Medical informatics computer applications in health care and biomedicine

Professor and Chair Department of Medical Informatics in Medicine Professor of Medicine and Computer Science Science Vanderbilt Clinic Columbia University College of Physicians and S

Health Informatics

(formerly Computers in Health Care)

Kathryn J Hannah Marion J Ball

Series Editors

Springer Science+Business Media, LLC

(formerly Computers in Health Care)

Series Editors

Kathryn J Hannah Marion J Ball Dental Informatics

Integrating Technology into the Dental Environment

L.M Abbey and J Zimmerman

Ethics and Information Technology

A Case-Based Approach to a Health Care System in Transition

J.G Anderson and K.W Goodman

Aspects of the Computer-Based Patient Record

M.J Ball and M.F Collen

Performance Improvement Through Information Management

Health Care's Bridge to Success

M.J Ball and J.V Douglas

Strategies and Technologies for Healthcare Information

Theory into Practice

M.J Ball, J.V Douglas, and D.E Garets

Nursing Informatics

Where Caring and Technology Meet, Third Edition

M.J Ball, K.J Hannah, S.K Newbold, and J.V Douglas

Healthcare Information Management Systems

A Practical Guide, Second Edition

M.J Ball, D.W Simborg, J.W Albright, and J.V Douglas

Healthcare Information Management Systems

Cases, Strategies, and Solutions, Third Edition

M.J Ball, C.A Weaver, and J.M Kiel

Clinical Decision Support Systems

Theory and Practice

E.S Berner

Strategy and Architecture of Health Care Information Systems

M.K Bourke

Information Networks for Community Health

P.F Brennan, S.J Schneider, and E Tornquist

Informatics for the Clinical Laboratory

A Practical Guide

D.F Cowan

(continued after index)

Edward H Shortliffe Leslie E Perreault

Professor and Chair

Department of Medical Informatics in Medicine

Professor of Medicine and Computer Science

Science Vanderbilt Clinic

Columbia University College of

Physicians and Surgeons

Columbia-Presbyterian Medical Center

New York, New York 10032-3720, USA

Gio Wiederhold, PhD, FIEEE, FACM,

FACMI

Emeritus Professor of Computer Science,

Electrical Engineering, and Medicine

Stanford University

Stanford, CA 94305-9040, USA

Series Editors:

Kathryn J Hannah, PhD, RN

Adjunct Professor Department

of Community Health Sciences

Faculty of Medicine

The University of Calgary

Calgary, Alberta, Canada T2N 4Nl

Library of Congress Cataloging-in-Publication Data

NewYork,NY USA

Lawrence M Fagan, MD, PhD, FACMI Senior Research Scientist and Associate Director, Stanford Medical Informatics Co-Director, Medical Information Sciences Training Program Director, Medical Informatics Short Course

Stanford University School of Medicine Stanford, CA 94305-5479, USA Marion J Ball, EdD

Vice-President, Clinical Solutions Healthlink, Inc

Baltimore, MD 21210, USA

and

Adjunct Professor Johns Hopkins University School

of Nursing Baltimore, MD 21205, USA

Medical informatics: computer applications in health care and biomedicine / editors, Edward H Shortliffe, Leslie E Perreault; associated editors, Gio Wiederhold, Lawrence M

Fagan-2nd ed

Includes bibliographical references and index

ISBN 978-1-4899-0517-8

1 Medical informatics 1 Shortliffe, Edward Hance II Health informatics

R858.M397 2000

Printed on acid-free paper

DOI 10.1007/978-0-387-21721-5

Originally published by Springer Science+Business Media, Inc in 2001

Softcover reprint of the hardcover 2nd edition 2001

All rights reserved This work may not be translated or copied in whole or in part without the written permission of the publisher Springer Science+Business Media, LLC, except for brief excerpts in con- nection with reviews or scholarly analysis Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known

or hereafter developed is forbidden

The use in this publication of trade names, trademarks, service marks and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject

to proprietary rights

(MPIMVY)

987

springeronline.com

Where is the wisdom we have lost in knowledge?

-T.S Elliot, "The Rock"

To the memory of Scott Blois, innovator, pher, and scholar, who showed us how to look be- yond technology to the concepts and perspectives that help to define the scientific underpinnings for the field of medical informatics

philoso-Series Preface

This series is directed to healthcare professionals who are leading the mation of health care by using information and knowledge Launched in 1988 as Computers in Health Care, the series offers a broad range of titles: some ad-dressed to specific professions such as nursing, medicine, and health adminis-tration; others to special areas of practice such as trauma and radiology Still other books in the series focus on interdisciplinary issues, such as the computer-based patient record, electronic health records, and networked healthcare systems

transfor-Renamed Health Informatics in 1998 to reflect the rapid evolution in the cipline now known as health informatics, the series continues to add titles that contribute to the evolution of the field In the series, eminent experts, serving as editors or authors, offer their accounts of innovations in health informatics In-creasingly, these accounts go beyond hardware and software to address the role

dis-of information in influencing the transformation dis-of healthcare delivery systems around the world The series also increasingly focuses on "peopleware" and the organizational, behavioral, and societal changes that accompany the diffusion of information technology in health services environments

These changes will shape health services in the new millennium By making full and creative use of the technology to tame data and to transform informa-tion, health informatics will foster the development of the know ledge age in health care As co-editors, we pledge to support our professional colleagues and the se-ries readers as they share advances in the emerging and exciting field of health informatics

Kathryn J Hannah Marion J Ball

vii

Preface

Just as banks cannot practice modem banking without financial software, and airlines cannot manage modem travel planning without shared databanks of flight schedules and reservations, it has become impossible to practice modem medi-cine without information technologies Health professionals recognize that a large percentage of their activities relates to information management-for example, obtaining and recording information about patients, consulting colleagues, read-ing the scientific literature, planning diagnostic procedures, devising strategies for patient care, interpreting results of laboratory and radiologic studies, or con-ducting case-based and population-based research It is complexity and uncer-tainty, plus society's overriding concern for patient well-being and the resulting need for optimal decision-making, that set medicine apart from many other in-formation-intensive fields Our desire to provide the best possible health and health care for our society gives a special significance to the effective organiza-tion and management of the huge bodies of data with which health profession-als must deal It also suggests the need for specialized approaches and for skilled scientists who are knowledgeable about both medicine and information technologies

Information Management in Biomedicine

Although the application of computers to biomedicine is recent, the clinical and research influence of medical-computing systems is already remarkably broad Clinical information systems, which provide communication and information-management functions, are now installed in essentially all healthcare institutions Physicians can search entire drug indexes in a few seconds, using the informa-tion provided by a computer program to anticipate harmful side effects or drug interactions Electrocardiograms often are analyzed initially by computer pro-grams, and similar techniques are being applied for interpretation of pUlmonary-function tests and a variety of laboratory and radiologic abnormalities Micro-processor systems routinely monitor patients and provide warnings in critical-care settings, such as the intensive-care unit or the operating room Both biomedical

ix

researchers and clinicians regularly use computer programs to search the ical literature, and modem clinical research would be severely hampered with-out computer-based data-storage techniques and statistical-analysis systems Ad-vanced decision-support tools also are emerging from research laboratories, are being integrated with patient-care systems, and are likely to have a profound ef-fect on the way medicine is practiced in the future

med-Despite this growing use of computers in healthcare settings and the resulting expansion of interest in learning more about medical computing, many health students and professionals have found it difficult to obtain a comprehensive and rigorous, but nontechnical, overview of the field Both practitioners and basic scientists are recognizing that thorough preparation for their professional futures requires that they gain an understanding of the state of the art in biomedical com-puting, of the current and future capabilities and limitations of the technology,

and of the way in which such developments fit within the scientific, social, and fmancial context of biomedicine In turn, the future of the medical-computing field will be largely determined by how well health professionals and other peo-ple are prepared to guide the discipline's development This book is intended to meet this growing need for well-equipped professionals

The first edition appeared in 1990 and has been used throughout the world

in courses on medical informatics Like the first edition, this new version vides a conceptual framework for learning about computer applications in med-ical care, for critiquing existing systems, and for anticipating future directions that the field may take In many respects, however, this new edition is very dif-ferent from its predecessor Most importantly, it reflects the remarkable changes

pro-in computpro-ing and communications that have occurred pro-in the past decade For example, the Internet was barely mentioned in the first edition, but it and the World Wide Web are now discussed in almost every chapter in light of their pervasive societal influence in recent years In addition, new chapters are in-cluded, and others have been deleted or revamped We include new chapters on bioinformatics, standards development, systems evaluation, technology assess-ment, legal topics, and ethical considerations The former chapter on "hospital information systems" has now given way to a discussion of enterprise comput-ing for integrated delivery networks Those who are familiar with the first edi-tion will find that the organization and philosophy are unchanged, but the con-tent is almost entirely new.!

This book differs from other introductions to the field in its broad coverage and in its emphasis on the field's conceptual underpinnings Our book does not presume that readers have a health-science or computer-science background, but

lAs in the first edition, the book tends to draw both its examples and its contributors from North America There is excellent work in other parts of the world as well, although vari-ations in healthcare systems, and especially in financing, do tend to change the way in which systems evolve from one country to the next The basic concepts are identical, how-ever, so the book is intended to be useful in educational programs in other parts of the world as well

it does assume that they are interested in a comprehensive summary of the field that stresses the underlying concepts, and it introduces technical details only to the extent that they are necessary to meet the principal goal It thus differs from

an impressive early text in the field (Ledley, 1965) that emphasized technical tails but did not dwell on the broader social and clinical context in which med-ical computing systems are developed and implemented

de-Overview and Guide to Use of This Book

This book is written as a text so that it can be used in formal courses, but we have adopted a broad view of the population for whom it is intended Thus, it may be used not only by students of medicine and of the other health professions but also by future medical-computing professionals as an introductory text as well as a text for self-study and for reference by practitioners The book is prob-ably too detailed for use in a 2- or 3-day continuing-education course, although

it could be introduced as a reference for further independent study

Our principal goal in writing this text is to teach concepts in medical

informat-ics-the study of biomedical information and its use in decision-making-and to illustrate them in the context of descriptions of representative systems that are in use today or that taught us lessons in the past As you will see, medical informat-ics is more than the study of computers in medicine, and we have organized the book to emphasize this point Chapter 1 first sets the stage for the rest of the book

by providing a glimpse of the future, defining important terms and concepts, scribing the content of the field, explaining the connections between medical in-formatics and related disciplines, and discussing the forces that have influenced re-search in medical informatics and its integration into medical practice

de-Broad issues regarding the nature of data, information, and knowledge vade all areas of application, as do concepts related to optimal decision-making Chapters 2 and 3 focus on these topics but mention computers only in passing They serve as the foundation for all that follows

per-Chapters 4 and 5 introduce the central ideas of computer hardware and ware that are important for understanding the applications described later Also included is a discussion of computer-system design, with explanations of im-portant issues for you to consider when you read about specific applications and systems throughout the remainder of the book

soft-Chapter 6 summarizes the issues of standards development, focusing in ticular on data exchange and issues related to sharing of clinical data This im-portant and rapidly evolving topic was not covered in our fIrst edition but war-rants inclusion here given the increasingly central role of standards in enabling clinical systems to have their desired influence on healthcare practices

par-Chapter 7 addresses the key legal and ethical issues that have arisen when health information systems are considered Then, in Chapter 8, the challenges associated with technology assessment and the evaluation of clinical information systems are introduced

Chapters 9 through 18 survey many of the key biomedical areas in which puters are being used Each chapter explains the conceptual and organizational issues in building that type of system, reviews the pertinent history, and exam-ines the barriers to successful implementations

com-Chapter 19 provides a historical perspective on changes in the way society pays for health care It discusses alternative methods for evaluating the costs and the benefits of health care and suggests ways in which fmancial considerations affect medical computing The book concludes in Chapter 20 with a look to the future-a vision of how informatics concepts, computers, and advanced com-munication devices one day may pervade every aspect of medical practice

The Study of Computer Applications in Medicine

The actual and potential uses of computers in medical care form a remarkably broad and complex topic Just as you do not need to understand how a telephone works to make good use of it and to tell when it is functioning poorly, however,

we believe that technical medical-computing skills are not needed by health ers who simply wish to become effective computer users On the other hand, such technical skills are of course necessary for individuals with career com-mitment to developing computer systems for medical environments Thus, this book will neither teach you to be a programmer nor show you how to fix a bro-ken computer (although it might motivate you to learn how to do both) It also will not tell you about every important medical-computing system or applica-tion; we shall direct you to a wealth of literature where review articles and in-dividual project reports can be found We describe specific systems only as ex-amples that can provide you with an understanding of the conceptual and organizational issues to be addressed in building systems for such uses Exam-ples also help to reveal the remaining barriers to successful implementations Some of the application systems described in the book are well established, even

work-in the commercial marketplace Others are just begwork-innwork-ing to be used broadly work-in biomedical settings Several are still largely confined to the research laboratory Because we wish to emphasize the concepts underlying this field, we gener-ally limit the discussion of technical implementation details The computer-science issues can be learned from other courses and other textbooks One ex-ception, however, is our emphasis on the details of decision science as they re-late to medical problem-solving (Chapters 3 and 16) These topics generally are not presented in computer-science courses, yet they playa central role in the in-telligent use of medical data and knowledge Sections on medical decision-making and computer-assisted decision support accordingly include more tech-nical detail than you will find in other chapters

All chapters include annotated Suggested Readings to which you can turn if you have a particular interest in a topic, and there is a comprehensive Bibliog-raphy at the end of the book We use boldface print to indicate the key terms of

each chapter; the definitions of these terms are included in the Glossary at the end of the book Because many of the issues in medical informatics are concep-tual, we have included Questions for Discussion at the end of each chapter You will quickly discover that most of these questions do not have "right" answers They are intended to illuminate key issues in the field and to motivate you to ex-amine additional readings and new areas of research

It is inherently limiting to learn about computer applications solely by reading about them We accordingly encourage you to complement your studies by seeing real systems in use-ideally by using them yourself Your understanding of system limitations and of what you would do to improve a medical-computing system will

be greatly enhanced if you have personal experience with representative tions Be aggressive in seeking opportunities to observe and use working systems

applica-In a field that is changing as rapidly as computer science is, it is difficult ever

to feel that you have knowledge that is completely current The conceptual sis for study, however, changes much more slowly than do the detailed techno-logical issues Thus, the lessons you learn from this volume will provide you with a foundation on which you can continue to build in the years ahead

ba-The Need for a Course in Medical

Computing Applications

Suggesting that new courses are needed in the curricula for students of the health professions does not increase your popularity If anything, educators and students have been clamoring for reduced lecture time, for more emphasis on small group

sessions, and for more free time for problem-solving and reflection A 1984 national survey by the Association of American Medical Colleges found that both medical students and their educators severely criticized the current emphasis on lectures and memorization Yet the analysis of a panel on the General Professional Education of the Physician [Association of American Medical Colleges, 1984] specifically iden-tified medical informatics, including computer applications, as an area in which new educational opportunities needed to be developed so that physicians would be bet-ter prepared for the practice of medicine The report recommended the formation

of new academic units in medical informatics in our medical schools, and quent studies and reports have continued to stress the importance of the field and the need for its inclusion in the educational environments of health professionals The reason for this strong recommendation is clear: The practice of medicine

subse-is inextricably entwined with the management of information In the past,

practi-tioners handled medical information through resources such as the nearest tal or medical-school library; personal collections of books, journals, and reprints; ftles of patient records; consultation with colleagues; manual office bookkeeping; and (all-too-often flawed) memorization Although all these techniques continue

hospi-to be valuable, the computer now offers new methods for fmding, filing, and ing information: online bibliographic-retrieval systems, including full-text publi-

sort-cations; personal computers, with database software to maintain personal mation and reprint files; office-practice and clinical information systems to cap-ture, communicate, and preserve key elements of the medical record; consultation systems to provide assistance when colleagues are inaccessible or unavailable; practice-management systems to integrate billing and receivable functions with other aspects of office or clinic organization; and other online information re-sources that help to reduce the pressure to memorize in a field that defies total mastery of all but its narrowest aspects With such a pervasive and inevitable role for computers in clinical practice, and with a growing failure of traditional tech-niques to deal with the rapidly increasing information-management needs of prac-titioners, it has become obvious to many people that a new and essential topic has emerged for study in schools that train medical and other health professionals What are less clear are how the subject should be taught and to what extent it should be left for postgraduate education We believe that topics in medical com-puting are best taught and learned in the context of health-science training, which allows concepts from both medicine and computer science to be integrated Medical-computing novices are likely to have only limited opportunities for in-tensive study of the material once their health-professional training has been com-pleted

infor-The format of medical-informatics education is certain to evolve as faculty members are hired to develop it at more health-science schools and as the em-phasis on lectures as the primary teaching method diminishes Computers will

be used increasingly as teaching tools and as devices for communication, problem-solving, and data sharing among students and faculty In the meantime, medical informatics will be taught largely in the classroom setting This book is designed to be used in that kind of traditional course, although the Questions for Discussion also could be used to focus conversation in small seminars and work-ing groups As resources improve in schools, integration of medical-computing topics into clinical experiences also will become more common The eventual goal should be to provide instruction in medical informatics whenever this field

is most relevant to the topic the student is studying This aim requires tional opportunities throughout the years of formal training, supplemented by continuing-education programs after graduation

educa-The goal of integrating medicine and computer science is to provide a anism for increasing the sophistication of health professionals so that they know and understand the available resources They also should be familiar with med-ical computing's successes and failures, its research frontiers, and its limita-tions so that they can avoid repeating the mistakes of the past Study of med-ical computing also should improve their skills in information management and problem-solving With a suitable integration of hands-on computer experience, computer-based learning, courses in clinical problem-solving, and study of the material in this volume, health-science students will be well prepared to make effective use of computer-based tools and information management in health-care delivery

mech-The Need for Specialists in Medical Informatics

As mentioned, this book also is intended to be used as an introductory text in programs of study for people who intend to make their professional careers in medical informatics If we have persuaded you that a course in medical infor-matics is needed, then the requirement for trained faculty to teach the courses will be obvious Some people, however, might argue that a course on this sub-ject could be taught by a computer scientist who has an interest in medical com-puting or by a physician who has taken a few computing courses Indeed, in the past, most teaching-and research-has been undertaken by faculty trained pri-marily in one of the fields and later drawn to the other Today, however, schools are beginning to realize the need for professionals trained specifically at the in-terfaces among medicine, computer science, and related disciplines such as sta-tistics, cognitive science, health economics, and medical ethics This book out-lines a first course for students training for careers in the medical-informatics field We specifically address the need for an educational experience in which computing and information-science concepts are synthesized with biomedical is-sues regarding research, training, and clinical practice It is the integration of the related disciplines that traditionally has been lacking in the educational opportu-nities available to students with career interests in medical informatics If schools are to establish such courses and training programs (and there are already a few examples of each), they clearly need educators who have a broad familiarity with the field and who can develop curricula for students of the health professions as well as of engineering and computer science

The increasing introduction of computing techniques into medical environments will require that well-trained individuals be available not only to teach students but also to design, develop, select, and manage the medical-computing systems of to-morrow There is a wide range of context-dependent computing issues that people can appreciate only by working on problems defined by the healthcare setting and its constraints The field's development has been hampered because there are rela-tively few trained personnel to design research programs, to carry out the experi-mental and developmental activities, and to provide academic leadership in medical computing A frequently cited problem is the difficulty a health professional and a technically trained computer scientist experience when they try to communicate with one another The vocabularies of the two fields are complex and have little over-lap, and there is a process of acculturation to medicine that is difficult for computer scientists to appreciate through distant observation Thus, interdisciplinary research and development projects are more likely to be successful when they are led by peo-ple who can effectively bridge the medical and computing fields Such profession-als often can facilitate sensitive communication among program personnel whose backgrounds and training differ substantially

It is exciting to be working in a field that is maturing and that is having a eficial effect on society There is ample opportunity remaining for innovation as new technologies evolve and fundamental computing problems succumb to the

ben-creativity and hard work of our colleagues In light of the increasing tion and specialization required in computer science in general, it is hardly sur-prising that a new discipline should arise at that field's interface with medicine This book is dedicated to clarifying the definition and to nurturing the effec-tiveness of that new discipline: medical informatics

sophistica-Edward H Shortliffe Leslie E Perreault

Acknowledgments

compre-hensive textbook on medical informatics, none of us predicted the enormity of the task we were about to undertake Our challenge was to create a multi-authored textbook that captured the collective expertise of leaders in the field yet was cohesive in content and style The concept for the book was first developed

in 1982 We had begun to teach a course on computer applications in health care

at Stanford University School of Medicine and had quickly determined that there was no comprehensive introductory text on the subject Despite several collec-tions of research descriptions and subject reviews, none had been developed with the needs of a rigorous introductory course in mind

The thought of writing a textbook was daunting due to the diversity of topics None of us felt he was sufficiently expert in the full range of important subjects for us to write the book ourselves Yet we wanted to avoid putting together a collection of disconnected chapters containing assorted subject reviews Thus,

we decided to solicit contributions from leaders in the respective fields to be resented but to provide organizational guidelines in advance for each chapter

rep-We also urged contributors to avoid writing subject reviews but, instead, to cus on the key conceptual topics in their field and to pick a handful of examples

fo-to illustrate their didactic points

As the draft chapters began to come in, we realized that major editing would

be required if we were to achieve our goals of cohesiveness and a uniform entation across all the chapters We were thus delighted when, in 1987, Leslie Perreault, a graduate of our training program, assumed responsibility for re-working the individual chapters to make an integral whole and for bringing the project to completion The fmal product, published in 1990, was the result of many compromises, heavy editing, detailed rewriting, and numerous iterations

ori-We were gratified by the positive response to the book when it fmally appeared and especially by the students of medical informatics who have often come to

us at scientific meetings and told us about their appreciation of the book

As the 1990s progressed, however, we began to realize that, despite our phasis on basic concepts in the field (rather than a survey of existing systems), the volume was beginning to show its age A great deal had changed since the

em-xvii

initial chapters were written, and it became clear that a new edition would be quired The original editors discussed the project and decided to redesign the book, solicit updated chapters, and publish a new edition Leslie Perreault by this time was a busy Director at First Consulting Group in New York City and would not have as much time to devote to the project as she had when we did the fIrst edition With trepidation, in light of our knowledge of the work that would be involved, we embarked on the new project

re-As before, the chapter authors have done a marvelous job, trying their best to meet our deadlines, putting up with editing changes that were designed to bring

a uniform style to the book, and contributing excellent chapters that nicely flect the changes in the fIeld in the last decade Weare all extremely apprecia-tive of their commitment and for the excellence of their work on behalf of the book and the fIeld

re-The completed volume reflects the work and support of many people in tion to the editors and chapter authors Particular gratitude is once again owed

addi-to Lyn Dupre, our developmental ediaddi-tor, whose rigorous attention addi-to detail is flected on every page We also appreciate the kindness and professionalism of Peter Gordon from Addison-Wesley, who worked with us to effect a smooth tran-sition in the transfer of the book to a new publisher for the second edition At Springer-Verlag we have been delighted to work with the responsible editors, fIrst with Bill Day and, more recently, with Nhora Cortes-Comerer Marion Ball and Kathy Hannah have also been extremely supportive in working with us to move this volume into their Springer-Verlag informatics series

re-Members of the administrative staff in our group at Stanford have also been remarkably supportive during what has at times been an exhausting and time-consuming process Rosalind Ravasio competently managed the administrative details so that we could attend to our writing and editing We are also grateful for the support of the Computing Resources Group at Stanford Medical Infor-matics Under the direction of Farhad Shafabaksh, the Computing Resources Group maintains a smoothly functioning computing environment that has been crucial to the production of this book in its electronic format and to the distrib-ution of chapters and revisions among the authors and editors via the World Wide Web

The unsung hero of the effort is our assistant, Barbara Morgan, who has dered the lion's share of the burden for online editing of the chapters, maintain-ing a system for keeping track of new versions, managing permissions and fIg-ures, and somehow keeping a good natured disposition throughout it all We are all sincerely grateful to Barbara for accepting this new set of responsibilities-clearly not in her job description-and doing a marvelous job in ushering this

shoul-volume to a successful completion Thank you, Barbara!

Edward H Shortliffe Leslie E Perreault

Edward H Shortliffe and Marsden S Blois

Medical Data: Their Acquisition, Storage, and Use 41

Edward H Shortliffe and G Octo Barnett

Medical Decision-Making: Probabilistic Medical Reasoning 76

Douglas K Owens and Harold C Sox

Essential Concepts for Medical Computing 132

Gio Wiederhold and Thomas C Rindfleisch

System Design and Engineering 180

Gio Wiederhold and Edward H Shortliffe

Standards in Medical Informatics 212

W Edward Hammond and James J Cimino

Ethics and Health Informatics: Users, Standards, and Outcomes 257

Kenneth W Goodman and Randolph A Miller

xix

CHAPTER 8 Evaluation and Technology Assessment 282

Charles P Friedman, Douglas K Owens, and Jeremy C Wyatt

UNIT IT MEDICAL COMPUTING APPLICATIONS

Computer-Based Patient-Record Systems 327

Paul C Tang and Clement J McDonald

Management of Information in Integrated Delivery Networks 359

Charles Safran and Leslie E Perreault

Public Health and Consumer Uses of Health Information: Education, Research, Policy, Prevention, and

William R Hersh, William M Detmer, and Mark E Frisse

Clinical Decision-Support Systems 573

Mark A Musen, Yuval Shahar, and Edward H Shortliffe

Computers in Medical Education 610

Parvati Dev, Edward P Hoffer, and G Octo Barnett

CHAPTER 20 The Future of Computer Applications in Health Care 697

Lawrence M Fagan and Edward H Shortlijfe

Bibliography 713

Glossary 749

Name Index 821

Subject Index 831

Contributors

Russ B Altman, MD, PhD, FACP, FACMI

Associate Professor of Medicine and of Computer Science, Stanford University School of Medicine, Stanford, CA 94305-5479, USA (russ.altman@stanford.edu)

Suzanne Bakken, RN, DNSc, FAAN, FACMI

Alumni Professor of Nursing and Professor of Biomedical Informatics,

Columbia University, New York, NY 10032, USA

(suzanne.bakken@dmi.columbia.edu)

G Octo Barnett, MD, FACP, FACMI

Professor of Medicine, Harvard Medical School, Director, Laboratory of Computer Science, Massachusetts General Hospital, Boston, MA 02114, USA (bamett.octo@mgh.harvard.edu)

Marsden S Blois, MD, PhD, FACMP

Formerly Professor of Medical Informatics and of Dermatology, University of California, San Francisco, CA, USA

Patricia Flatley Brennan, RN, PhD, FAAN, FACMI

Lillian S Moehlman Bascom Professor, School of Nursing and College of Engineering, University of Wisconsin-Madison, Madison, WI 53706, USA (pbrennan@engr.wisc.edu)

James F Brinkley, MD, PhD, FACMI

Research Professor of Biological Structure, Computer Science and ing, and Medical Education and Biomedical Informatics, University of Wash-ington, Seattle, Washington 98195, USA (brinkley@u.washington.edu)

Engineer-James J Cimino, MD, FACP, FACMI

Professor of Biomedical Informatics and Medicine, Columbia University lege of Physicians and Surgeons, Columbia-Presbyterian Medical Center, New York, New York 10032, USA (James.Cimino@columbia.edu)

Col-tDeceased

xxiii

William M Detmer, MD, MS, FACP

President and Chief Executive Officer, Unbound Medicine, Inc.,

Char-lottesville, V A 22902, Clinical Assistant Professor, Department of Health Evaluation Sciences, University of Virginia, Health Sciences Center, Char-lottesville, VA 22908, USA (bdetmer@virginia.edu)

Parvati Dev, PhD, FACMI

Director, SUMMIT Laboratory, and Senior Research Scientist, Stanford Medical Informatics, Stanford University School of Medicine, Stanford,

CA 94305-5466, USA (parvati.dev@stanford.edu)

Alain C Enthoven, PhD

Senior Fellow, Institute for International Studies, Center for Health Policy, Marriner S Eccles Professor of Public and Private Management (Emeritus), Graduate School of Business, Stanford University, Stanford CA 94305-5015, USA (enthoven_alain@gsb.stanford.edu)

lAwrence M Fagan, MD, PhD, FACMI

Senior Research Scientist and Associate Director, Stanford Medical

Informatics, Co-Director, Biomedical Information Sciences Training Program, Director, Medical Informatics Short Course, Stanford University School of Medicine, Stanford, CA 94305-5479, USA (fagan@smi.stanford.edu)

Andrew Friede, MD, MPH

Vice President for Health Mfairs, Analytical Sciences, Inc., Atlanta GA 30329 (afriede@asciences.com)

Charles P Friedman, PhD, FACMI

Professor of Medicine and Director, Center for Biomedical Informatics, versity of Pittsburgh, Pittsburgh, PA 15213-2582, USA (cpf@cbmi.upmc.edu)

Uni-Mark E Frisse, MD, MS, MBA, FACMI

Vice President, First Consulting Group, Chicago, IL 60606 (mfrisse@fcg.com)

Alan M Garber, MD, PhD, FACP

Professor of Medicine, Economics, and Health Research and Policy, Director, Center for Primary Care and Outcomes Research, Director, Center for Health Policy, Stanford University, Stanford, CA 94305-6019, USA

(garber@stanford.edu)

Reed M Gardner, PhD, FACMI

Professor and Chair, Department of Medical Informatics, University of Utah, Co-Director of Medical Informatics, LDS Hospital, School of Medicine, Salt Lake City, UT 84132, USA (reed.gardner@hsc.utah.edu)

Kenneth W Goodman, PhD

Director, Forum for Bioethics and Philosophy, University of Miami

Miami, FL 33101, USA (kwg@cs.miami.edu)

Robert A Greenes, MD, PhD, FACR, FACMI

Professor of Radiology, Harvard Medical School, Professor of Health Sciences and Technology, Harvard-MIT Division of Health Sciences and Technology, Professor of Health Policy and Management, Harvard School of Public Health, Radiologist, Brigham and Women's Hospital,

Director, Decision Systems Group, Brigham and Women's Hospital, Boston,

MA 02115, USA (greenes@harvard.edu)

W Edward Hammond, PhD, FACMI

Adjunct Professor, Fuqua School of Business, Professor Emeritus, Department

of Community and Family Medicine, Professor Emeritus, Department of medical Engineering, Pratt School of Engineering, Duke University, Durham,

Bio-NC 27710, USA (hammoOO1@mc.duke.edu)

William R Hersh, MD, FACP, FACMI

Professor and Head, Division of Medical Informatics & Outcomes Research, Oregon Health & Science University, Portland, OR 97201, USA

(hersh@ohsu.edu)

Edward P Hoffer MD, FACP, FACC, FRCP(C), FACMI

Associate Professor of Medicine, Harvard Medical School, Senior Scientist and Assistant Director, Laboratory of Computer Science, Massachusetts Gen-eral Hospital, Boston, MA 02114, USA (ehoffer@partners.org)

Clement J McDonald, MD, FACP, FACMI

Director, Regenstrief Institute, Regenstrief Professor of Medical Informatics, and Distinguished Professor of Medicine, Indiana University School of Medicine, Indianapolis, IN 46202, USA (cmdonald@regenstrief.org)

Randolph A Miller, MD, FACMI

Professor and Chairman, Division of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN 37232-8340, USA

(randy miller@mcmail.vanderbilt.edu)

Mark A Musen, MD, PhD, FACP, FACMI

Professor of Medicine and of Computer Science, Head, Stanford Medical Informatics, Stanford University School of Medicine, Stanford, CA 94305-

5479, USA (musen@stanford.edu)

Judy G Ozbolt, PhD, RN, FAAN, FACMI

Independence Foundation Professor of Nursing and Professor of Biomedical Informatics, Vanderbilt University, Nashville, TN 37240-0008, USA

Gudy ozbolt@mcmail.vanderbilt.edu)

Leslie E Perreault, MS

Consultant, New York, NY USA (perreault@smi.stanford.edu)

Thomas C Rindfleisch, MS, FACMI

Director Emeritus, Lane Medical Library, Stanford University School of Medicine, Stanford, CA 94305-5123, USA (rlndfieisch@stanford.edu)

Charles Safran, MD, MS, FACP, FACMI

Associate Clinical Professor of Medicine, Harvard Medical School, Chief ecutive Officer, Clinician Support Technology, Newton, MA 02459, USA (csafran@harvard.edu)

Ex-M Michael Shabot, MD, FACMI

Associate Director of Surgery and Director, Surgical Critical Care, Sinai Medical Center, Los Angeles, CA 90048, USA (shabot@csmc.edu)

Cedars-Yuval Shahar, MD, PhD

Associate Professor, Information Systems Engineering, Head, Medical matics Research Center, Head, Graduate Program, Information Systems Engi-neering, Ben Gurion University, Beer-Sheva 84105, Israel

Infor-(yshahar@bgumail.bgu.ac.il)

Edward H Shortliffe, MD, PhD, MACP, FACMI

Professor and Chair, Department of Biomedical Informatics, Professor of cine and of Computer Science, Deputy Vice President for Information Technol-ogy, Health Sciences Division, Columbia University, New York, NY 10032-

Medi-3720, USA (Edward.Shortliffe@columbia.edu)

Sara J Singer, MBA

Lecturer, Public Policy Program, Executive Director, Center for Health Policy, Senior Research Scholar, Institute for International Studies, Stanford Univer-sity, Stanford, CA 94305-6019, USA (sara.singer@stanford.edu)

Harold C Sox, Jr., MD, MACP

Editor, Annals of Internal Medicine, American College of Physicians, phia, PA 19106, USA (HSox@mail.acponline.org)

Philadel-Paul C Tang, MD, FACP, FACMI

Associate Clinical Professor of Medicine, University of California, San cisco, CA, USA and Chief Medical Information Officer, Palo Alto Medical Foundation, Palo Alto, CA 94301, USA (tang@smi.stanford.edu)

Fran-Gio Wiederhold, PhD, FIEEE, FACM, FACMI

Professor Emeritus of Computer Science, Electrical Engineering, and

Medicine, Stanford University, Stanford, CA 94305-9040, USA

(gio@DB.Stanford.edu)

Jeremy C Wyatt, DM(Oxon), FRCP(UK), FACMI, MB BS(Lon)

Professor, Medical Informatics, Academic Medical Centre, University of terdam, The Netherlands (uctqjwy@ucl.ac.uk)

Ams-Unit I

Recurrent Themes in

Medical Informatics

Chapter 1

The Computer Meets Medicine and Biology: Emergence of a Discipline 3

Edward H Shortliffe and Marsden S Blois

Chapter 2

Medical Data: Their Acquisition, Storage, and Use 41

Edward H Shortliffe and G Octo Barnett

Chapter 3

Medical Decision-Making: Probabilistic Medical Reasoning 76

Douglas K Owens and Harold C Sox

Chapter 4

Essential Concepts for Medical Computing 132

Gio Wiederhold and Thomas C Rindfleisch

Chapter 5

System Design and Engineering 180

Gio Wiederhold and Edward H Shortliffe

Chapter 6

Standards in Medical Informatics 212

W Edward Hammond and James J Cimino

Chapter 7

Ethics and Health Informatics: Users, Standards, and Outcomes 257

Kenneth W Goodman and Randolph A Miller

Chapter 8

Evaluation and Technology Assessment 282

Charles P Friedman, Douglas K Owens, and Jeremy C Wyatt

1

The Computer Meets

Medicine and Biology:

Emergence of a Discipline

After reading this chapter, you should know the answers to these questions:

• Why is information management a central issue in biomedical research and clinical practice?

• What are integrated information-management environments, and how might

we expect them to affect the practice of medicine and biomedical research in coming years?

medical informatics, clinical informatics, nursing informatics, bioinformatics,

and health informatics?

• Why should health professionals and students of the health professions learn about medical-informatics concepts and informatics applications?

• How has the development of minicomputers, microprocessors, and the net changed the nature of biomedical computing?

Inter-• Row is medical informatics related to clinical practice, biomedical ing, molecular biology, decision science, information science, and computer science?

engineer-• How does information in clinical medicine and health differ from information

in the basic sciences?

• How can changes in computer technology and the way medical care is fmance.d influence the integration of medical computing into clinical practice?

1.1 Integrated Information Management:

Technology's Promise!

After scientists had developed the ftrst digital computers in the 1940s, society was told that these new machines would soon be serving routinely as memory devices, assisting with calculations and with information retrieval Within the

IPortions of this section are adapted from a paper presented at Medinfo98 in Seoul, Korea (Shortliffe, 1998c)

tDeceased

3

next decade, physicians and other health workers had begun to hear about the dramatic effects that such technology would have on medical practice More than four decades of remarkable progress in computing have followed those early pre-dictions, and many of the original prophesies have come to pass Stories regarding the "information revolution" fill our newspapers and popular magazines, and to-day's children show an uncanny ability to make use of computers as routine tools for study and entertainment Similarly, clinical workstations are now available

on hospital wards and in outpatient offices Yet many observers cite the care system as being slow to understand information technology, to exploit it for its unique practical and strategic functionalities, and to incorporate it effectively into the work environment Nonetheless, the enormous technological advances

health-of the last two decades-personal computers and graphical workstations, new methods for human-computer interactions, innovations in mass storage of data, personal digital assistants, the Internet and the World Wide Web, wireless communications-have all combined to make routine use of computers by all health workers and biomedical scientists inevitable A new world is already with

us, but its greatest influence is yet to come This book will teach you both about our present resources and accomplishments and about what we can expect in the years ahead

It is remarkable to realize that the first microprocessors, enabling personal computers, did not appear until the late 1970s, and the World Wide Web dates only to the early 1990s This dizzying rate of change, combined with equally pervasive and revolutionary changes in almost all international healthcare systems during the past decade, makes it difficult for healthcare planners and in-stitutional managers to try to deal with both issues at once Yet many observers now believe that the two topics are inextricably related and that planning for the new healthcare environments of the twenty-first century requires a deep under-standing of the role that information technology is likely to play in those envi-ronments

What might that future hold for the typical practicing clinician? As we shall discuss in detail in Chapter 9, no clinical computing topic is gaining more at-tention currently than is the issue of electronic medical records (EMRs) Healthcare organizations are finding that they do not have systems in place that allow them to answer questions that are crucially important for strategic planning and for their better understanding of how they compare with other provider groups in their local or regional competitive environment In the past, administrative and financial data were the major elements required for such planning, but comprehensive clinical data are now also important for institu-tional self-analysis and strategic planning Furthermore, the inefficiencies and frustrations associated with the use of paper-based medical records have be-come increasingly clear (Dick & Steen, 1991 [revised 1997]), especially when inadequate access to clinical information is one of the principal barriers that clinicians encounter when trying to increase their efficiency in order to meet productivity goals for their practices

1.1.1 Electronic Health Records:

Anticipating the Future

Many healthcare institutions are seeking to develop integrated clinical tions These are single-entry points into a medical world in which computational tools assist not only with clinical matters (reporting results of tests, allowing di-rect entry of orders by clinicians, facilitating access to transcribed reports, and

worksta-in some cases supportworksta-ing telemedicworksta-ine applications or decision-support functions) but also with administrative and fmancial topics (tracking of patients within the hospital, managing materials and inventory, supporting personnel functions, man-aging the payroll, and the like), research (e.g., analyzing the outcomes associ-ated with treatments and procedures, performing quality assurance, supporting clinical trials, and implementing various treatment protocols), scholarly infor-mation (e.g., accessing digital libraries, supporting bibliographic search, and pro-viding access to drug information databases), and even office automation (pro-viding access to spreadsheets, word processors, and the like) The key idea, however, is that at the heart of the evolving clinical workstation lies the medical record in a new incarnation: electronic, accessible, confidential, secure, accept-able to clinicians and patients, and integrated with other types of nonpatient-specific information

Inadequacy of the Traditional Paper Record

The paper-based medical record is woefully inadequate for meeting the needs of modem medicine It arose in the nineteenth century as a highly personalized "lab notebook" that clinicians could use to record their observations and plans so that they could be reminded of pertinent details when they next saw that same pa-tient There were no bureaucratic requirements, no assumptions that the record would be used to support communication among varied providers of care, and few data or test results to fill up the record's pages The record that met the needs

of clinicians a century ago has struggled mightily to adjust over the decades and

to accommodate to new requirements as health care and medicine have changed Difficulty in obtaining information, either about a specific patient or about a general issue related to patient management, is a frustrating but common occur-rence for practitioners With increasing pressures to enhance clinical productiv-ity, practitioners have begun to clamor for more reliable systems that provide facile, intuitive access to the information they need at the time they are seeing their patients The EMR offers the hope for such improved access to patient-specific information and should provide a major benefit both for the quality of care and for the quality of life for clinicians in practice

Despite the obvious need for a new record-keeping paradigm, most tions have found it challenging to try to move to a paperless, computer-based clinical record (see Chapters 9 and 10) This observation forces us to ask the questions "What is a health record in the modem world? Are the available prod-

organiza-ucts and systems well matched with the modem notions of a comprehensive health record?" Companies offer medical-record products, yet the packages are limited in their capabilities and seldom seem to meet the full range of needs de-fined within our complex healthcare organizations

The complexity associated with automating medical records is best ated if one analyzes the processes associated with the creation and use of such records rather than thinking of the record as an object that can be moved around

appreci-as needed within the institution For example, on the input side (Fig 1.1), the medical record requires the integration of processes for data capture and for merg-ing information from diverse sources The contents of the paper record have tra-ditionally been organized chronologically-often a severe limitation when a clin-ician seeks to find a specific piece of information that could occur almost anywhere within the chart To be useful, the record system must make it easy to access and display needed data, to analyze them, and to share them among col-leagues and with secondary users of the record who are not involved in direct patient care (Fig 1.2) Thus, the computer-based medical record is best viewed not as an object, or a product, but rather as a set of processes that an organiza-tion must put into place, supported by technology (Fig 1.3) Implementing elec-tronic records is inherently a systems-integration task; it is not possible to buy a

FIGURE 1.1 Inputs to the medical record The traditional paper medical record is created

by a variety of organizational processes that capture varying types of information (notes regarding direct encounters between health professionals and patients, laboratory or radi-ologic results, reports of telephone calls or prescriptions, and data obtained directly from patients) The record thus becomes a merged collection of such data, generally organized

in chronological order

FIGURE 1.2 Outputs from the medical record Once information is collected in the tional paper medical record, it may be provided to a wide variety of potential users of the chart These users include health professionals and the patients themselves but also a wide variety of "secondary users" (represented here by the individuals in business suits) who have valid reasons for accessing the record but who are not involved with direct patient care Numerous providers are typically involved in a patient's care, so the chart also serves

tradi-as a means for communicating among them The mechanisms for displaying, analyzing, and sharing information from such records results from a set of processes that often vary substantially across several patient-care settings and institutions

medical-record system for a complex organization as an off-the-shelf product Joint development is crucial

The Medical Record and Clinical Trials

The arguments for automating medical records are nicely summarized in ters 2 and 9 and in the Institute of Medicine's report on computer-based patient records (Dick & Steen, 1991 [revised 1997]) One argument that warrants em-phasis is the importance of the electronic record in supporting clinical trials-

Chap-experiments in which data from specific patient interactions are pooled and alyzed in order to learn about the safety and efficacy of new treatments or tests and to gain insight into disease processes that are not otherwise well understood Medical researchers are constrained today by clumsy methods for acquiring the data needed for clinical trials, generally relying on manual capture of informa-tion onto datasheets that are later transcribed into computer databases for statis-

an-FIGURE 1.3 Complex processes demanded of the record As shown in Figures 1.1 and

both gather information to be shared and then distribute that information to those who have valid reasons for accessing it Paper-based documents are severely limited in meet-ing the diverse requirements for data collection and information access that are implied

by this diagram

tical analysis (Fig 1.4) The approach is labor intensive, fraught with nities for error, and adds to the high costs associated with randomized prospec-tive research protocols

opportu-The use of EMRs offers many advantages to those carrying out clinical search Most obviously, it helps to eliminate the manual task of extracting data from charts or filling out specialized datasheets The data needed for a study can

re-be derived directly from the EMR, thus making research data collection a product of routine clinical record keeping (Fig 1.5) Other advantages accrue as well For example, the record environment can help to ensure compliance with

by-a reseby-arch protocol, pointing out to by-a cliniciby-an when by-a pby-atient is eligible for by-a study or when the protocol for a study calls for a specific management plan given the currently available data about that patient We are also seeing the develop-ment of novel authoring environments for clinical-trial protocols that can help to

Medical

Data Sheets

Clinical Trial Design

• Definition of data elements

tri-ensure that the data elements needed for the trial are compatible with the local EMR's conventions for representing patient descriptors

1.1.2 Recurring Issues That Must Be Addressed

There are at least four major issues that have consistently constrained our efforts

to build effective patient-record systems: (1) the need for standards in the area

of clinical terminology; (2) concerns regarding data privacy, confidentiality, and security; (3) challenges of data entry by physicians; and (4) difficulties associ-ated with the integration of record systems with other information resources in the healthcare setting The first of these issues is discussed in detail in Chapter

6, and privacy is one of the central topics in Chapter 7 Issues of direct data try by clinicians are discussed in Chapter 2 and again in Chapter 9 and through-out many other chapters in this book as well In the next section we examine re-cent trends in networking and ask how communications are changing the way in

en-Medical Record System

Clinical TriaJ Design

com-in Figure 1.4 are thereby eliminated In addition, the interaction of the physician with the medical record permits two-way communication, which can greatly improve the quality and efficiency of the clinical trial Physicians can be reminded when their patients are el-igible for an experimental protocol, and the computer system can also remind the clini-cians of the rules that are defined by the research protocol, thereby increasing compliance with the experimental plan

which the patient-care record can be better integrated with other relevant mation resources and clinical processes that are currently fragmented and poorly coordinated

infor-1.1.3 Integrating the Patient Record

with Other Information Resources

Experience has shown that physicians are "horizontal" users of information nology (Greenes & Shortliffe, 1990) Rather than becoming "power users" of a narrowly defined software package, they tend to seek broad functionality across

tech-a wide vtech-ariety of systems tech-and resources Thus, routine use of computers, tech-and of EMRs, will be most easily achieved if the computing environment offers a crit-

ical mass of functionality that makes the system both smoothly integrated and useful for essentially every patient encounter

With the introduction of networked systems within our healthcare tions, there are new opportunities to integrate a wide variety of resources through single clinical workstations (see Chapter 10) The nature of the integration tasks

organiza-is illustrated in Figure 1.6 in which various workstations are shown at the upper left (machines for use by patients, clinicians, or clerical staff) connected to an

enterprise-wide network or intranet In such an environment, diverse clinical,

financial, and administrative databases all need to be accessed and integrated, typically by using both networks to tie them together and a variety of standards for sharing data among them Thus the clinical data repository has developed as

an increasingly common idea This term refers to a central computer that ers and integrates clinical data from diverse sources such as the chemistry and microbiology laboratories, the pharmacy, and the radiology department As is suggested in the diagram, this clinical database can provide the nidus for what will evolve into an EMR as more and more clinical data become available in electronic form and the need for the paper documents shrinks and eventually vanishes

gath-Enterprise Network

j

Administrative

systems (e.g., Admissions,

Discharges, and Transfers)

FIGURE 1.6 Networking the organization The enterprise intranet is a locally controlled

network that extends throughout a healthcare system It allows specialized workstations

to access a wide variety of information sources: educational, clinical, fmancial, and ministrative An EMR emerges from such an architecture if a system is implemented that gathers patient-specific data from multiple sources and merges them for ease of access by users such as those illustrated in Figure 1.2 Such systems are often called clinical data repositories, particularly if they do not yet contain the full range of information that would

ad-normally occur in a medical record

Another theme in the changing world of health care is the increasing ment in the creation of clinical guidelines and pathways (see Chapter 16), gen-erally in an effort to reduce practice variability and to develop consensus approaches to recurring management problems Several government and profes-sional organizations, as well as individual provider groups, have invested heav-ily in guideline development, often putting an emphasis on using clear evidence from the literature, rather than expert opinion alone, as the basis for the advice Despite the success in creating such evidence-based guidelines, there is a grow-ing recognition that we need better methods for delivering the decision logic to the point of care Guidelines that appear in monographs or journal articles tend

invest-to sit on shelves, unavailable when the knowledge they contain would be most valuable to practitioners Computer-based tools for implementing such guide-lines, and integrating them with the EMR, present a potential means for making high-quality advice available in the routine clinical setting Many organizations are accordingly attempting to integrate decision-support tools with their nascent electronic record systems

Rethinking Common Assumptions

One of the first instincts of software developers is to create an electronic version

of an object or process from the physical world Some familiar notion provides the inspiration for a new software product Once the software version has been developed, however, human ingenuity and creativity often lead to an evolution that extends the software version far beyond what was initially contemplated The computer can thus facilitate paradigm shifts in how we think about such fa-miliar concepts

Consider, for example, the remarkable difference between today' s word sors and the typewriter that was the original inspiration for their development Although the early word processors were designed largely to allow users to avoid retyping papers each time a minor change was made to a document, the word processors of today bear little resemblance to a typewriter Consider all the pow-

proces-erful desktop-publishing facilities, integration of figures, spelling correction, grammar aids, and the like Similarly, today's spreadsheet programs bear little resemblance to the tables of numbers that we once created on graph paper Also consider automatic teller machines and their facilitation of today's worldwide banking in ways that were never contemplated when the industry depended on human bank tellers

It is accordingly logical to ask what the health record will become after it has been effectively implemented on computer systems and new opportuni-ties for its enhancement become increasingly clear to us It is unlikely that the computer-based health record a decade from now will bear much resemblance

to the antiquated paper folder that still dominates many of our healthcare vironments One way to anticipate the changes that are likely to occur is to consider the potential role of wide-area networking and the Internet in the record's evolution

en-Extending the Record Beyond the Single Institution

In considering ongoing trends in information technology that are likely to make changes inevitable, it would be difficult to start with any topic other than the In-ternet The Internet began in 1968 as a United States research activity funded by the Advanced Research Projects Agency (ARPA) of the Department of Defense Initially known as the ARPAnet, the network began as a novel mechanism for allowing a handful of defense-related mainframe computers, located mostly at academic institutions or in the defense industry, to share data files with each other and to provide remote access to computing power at other locations The notion of electronic mail arose soon thereafter, and machine-to-machine elec-tronic mail exchanges quickly became a major component of the network's traf-fic As the technology matured, its value for nonmilitary research activities was recognized, and by 1973 the first medically related research computer had been added to the network (Shortliffe, 1998b)

During the 1980s the technology began to be developed in other parts of the world, and the National Science Foundation took over the task of running the principal high-speed backbone network in the United States The first hospi-tals, mostly academic centers, began to be connected to what had by then be-come known as the Internet, and in a major policy move it was decided to allow commercial organizations to join the network as well By April 1995 the Inter-net in the United States had become a fully commercialized operation, no longer depending on the U.S government to support even the major backbone connec-tions Many people point to the Internet as a superb example of the facilitating role of federal investment in promoting innovative technologies The Internet is

a major societal force that arguably would never have been created if the research and development, plus the coordinating activities, had been left to the private sector

The explosive growth of the Internet did not occur until the late 1990s, when the World Wide Web (which had been conceived initially by the physics com-munity as a way of using the Internet to share preprints with photographs and diagrams among researchers) was introduced and popularized The Web is highly intuitive, requires no special training, and provides a mechanism for access to multimedia information that accounts for its remarkable growth as a worldwide phenomenon

The Internet Society2 reports data on the growth of the international ing infrastructure, and they indicate that by 1998 there were close to 50 million host computers connected to the Internet worldwide (not counting all the end-user machines used to access the network) These machines were spread across

network-more than 2 million host domains-unique corporate or institutional addresses that often account for mUltiple hosts Given the current exponential growth, we should anticipate close to 200 million Internet host machines in the near future

2See http://www.isoc.org/

Such growth is not being ignored by the telecommunications industry Data from MCI-Worldcom, Inc., one of the companies that is actively involved in the growth of high-speed networking, indicate that they experienced a 5,600 percent growth in Internet-related traffic between October 1994 and January 1996.3 Be-tween 1996 and 1999, growth was compounding at approximately 15 percent per month, which naturally led to questions about when market saturation would oc-cur Particularly impressive was the projection that MCl's Internet traffic would exceed their voice telephony, in terms of relative use of MCl's network capac-ity, by the year 2001 The business implications of such data for the telecom-munications industry are clear

The societal impact of this communications phenomenon cannot be overstated, especially given the international connectivity that has grown phenomenally in the past 5 years Countries that once were isolated from information that was im-portant to citizens, ranging from consumers to scientists to those interested in political issues, are now finding new options for bringing timely information to the desktop machines of individuals with an Internet connection

There is accordingly a major upheaval underway in the telecommunications industry, with companies that used to be in different businesses now finding that their activities and technologies are beginning to merge In the United States, legislation was passed in 1996 to allow new competition to develop and new in-dustries to emerge The full implications remain to be seen, especially as the business arguments for investments by the industry are refined and we see just how quickly changes will occur There is, however, already ample evidence of the merging of technologies such as cable television, telephone, networking, and satellite communications High-speed lines into homes and offices are increas-ingly available, and inexpensive mechanisms for connecting to the Internet with-out using a computer have also emerged The impact on all individuals is likely

to be great and hence on our patients and on their access to information and to their healthcare providers Medicine cannot afford to ignore these rapidly oc-curring changes

Envisioning the Enterprise Internet

Although we should always expect a medical record to be populated with data about a specific patient, in the electronic implementation of records we also ex-pect to find data regarding populations of patients, integrated access to the bio-

medical literature, and interactive environments for offering clinical guidelines

or frank consultative advice We envision a world in which the enterprise tranet of Figure 1.6 is seamlessly connected to the full Internet beyond, with in-tegrated access to a wide variety of information sources that are geographically distributed well beyond our local institutions (Fig 1.7) To the extent that an in-dividual's medical records are maintained in compatible electronic formats at all 3Data are taken from a presentation delivered by an MCI-Worldcom corporate vice pres-ident and Internet visionary, Vint Cerf, at the National Library of Medicine in March 1997

Vendors or Various

Types (e.g.,

Pharmaceutical companies)

FIGURE 1.7 Moving beyond the organization The enterprise Internet is the integration

of an organization's intranet (Fig 1.6, encapsulated in the box here labeled "Local Health System") with the full potential of the worldwide Internet Both providers and patients increasingly access the Internet for a wide variety of information sources and functions suggested by this diagram (see text)

the institutions where they have been seen, the Internet provides the potential of creating "virtual medical records," the electronic compilation of a patient's health data from all the settings in which they have been seen Although such a con-cept raises important issues regarding patient data privacy and confidentiality, there are technical and policy measures that can be taken to help to ensure that such virtual records are kept secure from prying eyes but can be made available

at times of medical need (see Chapter 7) (National Research Council, 1997)

Implications for Patients

As the number of Internet users grows (estimates suggest that there were more than 50 million users in the United States alone by 1999), it is not surprising that increasing numbers of patients, as well as healthy individuals, are turning to the Internet for health information (see Fig 1.7) It is a rare North American physi-cian who has not encountered a patient who comes to an appointment armed with

a question, or a stack of laser-printed pages, that arose due to medically related searches on the World Wide Web The companies that provide search engines for the Internet report that medically related sites are among the most popular ones being explored by consumers As a result, physicians and other care providers must be prepared to deal with information that patients discover on the Web and bring with them when they seek care from clinicians Some of the in-formation is timely and excellent; in this sense physicians can often learn about innovations from their patients and will need to be increasingly open to the kinds

of questions that this enhanced access to infonnation will generate from patients

in their practices On the other hand, much of the health infonnation on the Web lacks peer review or is purely anecdotal People who lack medical training can

be misled by such infonnation, just as they have been in the past by printed fonnation in books and magazines dealing with fad treatments from anecdotal sources In addition, some sites provide personalized advice, often for a fee, with all the attendant concerns about the quality of the suggestions and the ability to give valid advice based on an electronic mail or Web-based interaction

in-In a more positive light, the new communications technologies offer clinicians creative ways to interact with their patients and to provide higher quality care Years ago medicine adopted the telephone as a standard vehicle for facilitating patient care, and we now take this kind of interaction with patients for granted

If we extend the audio channel to include our visual sense as well, the notion of telemedicine emerges Although there are major challenges to be overcome be-fore telemedicine is likely to be widely adopted for direct patient care (Grigsby

& Sanders, 1998), there are specialized settings in which it is already proving to

be successful and cost effective (e.g., international medicine, teleradiology, and video-based care of patients in state and federal prisons) The challenges are largely regulatory and fiscal before telemedicine will be widely adopted

A potentially more practical concept in the short tenn is to use computers and the Internet as the basis for communication between patients and providers For example, there has been rapid growth in the use of electronic mail as a mecha-nism for avoiding "telephone tag" and allowing simple questions to be answered asynchronously (the telephone requires synchronous communication; electronic mail does not) More exploratory, but extremely promising, are communications methods based on the technology of the World Wide Web For example, there are young companies that work with managed care organizations and healthcare systems to provide Web-based facilities for disease management Patients log in

to a private Web site, provide infonnation about the status of their chronic ease (e.g., blood glucose readings in diabetes), and later obtain feedback from their physician or from disease managers who seek to keep the patients healthy

dis-at home, thereby decreasing the need for emergency-room or clinic visits

Requirements for Achieving the Vision

Many of the concepts proposed above depend on the emergence of an Internet with much higher bandwidth and reliability decreased latency and financial models that make the applications cost effective and practical Major research efforts are underway to address some of these concerns, including the federal Next Generation Internet activity in the United States.4 In addition, academic institutions have banded together in a consortium designed to create new test beds for high-bandwidth communications in support of research and education