The data on levels of disease and death show conse-
quences of missed opportunities to reduce risk (if applicable) in the past. It is also
not possible to draw direct inferences on ‘attributable e¡ects’ as there may be
di¡erent populations involved and a time lag between the interventions and
the outcomes. However, such an approach does bring disparate but related
items of information together in a meaningful way.

The current state of health outcomes

assessment in the NHS using routine data sets
Only a few of these aspects are covered currently by comparative indicators
available at national level. Both the Department of Health’s NHS Performance
Indicators4 and the Compendium of Clinical and Health Indicators 20015 contain
examples of health outcome indicators produced by use of existing health ser-
vices data, as follows:

. e¡ective action (hip replacements)

. avoidable adverse events (hospital re-admissions, diabetes complications,
respiratory infections)
. restoration of function (timely discharge from hospital)
. premature death (deaths after hospital treatment).

Many others are feasible (stroke admissions, gastroenteritis in children). The

clinical and health outcomes knowledge base2 shows how some 150 existing
 Information for the assessment of health outcomes 95

indicators with comparative data may be presented within 29 condition-speci¢c

(for example, cancer, diabetes) health outcome overviews, along the lines
shown in Figure 8.1. What these charts do show, however, is that coverage is
very patchy and has numerous gaps.
These concepts have also been encapsulated in the NHS performance assess-
ment framework.4 There are two versions of the performance assessment
framework for the NHS: one that relates to health authorities and another that
relates to provider organisations. There are six areas of performance in the
health authority version of the performance assessment framework, aimed at
giving a balanced view of performance. Five of these six areas relate to health
outcomes. The ‘health improvement’ area re£ects the over-arching aims of
improving the overall health of the population and reducing health inequalities
that are in£uenced by many factors reaching well beyond the NHS. The
‘e¡ective delivery of appropriate healthcare’ area recognises that care must
be e¡ective, appropriate and timely, and must comply with agreed standards.
Indicators in this area may act as a proxy for health outcomes. For example, if
there is good scienti¢c evidence that vaccination is likely to prevent infections
then a measure of the coverage of a vaccination programme is an indicator of
future health outcomes in terms of infections avoided. ‘Fair access’ recognises
that the NHS contribution must begin by o¡ering fair access to health services
in relation to people’s needs. This overlaps with e¡ective delivery in that the
measures chosen are related to e¡ective processes but act as a proxy for equity
of outcomes. The ‘patient/carer experience’ area re£ects the way in which
patients and their carers experience and view the quality of the care they
receive, including outcomes. The ‘health outcomes of NHS care’ area covers
measures that assess the direct contribution of NHS care to improvements in
overall health.
The speci¢c indicators within each area and sub-area will vary from time to
time, based on changing policies and priorities. This implies that the main
requirement for data is £exibility to meet a variety of needs, often di⁄cult to
forecast in advance. Below we see how the vision of Edith Ko«rner contributed
to achievements in the assessment of health outcomes within the NHS, based
on routine health services data.

The contribution of Ko«rner-inspired principles

and procedures to what is possible now
The ¢rst report of the Steering Group on Health Services Information, on hospi-
tal clinical activity, highlighted the following principles and procedures which
have had a direct impact on creation of a data set for the kind of outcomes
assessment described above:
96 Vision and value in health information

. The report was concerned with information for health services management.
. Data should be collected because they are essential for operational purposes.
. User-oriented information yields bene¢ts to those who collect it and thus pro-
vides incentives for accuracy and expedition.
. There should be ongoing updating of data speci¢cation.
. The logical consequences of these principles are that:
^ central returns should be a by-product of local data
^ local data should be a by-product of operational processes
^ a minimum data set for central purposes represents a compromise between
what is desirable, feasible and practical
^ standard de¢nitions and classi¢cations are needed to enable comparison
^ there should be the potential for additional data collection to supplement
the minimum speci¢ed.
The hospital episodes statistics (HES) data set,6 which evolved after this report
and replaced the previous hospital inpatients enquiry (a 10% sample of regional
hospital activity data), embodies these principles and procedures in intent, if not
always in practice. HES contain personal, medical and administrative details of
all patients admitted to, and treated in, NHS hospitals in England. The records for
the database (around 12 million annually) are collected from all hospital NHS
trust providers of admitted patient care. There is at least one record for each
patient’s stay in hospital. The following examples illustrate the current use of
HES for outcomes assessment, based on analyses undertaken by the National
Centre for Health Outcomes Development for publication as part of the NHS
Performance Indicators4 and the Compendium of Clinical and Health Indicators.5

E¡ective delivery of appropriate healthcare as a

proxy for health outcome
There is research-based evidence to show that hip replacement surgery, if used
appropriately, leads to improvements in the quality of life through pain relief
and improvement in mobility. The NHS does not collect data routinely on qual-
ity of life outcomes among the patients treated within it. However, data on the
numbers of operations carried out may be used as a proxy. The operations are
carried out in hospital and postcode data from patient records in HES may be
used to relate each patient to a health authority of residence and hence calcu-
late population rates of these operations. Figure 8.2 shows that the operation
rate among women over 65 years old, resident in health authority areas in
England varies threefold. HES are limited in that they do not have data from
the private healthcare sector where patients pay for the operations privately.
Nor does it have data on criteria for selection of patients for surgery, thresh-
olds for treatment, appropriateness of the operation for individual patients,
 Information for the assessment of health outcomes 97

local waiting list initiatives and local need. It is unlikely, however, that these
would explain the levels of variation and the pattern observed.

Figure 8.2 Hospital procedures: primary hip replacement. Directly standardised rates for
females (aged 65þ), ¢nancial year 1999^2000.

Potentially avoidable adverse health events

Lower limb amputations among patients with diabetes may re£ect long-term
failure of control of diabetes. Ideally, an outcome indicator should be based on
the proportions of patients with diabetes who develop such complications.
However, there are currently no routine NHS data sets that enable such long-
term monitoring. As a proxy, data on patients undergoing amputations in hos-
pital, where diabetes is mentioned as a diagnosis, have been used alongside data
on whole populations of health authorities, to develop an indicator. Figure 8.3
shows that there is some threefold variation in the population rates of such
amputations. The indicator is constrained because the rate is currently based
on whole populations of health authorities, not all patients with diabetes. How-
ever, while some of the variation may be explained by di¡erences in the number
98 Vision and value in health information

of people with diabetes in each health authority area, amputations still re£ect
adverse outcomes as they are potentially avoidable.

Figure 8.3 Hospital procedures: lower limb amputations in diabetic patients. Directly
standardised rates for people of all ages, ¢nancial year 1999^2000.

Indicators on emergency re-admission to hospital within 28 days of previous

discharge have been part of the clinical outcomes section of the NHS Performance
Indicators since 1999. These have involved creative use of HES. The basic
assumption behind the indicators is that patients are deemed ¢t for discharge at
the time of discharge and a return to hospital as an emergency may re£ect
unplanned, unforeseen adverse events, some of which may be potentially pre-
ventable. There are exceptions to this and in some specialities, for example
obstetrics, mental health, conditional discharge and re-admission may be part
of planned practice. A patient may be re-admitted to another hospital in another
part of the country and thus measurement of re-admissions requires linkage of
records for the same patient. This is achieved by assuming that an exact match
on three variables between HES records ^ date of birth, sex and postcode ^
represents the same patient. The technical details of doing this (subject to
ongoing re¢nement and improvement) are described elsewhere.4 A variety
of indicators has been produced, including ones covering all discharges, either
for patients of all ages or speci¢c age groups (children, older people). In addition,
indicators have also been produced for individual diagnoses (discharges follow-
ing emergency admission with hip fracture, stroke). All of these have shown
 Information for the assessment of health outcomes 99

substantial and persistent statistically signi¢cant variation in re-admission

rates between ‘like’ hospital trusts, for example small/medium acute, large
acute, teaching hospitals, etc. There are a number of constraints with these indi-
cators that are described in the documentation published with the data. How-
ever, the indicators raise issues about outcomes which need further local
investigation and explanation.
HES has also been used to assess outcomes of care outside hospital, by proxy.
The 2002 NHS Performance Indicators had an indicator on population rates of
emergency admission to hospital for lower respiratory infections among chil-
dren. Successful prevention, timely diagnosis and treatment may help children
to avoid developing problems su⁄ciently severe to require emergency admis-
sion to hospital. The indicator shows fourfold variation in rates of hospital
admissions for such infections, with rates in many health authorities signi¢-
cantly higher than the national average.

Restoration of health-related function and

quality of life
HES does not have patient level data on health-related function and quality of
life. Two indicators were developed to assess this indirectly.4 The indicators
measure the proportion of patients, admitted to hospital as an emergency with
hip fractures and strokes, who were discharged to their usual place of residence
within a speci¢ed time period (28 days for hip fracture, 56 days for stroke).
In the absence of routine data on patient levels of function and wellbeing, a
return to usual residence following admission with these conditions may act
as a proxy for successful outcome of rehabilitation. The category of accommo-
dation as coded in HES is used as a proxy for place of residence. Although the
proportion of those who return to pre-admission category of accommodation
will depend partly on the availability of support at home and the quality of com-
munity services, a change in the category of accommodation may suggest an
important change in functional ability and health status. There are substantial
variations between ‘like’ populations and ‘like’ NHS hospital trusts in the pro-
portions who return to their usual residence. Production of this indicator
requires linkage of individual patient episodes into continuous inpatient spells,
spanning hospitals across the country. This enables assessment based on the
full patient stay, including transfers to other hospitals for rehabilitation before
¢nal discharge.

Potentially avoidable premature death

A range of indicators has been produced since 1999 on deaths within 30 days
following either an operative procedure in hospital or admission with certain
100 Vision and value in health information

diagnoses.4 The indicators have been generic (for example, all surgery) as well
as operation- (for example, heart by-pass) and condition- (for example, stroke)
speci¢c. All have shown variation between ‘like’ populations and ‘like’ NHS
hospital trusts. In order to measure deaths within a certain time period, it has
been necessary to link HES data between trusts as well as with death registra-
tion data from the O⁄ce for National Statistics. The former enables inclusion of
deaths that may occur in a di¡erent NHS hospital trust after a transfer. The
latter enables inclusion of deaths in the community after discharge from hospi-
tal. The techniques used for linkage are described elsewhere.4 The following
results from data at England level for calendar year 2001 show the importance
of doing this:

. Of 28 044 deaths after surgery (non-elective admissions), 22 344 occurred

within the trust of operation, 851 in another trust and 4849 after discharge
from hospital.
. Of 4523 deaths after emergency admission with a hip fracture, 3731
occurred within the trust of admission, 121 in another trust and 671 after
discharge from hospital.
. Of 17 629 deaths after emergency admission with a stroke, 16 787 occurred
within the trust of admission, 273 in another trust and 569 after discharge
from hospital.

Such linkage was constrained because of invalid and missing data in the ¢elds
used for linkage, that is, date of birth, sex and postcode, but more sophisticated
linkage methods involving the NHS number as well have been used more
recently. Linkage of data between years to produce more robust indicators
should also be possible in the future.

Current constraints due to failure to follow

Ko«rner-inspired principles and procedures
The above examples show what is possible, by way of production of information
for management, through the creative use of data collected based on Ko«rner
principles and procedures. There are, however, some constraints with such
use of data, not owing to problems with the principles and procedures them-
selves but due to failure to follow them. For example, the National Centre for
Health Outcome Development was asked to undertake a feasibility study of pro-
ducing an indicator on transfer of pregnant women during labour (as an indi-
cator of potentially avoidable adverse events). This required the use of special
additional data within HES on mothers and babies. The data for this indicator
should be available in the ‘baby tail’ of the records, one for each delivery.
However, the National Centre for Health Outcome Development found that
 Information for the assessment of health outcomes 101

20% of the baby tail records had missing data and advised that this indicator
was not feasible.
Variation in indicator values between hospitals and health authority popula-
tions may be due to di¡erences in the types of health conditions and operations
carried out, severity of conditions, seriousness of operations and socio-economic
characteristics of the patients and populations. HES do not have su⁄cient infor-
mation within them to enable adjustment of the indicators to allow for these
explanatory factors. A way round this is to compare ‘like’ with ‘like’ as far as is
possible. For these purposes, hospitals have often been grouped into clusters,
for example small and medium acute, teaching, specialised community, and
likewise health authorities have been grouped into clusters based on the socio-
economic characteristics of their populations. The Ko«rner principle that there
should be ongoing review of data sets should help to address this shortcoming
over time.
In 1997 the then NHS Executive published a consultation document on
potential clinical indicators for the NHS.7 Among the suggestions were several
indicators of complications in hospital following surgery, for example pulmon-
ary embolism (blood clots in the lung), organ damage, wound infections, cen-
tral nervous system problems, etc. The conclusion following the consultation
was that while clinically meaningful as outcome indicators (that is, re£ecting
potentially avoidable adverse events), the coding of clinical data within HES
was insu⁄ciently complete and accurate to make these indicators robust. The
Ko«rner principle that there should be ongoing updating of data sets should help
to address this shortcoming over time.
There have been improvements to the quality of HES data in recent years. The
clinical indicators in the NHS Performance Indicators set are accompanied by a
detailed assessment for each NHS hospital trust separately, of the completeness
of data in all the HES ¢elds used for each indicator. At England level, the percen-
tage of episodes in 1998^1999 that had missing or invalid data in the ¢elds
used for the deaths indicators was 5.46 (ranging by trust from 0.06 to 42.5).
In 2000^01 this changed to 1.34 (range 0^32.5). In relation to the K˛rner
principles, this has happened not because those collecting data at local level
¢nd the data useful, but because those producing management information at
central level have exposed de¢ciencies and made improvements in data comple-
teness mandatory. Trusts are now held accountable for data quality as well as
for the results shown by the indicators.

Applying Ko«rner-inspired principles and

procedures to ¢lling in gaps
The examples above show that routine national data are severely limited in
terms of ability to measure health outcomes precisely. However, creative use of
102 Vision and value in health information

the data has shown that expectations of health outcomes are not being met opti-
mally by the NHS and has raised issues for further investigation. National data
will remain limited as it is unlikely that routine data systems will ever be su⁄-
ciently detailed and comprehensive to cover all needs and eventualities
(although there is room for improvement). However, national comparative
indicators should act as a trigger for further work . A number of case studies
of follow-up work undertaken by health authorities, following publication of
national comparative data on population health outcomes, show the value
of this.2 For example, She⁄eld, which had high death rates from coronary heart
disease, undertook an angina survey which identi¢ed a mismatch between need
and service access and led to new service policies. In order to support such local
work, 10 working groups, set up by the Department of Health, advised on
health outcomes assessment for each of 10 health topics, for example asthma,
stroke, and made suggestions for potential health outcome indicators, some of
which may require new data collection.2 Alongside these reports, advice on
how to access and use instruments for the measurement of health or health-
related quality of life, and information on national clinical databases concerned
with more detailed risk-adjusted health outcome indicators are available via the
knowledge base.2
Table 8.1 shows the kinds of indicators that would be needed to provide a
more complete picture of health outcomes for coronary heart disease. The indi-
cators in square brackets are ones that cannot be produced as comparative
indicators at national level by use of current routine data. It is remarkable how
little can be monitored using routine data at present, given that heart disease is
a common health problem and one of the largest single causes of premature
death in England. HES cover hospital inpatient care only and are thus limited
in terms of coverage of health outcomes. Comprehensive assessment of health
outcomes in the context of heart disease would require data on risks, levels of
disease, adverse events, quality of life, health promotion, primary care, outpati-
ent care, ambulances, accident and emergency and community care. For exam-
ple, data from the Health Survey for England show that a proportion of patients
with high blood pressure either do not know that they have it or are inade-
quately treated, with variations between NHS regions.2 Such indicators could
be produced from routine data from the various settings listed, if collected along
the same lines as HES. It may even be necessary to link data between settings.
For example, production of meaningful health outcome indicators re£ecting
shared care for heart disease would require the linkage of primary care, hospital
outpatient and hospital inpatient data.
Although much of the discussion has focused on central returns and national
comparative data, the principles and procedures elicited from the Ko«rner reports
did cover local data collection as well. They could be applied to cover data collec-
tion for many of the outcome indicators suggested in Table 8.1. Data on the
number of patients who get to hospital fast and bene¢t from early e¡ective
 Information for the assessment of health outcomes 103

Table 8.1 Health outcome indicators in the context of the NHS performance assessment
framework ^ example coronary heart disease

Performance assessment area and related indicators

^ existing [examples of potential]

Health improvement: the overall health of populations, re£ecting social and environmental factors
and individual behaviour as well as care provided by the NHS and other agencies

. Success in reducing level of risk to health

^ mean systolic blood pressure
^ mean diastolic blood pressure
^ mean weight of adults
^ mean body mass index (BMI)
^ adults who are overweight
^ adults who are obese
^ smoking of cigarettes, pipes or cigars
^ serum cotinine level above 20 ng/ml
^ alcohol consumption above recommended levels
. Success in reducing level of disease and impairment
^ [coronary heart disease]
. Success in reducing adverse consequences (of intervention)
^ [side e¡ects of risk reduction]
. Success in reducing adverse consequences (of lack of intervention)
^ [heart attacks]
. Success in restoring function and improving quality of life
^ [time o¡ work]
. Success in reducing premature deaths
^ deaths from coronary heart disease
^ deaths from heart disease ^ other

Fair access: the fairness of the provision of services in relation to need on various dimensions:
geographical; socio-economic; demographic (age, ethnicity, sex); and care groups

^ [surgery for heart disease by gender, socio-economic status, age, ethnicity]

E¡ective delivery of appropriate healthcare: the extent to which services are: clinically e¡ective
(evidence-based); appropriate to need; timely; in line with agreed standards; provided according to
best practice service organisation; and delivered by appropriately trained and educated sta¡

. Proactive intervention to avoid risk to health

^ [health promotion initiatives on healthy eating]
. Timely detection of risk; removal/reduction of existing risk to health
^ [smoking cessation services]
. Timely interventions to detect and treat disease or ill-health
^ [surgery for heart disease]
. Late interventions to minimise consequences of disease or ill-health
^ [rehabilitation after a heart attack]

E⁄ciency: the extent to which the NHS provides e⁄cient services, including: cost per unit of care or
outcome; productivity of capital estate; and labour productivity

^ [Cost of surgery]
104 Vision and value in health information

Table 8.1 (continued )

Performance assessment area and related indicators

^ existing [examples of potential]

Patient/carer experience: patient or carer perceptions on the delivery of services, including:

responsiveness to individual needs and preferences; the skill, care and continuity of service
provision; patient involvement, good information and choice; waiting times and accessibility; the
physical environment; and the organisation and courtesy of administrative arrangements
. Patient experience
^ [waiting time for consultation and heart disease surgery]
. Carer experience
Health outcomes of NHS care: the direct contribution of NHS care to improvements in overall health
. Success in reducing level of risk to health
^ [treated and controlled high blood pressure]
. Success in reducing level of disease and impairment
^ [prevention of subsequent heart attacks]
. Success in reducing adverse consequences (of intervention)
^ [complications after surgery for heart disease]
. Success in reducing adverse consequences (of lack of intervention)
^ [time from symptoms to arrival at hospital and thrombolysis]
. Success in restoring function and improving quality of life
^ [quality of life six months after a heart attack]
. Success in reducing premature deaths
^ [sudden deaths caused by heart attacks]

Note: Comparative data at national level are not currently available for examples in [ ].

emergency treatment (thrombolysis) following a heart attack would show local

e⁄ciency and e¡ectiveness and could be produced as a by-product of linked
operational ambulance and accident/emergency data collection systems.

Conclusions
Sadly, existing data and comparative indicators to monitor health outcomes
within the NHS only scratch the surface of what is needed, as set out here. This
is because of the complexity of the methods as well as a lack of suitable routine
data. The examples presented, however, show what can be achieved through
the creative use of routine data. Many of these examples are only possible
because of the vision articulated in the reports of the Ko«rner steering group and
the evolution of national data systems based on that vision. Extension of that
vision to the creation of other data sets at national level could ¢ll in many of
the gaps in the ability of the NHS to monitor health outcomes. Further applica-
tion of the principles and procedures for data collection highlighted in the early
 Information for the assessment of health outcomes 105

Ko«rner reports could be applied at local level, to supplement national assessment

and provide a true comprehensive picture of what is being achieved. Without
that, the real return on the vast NHS expenditure will remain in the realms of
speculation. The current limited information does show that achievements are
far from optimal.

Acknowledgement
The work of the National Centre for Health Outcomes Development is funded by
the Department of Health. All views expressed are those of the author and not
necessarily of the Department of Health.

References
1 Pearson M, Goldacre M, Coles J et al. (eds) (1999) Health Outcome Indicators: asthma.
Report of a working group to the Department of Health. National Centre for Health Out-
comes Development, Oxford.
2 Lakhani A (2000) Assessment of clinical and health outcomes within the National Health
Service in England. In: D Leadbeter (ed) Harnessing O⁄cial Statistics. Radcli¡e Medical
Press, Oxford.
3 Department of Health (2003) (www.doh.gov.uk)
4 Department of Health (2002) NHS Performance Indicators: February 2002. (www.doh.
gov.uk/nhsperformanceindicators/2002/)
5 Lakhani A and Olearnik H (eds) (2001) Compendium of Clinical and Health Indicators, 2001.
National Centre for Health Outcomes Development, London.
6 Statistics Division 2HES (2000) Hospital Episode Statistics: HES the book. Department of
Health, London.
7 Central Health Outcomes Unit and Statistics Division 2HES (1997) Clinical Indicators for
the NHS (1994^95): a consultation document. NHS Executive, Leeds.
CHAPTER 9

Principles and purpose for child

health informatics
Mitch Blair and Michael Rigby

Introduction
If we de¢ne child health as referring to both preventive and curative care
delivered from birth to 19 years of age, and informatics as the application of
technology and systems to information handling, then it is in the ¢eld of child
health in the UK where we have seen one of the most successful widespread
applications of informatics to healthcare delivery. However, this lead is in
danger of being destroyed, and the current bene¢ts lost, as a perverse e¡ect of
current and supposedly more integrated strategies.
There are three driving forces behind this. This ¢rst is the move towards gen-
eric integrated record systems in all settings ^ primary, community and sec-
ondary care; this is a legitimate goal provided there are clear objectives and
safeguards, but at present children stand to be disadvantaged as there appears
to be a generic adult focus.1,2 At the same time, existing child health infor-
matics applications are undervalued because their past success has reduced
their pro¢le in an era of de¢cit-correcting strategic informatics investment.
The second is the anxieties raised about the con¢dentiality and alternative
possible uses of large electronically held record systems. However, this type of
system has been proved to have been secure and to have bene¢ted children’s
health in the past, especially in the areas of immunisation and child health
screening, and should continue to do so in the future.3 ^5
Third, the experienced NHS bodies best placed to operate such population-
based systems, such as health authority and regional NHS bodies, are subject
to regular change and destabilisation.6 NHS reforms are not conducive to the
operational stability necessary to create established procedures or the build-up
of valuable databases. In order to address these issues positively and construc-
tively, promoting health gain bene¢ts whilst setting an ethical and practical
108 Vision and value in health information

framework, the Royal College of Paediatrics and Child Health (formerly the
British Paediatric Association) worked with partner professional organisations
involved in child health to foster a constructive dialogue in the context of pro-
fessional and national policy issues, and to promote clear principles, through
an interprofessional forum on child health informatics, co-chaired by the
authors. Through these discussions, key principles have emerged, and these
are reported here.

Background and context

British child health informatics ^ a history
of bene¢ts
Whilst the collection, recording and communication of relevant information is
central to all clinical work, the number of dispersed professionals from a number
of di¡erent agencies contributing to the health and related care of children, and
the involvement of their parents as well, makes child health information an
important and challenging area. The ¢rst signi¢cant developments in child
health informatics to address these issues were made 30 years ago, with inno-
vation in West Sussex involving child registration and immunisation,7,8 with
subsequent extension to developmental surveillance9 and school health in
Cheshire. Even from those early days health gain bene¢ts were proven,10
including greater social equity.11 The steady uptake of such systems, at least
for immunisation functions, coupled with the uni¢cation of the NHS in 1974,
stimulated the development of the national Child Health System,12 hitherto
one of the most successful and widespread patient-based electronic record
systems in the UK, and possibly in any country (covering up to 60% of the chil-
dren born in England, and all children born in Wales and Northern Ireland).
As a result of the widespread use of these computer-based records systems bene-
¢cial initiatives such as the monitoring and improvement of immunisation
coverage,13 or audit of screening programmes,14 have been possible. Edith
K˛rner recognised, and directly involved, this successful experience in the
work of her steering group community services,15 though the child health
system has consistently su¡ered the same central ambivalence to national inno-
vation as was encountered by the steering group.16 Children have also been
bene¢ciaries in the development of computerised maternity information sys-
tems, insofar as these generate records of the neonate, including obstetric
history, to a structured format, allowing for improved communication of infor-
mation in the critical ¢rst few weeks of life.
 Principles and purpose for child health informatics 109

Current informatics policy direction

However, in recent years the potential and a¡ordability of informatics applica-
tions in general have increased considerably, and thus in turn so have the
realistic and legitimate expectations of health professional sta¡ and managers to
be able to harness health computing. The focus, rightly, is on integrated infor-
matics systems, co-ordinated data capture, and optimised feedback of derived
information, utilising modern technology. Each of the four home countries of
the UK has its own strategic approach to health information systems, but Eng-
land in particular was early to place emphasis on integrated patient-based
records for all clients in community health.17
Simultaneously, changes in the structure and management of the NHS have
led to new information requirements, including information to support health
commissioning and reimbursement. However, even before the necessary and
proposed information system developments of one reorganisation can be imple-
mented, the next round of changes is probably going to be announced. The
constant rounds of change to support the (in principle) legitimate good of mod-
ernisation do have the serious side e¡ect of undermining the information
system stability needed to underpin the desired better services.

The resultant threats to child health systems

An adult model strategic philosophy
Thus, this important lead in child health informatics is perversely threatened by
recent much-needed moves to strengthen health record and health informatics
applications, seeking to address acknowledged de¢ciencies.18 Moves to imple-
ment generic hospital information systems, bring person-based record keeping
into community health, and extend primary care computing, are each being pur-
sued by means of systems which treat all records generically based on the adult
model, and principally focused on treatment of illness. This is re£ected in the
information strategies and requirements to support the national service frame-
works ^ that started o¡ with a largely ‘disease’-focused approach.19^21 Moves
toward better management of resources adopt a case-mix approach, which does
not e¡ectively support preventive or personally customised longer-term services.
Possibly most seriously, projects to redress the past under-development of com-
munity health applications have assumed that children’s community services
and record needs are the same as for adults,22 when this is not the case. Thus,
although child health was the service that pioneered community health infor-
matics, with proven bene¢ts,7^11 because of subsequent under-recognition of
110 Vision and value in health information

the issues and past successes this is the very client group which will lose out
through latest developments.

Underappreciation of existing child

health systems
Paradoxically, whilst improved and sophisticated strategic components are
being advocated in areas where information capabilities are lacking at present,
long-serving systems are being used suboptimally. This is largely because the
traditional organisational structure of service delivery causes a focus only on
the information interests of the individual provider, at the expense of building
up a holistic picture of the whole child (and total child population), and although
the moves towards electronic patient records and electronic health resources
have much to commend them, they are a long way o¡ universal application,
and appear to have little focus on the special issues of preventive child health.
Furthermore, because of the inevitable pressure on scarce resources for infor-
mation systems, healthcare providers have been least interested in investing
further money in developing a system (or informatics area) which has been
perceived to operate satisfactorily for 25 years, yet which is threatened organi-
sationally. This lack of necessity for investment in updating a satisfactory
system, coupled with operational drift because of the passage of time (particu-
larly through sta¡ turnover not being matched with formal induction and
maintenance training) since ¢rst implementation together with lack of continu-
ity of ownership and appreciation with repeated organisational change, has
resulted in lack of appreciation of the availability and potential value of existing
child health databases.23

Potential risks and abuses of electronic

record systems
At the same time, there have been concerns about the possibility of breach
of con¢dentiality from new types of networked databases.24^26 Whilst child
health electronic records have not been reported victims of any such abuse in
the last 30 years, it is nevertheless entirely proper that adherence to full appro-
priate con¢dentiality measures should be ensured, but without inadvertent
detrimental e¡ects.27,28

Recognition of child health issues

These direct and indirect threats mean that a lead area in health informatics is
in danger of losing its position, and thereby a priority client group risks being
 Principles and purpose for child health informatics 111

put at a disadvantage by the application of generic solutions that do not meet

the particular needs of children. Through interprofessional discussion the
unique particular aspects of e¡ective child health services, the resultant special
requirements for record keeping, and the use when necessary of information
sharing, have each been identi¢ed in a way that ¢ts with appropriate infor-
matics application.

Multiple record subjects

First, others are directly involved in the child’s health. Children are seen and
treated with the support and involvement of their parents, or of a person acting
in loco parentis. Indeed, the fact that a natural parent is not the supporting adult
is itself important, and very sensitive, health information. Clinical information
about the mother in pregnancy and the delivery, and genetic information about
both biological parents, may be an intrinsic part of the child’s clinical history,
yet this is primarily con¢dential information about other people.

Unique data structures

Second, some health monitoring or healthcare delivery needs of children (and
their parents) are unique. Developmental and growth problems are best identi-
¢ed early, often requiring sequential recording of de¢ned data sets. These sets
must therefore be compiled to clear structures and meaning, in a way which
also allows comparison over time, for instance by calculation of growth velocity
or completeness of a particular screening or immunisation programme.

Child-centred inter-agency collaboration

Not infrequently, a child’s education and social development are in£uenced by
health factors, and an agreed sharing (and updating) of key factors is desirable.
Where interagency support is established as necessary to support or protect
children with particular needs or risks, sharing of changes to patterns is particu-
larly important ^ review of reports into service failures shows that breakdown
in interagency communication and recording is a major avoidable cause.29
Thus common meaning, and strictly controlled yet totally e¡ective information-
sharing protocols, are essential.

Mandatory data £ows

Lastly, there are some information requirements and speci¢c data £ows (most
being backed by statute or regulation) that are speci¢c to children. In particu-
lar, the Public Health Act 1926 (as incorporated into subsequent legislation)
112 Vision and value in health information

prescribes statutory responsibilities and data £ows, whereas the Children Act
1989 and the Education Act 1996 each place requirements for interagency
working and for the controlled sharing of certain items of information in
relation to child protection and special educational needs, respectively. Addi-
tionally, there are long-standing requirements for notifying birthweight, and
congenital malformations observable at birth, to the O⁄ce for National Statis-
tics,30,31 and health visitors and general practitioners need rapid noti¢cation of
neonatal discharges regardless of the locality of birth of the child.

Child health recording requirements

There are particular requirements and opportunities for child health infor-
matics. Interprofessional (and thus child-focused) study has identi¢ed these
requirements as:

. Foundation-setting ^ new records created after birth are intended to last well
into a lifetime ^ these include community health records, primary care
records and, where necessary, hospital records; therefore it is important
that these foundation records are accurate and meaningful.
. Timeliness ^ activities happen extremely fast in the ¢rst eight weeks of life:
neonatal examination, selective immunisation with BCG, midwife and
health visitor clinical visits and primary care-based screening and immuni-
sation. Information needs to travel with speed and with no loss of accuracy
or con¢dentiality, but also without undue cost.
. Stability in stress ^ parents (particularly ¢rst-time parents) will ¢nd the birth
and neonatal period hectic, challenging to their lifestyle and probably stress-
ful; they may not retain information given to them, and some information
they supply later may be inaccurate, therefore the record system itself must
act as a focus for accuracy and stability. The introduction of the parent-held
child health record for nearly all births in the UK has been one successful
response.32,33
. Changes of identity ^ at birth, children’s forenames may not be known or for-
mally recorded, whilst the surname recorded at the time of birth may not be
that chosen for common use thereafter; record systems must be robust
enough to cope with this possible change without loss of data or interruption
to access (while also being able to accommodate intended complete changes
of identity through adoption). The issue of the NHS number at birth will help
greatly in tracking infants in these ¢rst few weeks, long after the new NHS
number is introduced.
. Statutory data requirements ^ there are particular requirements of legislation
or regulation including noti¢cation to registrars of births, to the O⁄ce for
 Principles and purpose for child health informatics 113

National Statistics concerning birthweight for all children and to the O⁄ce
for National Statistics for all children with congenital malformations obser-
vable at birth, whilst the system of noti¢cation of births to proper o⁄cers is a
statutory requirement.
. Cross-boundary working ^ a signi¢cant proportion of births (possibly 30% in
some localities) occur outside the health catchment area of residence, requir-
ing rapid information transfer; similarly, many secondary and tertiary care
services for ill children are outside their immediate district of residence,
yet those providing care need comprehensive and accurate information.
Diagnostic support to preventive services such as phenylketonurial (PKU)
screening are often provided regionally, giving major information £ow
problems when strict timeliness is clinically essential.
. Cross-agency working ^ services for children frequently involve social ser-
vices, and possibly voluntary bodies, to complement healthcare. Education
services are involved with all children, including those of pre-school years,
and later in life individual schools also have a key role, often without appre-
ciating the importance of health factors.34,35 Information systems need to be
able to support and learn from partner agencies, but within a sound ethical
framework ^ this issue is not unique to child health but has particular
special dimensions, not least the involvement of education services.
. Unique data sets ^ data sets concerning children are far from being age-
group-speci¢c versions of generic data sets ^ in the clinical domain, infor-
mation about the antenatal history, labour and immediate postnatal period
consist of unique structures, whereas other special information recording
requirements include parental consent to or refusal of procedures such
as immunisation. Developmental surveillance records also need a special
structure, as does information relating to educational need and the needs of
children under the Children Act 1989.
. Added value of data ^ some types of data gain added value by being recorded in
a common format in a common record ^ these include height and weight
data, which, when recorded serially, enable calculation of growth velocities
against norms, hearing test results and assessments, and information about
unscheduled hospital and other clinic attendances that may show patterns
if drawn together. For example, repeat non-attendance at hospital out-
patients, with a reduced uptake of immunisation and screening coverage,
may alert the practitioner to possible neglect.
. Record of professional interpretation ^ certain clinical items may be capable of
recording by standard terminology but with the result that this would lose
the signi¢cant value that can be added either by context or by professional
interpretation, for instance a bruise or a burn is not itself unusual or even
worthy of recording, but if repeated, if in an unusual position for an acciden-
tal cause, or at variance with the explanation given by the parent, may
become highly signi¢cant.
114 Vision and value in health information

. Record as a tool for advocacy ^ the health record may frequently support advo-
cacy for children, who in the early years are unable to give their own clinical
history ^ this may range from the value that can be gained from the hospital
paediatrician in seeing the full developmental history, as ascertained in a pri-
mary care setting, through to seeking to compensate for low motivation of
some parents in presenting the child for preventive services, or to identify
the aggregate of the full range of treatment obtained by presentation at an
intentionally dispersed range of settings. Furthermore, for particular
groups of vulnerable children ^ such as travellers, refugees, those looked
after by the local authority or in temporary accommodation ^ it is often
very di⁄cult for the individual health professional to attain the whole pic-
ture without a comprehensive health record.
. Record partnership ^ the existence of an integrated child health system per-
sonal record for the child has a symbiotic relationship with parent-held
records: the central record will ensure the input to the parent-held record is
complete, whilst thereafter acting as a backstop should the parent-held
record not be presented at consultation.
. Transient information ownership ^ in infancy children are not responsible for
their own health information, and a parent or guardian discharges this
responsibility, which passes to children as they develop the appropriate
maturity in line with the legal ruling from the Gillick case;36 thus the pro-
prietary interest in (and access to) the record changes from an adult advocate
to the data subject at a period which is not temporally de¢ned.

Child health informatics principles

These information requirements set particular challenges. On one hand, they
are not adequately met by strategic initiatives in England (and probably not
elsewhere). On the other, if met in an unstructured way they could be perceived
to encourage casual sharing of sensitive data in an unethical way, in line with
the worst fears of those with anxieties as to the abuse of electronic health
databases.24 ^26 This is certainly not the intention ^ child health indeed is
highly protective of its good record in this respect, and wishes to promote the
highest ethical standards, both intrinsically to protect individuals, and also as
an essential prerequisite for the consumer trust needed for e¡ective services.
Aware of these unique requirements and expectations of child health infor-
matics, and at the same time very aware that existing national initiatives spon-
sored by health departments in the UK are not fully addressing (and in some
cases explicitly not recognising) these issues, interprofessional discussions
have identi¢ed the following child health informatics principles (which apply
 Principles and purpose for child health informatics 115

to information held in any medium, paper-based or electronic). These are based

upon professional practice, and re£ect earlier policy analysis literature.37,38
They are shown in the box below.

Child health informatics principles

1 Purposefulness: Personal health information about children should

only be collected for the purpose of delivering care (preventive, sup-
portive or therapeutic).
2 Enhancement of care: Data should only be transmitted between health-
care providers (or between agencies) if this will enhance care.
3 Openness and transparency: Young persons, parents or legal guardians
should have an interest in, and ownership of, the health information
recorded about them, and the data items shared as necessary, except
in exceptional circumstances.
4 Third party interests: Child health records raise particular problems
with regard to third parties’ identity, involvement and clinical history;
speci¢c care must be taken in cases of separation, changed partners,
family con£ict and other special circumstances, and over the potential
for rapid changes to these facts.
5 Unimpeded transmission: Information should £ow ahead of, or with,
the child when authorised by a health professional in the interest
of the child for continuity and completeness of care, and normally
with the consent of the child or parent.
6 Comprehensiveness: Information sets should be comprehensive, which
means that a link with prenatal information should be the norm;
health information should normally set children in the context of
their home and family.
7 Positive support of health and development: Health records for children
should support the healthy and full development of individuals to the
maximum of their potential; sequential observations should be linked
so as to identify actual or potential risks of compromised health or
development.
8 Health outcome and health gain: Both for the bene¢t of the individual and
so as to increase the overall e¡ectiveness of services, regular review
must be undertaken to identify health outcome of support to the indi-
vidual, and the overall health gain to the child population.
9 Ethical sharing: Ethical and open means of sharing appropriate data
with partner agencies outside health is frequently important in the
child’s interest, but requires explicit and clear protocols and the shar-
ing of objectives and methods by health, education and social services
116 Vision and value in health information

for an individual child; if information is being exchanged between

two or more agencies with di¡erent approaches to con¢dentiality,
the information must be handled in accordance with the most restric-
tive principles.
10 Ethical informatics context: Data should be recorded, stored, trans-
mitted and used according to sound informatics principles; the eight
principles within the Data Protection Act 1984, and use of agreed
terms, are of particular importance.

Rationale
The intention behind these principles is to give clear guidance to individual
system users in daily practice, to trusts and general practices in the use
and the selection of systems, and to system developers. It is recognised that
given the sensitivity of the topic, it is unreasonable for all to have to discover
robust and ethical solutions. These principles are based upon professional prac-
tice principles, e¡ective discharge of the duty of care, and the need to protect the
interests of children ^ both for e¡ective and con¢dential services, and for qual-
ity assurance and service development based on robust empirical analysis.
The principles to do not give scope for indiscriminate data sharing or data
aggregation, but nor do they cut across the scope of local ethical committees
to authorise responsible research. They seek to re£ect legal and mandatory
requirements in a client-based and practice-relevant way. They balance the
need for complete information if healthcare delivery is to be e¡ective, with recog-
nition of the heightened sensitivity of more comprehensive information sets.

Enablement
If these principles are to underpin and enable the restoration of sound child
health informatics as a key underpinning plank of local and national health
information systems, two sets of action are needed. One is the widespread use of
these principles by all interests ^ this should range from individual health pro-
fessionals looking for guidance in a speci¢c situation, through to information
managers and senior executives establishing corporate strategic frameworks
and audited operational policies. The other is the development of supportive
informatics initiatives at national level to enable an active child health infor-
matics environment, for instance by the agreement of common terms in areas
where they do not yet exist, such as recording of disability and mental health
status. Professional bodies could also strengthen practical aspects of educa-
tional programmes.
 Principles and purpose for child health informatics 117

The future imperative

There is a danger of thinking that 25 years’ practical experience of running
population-based child health information systems is merely a quarter of a
century of history rooted in an earlier century. In mere technology terms this is
largely true ^ computing and communications technologies have moved on
radically since the national Child Health System was ¢rst created, based
on even earlier systems of technology which are now redundant. However, pro-
gress cannot and should not be measured merely in terms of the rate of advance of
new policies based on new technologies ^ abounding principles need to endure.
The current march to integrated electronic health records, and much more
integrated dependence upon electronic recording and delivery support systems
in healthcare, is to be commended. However, if the price of that progress is the
loss of the practical experience, carefully de¢ned principles, well-honed tools
and clearly de¢ned positive bene¢ts of child health computing, it will be in that
regard a false progress and society and its children will be the losers.
Neither UK society, nor indeed any other supposedly developed and civilised
societies, can be complacent about either the societal or health sector approaches
to meeting the needs of children. Not only do serious problems regarding child
abuse, and health service delivery failures, continue to arise, but established
successes such as high immunisation rates can be threatened by organisational
and service development principles which are continually under the pressure
of change.
On the one hand, within England there are desirable initiatives such as the
development of a national service framework for children. This is ¢rmly
intended to protect and improve services to children, but it is important to
ensure that the information initiatives which have been well-developed are
similarly understood and supported as part of that focus. The UK is one of the
many countries which is committed to supporting the United Nations Conven-
tion on the Rights of the Child,39 which not least has speci¢c clauses relating to
the right to e¡ective healthcare. To lose the bene¢ts of the speci¢c innovations
in child health computing as part of the move towards a generic (and inadver-
tently adult-focused) electronic record environment could be construed as
directly incompatible with honouring that convention.

Conclusion
There is considerable danger that the development of person-based community
and other electronic health records will fail to recognise the particular needs ^
and potential bene¢ts ^ of child health informatics, while stimulating over-
protective safeguards against possible general abuse which would actively
118 Vision and value in health information

disadvantage children.27 Recent commendable attempts to make good previous

de¢ciencies in information policy perversely may destroy past gains in this ¢eld,
rather than making progress from the strong baseline. This has stimulated the
professional bodies concerned to identify issues, principles and enabling factors
which are desirable to address and promote in this regard.
It is intended that consideration and adoption of these principles will inform
national policy and development of techniques (particularly terms and out-
comes measures); assist system developers in producing more relevant yet
sound products; give healthcare organisations a baseline against which to
specify and fund information systems in provider organisations; give clear
guidance to practitioners; form an important contribution to de¢ning the
benchmark of ‘reasonable’ practice in a technical area; and above all reassure
patients and the public. In a fast-moving NHS modernisation and information
investment era, these principles do not lose any cogency or relevance ^ rather,
they should provide sound benchmarks to protect the special interests of chil-
dren’s health.

References
1 Dick RS and Steen EB (eds) for Institute of Medicine (1991) The Computer-based
Patient Record ^ an essential technology for health care. National Academy Press, Wash-
ington DC.
2 NHS Executive (1998) Information for Health. NHS Executive, Leeds.

3 Rigby M (1982) Keeping tabs on immunisation. Health and Social Services Journal
92: 1104^5.
4 Rigby M (1983) Clean bill of health for module. Health and Social Services Journal
93: 1444^5.
5 Rigby M (1985) Child health comes of age. British Journal of Healthcare Computing
2: 13^15.

6 Smith J, Walshe K and Hunter D (2001) The ‘redisorganisation’ of the NHS. BMJ 323:
1262^3.
7 Galloway T McL (1963) Management of vaccination and immunisation procedures by
electronic computer. Medical O⁄cer 109: 232.

8 Saunders J (1970) Results and costs of a computer assisted immunisation scheme.

British Journal of Social and Preventive Medicine 24: 187^91.
9 Chesham I, Rigby MJ and Shelmerdine HR (1975) Paediatric screening. Health and Social
Services Journal 85: 293^4.
10 Bussey AL and Harris AS (1979) Computers and the e¡ectiveness of the measles vacci-
nation campaign. Community Medicine 1: 29^35.
 Principles and purpose for child health informatics 119

11 Bussey AL and Holmes BS (1978) Immunisation levels and the computer. Lancet i: 450.
12 Rigby MJ (1987) The National Child Health computer system. In: A Macfarlane (ed) Pro-
gress in Child Health Vol 3. Churchill Livingstone, Edinburgh.
13 Begg N, Gill ON and White JM (1989) COVER (Cover of Vaccination Evaluated
Rapidly) ^ description of the England and Wales scheme. Public Health 103: 81^9.
14 Streetly A, Grant C, Pollitt RJ et al. (1994) Variation in coverage by ethnic group of neo-
natal (Guthrie) screening programme in south London. BMJ 309: 372^4.
15 Steering Group on Health Services Information (1984) Fifth Report to the Secretary of
State. HMSO, London.
16 Rigby M (1998) Information in child health management. In: M Rigby, EM Ross and
NT Begg Management for Child Health Services. Chapman & Hall, London.
17 NHS Management Executive Information Management Group (1993) An Information
Management and Technology Strategy for the NHS in England ^ getting better with informa-
tion. Department of Health, Leeds.
18 Audit Commission (1995) For Your Information ^ a study of information management and
systems in the acute hospital. HMSO, London.
19 Department of Health (1999) National Framework for Mental Health. Modern standards and
service models for mental health. Department of Health, London.
20 Department of Health (2000) National Service Framework for Coronary Heart Disease.
Department of Health, London.
21 Department of Health (2001) Diabetes National Service Framework. Department of
Health, London.
22 Bhuptani B, Collier B, Hull M et al. (1993) Community Information Systems for Providers
(CISP) System Overview. NHS Management Executive, Leeds.
23 Rigby MJ and Nolder D (1994) Lessons from a child health system on opportunities and
threats to quality from networked record systems. In: F Roger-France, J Noothoven van
Goor and K Staer-Johansen (eds) Case-based Telematic Systems Towards Equity in Health
Care (Studies in Health Technology and Informatics Vol. 14). IOS Press, Amsterdam.
24 Anderson R (1996) Clinical system security: interim guidelines. BMJ 312: 109^11.
25 Anderson RJ (1996) Security in Clinical Information Systems. British Medical Association,
London.
26 Jenkins S (1997) Information and the NHS (for me or for them?). In: R Anderson (ed)
Personal Medical Information ^ security, engineering, and ethics. Personal information
workshop, Cambridge, UK, 21^23 June 1996, Proceedings. Springer-Verlag, Berlin.
27 Rigby M (1997) Keeping con¢dence in con¢dentiality: linking ethics, e⁄cacy, and
opportunity in health care computing ^ a case study. In: R Anderson (ed) Personal Med-
ical Information ^ Security, Engineering, and Ethics. Personal information workshop,
Cambridge, UK, 21^23 June 1996, Proceedings. Springer-Verlag, Berlin.
28 Roberts R, Thomas J, Rigby M et al. (1997) Practical protection of con¢dentiality in
acute care. In: R Anderson (ed) Personal Medical Information ^ security, engineering,
120 Vision and value in health information

and ethics. Personal information workshop, Cambridge, UK, 21^23 June 1996, Proceed-
ings. Springer-Verlag, Berlin.
29 Noyes P (1991) Child Abuse ^ a study of inquiry reports 1980^1989. HMSO, London.
30 Alberman E (1998) Analysing and promoting the health of children. In: M Rigby,
EM Ros and NT Begg. Management for Child Health Services. Chapman & Hall, London.
31 Macfarlane A (2000) Birth and maternity statistics. In: D Leadbeter (ed) Harnessing O⁄-
cial Statistics (Harnessing Health Information Series). Radcli¡e Medical Press, Oxford.
32 Sa⁄n K (1998) Working with parents. In: M Rigby, EM Ross and NT Begg. Management
for Child Health Services. Chapman & Hall, London.
33 Macfarlane A and Sa⁄n K (1990) Do general practitioners and health visitors like
‘parent held’ records? British Journal of General Practice 41: 249^51.
34 Polnay L (1995) Health Needs of School Aged Children ^ report of a joint working party of the
British Paediatric Association. British Paediatric Association, London.
35 Hall D (ed) (1996) Health of All Children, Report of the Third Joint Working Party on Child
Health Surveillance. Open University Press, Buckingham.
36 Times Law Report (1985) Gillick v West Norfolk and Wisbech Health Authority. The
Times, 21 October.
37 Rigby MJ (1981) Child health ^ a time for better understanding? Health Trends 13: 97^9.
38 Rigby MJ, Tiplady P and Osborne D (1989) Principles, ethics and law in preventive child
health. In: B Barber, D Cao and D Quin et al. Medinfo ’89, Proceedings of the Sixth Confer-
ence on Medical Informatics, Beijing and Singapore 1989. North Holland, Amsterdam.
39 United Nations (1990) Convention on The Rights of the Child. United Nations, New York.
CHAPTER 10

New methods of documenting

health visiting practice
June Clark

Preamble
I confess to being a hoarder. In a box of papers that I hadn’t touched for years,
there it was . . . Steering Group on Health Services Information. Fifth Report to the
Secretary of State: a report on the collection and use of information about services for
and in the community in the National Health Service. And tucked inside it was a
yellowed paper typed on an old-fashioned typewriter.

9 May 1983
Dear Ms Clark,
As a regular reader (and admirer) of your bi-weekly column, I found the cri
de coeur in your last article of particular interest. And a matter for concern.
We published in April of this year an Interim Report about information
needed by the community health services and we enjoined health authori-
ties to give the report the widest possible circulation among sta¡ delivering
those services in order that they may respond to our invitation to comment
before our recommendations become ¢nal. It is obvious that the report did
not reach you.
I now enclose a copy and hope that you will rejoice in the proposals we
make for a computer based system . . . I shall put you at the top of the list
of future users!
(You understand, of course, that our progress in this ¢eld will depend on
many factors: the view of the Computer Policy Committee, of authorities,
of sta¡ and, not least, of Ministers. So we have some way to go.)
Best regards,
Yours sincerely,
E K˛rner
122 Vision and value in health information

At the time I was working as a health visitor in Reading and I was also writing a
regular column called Nursing Matters in a (then) new journal called The Health
Services. I cannot now ¢nd the article, and I cannot remember exactly what I
wrote, but I have a pretty good idea, for it seems sometimes that I have been
struggling with the same issue for all of the 35 years I have been in and
around health visiting: why are our documentation and information systems
for health visiting so awful?
I re-read K˛rner 5 (as it came to be called) with a mixture of sadness and
anger. Anger because, almost 20 years on, the situation in documentation in
community health services is at least as bad as K˛rner found it. Sadness because
over the years the name K˛rner became a dirty word in health visiting and
‘K˛rner statistics’ became the be“ te noire of community information systems.
It was a fate she did not deserve. I am reminded of a comment by the late Sir Roy
Gri⁄ths, who su¡ered a similar fate in respect of his proposals for introducing
general management into the NHS: ‘Don’t blame me,’ he said, ‘I didn’t imple-
ment the **** thing!’

A personal journey
It was at exactly the time that Edith K˛rner’s steering group began its work that
my own interest in documentation and information systems was beginning.
In 1981 I went back to practice as a health visitor after several years out doing
research and bringing up my children. Two incidents triggered what has now
become ‘my obsession’.
Before taking up my new job, I received a Council of Europe Fellowship to look
at child health services in Denmark and Finland. For one who had been brought
up, as we were in those days, to believe that health visiting was unique to the UK,
and that the way we did things was of course the only right way, it was a real eye-
opener; many of the things I learnt then transformed my thinking about primary
healthcare and changed my practice as a health visitor for ever after.
I remember one visit in particular that I made with a health visitor in Finland.
It was to a young mother with a baby of about three months. After the usual
greetings, the mother undressed the baby while the health visitor got out the
baby’s health record ^ an A4 folder (we were still using those dreadful A5
cards at that time), which she opened at the centile weight charts. The mother
already had out the little book which was her parent-held child health record (an
idea which was only just dawning then in the UK). The health visitor weighed
the baby and recorded the weight both on her own chart and in the mother’s
book. I could see that the line of the baby’s weight was below the printed curve,
and I could see that the health visitor looked worried. They talked for a while ^
of course I could not understand the language, but I could guess the discussion.
Then the mother left us for a moment to fetch ^ guess what ^ another little
 New methods of documenting health visiting practice 123

book which was her own record from when she was a baby! They compared the
two charts, they laughed, they talked a bit more, the health visitor wrote in both
the records, then the mother dressed the baby as the health visitor watched, and
the visit was ended.
When we got outside I asked the health visitor to tell me about what had been
going on. She said that they had talked about the baby’s weight and they had
made some decisions about changing the baby’s feeding, but she was not wor-
ried any more because the mother’s own weight pattern had been exactly the
same when she was a baby. I said that I had never seen records used in this
way ^ remember this was 1981 ^ but what a great way to teach! The health
visitor was surprised and asked me what we did, and I told her.
‘Well,’ I said, ‘we aren’t supposed to take records out of the o⁄ce, but of
course if I have four visits to do in a morning I can never remember which
baby has had which immunisations, so I do take the records with me. And
then when I park the car (yes I had moved on from a bicycle!) I look at the
records of the children I am just about to see, and then I do the visit, and then
back in the car, I jot down just the most important things so that I won’t forget,
and then I ¢ll in the rest when I get back to the o⁄ce.’
Her eyes got wider and wider, and then she said ‘You mean you write down
things about people without them knowing? But that’s unethical!’ It was like a
blow between the eyes. I had honestly never thought of it that way before.
Some weeks later I started my new job, taking over a caseload where the pre-
vious health visitor had left about three months previously. On my ¢rst day, at
the top of the pile of urgent visits, there was a request to visit an Asian family
with a mentally handicapped boy of four, who should be starting school that
week, but the forms had still to be completed. When I visited the family at
home, I found a mother who spoke no English, holding the boy on her lap
while feeding him from a bottle. The 12-year-old daughter who acted as inter-
preter told me how proud they were of her little brother because he was the only
boy in the family and she was the youngest of four daughters. I am sure that I
don’t have to spell out the implications.
I went back to the o⁄ce and consulted the child’s records. There was a pile six
inches high, including letters from various agencies, and one form about immu-
nisations from the Child Health Computer System which was then just begin-
ning. But since they were totally unstructured, after an hour’s struggle I still
could not ¢nd who had done what when, or what the family had been told or
had understood about the child’s handicap, or what had been planned for him.
So began a period of four years’ practice as an ordinary health visitor again,
in which I started to think about records and tried to do something about my
own. No one was thinking about computers as things clinicians might use for
record-keeping ^ ‘the computer’ was the thing that sent out these forms about
immunisations that you had to ¢ll in and send back to somewhere unknown,
and they were always wrong anyway.
124 Vision and value in health information

I did not know anything about computers either, but I could already see how
useful proper records might be for several purposes:
. for continuity of care (taking over from or sharing care with other workers)
and to remind myself when the interval between visits might be several
months
. as a tool for teaching and for sharing, and for what we now call contracting
with clients
. for managing my caseload
. and then when, a few months later, I spent the days between Christmas and
New Year struggling to do what were then called ‘end of year stats’, I realised
that if only I had recorded these things as I went along, all I would need to
do now was some simple adding up to produce what I now know is aggre-
gated data.
A year later I changed jobs again, into my ¢rst senior management post ^ as
Director of Community Nursing Services for West Lambeth Health Authority.
And on my ¢rst day, on my desk was the con¢dential internal report into the
death of Jasmine Beckford ^ one of the ¢rst of the many children who have died
because the information that could have identi¢ed risks and might have saved
their lives was held by di¡erent workers in di¡erent places and was not shared.
Of course it was not very long before I found how di⁄cult it was to make the
decisions about planning and resource allocation that I was now required to
make when I did not have the information I needed. I knew that the information
I needed was there in the heads, and possibly in the records, of the health visi-
tors and district nurses who were my sta¡, but there was no way I could get at
it without doing massive special investigations ^ for which I had neither time
nor resources.
The K˛rner recommendations were just beginning to be implemented:
As an indicator of the volume of the service provided, district management
needs to know how many contacts took place during a speci¢ed period
within each community patient care programme. A contact is an event in
which a patient is seen by a nurse face-to-face and may be either a clinic
attendance or a domiciliary visit. We recommend that all face-to-face con-
tacts be recorded. (Fifth Report, paragraph 6.13)1
If only I had known then what I know now about the need to link interventions
with the problems to which they referred! To be fair, K˛rner herself went on
to say:
We recognise that, in workload terms, contacts with patients di¡er greatly
. . . A simple count of contacts cannot be used as a measure of work done
either by individual members of sta¡ or sta¡ groups. (Fifth Report para-
graph 6.15)1
 New methods of documenting health visiting practice 125

But the damage was done. That is exactly how the new managerialism of the
time measured performance. As W Barker, whose own work, paradoxically,
was later to experience the same kind of misinterpretation and misuse, wrote
in 1992 in an article entitled ‘Measurement of NHS service provision: activity
levels or outcomes?’:

The more hyperactive the service the better it is seen to perform . . . For the
past decade thinking has been dominated by the Orwellian view of activity:
on Animal Farm four legs were good, two legs were bad. For the NHS four
operations are good, two are bad. Forty patients at surgery are good,
twenty patients are bad. It matters not if a successful GP practice has a
strong health promotion focus so that the other twenty are doing more
exercise and eating better, and thus have less need of the GP. That is not
how health care is measured today.2

My own earlier research3 had already identi¢ed the potential for converting
actual times to average times to target times, and using these as a measure of
health visiting performance described by Young:

Such target times can thus be estimated for each health visiting activity; if
each nurse records weekly the number and type of visits and clinics she
carried out, the administrative sta¡ can estimate a weekly target time for
each nurse. If the nurse also records her actual weekly time on duty, the
weekly target time expressed as a percentage of the actual weekly time
can be returned to each nurse as her performance ratio for that week . . .
The performance ratios could engender a spirit of competition among the
members of a nursing group, but there is no harm in this. As far as the
administrative sta¡ is concerned, the ratios would provide a measure of
overall nursing performance.4

The report that establishing performance targets was followed by an increase of

38% in the amount of visiting (without any change in clinic and school ses-
sional work), taken in conjunction with what Dawtrey in 1977 called the
‘weekly liar’, perhaps demonstrates the na|« vety of this approach.5 These studies
preceded, and were part of, the impetus for the K˛rner work, but, sadly, the
K˛rner approach exacerbated rather than alleviated it.
By 1991 I had left the health service altogether for higher education (a fugi-
tive from the 1990 round of NHS reforms), but quite independently, I became
involved in a project which, although I did not know it at the time, was to be
the bridge between my previous concern with documentation and information
systems as a practising health visitor, and the work I am doing now.
It was a project headed by the International Council of Nurses to develop
an International Classi¢cation for Nursing Practice ^ a kind of International
126 Vision and value in health information

Classi¢cation of Diseases (ICD) for nursing. The American Nurses Association

had brought to the International Council of Nurses the problem that in their
system of billing for patient care, nursing was not getting proper recognition,
¢nancial or otherwise, because information about what nurses do for what con-
ditions, and with what outcomes, was not being included in the computerised
information systems and the large databases that were being used for billing
and increasingly to meet the new imperatives of identifying and measuring
value for money and outcomes of care (it was to be another ¢ve or six years
before such imperatives crossed the Atlantic). One reason for this problem was
the lack of a standardised language to describe nursing phenomena analogous
to the languages such as ICD and Procedures Classi¢cations that were in use
in medicine.
As we said in our ¢rst publication about the project, ‘If we cannot name it, we
cannot control it, ¢nance it, teach it, research it, or put it into public policy’.6
Over the next three or four years I was catapulted into the world of health
informatics and information management, standardised languages and clinical
information systems, and the electronic patient record. I discovered the North
American Nursing Diagnosis Association (NANDA), the work in language and
classi¢cation systems going on in the University of Iowa, the work of the Amer-
ican Nursing Association on standardised terminologies for nursing and the
Nursing Minimum Data Set. Meanwhile in the UK I began to become involved
in the work of the NHS Centre for Coding and Classi¢cation and the develop-
ment of the Read Codes and I joined the fringes of the Nursing Terms project.
In 1996 I was awarded a Winston Churchill Fellowship to investigate the
development of standardised languages and clinical information systems for
nursing in Europe, and a Florence Nightingale Scholarship to do the same in
the USA. That sabbatical changed my world view as dramatically as had that
visit to Scandinavia 15 years previously.
On my travels around Europe I learnt about the Swedish legislation on medi-
cal and nursing documentation and Ehnfors’ work there developing of the VIPS
system;7 about the Belgian Minimum Data Set ;8 and in the Netherlands about
William Goosen’s work on developing a reference model for nursing informa-
tion9 that shows not only how the information system supports clinical, man-
agement and policy decisions, but also how the data about patients recorded by
nurses as an integral part of their clinical practice (because it is the basis of their
clinical decisions) also constitutes the building blocks of the aggregated data
that are required for management and policy decisions.
But in the USA I found that things had moved much further. There was
tremendous interest in the development of structured and standardised nursing
documentation, tremendous investment in IM&T. Lots of places were using
computerised patient records, for a variety of purposes, with bedhead terminals
in hospitals and laptops in the community. Nurses everywhere recognised the
importance of their records and the signi¢cance of nursing diagnosis and
 New methods of documenting health visiting practice 127

the nursing minimum data set for aggregated information derived from clinical
records. And researchers were using these data for identifying and measuring
the outcomes of clinical practice using new techniques of mining the data held
in databases.
In my mind, the pieces of the jigsaw puzzle were coming together. The key
pieces ^ the lessons I had now learnt ^ were already in place:
. the importance, and potential, of records as a source of information
. the need for structure, standardisation and linkage, and how this could be
achieved
. the way that management information could be derived from clinical records
. the way that databases could be used for research purposes
. and, in particular, an exciting new approach to the identi¢cation and mea-
surement of nursing outcomes, which avoided the problems of the rando-
mised controlled trial.

The ¢nal piece of the jigsaw was my appointment to my present post as Profes-
sor of Community Nursing in Swansea in 1997 ^ back to my roots not only in
terms of geography and culture, but also back to health visiting and child
health. I wanted to work on identifying and measuring the outcomes of nursing
practice, using what I had learnt about health informatics and clinical infor-
mation systems. And the obvious place to start, given my clinical roots and
previous research, was with health visiting records.

My vision
Like Martin Luther King, I have a dream:

. Decision support: I want all health visitors to be able to use their records in the
way I saw that health visitor use them in 1981 in Finland, now not with an
A4 paper folder, but with an electronic notebook. I want them to be able,
during visits, to call onto the screen everything they need to know to make
their assessments and to do what they need to do. That means access not only
to things like each child’s immunisation record, but other information
entered by other agencies, for example hospital admissions, and to be able
to see this alongside relevant information about other members of the family.
. Sharing information with parents: I want health visitors to be able to share this
with the family, to use it for negotiating plans. When health visitors need to
teach about something I want them to be able to pull up diagrams and the
latest evidence and clinical guidelines. I want them to be able to call up and
print o¡ the relevant information for the client on the spot. If they need to
refer or to make an appointment I want them to be able to do it there and
then by email.
128 Vision and value in health information

. The electronic health record: I want all the relevant data to be stored in the child
health system as the electronic patient record that is the beginning of the
child’s electronic health record, and I want mothers to have and to hold a
parent-held child health record that is used by all the agencies that are con-
cerned with their children.
. I want health visitors to record what they do, preferably on site, using stan-
dardised language that the computer system will code and store so that later
they can review and evaluate their own practice, and so that their data can
be aggregated with that of other health visitors to form the database that will
provide managers and policy-makers with the information they need for
policy and ¢nancial decisions.

Everything on my wish list is technically possible now. Most of it I have seen

working. The hardware is already available, and getting better and cheaper
every day. We have the expertise to develop the software. We now have the
knowledge about clinical practice to work with the systems developers to
avoid the mistakes and limitations which have given K˛rner such a bad name.
The bad news is that soon after I arrived in Swansea I went to a local study
day on documentation for health visitors, and I found that little had changed
since my e¡orts of the early 1980s. Health visitors were still spending up to
three hours a day on ¢lling in multiple forms about every child they visited;
still providing statistics for managers which were of little relevance to them
and which they resented as a waste of time; still complaining that they could
not get back any of the information they put into the child health system.
Worse, the burden was dramatically increased by new, legalistically driven
demands for documentation about child protection. The concepts of minimum
data sets, standardised languages and such like ^ well documented in the
K˛rner reports 15 years before ^ were unknown, and the idea that records
might actually be useful and not just a necessary but time-consuming chore
had not entered anyone’s head.

The Swansea ‘Omaha project’

My response was to develop what has become known as the ‘Omaha project’.
This is a developmental project that began in 1998, the original purpose of
which was to ¢nd a way of identifying and measuring the outcomes of nursing
(health visiting) practice through the retrospective analysis of patient records.
At this time health visitors, even more than other groups, were under great
pressure to demonstrate cost-e¡ectiveness through the demonstration of out-
comes.10 Examination of existing records showed that the health visitors’
records consisted of unstructured narrative that could not easily be analysed
for this purpose. Examination of the information available about community
 New methods of documenting health visiting practice 129

health services11 showed that, 20 years after K˛rner, NHS trusts in Wales had
almost no information about what health visitors were doing, how they were
doing it, for what clients, for what problems ^ let alone with what results.
In the trust where the project has taken place, health visitors had no access to
computers, and very few had even basic skills in IT or information management.
The trust had been using a K˛rner-based system that had been found to be non-
‘2000-compliant’, and had been abandoned. Health visitors were recording the
same data, however, on paper forms; altogether 38 di¡erent forms were found to
be in use. Four years later this situation remains unchanged. Edith K˛rner
would be horri¢ed.
The emphasis of the project therefore shifted into the development of a struc-
tured documentation system that would enable the kind of analysis necessary
to achieve our original goals to be achieved. Four years on, we have achieved:

. A structured (person-based, problem-oriented) system for documenting

patient encounters, that can:
^ provide an epidemiology of patient problems or health issues discussed by
health visitors with their clients, which can be presented at the di¡erent
levels required by individual practitioners in order to prioritise their
work, and by managers to plan services
^ identify the professional interventions made in response to the various
issues and problems
^ identify outcomes in terms of changes in the status/severity of the pro-
blem, and clients’ knowledge and behaviour.
. A standardised terminology for expressing problems and health issues, that:
^ is multi-axial
^ is compatible with the emerging SNOMED-CT terminology
^ includes de¢nitions
^ has been validated in use.
. A (still fairly primitive) computerised version, written as a relational data-
base, that:
^ links the record of the patient encounter to other information about the
client (individual or family), such as previous encounters, demographic
data, standardised assessments
^ enables the individual practitioner to review work with individual clients
^ enables the production of aggregated data for statistical returns, caseload
review, etc.

The Omaha system

The project has come to be called the ‘Omaha project’ because it uses the docu-
mentation system that was developed during the 1970s by the Visiting Nursing
130 Vision and value in health information

Association of Omaha, Nebraska, USA, supported by a series of grants from

the US Department of Health and Human Services, as a method of describing
and measuring client problems, nursing actions and client outcomes. 12 The
Omaha system is a structured problem-oriented documentation system that
includes and relates the three elements of the ‘problem’ (the focus of the inter-
vention), the intervention and change in the client’s knowledge, behaviour or
health status. Each of these elements is expressed in a standardised and coded
terminology. The terminology consists of a problem classi¢cation scheme of 44
nursing diagnoses arranged in four domains (environmental, physiological,
psycho-social and health-related behaviours), an intervention classi¢cation
scheme of four categories (assessment, teaching, case management and proce-
dures) and an outcomes rating scale that is used to score the client’s knowledge,
behaviour and health status. Each element is de¢ned and coded, and the termi-
nology has been recognised for inclusion in the National Library of Medicine’s
Uni¢ed Medical Language System and will therefore be incorporated into the
developing SNOMED-CT terminology. The Swansea project has modi¢ed and
validated the terminology for use in health visiting while retaining compati-
bility with SNOMED-CT. The system can be used in either a paper-based or a
computerised format, as a stand-alone system or (more commonly) as part of
an integrated information system. Its validity and reliability have been exten-
sively tested,12 and it is used in several countries, although the current project
is the ¢rst time it has been used in the UK.

The Swansea project

The project began in 1998 with a small group of six volunteer health visitors
and is still continuing.
In Phase 1, a pilot study undertaken over a period of three months from
January to March 1999, 17 health visitors recorded their encounters with
families with new babies by use of a specially designed encounter record
(described below). The study was small in scale but the results were encouraging,
and the group was awarded a Community Practitioners and Health Visitors’
Association/Department of Health Centenary Award of »8000, which enabled
one member of the project team to be released for one day per week to co-ordinate
a larger and more systematic study.
Phase 2 began in November 1999 and ¢nished at the end of July 2000. In this
phase 27 health visitors recorded 769 encounters with a wider variety of client
groups (205 families), including older people and families with child protection
problems, over a period of nine months. Anonymous data from the encounter
records were entered into an ‘Access’ database for analysis. Evaluation also
 New methods of documenting health visiting practice 131

included a questionnaire to the participating health visitors and to a control

group of non-participating health visitors, as well as a focus group held at the
end of the study.
In Phase 3 a local Sure Start team (a project to provide intensive parenting
support to vulnerable families) used the system to record their encounters
with clients.
Phase 4 (the Less Paperwork Better Information Project) ran from August
2001 to July 2002. This was a 12-month project funded by the National Assem-
bly for Wales’ ‘Invest to Save’ budget, to develop a computerised prototype
or demonstrator. The participants were the health visiting team (two health
visitors and one sta¡ nurse) attached to one general practice. In addition to
recording the encounter by use of the Omaha system, the system enabled the
practitioners to upload onto their laptop computers data from the national
child health system about each of the children they were to visit, and a link
to the local hospital enabled messaging about children on the health visitors’
caseload between the health visitors and the paediatric liason health visitor. In
addition to enabling point of contact recording of the encounter, this phase
therefore prototyped the use of data to support decision-making and linkage
between the community and hospital systems.

Identifying and measuring outcomes

Identifying and measuring outcomes involves a number of conceptual and
methodological problems. In the UK the tendency has been to select service out-
comes that are easy to measure, such as length of stay, use of other services or
patient satisfaction, rather than clinical outcomes, which are often di⁄cult to
measure. Measuring the e¡ects of a particular treatment on a disease can be
achieved using the ‘gold standard’ of the randomised controlled trial. These
approaches are di⁄cult or impossible in preventive health services. Most of the
literature about measuring outcomes in preventive and public health services
focuses on the ‘big picture’ of mortality and morbidity rates using the methods
of epidemiology. But the World Health Organization (WHO), among others, has
pointed out that in the end quality depends on individual practitioners, and that
individual practitioners therefore need feedback on the quality and e¡ectiveness
of their own work.13 This can only be achieved if clinical practitioners get back
information about their own practice, so that they can see what works and
what needs to be improved.
Donabedian has de¢ned an outcome as:

not simply a measure of health, well-being, or any other state. Rather it is a

change in status con¢dently attributable to antecedent care.14
132 Vision and value in health information

In order to measure an outcome, therefore, three elements are needed:

. the pre-existing condition

. the intervention
. a measure of change in the condition.

In nursing (or health visiting) this means:

. a nursing diagnosis
. a nursing intervention
. a change in the client’s knowledge, behaviour or health status related to the
nursing diagnosis.

Diagnosis
Many people ¢nd the term ‘nursing diagnosis’ di⁄cult because of the idea that
‘diagnosis’ refers only to diseases and to a process that is peculiar to medicine.
However, the process of diagnosis (that is, identi¢cation and labelling of the
‘problem’) is used by many other professionals, and the diagnostic label which
is the end-point of the process refers more generally to the ‘condition’ that the
professional treats: one of the key di¡erences between medicine and nursing is
the ‘conditions’ that each profession diagnoses and treats.15 In medicine or hos-
pital nursing the diagnostic label usually implies a problem. More commonly in
health visiting, which is concerned with normal healthy people, it is a topic
to be discussed as part of health promotion. In the Omaha project the term
‘nursing diagnosis’ is used to describe the topic or health-related issue that is
the focus of a health visiting intervention.

Intervention
The intervention is what the health visitor ‘does’ about the nursing diagnosis
that has been identi¢ed. In the case of a drug or a speci¢c treatment, it is easier
to describe and measure the intervention. Health visiting interventions, such as
teaching and support, are far more intangible and di⁄cult to measure.

Client outcomes
In the acute sector outcomes can usually be measured by determining whether
a disease has been cured or symptoms have improved. In health visiting the
measurement of change is more problematic: in the absence of a de¢ned
 New methods of documenting health visiting practice 133

‘problem’, a more appropriate measure of change may be improved knowledge

or behaviour, and even in these ¢elds, change may be small and may not be seen
for a long time.
Moreover, the three elements are interdependent. Each can be properly
understood only in relation to the others (see Figure 10.1). Too often in the
past, attempts have been made to evaluate health visiting by the measuring of
just one of these elements, usually interventions. This was the key criticism
of the K˛rner approach.

Patient

Interventions

Diagnosis Outcomes

Data

Figure 10.1 The relationship between diagnoses, interventions and outcomes in the
client record.

The client record as a source of data

In a randomised control trial these elements are deliberately set up as part of the
study. But in fact these elements already exist, or should exist, in the client
record. The problem is that sometimes they are not recorded, or if the record is
in the form of an unstructured narrative they are di⁄cult to ¢nd.
Finding and specifying the three elements and the relationships between
them is especially di⁄cult in health visiting. The topics that are the focus of
intervention cover a wide range, including both health and social issues, and
do not necessarily focus on problems. Sometimes they relate to individuals,
sometimes to the family as a unit and sometimes to whole communities. The
timescale for identifying change can be very long, spread over many episodes
of care, with some issues taking many months, or more commonly years,
before they are resolved. The problem of attribution is di⁄cult to determine in
any healthcare setting, but becomes even more complex in a preventative ser-
vice such as health visiting.
Records must be structured so that the three elements can be related, and the
concepts must be expressed in a standardised language so like is compared with
134 Vision and value in health information

like. This involves a completely new way of documenting health visiting prac-
tice, but it is exactly what the Omaha system is designed to do.

The Swansea Omaha project encounter record

The core of the system, as used in the Swansea project, is the encounter record.
In the computerised prototype the encounter module is linked to a registration
module, which records identi¢cation and demographic data, and an assess-
ment module (not yet developed), which provides for assessments, including
standardised assessment tools. The encounter record is a continuous record of
health visitors’ encounters with a family. It replaces the traditional ‘progress
notes’. Both paper and computer versions incorporate three sections:

. service details
. topic or nursing diagnosis
. interventions.

The service details recorded include the date, type of contact (for example, home
visit, telephone contact, clinic attendance or signi¢cant street contact), and
a code for Patient’s Charter compliance. The computerised version includes
who is present during the encounter (important for child protection) and
space for emerging performance indicators, which will replace Patient’s Char-
ter compliance.
The second section of the encounter record concerns the nursing diagnosis or
focus of intervention. Health visitors record each topic, identifying the term and
the code from the problem classi¢cation scheme. Against each topic they
record the ‘Bearer’ ^ the client to whom the topic relates (for example, a topic
such as ‘nutrition’ might relate to an individual or to the family as a whole).
Next, health visitors judge whether the topic constitutes an actual problem
(P), an identi¢ed risk (R), or no problem (that is, a topic for health promotion)
(N). The acronym PRN has proved easy to remember and code, and enables
the record to show the extent of health visitors’ health promotion work, which
traditional documentation rarely captures. Detail is provided in the the paper
version by a second level of classi¢cation, and in the computerised version by
drop-down lists of signs and symptoms for problems, risk factors in the case of
risks and by provision for free text.
Next health visitors score and record the client’s knowledge (K), behaviour
(B) and health status (S) (acronym KBS) in relation to the particular topic, by
use of a simple ¢ve-point Likert scale. In Phase 2, participants also recorded
the client’s level of coping (C) as part of a subproject to test the relevance of
coping as a possible outcome indicator. As topics and the client’s responses are
recorded in successive visits, changes in the PRN and KBS coding provide the
 New methods of documenting health visiting practice 135

measure of change that constitutes the outcome and enables a decision to be

made about whether an intervention is working.
Next health visitors record their interventions in relation to each topic. The
original version has four broad categories:

. Assessment (called ‘surveillance’ in the original system, and ‘monitoring’ in

the computerised version, but de¢ned to include assessment and monitoring).
. Teaching (de¢ned to include guidance and support). In the computerised ver-
sion this category is divided into two: teaching, and support and counselling.
. Co-ordination (called ‘case management’ in the original system, but de¢ned
to include activities such as mobilising resources, arranging referrals, etc.).
. Procedures and treatments (less used by health visitors than by district nurses,
but de¢ned in Phase 2 of the project to include speci¢c developmental test-
ing, etc.).
An intervention may contain one, two, three or all four of these categories, and
can be expanded by a short free-text description.
In the computerised version the intervention section is expanded, using the
SNOMED-CT approach, to incorporate details of the focus and the method used
(for example under ‘monitoring’ there is a drop-down list of screening proce-
dures).

Analysing the initial data

Phase 2 enabled the creation of a small database that was used to test the feasi-
bility of analysing the data in ways which would demonstrate the usefulness of
the system, in particular the identi¢cation and measurement of outcomes. The
database can be analysed at any level, for example the individual client, the indi-
vidual practitioner, a particular locality. Because of the small size of the data-
base, and the limited purpose of demonstrating the potential of the system
rather than actual use, the data presented here are aggregations for the whole
database.

Epidemiology, needs assessment and

service planning
Simple frequency distributions of the problems or issues discussed provide epi-
demiological data for needs assessment and service planning. Analysis of the
most frequently recorded topics (or nursing diagnoses) revealed di¡erent
trends emerging in di¡erent client groups. Tables 10.1, 10.2, and 10.3 show
the topics most frequently discussed in encounters with di¡erent kinds of clients.
136 Vision and value in health information

Table 10.1 Topics most frequently discussed in contacts with families beginning at the
time of the six-month visit

Topic No of occurrences Frequency of No of occurrences

occurrence (%) as percentage of all
(n ¼ 89) topics recorded
(n ¼ 167)

Development screening 26 29.2 15.6

Hearing screening 24 27.0 14.4
Vision screening 13 14.6 7.8
Immunisations 7 7.9 4.2
Weaning 6 6.7 3.6
Postnatal depression 5 5.6 3.0
Home or garden safety 5 5.6 3.0

Table 10.2 Topics most frequently discussed in families identi¢ed by health visitors as
‘cause for concern’

Topic No of Frequency of Percentage of all

occurrences occurrence (%) topics recorded
(n ¼ 144) (n ¼ 199)

Weight in relation to centile 9 6.3 4.5

Home or garden safety 7 4.9 3.5
Partner relationships 7 4.9 3.5
Immunisations 7 4.9 3.5
Managing stress 6 4.2 3.0
Development screening 6 4.2 3.0
Failure to carry out important 5 3.5 2.5
aspects of care

Table 10.3 Topics most frequently discussed with older people

Topic Count of topic Frequency of No of occurrences as

occurrence (%) percentage of all
(n ¼ 127) topics recorded
(n ¼ 225)

Range of mobility 20 15.7 8.9

Bene¢ts 15 11.8 7.6
Assuming new role 12 9.5 5.3
Urinary incontinence 10 7.9 4.4
Home or garden safety 9 7.1 4.0
Identifying or mobilising services 9 7.1 4.0
Social contact or social isolation 9 7.1 4.0
 New methods of documenting health visiting practice 137

The numbers are small, but this analysis does show the potential of the system:
if these data were recorded by all health visitors on all families in their caseload,
it would give us an epidemiology of health visiting ^ identifying the kinds of
trends and patterns that are familiar to medical epidemiology, not of diseases
but of the kind of problems and issues health visitors deal with, which could be
used for health needs assessment and service planning.
Each nursing diagnosis or topic discussed is modi¢ed by the code P, R or N.
When the distribution of PRN codes across an aggregation of all topics discussed
is examined, it can be seen that health visitors more frequently identify items for
health promotion (that is, N and R) than actual problems (P) as shown in Table
10.4. Previously, this important aspect of the health visitor’s role was rarely
documented. Certain topics, however (such as bene¢ts, partner relationships,
temper tantrums) were usually identi¢ed as problems, as shown in Table 10.5.

Table 10.4 Distribution of PRN across all topics

Problem Risk No problem

Frequencies of nursing diagnoses 422 (36%) 241 (20%) 511 (44%)

Table 10.5 Distribution of PRN for selected topics

Nursing diagnosis Problem Risk No problem

Development 9 (9%) 3 (3%) 85 (88%)

Weight 10 (11%) 19 (21%) 63 (68%)
Immunisations 5 (8%) 1 (2%) 53 (90%)
Bottle feeding 8 (18%) 11 (24%) 26 (58%)
Breast feeding 6 (15%) 11 (28%) 33 (57%)
Postnatal depression 17 (59%) 10 (34%) 2 (7%)
Safety 10 (40%) 9 (36%) 6 (24%)
Depression 16 (73%) 5 (23%) 1 (4%)
Bene¢ts 10 (50%) 6 (30%) 4 (20%)

Interventions
The encounter record records the interventions that health visitors undertake.
As health visitors perform very few ‘hands on’ tasks, those outside the profes-
sion ¢nd it very hard to understand exactly what it is that health visitors
do, and health visitors themselves ¢nd it very hard to articulate their work.
As explained earlier, each topic may provoke one or more than one interven-
tion. Weighing a baby, for example, would usually include all four categories.
The physical act of weighing the baby would be recorded as a procedure (P); the
138 Vision and value in health information

health visitor would judge whether the weight gain was appropriate (A);
the mother might be counselled about the weight gain (T), and a decision might
be made to weigh the baby at clinic next week (C). Table 10.6 shows the
frequency of interventions across all topics recorded.

Table 10.6 What do health visitors do?

Intervention No of occurrences Percentage of topic records for

which this intervention was
recorded (n ¼ 2179)

Assessment 2126 97.6

Teaching/guidance/counselling 2000 91.8
Co-ordination 1985 91.1
Procedures 673 30.9

Almost all topics include ongoing assessment or monitoring (A), which is one of
the most important but least recorded aspects of the health visitor’s role. This
category is followed closely by teaching and guidance (T). A lot of health visi-
tors’ time is also spent on co-ordinating care (C), that is, referring and liasing
with other agencies. The ¢gure for procedures is low, and in reality is much
lower than the ¢gure suggests because in this phase health visitors were asked
to record all ‘protocol’ visits, such as birth visits, development checks and
screening checks as procedures. This was a pragmatic decision because of
the complexity of the records required by the trust for such visits. Other than
‘protocol’ visits, health visitors were asked to record P for any ‘hands on’ activ-
ity, such as demonstrating how to make up a feed or clean a sticky eye as opposed
to just discussing it.

Outcomes
Figure 10.2 shows the changes in the status of the problem and in the client’s
knowledge and behaviour over the database as a whole. This is the ¢rst quanti-
tative evidence that health visitors ‘make a di¡erence’.

3.6
Mean score

3.4
3.2
3.0
2.8
Knowledge Behaviour Status Coping
P < 0.01 P < 0.01 P < 0.005 P < 0.01
Figure 10.2 Change in mean outcome rating score for all topics
 New methods of documenting health visiting practice 139

Tables 10.7 and 10.8 show one example taken from the database at a level
which individual practitioners might use to monitor their work with a particu-
lar client.

Table 10.7 Mother X Topic: Social isolation (Code 0601). Database

example: change in pre-existing conditions

Date 24/11/99 24/12/99 21/01/00

P P R

Table 10.8 Mother X Topic: Social isolation (Code 0601). Database

example: Change in client’s response

24/11/99 24/12/99 21/01/00

Knowledge 3 3 3
Behaviour 2 2 4
Status 3 4 4
Coping 3 3 5

The topic being considered is social isolation in a young mother who is the
subject of domestic violence and lives in overcrowded conditions with ¢nancial
problems. Each of these issues would be recorded separately on the encounter
record, but in Tables 10.7 and 10.8 only the topic of social isolation is illu-
strated. It can be seen that on successive visits the Ps (problems) have changed
to Rs (risk of problems), showing an improved outcome for this mother in rela-
tion to the topic of social isolation. In another family, or for a di¡erent topic, it
may be possible to see the Ps change to Rs and then to Ns (no problems), demon-
strating that a problem had been completely resolved.
Table 10.8 shows the levels of knowledge (K), behaviour (B), status (S) and
coping (C) recorded for the same mother and the same topic of social isolation
over the same timespan. This gives a more sensitive measure of the change in
outcome than PRN, which helps to direct future interventions. For example,
it highlights whether interventions need to be focused on knowledge de¢cit or on
providing support to assist in coping skills. In this particular case it can be seen
that on 24 November 1999 the lowest score was 2 for behaviour (B). The free
text showed that this young mother had isolated herself from her family and
friends because of the domestic violence. The health visitor therefore directed
her intervention at improving the behaviour, and by 21 January 2000 this
score had risen to 4. The free text showed that after counselling by the health
visitor the mother renewed contact with her family and friends. As a result, her
coping skills had also increased to a maximum score of 5. The health visitor had
140 Vision and value in health information

been unable to take away the risk of domestic violence, but had considerably
increased this young mother’s ability to cope with the situation she found herself
in. This is an outcome that the health visitor would have been aware of before
using the Omaha system, but would have found impossible to demonstrate.
Because the Omaha system is a structured and standardised system, it is pos-
sible to link these improved outcomes to particular interventions. At the level of
individual practitioners it is therefore possible to decide whether an interven-
tion is working or whether it would be helpful to try a di¡erent approach. In a
larger database the aggregation of data should be able to show, for example,
whether levels of behaviour (B) in relation to breast feeding improved following
a particular campaign, although very large numbers would be needed to allow
the statistical analysis required to demonstrate acceptable levels of probability.

Acceptability of the system to health visitors

A questionnaire was sent at the end of Phase 1 to all the health visitors who had
participated in the project. This was analysed in its own right, but was also used
as a pilot for the questionnaire sent at the end of Phase 2 to all the health visitors
who had participated in that phase of the project and to a control group of health
visitors who had not done so. This section relates to the Phase 2 questionnaire.
Twenty-seven questionnaires were sent to the participating health visitors
and 20 were returned, a response rate of 74%. Sixty-nine questionnaires were
sent to health visitors in the same area who had not participated in the project,
and 44 were returned. Health visitors who were using the Omaha system, and
those who were not, were ¢rst asked about their traditional record system.
Then those who were using the Omaha system were asked how it compared
with their traditional record system for a variety of factors. The full results are
not included here, but Table 10.9 summarises the results of one key question.

Table 10.9 How useful is the Omaha system, compared with your present system, for
achieving the following purposes?

Traditional About the Omaha system Missing

better same better
(%) (%) (%) (%)

Clinical decision-making 0 30 65 5
Creating a caseload pro¢le 0 25 60 15
Identifying and measuring 0 5 90 5
outcomes
Allocation of resources 0 50 45 5
Teaching tool 0 10 80 10
Clinical governance 0 25 70 5
 New methods of documenting health visiting practice 141

The health visitors overwhelmingly considered the Omaha system to be far

better than their traditional record-keeping systems.

Future developments
This is a very small scale project undertaken with very little funding support
(a total of »58 000 over ¢ve years), but with an enormous amount of enthu-
siasm and goodwill from its participants. Its signi¢cance lies in the fact that the
developmental work that has been done on a small scale is a prerequisite for
the successful development of any computerised information system and espe-
cially for the development and implementation of an electronic health record.
We have learnt a great deal about the clinical decision-making that is the core
of health visiting practice (as of any professional practice) and how this can be
captured in documentation and clinical information systems; for example, we
have found that use of a structured documentation system changes clinical
practice and improves the focus and speci¢city of the encounter. Perhaps most
important of all, we now understand, and can specify, the requirements for a
system that will support practice as well as meeting management require-
ments ^ and thus overcome the main criticism of the K˛rner approach.

Endpiece ^ visions and progress

A few months ago the gasman came to service my central heating boiler. He put
on my kitchen table a little black bag, which he unzipped to reveal a laptop,
a baby printer and a mobile phone. Before examining the boiler he pulled up the
screen headed ‘History’ and checked with me when the boiler was last serviced,
whether there were any previous problems and so on. On the next screen,
‘Assessment’, he ¢lled in, in boxes and free text, what he found when he exam-
ined the boiler. A widget at the back of the boiler was worn, and he wanted to
explain to me why I needed it replaced, so he pulled up a 3-D diagram of the back
of the boiler, which I could not see. We agreed that I needed the new part, so he
pressed a key to order it from the central depot and said he would be back to ¢x it
in a week’s time. When everything was done, he pressed a button on the printer,
and put into my hand the printed summary of what we had done, for me to put
into ^ of course ^ my parent-held gas boiler record.
Now if British Gas can do that for every service engineer, why can’t the NHS
do it for every health visitor in Wales?

References
1 Steering Group on Health Services Information (1984) Fifth Report to the Secretary of
State. HMSO, London.
142 Vision and value in health information

2 Barker W (1992) Measurement of NHS service provision: activity levels or outcomes?

Radical Statistics 21^9.
3 Clark J (1981) What Do Health Visitors Do? Royal College of Nursing, London.
4 Young WC (1971) A work study of nursing sta¡ in a health department. Health Bulletin
29: 154^61.
5 Dawtrey E (1977) The Health Visitor in Primary Care. Unpublished MSc thesis. Medical
Architecture Research Unit, North London Polytechnic.
6 Clark J and Lang N (1992) Nursing’s next advance: an International Classi¢cation for
Nursing Practice. International Nursing Review 29: 109^17.
7 Ehrenberg A, Ehnfors M and Thorell-Ekstrand I (1996) Nursing documentation in
patient records: experience of use of the VIPS model. Journal of Advanced Nursing
24: 853^67.
8 Sermeus W and Delesie L (1994) The registration of a nursing minimum data set in Bel-
gium: six years of experience. In: S Grobe and E Pluyter-Wenting (eds) Nursing Infor-
matics: an international overview for nursing in a technological era. Elsevier, Amsterdam.
9 Goosen W, Epping P and Abraham I (1996) Classi¢cation systems in nursing: formalis-
ing nursing knowledge and implications for nursing information systems. Methods of
Information in Medicine 35: 59^71.
10 Roberts C (1996) The proof of the pudding. Health Service Journal 14: 27.
11 Clark J (2000) A Review of Health Visiting and School Nursing Services in Wales. National
Assembly for Wales, Cardi¡.
12 Martin K and Scheet N (1992) The Omaha System: applications for community health nur-
sing. WB Saunders, Philadelphia.
13 World Health Organization (1999) Health 21: the health for all policy framework for the
WHO European Region. World Health Organization, Copenhagen.
14 Donabedian A (1988) The quality of care: how can it be measured? JAMA 260: 1743^48.
15 American Nurses Association (1980) Nursing: a social policy statement. American Nurses
Association, Kansas City.
CHAPTER 11

Using information for

public benefit
Peter Tiplady

Introduction
Modern medical practice is dominated by information. Communication between
healthcare sta¡ and patients generates information. Computers and IT process
and distribute information that is central to healthcare. Information is what we
use to make decisions, and good decisions need information about patients, as
well as medical knowledge. The amount of information we need is quite stagger-
ing and doctors and nurses spend about two hours every day and about 15% of
their budgets on gathering and managing information.
We are only now waking up to the potential of IT and electronic communica-
tions to make decision-making easier. We are only now beginning to realise the
full potential of sharing this information openly between the healthcare profes-
sions, on the one side, and the consumers of healthcare, on the other. We are
now at the point of realising the dream of providing equal access to information
for clinicians, managers and consumers.

The problems
In the last few years the speciality of informatics has led these developments, but
it has to be said that information management in the health service has not had a
good record. There has been a poor return on investment in information. The
capital costs of installing computer systems in hospitals and surgeries are sub-
stantial, and there is no doubt that in the past money has not been spent wisely.
Annual maintenance is high; sta¡ training has often been poor; upgrades are
expensive and theft of hardware can be a problem. It is sometimes unclear what
bene¢ts have resulted from all of this investment, and evaluation of information
systems in the NHS has not been good.
144 Vision and value in health information

It is not just hardware that disappears from hospitals, data disappear too.
When information about patients or healthcare sta¡ falls into the wrong hands,
either by accident or by deliberate theft, patients become unhappy about their
personal data being stored in the computer, sta¡ can imagine that management
is spying on them and there is a resulting loss of quality of data until there is a
possibility of bad decisions being made.
People need to be able to rely on information systems, and failure can cause
clinical problems. However, total system failures are very uncommon, but
many health service systems are unreliable and have not been developed with
the same rigorous approaches that are usually made in other systems where
safety is critical, such as air tra⁄c control.
There are issues here, too for the developing world; the increasing costs of IT
could widen the gap between rich and poor countries. Perversely, developments
in the Western world could then hinder, or even deny access to information to
those in the developing world. At a basic level, the infrastructure to manage
information may not be present, and even when it is, the costs of processing
and disseminating information may be too great for local economies to bear.
In the West, the availability of huge amounts of health information tempts
quasi epidemiologists to go on ‘¢shing expeditions’, particularly in the area
of clinical outcome studies, and this kind of unscienti¢c approach brings epi-
demiology into disrepute.

The bene¢ts
After such a long list of areas where information and computers have had a bad
track record in the past, we need to look at examples where there are bene¢ts.
Some of these developments have been hindered because of our inability to
demonstrate bene¢t, and in the context of a continuing debate on the con¢den-
tiality of health information, it is right that this question has been asked.
Of particular signi¢cance to healthcare personnel, in particular clinicians, is
the development of relatively inexpensive workstations. The availability of an
o⁄ce suite of software to do word processing, spreadsheet analysis, presenta-
tions and database management has empowered healthcare professionals to
access and manage data in a way that was unimaginable only 20 years ago.
Patient-based information systems with access on a ‘need to know’ basis are
now commonplace. Data can be added to, and analysed, in ways to match indi-
vidual clinicians’ needs and can be developed into management information
systems. There are new ways of entering data, which speed up the process for
professionals, such as optical character recognition and speech recognition.
These developments further ‘ownership’ of data, and can only lead to improve-
ments in accuracy and comprehensiveness.
 Using information for public benefit 145

Health information and how it is used

The way data about individual patients are now generally stored across several
sites poses challenges to their use for public health purposes. Although most
systems provide more than adequate access to professionals, it is perhaps only
by extending this access to the patients themselves that the highest standards of
accuracy could be achieved. The healthcare system tends to have little enthu-
siasm for sharing data with patients, and patients themselves are increasingly
concerned about how their data are handled. In a sense, these concerns are
justi¢ed when one brings to mind the way the information about individual
patients is passed around between healthcare providers, with insurance com-
panies, with researchers, with national statistical o⁄ces, with prescription
bureaux and the like.
Ordinary members of the public are now managing their bank accounts, their
telephone bills, their investments and their purchasing over the internet. They
are increasingly turning to on-line sources for information about their health,
and consumers will want to see the same access and control for their own med-
ical records. There need to be clear standards developed to guard patients’ priv-
acy and those systems should be developed involving patients themselves.
The development of the internet and the world wide web has had a dramatic
impact on health informatics. Professionally oriented databases, such as MED-
LINE and PubMed, have transformed the way health professionals work. It is
perhaps no coincidence that alongside IT developments, evidence-based clinical
practice has become the norm. Professionals need almost instant access to the
most up-to-date information, and standard textbooks can no longer meet this
requirement. Their advice about treatment, prognosis and prevention is fre-
quently out of date before the book is even published. Doctors get better at
diagnosis and judgement as they get older, but their knowledge increasingly
falls behind. The amount of time available to seek out information for the multi-
tude of clinical problems with which they are presented each day gets less and
less. There is simply not enough time in the day to do all of these jobs.
Recent developments in informatics have supported an evidence-based
approach to clinical medicine. There are now well developed strategies for track-
ing down evidence and assessing its quality. Systematic reviews of research and
short summaries, such as the Cochrane Collaboration, make sense out of
research and lead to better decisions. Most importantly, search strategies and
extraction of evidence can now be done through the internet, from wherever
one is located in the world.
The same sources of information are available to consumers of healthcare.
In the past, health informatics were focused on the needs of the provider. This
focus is now moving towards the consumer, a process which is driven by the
development of evidence-based clinical practice and the recognition that
patients and their carers should have equity of access to information. In Western
146 Vision and value in health information

healthcare systems, the move towards more consumer-focused information

bases is partly the result of e¡orts to control rising healthcare costs by empower-
ing patients to participate in decisions a¡ecting their own health. Patients are
now very willing to take on more responsibility for their own health, and
the ready availability of information, particularly on the world wide web, is
accelerating this change. The new public health agenda emphasises the health
of populations as opposed to individuals, and the bene¢ts of primary prevention
and the availability of information to both providers and consumers will lead to
a new partnership for health, in which consumers use IT to gain access to infor-
mation about their own health and control their own healthcare utilising
scarce resources more e⁄ciently.
It is more than possible that these developments will lead to ‘cyber doctors’,
whose practice is entirely located within cyberspace. There will be a major
challenge in regulating the professional practice of a doctor who may not
be human. Already in the UK we have NHS Direct, which uses computer algo-
rithms managed by trained sta¡, to o¡er a ¢rst point of contact to patients.
Patients are becoming more and more con¢dent in handling information
sources like this, and will increasingly expect immediate access to every kind
of information about them or their health.
Every clinician will be familiar with patients or their representatives who
come into the clinic with a whole ¢le of information ‘o¡ the web’. Whilst doc-
tors, in particular, often ¢nd this a threatening process, there are real bene¢ts in
patients having access to the same information. Doctors should always remem-
ber that patients just might ¢nd some information relevant to their condition
that they have not come across! The na|« ve consumer might be seduced by
information found on the internet, assuming it to be accurate, up-to-date and
valid. This, of course, is far from true, and currently there is no regulation of
information on the internet to guarantee safety and e⁄ciency. There are many
websites devoted to promulgating information about particular health prob-
lems or diseases. The information is frequently unmoderated and lacks peer
review. Evidence may be presented without any assessment of its quality and
often in a dogmatic style, which gives a false impression of veracity and author-
ity. Organisations like the British Medical Association in the UK are help-
ing enormously by publicising good websites and it is essential that this kind
of evaluation is conveyed to consumers. In addition to that, there should be
a system of regulating health informatics on the world wide web, probably
of a voluntary nature, requiring ^ among other things ^ websites to identify
authors, their institutions and published peer reviewed articles.
Consumerism in health information empowers patients by giving them
access to and control of information. This should include information on their
own personal health, such as current diagnostic information, pathology and
laboratory results, risk pro¢le, details of prescriptions and untoward e¡ects
 Using information for public benefit 147

and so on. Giving patients their medical records is not new, and the idea of a
patient-held medical record o¡ers some signi¢cant advantages.
The issue of con¢dentiality meant that in the past, patients did not have
access to their medical records. Current UK legislation gives patients access to
records made after November 1991 and before that time, if it is necessary for the
patient to understand what was written later. There are occasions when access
is denied because the doctor feels that some information may be harmful to the
patient. Patient-held records circumvent most of the issues about con¢dential-
ity and also have enormous practical advantages to both doctors and patients.
Patient access to their own medical records could be achieved either through
the internet or through smart cards, or through a combination of both. The
smart card would be able to store personal information directly, and would also
serve as an electronic key to provide access to information stored elsewhere.
Users will be able to ¢nd and read their medical records wherever they are, and
ensure that information is accurate and timely.
The internet is a vast treasure house of information, but to achieve its full
potential users should be pointed in the direction of good-quality information,
and websites should instruct them how to assess the quality of information. The
internet is notoriously di⁄cult to control, and indeed one of its main attractions
is that it is ‘uncontrolled’. However, ensuring quality is an urgent priority and
cannot be achieved easily. Consumers must be educated in how to phrase ques-
tions and search for information to answer those questions. This education
could be built into websites; it is unlikely that many people would enrol on a
course when they feel that ¢nding information on the internet is so easy. Provi-
ders of health information need to regulate themselves and an independent
assessor should evaluate their information and issue some kind of ‘seal of
approval’. A website having this cachet should then be guaranteed high-quality
information. Lastly, there need to be sanctions available to stop the distribution
of information that is of such poor quality that it could be harmful.
Mention was made earlier of problems in health information in developing
countries. There is a well-known story of how Bill Clinton, when President
of the USA, visited a village health centre in India. He watched a woman
access the world wide web on a computer and get information on how to care
for her baby. Writers have retold this story as an illustration of the possibility
that this woman would then get better healthcare for her baby because of the
availability of information on the internet. This is far from a common situation
in developing countries, and on a recent visit to Nigeria I met very few public
health colleagues who had any access to the internet themselves, let alone
patients. The population of Africa is about 700 million, and less than one mil-
lion have access to the internet, and eight out of 10 of these individuals are in
South Africa. This is probably more representative of the informatics situation
in the developing world than what President Clinton saw in India.
148 Vision and value in health information

Despite the potential value of health informatics in reducing inequalities in

the developing world, the sad truth is that the digital divide is more dramatic
than any other health inequality. There are very substantial ¢nancial barriers
to the use of the internet in the developing countries, and in Nigeria a subscrip-
tion to a service provider and telephone costs are in the region of »500 per
annum. In a country where the average earnings are less than a dollar a day,
this is a barrier that for most people cannot be breached.
In Nigeria I met professional people delivering primary care of high quality
who had no access to literature, no textbooks, no system of recording or analys-
ing information and no method of measuring their progress towards targets.
In this sort of environment, public health achievement will be slow. No amount
of advice or support from the rich West will make much di¡erence when health-
care professionals and consumers are starved of information. There are some
breathtaking IT developments, which could make access to information world-
wide. Satellite technology is relatively inexpensive and has already been
established in 30 countries, including Asia and Africa. The SatelLife project is
supported by the World Bank and is even going beyond communications with
sponsorship of a training centre in Kenya where professionals can be taught
how to use information technology.
The Untied Nations Millennium Assembly proposed that the right of univer-
sal access to information and communication services should be added as a new
component of the United Nations principles and conventions on human rights
and developments. That right is our responsibility, and it can only be implemen-
ted through a truly global initiative to improve access to the internet.

And Edith K˛rner?

Edith K˛rner was chair of a steering group that was to re-design the informa-
tion-gathering system of the NHS and its management use. Sadly, her work
was never allowed to be completed, and those of us who worked with her are
very much aware that she is remembered in the current NHS largely for a set
of data collection forms widely regarded to be unhelpful and tedious; that was
never her objective. These are not suitable memorials for Edith K˛rner. If she
had been allowed to complete her work her steering group would by now have
moved into developing informatics and information services along the lines that
have been outlined in this brief essay. Twenty years on, it is possible to speculate
on how many of these proposals would have been well and truly sorted out if she
had been allowed to pursue her vision and goal.
CHAPTER 12

Globalisation or localisation:
common truths or
local knowledge?
Michael Rigby

The work of Edith K˛rner was commissioned speci¢cally within the UK, and
indeed within England. Consequently, most of the essays in this book are
rooted in issues of the English healthcare system.
But we are currently in an era of globalisation. Communications, mass media
and consequent awareness and expectations mean that news and ideas, and
indeed expressed consumer views, travel rapidly around the world. Much of
healthcare claims to be evidence-based.1^ 4 So, is not the development of locally
designed information systems parochial and perverse? Did not Edith K˛rner’s
work implicitly draw upon international experience, and are not her principles
and her learning points generic and universally applicable? If her work was so
seminal, should we not be holding up her guiding principles as universal ideals?
These questions raise a number of important issues. The ¢rst of these is
the paradox, and the tension, between healthcare delivery and its supposed
scienti¢c underpinning. There is a generic desire by the range of stakeholders
from consumers to health professionals to base healthcare delivery upon best
scienti¢c knowledge and re¢ned expertise. Yet at the same time all stakeholders
want healthcare delivery to be localised, both culturally and organisation-
ally ^ and healthcare information systems are integral to healthcare delivery
systems. With the combination of increasingly specialised clinical services and
increasing exchange of reference information through modern communica-
tions, this focus on localised delivery is looking increasingly di⁄cult, yet is
becoming increasingly reinforced.
It is an interesting phenomenon worldwide that healthcare is always a
responsibility of devolved levels of government. This is true in the UK, where
contrary to international and indeed local indigenous consumer belief, there is
150 Vision and value in health information

no such organisation as the ‘British National Health Service’. Within the UK

healthcare is devolved to the four home countries of England, Northern Ireland,
Scotland and Wales, and with the recently increased amount of devolution
those health systems are further diverging, to the point of causing tension
because of the di¡erences on speci¢c policy issues. In federal countries such as
Canada and Australia the same principle has always applied, with health-
care being the responsibility of provincial or state governments. Nor is this a
British Commonwealth phenomenon ^ the same principle of devolved health-
care delivery exists for instance in the USA, where it is a state responsibility;
Germany, where health is delivered by the L8nder; Sweden, where counties
are responsible; and Finland, where municipalities are responsible. This under-
pins the perceived value of localisation as an important consumer and political
principle in healthcare, despite the increasing recognition that many principal
therapeutic services, such as trauma and oncology, as well as regional spe-
cialities, are best delivered for catchment populations of millions rather than
thousands. In the generalist view, sensitivity superimposes on science; commu-
nity and culture a¡ect clinical principles.
And that highlights a second fundamentally important point, that scienti¢c
truth in healthcare delivery is less universal than is often assumed. And here
the consumer paradox works in the reverse direction ^ the public hears (often
through the mass media) of the latest drug development, or surgical innovation,
in another country and instantly demands access to it, ahead of any veri¢cation
of claims, or consideration of the therapeutic e¡ects or the ethical implications
within their own local situation.

The importance of local context

Above all, health is a personal issue and a personal demand. But despite the self-
survival instinct, which usually assumes that biggest pro¢le means best evi-
dence, in fact a range of supposedly objective scienti¢c facts have very di¡erent
meaning, values, and application at national or local level.
This is an under-recognised issue, which causes di⁄culty when well-inten-
tioned philanthropists or organisations seek to make health knowledge available
internationally. Scienti¢c ‘fact’ from, say, North America or Europe may not
be universally applicably within those global regions, and even less applicable
in other parts of the world. The proponents of ‘knowledge’ transfer argue that
there are empirical facts that can be grouped together into information
and knowledge, which should then be o¡ered on a wider basis to less well-o¡
economies in the interests of the common good. But in fact comparatively few
scienti¢c health facts are truly universal. Health and health facts are very per-
sonal to the complexity of the human body, to society, to health systems and to
cultural values. Thus, any ‘scienti¢c’ fact requires assessed contextualisation.
 Globalisation or localisation? 151

Any such ‘fact’, whether it is scienti¢cally based, such as a drug response, or

more related to care delivery, such as the most e¡ective programme of a pattern
of care, needs to be contextualised into the local setting because a range of fac-
tors need to be take into account to give local context. These can be classi¢ed as
including the following.

Genetic di¡erences
Not all groupings within the human race have the same physiological or
genetic make-up. Di¡erent diseases or illness vulnerabilities a¡ect speci¢c
groups, often in ways that are not understood, such as the di¡erence in the inci-
dence of cystic ¢brosis between Finland and its near neighbours, or the fact that
the pro¢le of diabetes in Sardinia replicates that of northern Europe rather than
southern. Other factors are well-known, such as vulnerability to sickle cell dis-
ease by those of Afro-Caribbean origin, with this trait following communities as
they relocate across the globe. Height and body mass also vary by grouping, and
so a whole range of tables from growth velocity of infants to pharmacological
absorption need to be adjusted for di¡erent ethnic groups. And these are just
the more obvious di¡erences that are known.

Climatic di¡erences
The human population is distributed from Artic environments to tropical ones;
however, much scienti¢c research is based in countries with temperate climates.
Many aspects of healthcare are a¡ected by meteorological factors, including
temperature and humidity, in£uencing issues ranging from the best prophylac-
tic approaches to the prevention of pressure sores or to pharmodynamics.
In turn, these variations will in£uence issues such as anticipated length of
stay. It is unsafe to assume that research or clinical trial data from one climatic
region apply optimally in other climatic regions without adjustment.

Cultural and societal di¡erences

Societies and cultures have di¡erent values, many of which are particularly
important and precious. These include the role of the extended family varying
from being highly active and involved in care to the extent that they take
up residence at hospitals and provide patient food, through to the belief in
the autonomy and independence of the individual and the expectation that the
health system will comprehensively care for their relative ^ with clear e¡ects
upon healthcare delivery patterns and resource assumptions. Other societal
152 Vision and value in health information

and cultural (including religious) beliefs will have major implications for a
range of health conditions from teenage pregnancy to the acceptability or
otherwise of organ transplants and blood transfusions. These all undermine
the assumption of ‘universal scienti¢c fact’ in clinical and treatment issues.

Health system
There are many types of health system, both with regard to the type of provision
being either by the public service, private sector or not-for-pro¢t organisations,
and as to whether payment is at the point of consumption, based upon insur-
ance or provided through a tax-funded ‘free’ public service. These variations in
health systems will have major e¡ects upon the information systems needed,
and will also a¡ect pro¢les of delivery patterns. Almost as signi¢cant are varia-
tions in the interface with other services, not least social welfare systems and
social care, but also including housing and education. A health system is not
an island, but a focus. And the information it uses needs to recognise the spe-
ci¢c local bridges to other services, and the amount and nature of the tra⁄c
over them.

Economic status
The healthcare that a country can a¡ord varies tremendously. What is a normal
service and reasonable consumer expectation in one country, such as for
instance a prosthetic hip replacement or open heart surgery, may be neither
physically possible nor economically a¡ordable in another country. At a more
mundane level, the availability and a¡ordability of drugs will vary tremen-
dously. Thus many forms of supposedly objective health information, from pre-
ferred prescribing patterns to outcome measures, as well as the treatment
pro¢les themselves, vary from country to country, and thus data from one
country may not only be irrelevant in another, but actually harmful in that it
will skew aspirations or interpretations.
Those issues concern the information contained within systems, the use of
supposedly ‘universal’ scienti¢c facts, and the application of comparative ana-
lyses. But what about health information systems themselves, the channels
and conveyors of such information? Are there not generic principles that can
apply here?
The answer to that must be mixed. And here the legacy of Mrs K˛rner, and its
under-appreciation, applies yet again. Many of the principles that her steering
group established apply universally, including the importance of timely and
accurate data, and the principle that understanding and relevance must come
from the operational levels upwards to the strategic levels. However, almost
 Globalisation or localisation? 153

everywhere the experience is that information systems are developed on pre-

cisely the opposite principle ^ they are designed and imposed on a top-down
basis, and the e¡ect is both to load and to disenfranchise the operational levels.
This is true in general right across the globe, from highly developed countries to
Paci¢c islands.5 Second, and linked to this, information systems are usually
designed and implemented by those with the resources ^ namely the central
level ^ with little thought about the views or experience of operational level
stakeholders, or concern about meeting their information needs. And, as Mrs
K˛rner discovered to her cost, central sponsors look for quick answers and
immediate solutions. Furthermore, they are reluctant to undertake evaluation,
and thus re¢nement and betterment, for a whole range of reasons, including
but not restricted to, the desire for minimised investment and also avoidance of
loss of credibility.6
So, if health information itself cannot be considered universal or global, and if
information systems need to be sensitive to local cultures and health systems, is
it a sad conclusion that the challenges addressed by Edith K˛rner have to be
replicated in every country? Is there not a place and a means of sharing vision?
Fortunately, there are a number of positive ideas that can be put forward, even
if the climate is not exactly welcoming because of the expediency and parochi-
alism so often faced.

Rays of light
First, not all countries take the short-term and minimal investment approach
that Edith K˛rner faced. Some countries recognise the importance of developing
health information systems, and for doing so in an open and consultative way,
with a strong move towards evidence-based and need-based systems. Exemplars
of this are Canada, with its strongly supported Canadian Institute for Health
Information,7 and the Commonwealth Government of Australia, with its devel-
opment of an evidence-based e-health system.8 This investment in evidence,
planning and consultation must surely be bene¢cial in the longer-term, and
the vibrancy of the reports produced by the Canadian Institute for Health
Information and Statistics Canada, and the credibility of Australian health infor-
mation policy documents, would give credence to this.
Second, it ought to be possible to have much better exchange of knowledge
and experience about the development of health information systems, and the
availability of comparable data. One positive example is within the European
Union, where the comparatively new health monitoring programme seeks to
provide common analyses of health problems, mortality and morbidity, and
health systems’ e¡ectiveness across all the member states. There are many prob-
lems to be faced in this, given the variety of cultures and health systems across
Europe, but it is an ambitious programme which is now well under way.9
154 Vision and value in health information

But on a wider basis, it is highly desirable to share experience of development

of information systems themselves, a subject that is currently more restricted to
laudable but probably ill-conceived (and sometimes not impartially motivated)
drives simply to transfer information systems. Rather than taking this simple
transplantation of systems, which will have the unfortunate e¡ect of little local
ownership and high probability of limited relevance, it would be much more
appropriate to help local opinion leaders and stakeholders to understand the
issues and seek to assemble their own structure from experience and compo-
nents available elsewhere ^ demand-led rather than donor-driven. This would
also reduce the risks of e-colonialism ^ the imposing by well-o¡ countries of
inappropriate information and systems on to the cultures of economically
lesser-developed nations through ease of electronic transfer of content without
regard to context.
This sharing of experience would probably best be discharged by seeking
to assemble learning sets of those who were seeking to develop information
systems, and to overlap those with learning sets of those willing to supply ideas
or systems but who did not have the same local knowledge of requirement and
feasibility. Such learning sets could be focused upon particular common interest
constituencies ^ for instance restructuring countries of Eastern Europe, middle-
income countries of Latin America or developing countries in Africa. By coming
together on a basis of mutual respect and willingness to learn on a principle of
equal standing between both parties (as opposed to any apparent assumption
of superior knowledge or status), progress could be made far more construc-
tively. This ought to avoid both the imposition of systems doomed to failure
from the beginning (of which there are too many expensive and resource-
diverting examples),10 and instead for each party to develop their thinking
based on what they learn across the table from the other.11

Developing Joint Supporter/

country/countries debate/ country/countries
learning set learning learning set

Figure 12.1 A framework for the sharing of experience.

 Globalisation or localisation? 155

A framework for this has been postulated,12 as shown in Figure 12.1.

A number of potential vehicles exist, not least working groups of the Interna-
tional Medical Informatics Association and its regional a⁄liates such as the
European Federation for Medical Informatics. The most challenging problem is
that of ¢nding resources, for either group. For developing or middle-income
countries the di⁄culty of their local experts coming together is self-evident.
Unfortunately, it is similarly di⁄cult for potential advisers from more a¥uent
countries to come together in working meetings if there is little possibility of
expenses being recouped by commercial sales.
This seems a prime opportunity for funding by appropriate philanthro-
pic bodies, whether these be traditional aid donor agencies, intergovernmental
bodies or large private philanthropic trusts. It will be far more e¡ective, because
of the associated commitment and thus the enduring nature of the resultant
health information systems, if investment is made into this type of learning set
and exchange of experience rather than simplistic donation of culturally and
organisationally extraneous systems.
Positive examples now exist of organisations seeking to make regionally or
culturally customised health information available, usually on the internet,
addressed to a particular locality. For example, a charitably underpinned sys-
tem enables irishhealth.com13 to publish a regular bulletin summing health
information matters relevant to the country of Ireland, speci¢c to its culture,
local health problems and healthcare delivery system as well as touching
on local health news topics of interest. Quite di¡erently, recognising that much
health information at a global level is not directly applicable to the beliefs and
cultures of the Arab world, a website has been developed in Saudi Arabia to
produce Arab health news.14 A third example is that of the Eastern Mediterra-
nean Regional O⁄ce of the World Health Organization (WHO), which is far
more proactive than the other regional o⁄ces in seeking to translate selected
published material into Arabic for local use.15

Conclusion
So having started with the question as to whether the insight into health infor-
mation as created by Edith K˛rner and her team, and the related e¡orts and
frustrations they encountered, have to be replicated in every country, the
reply must be that there are better ways than this expensive duplication and
frustration. However, there is also the dangerous paradox that attempts to
share this learning run the great danger of creating the fundamental mistake
of turning into global, and top-down imposed, ideas and solutions, when what
is actually needed is systems and information that meet local need and respect
156 Vision and value in health information

the issues and values of local healthcare delivery. And in meeting that conclu-
sion it must be reiterated that this was a fundamental principle of Edith K˛rner
herself. So if the adverse learning points of the K˛rner experience in developing
healthcare information systems are to be avoided, the key underlying principle
of Edith K˛rner ^ the focus on the needs of local healthcare delivery ^ must be
the driving force. In the modern global communications century, the dangers as
well as the bene¢ts of instant global health information communication must be
assessed in any situation and setting.16 That suggests three conclusions.

Mutual learning
First, any attitude of proclaiming answers or knowledge from one country to
another, and particularly from an a¥uent global region to areas less well o¡,
must be avoided as not only demeaning, but likely to be irrelevant, ine¡ectual
and thus wasteful of resources.17 Rather, a context of mutual learning needs to
be developed so that local opinion can develop local systems, but based on
sound principles and knowledge gained from open discussion and exchange.
Second, any form of information system or information itself needs to be made
appropriate for the local setting. Sultan Bahabri, founding Chief Executive of
the International e-Health Association, coined the phrase ‘glocalisation’ for
this.18 This term neatly encapsulates the need to take global information and
render it appropriately local. But that too requires resources.
Third, there is no real leadership or vehicle for this localisation process.
Though most informed commentators see information as vital to the healthcare
process, it is unfortunately the situation that the WHO does not provide strong
leadership in this ¢eld, though individual programmes, and some of the regional
o⁄ces, have particular exemplary programmes and strengths. Indeed, some
WHO initiatives, such as the HINARI programme of low-cost electronic publish-
ing access to the third world, run the risk of making external and potentially
inappropriate information more available in developing countries than appro-
priate local information.19

Global vision of localised systems

Thus there is a need for an approach which brings together the di¡erent types
and foci of stakeholders, appreciation of their individual values, and an accumu-
lated body of currently scarce evidence as to best types and content of di¡erent
types of information and media for its supply, so as to develop policies and initia-
tives. This information and input needs to be from developing countries, and
supportive agencies, as much as from the Western world or donors.20^23
 Globalisation or localisation? 157

The International Medical Informatics Association seeks to provide a frame-

work for automated information experience and policy, but has practical
di⁄culty in reaching the underdeveloped and developing worlds. Bodies such as
the World Trade Organization and International Telecommunications Union
seek to regulate health information activities within those aspects that fall to
their own competence. But overall there is a lack of vision of the K˛rner style
at international level. There are no strong vehicles for funding organisational
learning as opposed to products, in a way that would be bene¢cial.
Within the UK, the fact that Edith K˛rner was an informed visionary has only
been learnt slowly over two decades, and the opportunities her work presented
in her adoptive country were not taken up at the time. As we move to an era of
seeking much greater worldwide understanding of common social and scienti¢c
issues, the same situation applies with regards to true global understanding of
health information systems. It is an important learning point from the visions
Edith K˛rner provided, and could be a ¢tting recognition of the deeper contribu-
tion she made, if understanding of her approach could be a catalyst to stimulate
development of appropriate vehicles for global understanding and resourcing
for the building of locally relevant and e¡ective health information systems in
countries where they are most needed.

References
1 Cochrane A (1972) E⁄ciency and E¡ectiveness: random re£ections on health services. Nuf-
¢eld Provincial Hospitals Trust and British Medical Journal, London; printed in revised
format, 1989.
2 Davido¡ F, Haynes B, Sackett D et al. (1995) Evidence-based medicine. BMJ 310: 1085^6.

3 Sackett DL, Richardson WS, Rosenberg W et al. (1996) Evidence-based Medicine: how to
practise and teach EBM. Churchill Livingstone, Edinburgh.
4 Roberts R (1999) Information for Evidence-based Care (Harnessing Health Information
Series No. 1). Radcli¡e Medical Press, Oxford.

5 Finau SA (1994) National Health Information Systems in the Paci¢c Islands: in search of
a future. Health Policy and Planning 9: 161^70.
6 Rigby M (2001) Evaluation: 16 powerful reasons why not to do it ^ and 6 over-riding
imperatives. In: V Patel, R Rogers and R Haux (eds) Medinfo 2001: Proceedings of the 10th
World Congress on Medical Informatics. IOS Press, Amsterdam.
7 (www.cihi.ca)

8 (www.health.gov.au/healthonline/welcome.htm)
9 (europa.eu.int/comm/health/ph/programmes/monitor/index^en.htm)
158 Vision and value in health information

10 Rosenow D (2000) Health Management Information Systems Theory and Reality in

Developing Countries ^ A Case Study from the Yemen. Dissertation in part ful¢lment
for the Degree of MBA (Health, Population and Nutrition), Keele University.
11 Chitalika J (2002) An Assessment of the Need for Computerization of Information Sys-
tems at District Level of the Health Sector in Tanzania. Dissertation in part ful¢lment for
the Degree of MBA (Health, Population and Nutrition), Keele University.
12 Rigby M (2000) Presentation to EFMI Working Group 10 at Medical Informatics Europe
Conference, Hanover.
13 (www.irishhealth.com)
14 (www.healthgulf.com)

15 (http://208.48.48.190/his.htm)
16 Rigby M (1999) The management and policy challenges of the globalisation e¡ect of
informatics and telemedicine. Health Policy 46: 97^103.
17 Rigby M (2000) And into the 21st century: telecommunications and the global clinic.
In: M Rigby, R Roberts and M Thick (eds) Taking Health Telematics into the 21st Century.
Radcli¡e Medical Press, Oxford.
18 Bahabri S (2002) From Telemedicine to e-Health ^ what’s next? Opening presentation
to First International e-Health Association Conference and Exhibition ‘e-Health 2001’,
Jeddah, Saudi Arabia.
19 (www.healthinternetwork.org)
20 Arunachalem S (1999) Informatics in clinical practice in developing countries: still
early days. BMJ 319: 1297.
21 Tan-Torres Edejer T (2000) Disseminating health information in developing countries:
the role of the internet. BMJ 321: 797^800.
22 Costello A and Zumla A (2000) Moving to research partnerships in developing coun-
tries. BMJ 321: 827^9.
23 Lam CLK (2000) Knowledge can £ow from developing to developed countries. BMJ
321: 830.
 Index

Academy of Royal Colleges, secondary care care planning, and patient records 57,
system speci¢cations 75 61^2
accident and emergency records project 7 CAS see complex adaptive systems
accountability 84^5 CASP see Critical Appraisal Skills
information requirements 85^6 Programme
internal control model 87 central data returns see hospital episodes
see also governance; risk management statistics (HES); minimum data sets
activity levels, as measure of service change management
performance 125 complex adaptive systems (CAS) 35^6
adverse events reporting 83 and knowledge driven paradigms 53^4
avoidable incidents and health outcome leadership needs 38^41
indicators 94, 97^9 Chantler, Cyril 18^19
advocacy role of patient records 57, 62^6 child health information systems 107^18
requirements of record systems 64 background and context 108^9, 122^4
American College of Pathologists, potential problems 109^10
SNOMED-CT collaboration 74 principles and recommendations
American health information 115^17
strategies 33^4 system requirements 111^14
government reports 35 child protection issues 63, 113, 117^18,
IOM reports 35, 60 124
amputations, as outcome indicator chronic illness
97^8, 98 impact on healthcare 37
appraisal, and clinical performance and IT assisted home monitoring 41
measures 72 civil service
audit membership of Steering Group
and Governance Standards 80 Committee 3
see also governance; health outcomes; see also political trends
risk management Clark, J 125
Australia health information strategies Clay, Trevor 20
153 clinical decision support systems (IT)
41^2, 73
Bahabri, S 156 clinical governance 87
Barker, W 125 see also governance
British Medical Association clinical information needs 27, 69^75
and access to health research accident and emergency records
information 146 project 7
on NHS information requirements 19 evidence-based best practice knowledge
systems 70^2, 145
Cadbury, A 84^5 hospital episode statistics 72, 104^5
Canada health information strategies 153 information overload 49, 145
160 Index

clinical standards, thesaurus development Critical Appraisal Skills Programme

and SNOMED-CT 74^5 (CASP) 51
clinical terms project 74 Crossing the Quality Chasm (IOM 2001)
see also headings project; language 34, 36
clinicians culture
attitudes to data capture and resistance to data capture 20
requirements 20^1, 22, 128 societal values and healthcare
attitudes to management 40 implications 151^2
leadership development 39^41
time spent on data capture 128, 143 data capture
‘clinicians in management’ initiative 2 aims and objectives 16^17
Cochrane collaboration 71, 145 and clinicians time 128, 143
coding, validity of HES returns 100 costs 21, 143^4
The Collection and Use of Information about impact of national service framework
Dental Services (1985) 8 targets 70
commercial sector in£uence 13, 59 limitations of discrete episodes 75, 104
community care generic record-keeping 75 and organisational instability 107
Compendium of Clinical and Health sta¡ attitudes 20^1, 22, 128
Indicators 94 Steering Group recommendations 16
complaints reporting 83 training needs 7, 16, 18
complex adaptive systems (CAS) 35^6 see also information management;
Computer Policy Committee see NHS minimum data sets
Computer Policy Committee (1981) Dawtry, E 125
computerisation see information technology Day, Christopher 8
(IT) systems death as outcome indicator
condition-speci¢c data-sets, and national see peri-operative deaths statistics
service framework targets 70 decision support systems (IT) 41^2, 73
con¢dentiality developing countries healthcare
child health information systems information needs 144, 147^8
107, 110 appropriateness of initiatives 148, 156
and integrated electronic records 70 demand-driven v. donor-led 154
legislation 147 mutual learning initiatives 154^7
Steering Group standards 7, 16 devolution of healthcare 149^50
and third parties 145 DHSS
consultant episodes 10, 72, 75 central data returns 6
consumer-focused systems 145^7 policy towards Steering Group’s
medical records as patient advocates recommendations 3, 10
57, 62^6 discharge statistics, as health-related
patient information needs 41, 70 function outcome 99
control systems disease speci¢c data
de¢ned 81^2 health outcome development work 104
see also governance; risk management and national service framework targets
controls assurance 80, 83 70, 71
website 88 district minimum data sets see minimum
see also governance data sets
cost e¡ectiveness measures 128 document management 52^3
cost implications scale of problems 49, 145
installation of IT systems 143^4 see also information management
Steering Group’s recommendations 21 Donabedian, A 131
 Index 161

education programmes genetic di¡erences 151

in developing countries 147^8 societal and cultural values 151^2
and leadership 39^41 system di¡erences 152
new IT systems 73 see also developing countries healthcare
Ehnfors, M 126 information needs; USA health
electronic patient records see information information strategies
technology (IT) systems; patient records Goosen, W 126
Elston, Mary Ann 20 government policy see NHS strategy and
emergency re-admissions performance policy; political trends
indicators 98^9 governance 79^89
encounter records see Omaha project background 81^5
epidemiological requirements The Governance Standard 79^81, 80
and networked patient record systems information requirements 86^8
63^4 intelligent accountability concept 84
and Steering Group’s remit 3 see also risk management
episode-based medical records 69^70, 72 Gowing, W 26
episodes of care project 75 Greenbury, R 84^5
European Union, health monitoring Gri⁄ths Report (1983) 20
programme 153
evidence adaptive clinical decision-support Hampel Committee report 84^5
systems 42 HC84(10) 17^18
evidence-based knowledge headings project 74^5
and advances in IT systems 41^2, 73, health authorities’ information
145 requirements
clinicians requirements 70^2 and performance assessment frameworks
and health policy requirements 33^4, 95
37 Steering Group’s recommendations
National electronic Library for Health 71 6, 7^8
validity concerns 71 health histories
The Expert Patient (DoH 2001) 47^8 and IT 41
see also patient records
Feachem, RGA 32 health needs assessment
¢nancial management standard 81 and IT evidence-based knowledge 41
Finland, nursing documentation 122^3 national outcome data 104
First Report to the Secretary of State (Steering NHS performance assessment
Group 1982) 9, 19 frameworks 95
Form SBH 203 (waiting list data) 2 and Steering Group’s remit 3, 27
Form SH 3 (annual facilities return) use of patient records 57, 63^4
1^2 health outcomes
Fostering Rapid Advances in Health Care as cost e¡ectiveness measure 128
(IOM 2002) 35 de¢nitions 92, 131
Fowler, Norman 9 disease speci¢c development work 104
From Facts to Figures 9 DoH/national centre 91^2
functional ability and discharge
generic record-keeping 70, 72 statistics 99
and child health information needs 107 and local information needs 104^5
globalisation of healthcare 149^57 measurement criteria requirements 132
climate in£uences 151 national comparative data 104
economic status 147^8, 152 and NHS performance frameworks 102^3
162 Index

health outcomes (continued ) Information for Health: a strategy for building

for nursing practice 127, 131^4, the national health information
138^40 infrastructure (National Committee on
omission from Steering Group’s remit Vital and Health Statistics 2001) 35
18^19, 27 Information for Health (NHS Executive 1998)
as ‘results’ 93 11, 75
routine data sets as proxy indicators information management
94^5, 96^100 aims and objectives 50^4
USA experience 126^7 of child health records see child health
health professionals see clinicians information systems
health status measures 27 cost implications of technological
and IT assisted home monitoring 41 advances 21, 143^4
Health Survey for England 104 developments 144
health visiting information systems of documents 52^3
121^41 and e¡ective leadership 39^40
background history 122^7 of evidence-based knowledge 41^2, 65,
development of encounter records 70^2
see Omaha project and globalisation of healthcare 149^57
see also nursing documentation knowledge paradigms 48
healthcare paradigms, and knowledge tasks and organisational instability 107
54 of patient records see patient records
healthcare processes, IOM guidance pre-1980 1^2, 6^8, 14
36^7, 37 problems 49, 69^70, 72, 73^4
healthcare strategies see NHS strategy and for promotion of best practice 48^9,
policy 70^2
HES see hospital episodes statistics and quality improvements 48, 52, 53^4
HINARI programme 156 for resource management 57, 71^2
hip replacement HES data, as quality of life role of libraries 49
indicators 96^7, 97 Royal Commission Report’s ¢ndings 1^2
hospital activity analysis systems and scope of healthcare knowledge 48^9
episodes of care project 75 time costs 143
post-1982 73, 75 top-down approach 152^3
pre-1982 1^2, 72 see also data capture; information
see also hospital episodes statistics (HES) technology (IT) systems; patient
hospital episodes statistics (HES) 73, 75, records
96 information technology (IT) systems
and appraisal of clinical performance 72 annual maintenance 143
causes of variations 101 bene¢ts 144
and health outcomes assessment community care 75
96^100 consumer-focused systems 145^7
limitations as health outcome indicators cost implications of advances 21, 143^4
104 decision support systems 41^2
validity concerns 72, 74^5, 100^1 and developing countries 144, 147^8
e-mail contact (home-based patients) 42
informatics, de¢ned 107 electronic patient records 48, 59^62,
Information for Action (Institute for Health 64^6, 110, 117
Services Management) 19 electronic prescribing 73
information governance 87 lessons from industry 13, 59
see also governance medical v. social models 64^5
 Index 163

need for central leadership 43 knowledge management 48^54

networks see networked IT systems aims and objectives 50
on-line ordering 73 developing improvement initiatives
primary care 73, 75 50^4
satellite technology 148 problems 49
secondary care 73, 75 tacit v. explicit 53
and social context 60^1, 65 see also information management
use in identifying best practice 42, knowledge paradigms (NHS), old v. new
70^2 47^8
web-based systems 53^4, 146^7 Knox, EG 25^6
workstation developments 144 K˛rner Klubs 7
see also data capture; patient records ‘K˛rner’ returns’
‘infostructure’ 32, 42^3 in£uence of DHSS 10
see also information management; limitations of data sets 23, 25^6, 27
information technology (IT) systems patient contacts 124^5
Institute of Medicine (IOM) (USA) 33^5, use as performance targets 125
37, 41 K˛rner, Edith
and IT reports 35, 60 appointment to steering group 2, 3^4
as model for UK 33 biographical details 3^5
intelligent accountability 84, 85 legacy and achievements 12, 16^18,
inter-agency collaborations 23^6, 28
and information needs 111 see also Steering Group on Health Services
see also generic record-keeping Information (NHS/DHSS)
internal control 83, 87
and board members’ responsibilities language
86^8 and clinical terms project 74
performance indicators 87^8 electronic record-keeping 70
see also governance; information European and American classi¢cation
management; risk management advances 126
International e-Health Association 156 headings project 74^5
International Medical Informatics in nursing practice 126
Association 155, 157 thesaurus development and SNOMED-CT
investment 23 74^5
child health information systems 110 leadership 38^41
IOM see Institute of Medicine development initiative proposals 39^41
Ireland, local information needs initiative and IT 40
155 Learning from Bristol (DoH 2002) 48
IT see information technology (IT) legal issues (record-keeping) 69
systems child health systems 112^13
patient access 147
Jenkin, Patrick 2, 9 Less Paperwork Better Information
Joint Consultants’ Committee, Project 131
representation on Steering Group 3 libraries 49
local needs v. central requirements 153^7
Kennedy report (Paediatric heart accountability and governance issues
surgery-Bristol) 48 84
King, David 4 and globalisation of healthcare 149^57
King’s Fund, Steering Group publications Steering Group’s recommendations
8^9 6, 25
164 Index

managers NHS Modernisation Agency

attitudes to data capture 21 implementation of accountability
IT and leadership 38^41 frameworks 84
Mark, AL 21 Leadership Centre 39
maternity services The NHS Plan (DoH 1995) 47
events capture (pre-1985) 1 NHS strategy and policy 31^44
software development project 7 adoption of a national ‘guidance’ body
Maxwell, Robert 9^10 (R/NAH) 32^5
mental health data capture common healthcare problems 52
pre-1985 1 development of healthcare process rules
and social context 60^1 35^7, 37
Merrison Commission (1979) 14 framework guidance and
minimum data sets 6^7, 16, 58 devolution 149^50, 156^7
costs of implementation 21 globalisation issues 149^57
implementation post-1985 9^10, 72, organisational instability and e¡ects on
75 information management 107
proposed changes 74 promotion of e¡ective leadership
subset central (DHSS) returns 7 38^41
see also hospital episodes statistics (HES) provision of evidence-based knowledge
Modernisation Agency see NHS systems 41^3
Modernisation Agency 1982 restructuring 1, 3
multi-professional record-keeping systems NHS/DHSS Steering Group on Health
see generic record-keeping Services Information see Steering Group
on Health Services Information
NANDA see North American Nursing North American Nursing Diagnosis
Diagnosis Association Association (NANDA) 126
National Academy of Sciences (USA) nursing documentation 121^41
33, 35 background (health visiting practices)
National Centre for Health Outcome 122^7
Development 91^2, 100^1 Finland 122^3
National electronic Library for Health 71 language and identi¢cation
national hospital enquiry database initiatives 125^6
(pre-1982) 1^2 Sweden 126
National Patient Safety Agency 83 see also patient records
national service framework targets 71
for children 117 Omaha project 128^41
and electronic patient records 70 achievements 129
Netherlands, nursing information American origins 129^30
models 126 design and methodology 130^5
networked IT systems outcomes/data analysis 135^40
and child protection issues 63, 113, participant acceptability scores 140^1
117^18 signi¢cance 141
generic record-keeping 70 on-line ordering systems 73
public health role 63^4 O’Neil, O 85
as scheduling aid 63 An Organisation with a Memory
NHS Computer Policy Committee (1981) 3 (DoH 2000) 47^8
NHS Direct 146 organisations
NHS Information Authority, SNOMED-CT and change management see change
collaboration 74 management
 Index 165

and information/knowledge management political trends

51^2 centralising in£uence 10
outcome measures see health outcomes; in£uence on stability of NHS 32^3
social outcome measures see also NHS strategy and policy
POMR see Problem Orientated Medical
pathology, lead role in IT developments Record
59 prescribing systems 73
patient contacts, as de¢ned by Fifth primary care
Report 124^5 exclusion from Steering Group remit 3
patient records record-keeping systems 73, 75
background and history 57^9 Primary Care Information for Modernisation
child health systems 108, 110, 115^16, Programme 75
117 Problem Orientated Medical Record
clinical standards 74 (POMR) 61
electronic record systems 48, 59^62, The Protection and Maintenance of
64^6, 70, 110, 117 Con¢dentiality of Patient and Employee
encounter records (Omaha project) 129, Data (1984) 8
134^41 Public Accounts Committee, review on risk
episode-based systems 69^70 management 83
in Finland 122^3 Public Health Observatory, and knowledge
function and purpose 72 management 54
generic systems 70, 72, 107 public health requirements
and health needs assessment patient access to health information
57, 63^4 146^7
limitations of paper-based systems 69 see also epidemiological requirements;
patient access and ownership 145 health needs assessment
as patients’ advocate 57, 62^6
progress in primary care 73 quality improvement
secondary care 73 and information/knowledge management
speciality-based systems 69^70 48, 52, 53^4
system requirements 64 and minimum data sets 75
see also information technology (IT) see also governance; health outcomes
systems quantitative data, as proxy for
patient-centred care, and IOM guidance outcomes 96^100
36^7
patients’ information needs 41, 70 R/NAH see Royal (National) Academy of
see also consumer-focused systems Health
performance assessment frameworks 95 randomised controlled trials, and outcome
performance indicators measurement 131, 133
and appraisal/revalidation 72 Rapid Access Chest Pain Clinic web
and health outcomes 94^5 folder 54
history of 2, 125 Read Codes 126
internal control systems 87^8 redesign rules (healthcare processes)
validity of records 1225 35^7
peri-operative deaths statistics Reports to the Secretary of State, Steering
hospital returns pre-1985 2 Group Publications 8
as outcome indicators 99^100 research governance 87
personal development strategies, and research reports, information £ow concerns
knowledge management 50^1 48^9
166 Index

Resource Allocation Working Party (RAWP) social care needs, representation in

report (1976) 14 records 60^1
resource prioritisation, and patient social outcome measures 27
records 57 speciality-based medical records 69^70
revalidation, and clinical performance and national service framework targets
measures 72 70
risk management 79^88 see also sector-based episodes
and accountability 84^5 Statement of Internal Control 82
background 81^3 Steering Group on Health Services
balancing risk/control 85^8 Information (NHS/DHSS) 2
board duties 86^8 achievements 12, 16^18, 24^6, 28
de¢nitions 83 critics views 23^4, 25^8
and freedom of choice 86 development of minimum data sets
and The Governance Standard 6^7, 16
79^83, 80 Fifth Report 121^2, 124^5
‘hard’ controls 83 key concerns and recommendations
and health outcome objectives 93, 94 6, 16^17, 96
safety concerns v. strategic issues limitations of remit 3, 26^7
82^3 local data requirements 6
‘soft’ controls 83 membership 3, 15
risk registers 88 pilot studies 7, 15^16
Robert Wood Johnson Health Policy policy implementation di⁄culties 9^10,
Fellowship Program 34 18^23
Royal Commission Report (1979) 1^2 political implications 23
Royal (National) Academy of Health publications 8^9
(R/NAH) 33^5 remit and terms of reference 2^3, 15
signi¢cance and outcomes 9^10,
safety 82^3 17^19, 20^6
and organisational responsibility 80^1 Swansea Omaha project see Omaha
SatelLife project 148 project
satellite technology 148 Sweden, and nursing documentation
Saudi Arabia, Arab information needs 126
initiative 155
secondary care To Err is Human (IOM 2000) 34
developments in clinical recording training
systems 75 in developing countries 147^8
see also hospital activity analysis systems for IT systems and data collection 73^4,
sector-based episodes 75 143
security for leadership/management 38^41
hardware thefts 143^4 Steering Group’s recommendations
of patient records 145 7, 16, 18
and web-based patient records 147 Transforming Health Care Through Information
see also con¢dentiality; safety Technology (President’s Information
service users information needs 41, 70 Technology Advisory Committee 2001)
Shifting the Balance of Power within the NHS 35
(DoH 2001) 84 treatment, and advocacy function of patient
SNOMED-CT clinical terms project 74 records 62
and Omaha system 130 Trust boards, risk management duties
SOAP records 61 86^8
 Index 167

trust issues (data reliability), managers/ and Rapid Access Chest Pain Clinic
clinicians perceptions 16, 21, 74 programme 54
Turnball report 81^2, 83 satellite technology 148
Weed, Laurence 61
United Nations Millennium Assembly 148 WHO
USA health information strategies 33^44 HINARI programme 156
government IT reports 35 and quality 131
IOM reports 35, 60 Wickings, I 2
Willcock, LP 21
VIPS system 126 Windsor, P 18, 26
waiting list statistics 73
Walk Don’t Run (1985) 4 Yates, J 2, 21, 22
web-based information Young, Barbara 19
and document management 53 Young, WC 125
patient access 146^7