The Internet and Health Communication: Experiences and Expectations [Hardcover ed.] 0761922326, 9780761922322

Table of contents :
Cover
Contents
Foreword
Preface
Part I --
An Overview of Experiences and Expectations
Chapter 1 --
The Internet and Health Communication: A Framework of Experiences
Chapter 2 --
The Future of the Internet in Health Care: A Five-Year Forecast
Part II --
Sources of and Experiences With Online Medical Information
Chapter 3 --
Consumer Use of Medical Information From Electronic and Paper Media: A Literature Review
Chapter 4 --
Assessments of Quality of Health Care Information and Referrals to Physicians: A Nationwide Survey. Chapter 5 --
Use of the Internet for Professional Purposes: A Survey of New Jersey PhysiciansChapter 6 --
Expectations and Experiences of Seeking Infertility Information via the Internet and the Telephone Directory
Part III --
Experiences Developing and Evaluating Health Information Sites
Chapter 7 --
Using the Web to Assist Communities in Public Health Campaign Planning: A Case Study of the REACT Project
Chapter 8 --
Evaluating a Federal Health-Related Web Site: A Multimethod Perspective on Medicare.gov. Chapter 9 --
A Pound of Cure: A Content Analysis of Health Information on Web Sites of Top-Ranked HMOsChapter 10 --
A Comparative Features Analysis of Publicly Accessible Commercial and Government Health Database Web Sites
Part IV --
Experiences of Online Health Communities and of Organizations Moving to E-Commerce
Chapter 11 --
Experiencing Empathy Online
Chapter 12 --
The Role of the Organization in the Success of Web-Based Continuing Medical Education Programs
Chapter 13 --
Improving Diabetes Care With Telecomputing Technology. Chapter 14 --
Web-Enabled Hospitals in the United States: Influences on Adoption ProcessesChapter 15 --
Competitive Collaboration in Australia's Pharmaceutical Industry
Part V --
Public Policy Experiences and Expectations
Chapter 16 --
ehealth: Federal Issues and Approaches
Chapter 17 --
Old Wine in Silicon Prescription Bottles: Some Legal Issues, Benefits, and Disadvantages Associated With Internet Pharmacies
Chapter 18 --
Networked Communication Practices and the Security and Privacy of Electronic Health Care Records
Chapter 19 --
Concluding Thoughts
Appendix
Index
About the Editors.

Citation preview

The Internet and Health Communication: Experiences and Expectations

Ronald E Rice

THE

INTERNET AND HEALTH

COMMUNICATION

With enduring love and gratitude, we dedicate this book to our parents: Eugene and Gloria Rice Raymond and Frances R. Katz

THE

INTERNET

Μ® HEALTH COMMUNICATION

Experiences and Expectations

Ronald E. Rice James E. Katz — Editors

/ Ä \ S a g e Publications, Inc. I

1 International Educational and Professional Publisher Thousand Oaks • London • New Delhi

Copyright © 2001 by Sage Publications, Inc. All rights reserved. No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the publisher.

For information:

®

Sage Publications, Inc. 2455 Teller Road Thousand Oaks, California 91320 E-mail: [email protected] Sage Publications Ltd. 6 Bonhill Street London EC2A 4PU United Kingdom Sage Publications India Pvt. Ltd. M-32 Market Greater Kailash I New Delhi 110 048 India

Printed in the United States of America Library of Congress Cataloging-in-Publication Data Main entry under title: The internet and health communication: Experiences and expectations / edited by Ronald E. Rice and James E. Katz. p. cm. Includes bibliographical references and index. ISBN 0-7619-2232-6 (cloth: alk. paper) ISBN 0-7619-2233-4 (pbk.: alk. paper) 1. Health—Computer network resources. 2. Internet (Computer network) 3. Medical telematics. 4. Communication in medicine. I. Rice, Ronald E. Π. Katz, James Ε. ΙΠ. Title. R859.7.E43156 2000 025.06'3621—dc21

01

02

03

04

Acquiring Editor: Editorial Assistant: Production Editor: Editorial Assistant: Typesetter/Designer:

05

10

00-010038

9

8

7

6

Margaret H. Seawell Heidi Van Middlesworth Claudia A. Hoffman Cindy Bear Janelle LeMaster

5

4

3

2

1

Contents

Foreword

ix

Stephen A. Schroeder Preface

xiii

P A R T I: A n O v e r v i e w of E x p e r i e n c e s a n d Expectations 1. The Internet and Health Communication: A Framework of Experiences

1

5

Ronald E. Rice 2. The Future of the Internet in Health Care: A Five-Year Forecast Robert Mittman and Mary Cain

47

PART I I : Sources of a n d E x p e r i e n c e s W i t h O n l i n e M e d i c a l Information 3. Consumer Use of Medical Information From Electronic and Paper Media: A Literature Review

75

79

Philip M. Napoli 4. Assessments of Quality of Health Care Information and Referrals to Physicians: A Nationwide Survey

99

Philip Aspden and James E. Katz 5. Use of the Internet for Professional Purposes: A Survey of New Jersey Physicians

107

Philip Aspden, James E. Katz, and Ann E. Bemis 6. Expectations and Experiences of Seeking Infertility Information via the Internet and the Telephone Directory

121

June Anigbogu and Ronald E. Rice

PART III: E x p e r i e n c e s D e v e l o p i n g a n d Evaluating Health Information Sites 7. Using the Web to Assist Communities in Public Health Campaign Planning: A Case Study of the REACT Project

145

147

John R. Finnegan, Jr., Deborah Alexander, Jason Rightmyer, Barbara Estabrook, Bernadette Gloeb, Melinda Voss, Laura Leviton, and Russell V. Luepker 8. Evaluating a Federal Health-Related Web Site: A Multimethod Perspective on Medicare.gov Sid J. Schneider, Joy Frechtling, Timothy Edgar, Barbara Crawley, and Elizabeth Goldstein

167

9. A Pound of Cure: A Content Analysis of Health Information on Web Sites of Top-Ranked HMOs Elizabeth

189

Witherspoon

10. A Comparative Features Analysis of Publicly Accessible Commercial and Government Health Database Web Sites

213

Ronald E. Rice, Michael Peterson, and Robert Christine

PART IV: E x p e r i e n c e s of O n l i n e Health C o m m u n i t i e s a n d of O r g a n i z a t i o n s M o v i n g to E - C o m m e r c e

233

11. Experiencing Empathy Online

237

Jennifer J. Preece and Kambiz

Ghozati

12. The Role of the Organization in the Success of Web-Based Continuing Medical Education Programs

261

Pamela S. Whitten, Matthew S. Eastin, and David Cook 13. Improving Diabetes Care With Telecomputing Technology

287

Richard L. Street, Jr., and Veronica K. Piziak 14. Web-Enabled Hospitals in the United States: Influences on Adoption Processes

309

Lauren B. Eder and Donald E. Wise 15. Competitive Collaboration in Australia's Pharmaceutical Industry Elizabeth More and G. Michael

McGrath

329

P A R T V: P u b l i c Policy Experiences a n d Expectations

351

16. ehealth: Federal Issues and Approaches

355

Cynthia Baut, Mary Jo Deering, and Leslie Hsu 17. Old Wine in Silicon Prescription Bottles: Some Legal Issues, Benefits, and Disadvantages Associated With Internet Pharmacies

385

Barry D. Bayer 18. Networked Communication Practices and the Security and Privacy of Electronic Health Care Records

393

James E. Katz and Philip Aspden 19. Concluding Thoughts

417

James E. Katz and Ronald E. Rice

Appendix

431

Index

433

About the Editors

449

About the Contributors

451

Foreword

T

he U.S. health care system produces two primary outputs: treatments and information. Despite having perhaps the best scientific apparatus in the world to develop new treatments, our health system lags far behind many other economic sectors in various information handling chores. Many of its shortcomings could be addressed, with greater efficiency and speed, through the smart application of new information technologies, particularly the Internet-based applications that are the subject of this fascinating volume. Information technology can be used to train health care workers in novel ways and keep the knowledge and skills of existing professionals up to date. It can be used to track disease patterns and educate the public on how to stay healthy and manage chronic disease. It can be used to improve the efficiency of clinical practice and to streamline the management of health care organizations. In fact, real dreamers can even envision a day when our nation's health care "system" could actually become a real system. The promise is great. But the reality falls far short. At a personal level, many of us are frustrated by the paperwork involved in insurance claims, and those of us who have had to handle the paperwork for an elderly parent with multiple chronic diseases know the true meaning of the word overwhelming. For their part, doctors must help some patients sort out reams of diverse— ix

χ

THE INTERNET AND HEALTH COMMUNICATION

even conflicting—Internet information. The paperwork demands on doctors and other providers from multiple payers and health systems continue to escalate. Hospitals can't share data with others in their own network, and the recent bankruptcy of a prestigious northeast health system, laid partly at the door of computer systems that couldn't talk to each other, shows how mission-critical this is. Dealing with health and medical information imposes some special burdens and questions that other social sectors don't face. First, of course, is the absolute need for privacy and confidentiality of patient records. No one who has spent time in a busy hospital clinic would assert that paper records are strictly confidential. Yet there is something very different about the risk of a pair—or two—of prying eyes, compared to the black-box ability of a computer to collect, analyze, disseminate, and link our personal information to others whom we cannot see and may never know. How will we protect confidentiality and privacy without posing impossible security burdens on caregivers? How will we share records across settings for patient convenience and safety? Second, a set of questions lies with various telemedicine applications. For example, it is possible to transmit right from the home bedside to health center amazingly clear, live pictures or data from monitoring equipment and immediately bring back to the patient and family the voice and face of the health professional many miles away. But, what is the loss to the patient in terms of human contact? How do these gains and losses balance out? How acceptable are high-tech communications approaches to the elderly? the confused? How do we collect and disseminate best practices? Again, some of the chapters in the present volume are a start. Third, a difficult question in Internet applications in health care has to do with consumer and patient education. One source predicts that 30 million U.S. adults will seek health information on the Web by 2001. But what will they find? Snake oil salesmen have been with us always, now they attract their customers through modems and animated GIFs, instead of brightly painted horse-drawn wagons. There is so much health information on the Web—some of it bad, some good, and some (like online support groups spanning the country, even the world) unreproducible by any other means. Although the exceedingly tidy among us would like to abolish health Web sites that don't adhere to scientific standards, that would be difficult. In fact, the sheer volume may make censorship unnecessary. It's so easy to get multiple opinions that patients may be able to recognize a consensus—and outlier opinions—more readily. At the same time, the

Foreword

yd

emergence of well-funded health sites, whose credentials are clear, ups the ante for the marginal ones. These sites, supported by online advertising, may be editorially influenced by commercial imperatives—another hazard to consumers. Last, although we see evidence that the "Digital Divide/ as it applies to individuals, may be shrinking because of programs offering free computers through schools and employers, programs to open libraries and school computer labs to community uses, and so on, the Divide will be hard to bridge for safety net health care providers that do not have the resources to purchase and support the hardware, software, and humanware that would let them interface with the increasingly automated medical record, referral, and reimbursement systems used by other providers. How can we help them get the equipment and trained staff to enter the electronic era? This is a survival issue for them, and the surrounding system becomes increasingly reliant on efficient data interchange. This volume of essays, each of which tackles a different type of problem or example of Internet applications in health care, is a start in answering such questions as these. The world is changing around us at an increasing pace. It is well past time for the health care sector—one seventh of our nation's economy—to start realizing the full benefits of information technology. Touchstones like this will help us get and keep our bearings in the fray. 7

Steven A. Schroeder, MD President The Robert Wood Johnson Foundation

Preface

I

n some ways, this book is an academic version of the classic movie, King Kong Meets Godzilla. King Kong is played by the health care sector. Central casting's choice is sound: The national health care sector can easily play the proverbial 800-pound gorilla that must not be ignored when he occupies a seat at important policy discussion tables. In the United States alone, health expenditures tip the scales at $1.3 trillion and are projected to balloon to a yet more impressive $2 trillion before the decade is out (U.S. Census Bureau, 1999). In fact, it is the largest sector in the U.S. economy. And the importance of health to Americans is not something that can be measured by money alone. It is a high priority for all Americans; simply, we cannot live without it. In the other corner is the Godzilla-like Internet. We could easily venture that the Internet is the first new mass medium in a half century. But we go farther. We propose that it, like Godzilla, is unprecedented. Godzilla, the reader may recall, is a not entirely seamless amalgamation of preexisting creatures, but is also endowed with a generous mixture of unprecedented and, one might add, commonsense-defying, powers. The Internet, too, is something that has both been seen before yet also has unexpected and commonsense-defying powers. In a few short years, the Internet has sprouted from a handful of kluged computer networks to a frenetic monster that embraces half the U.S. population. As such, it is sui generis, the xiii

xiv

THE INTERNET AND HEALTH COMMUNICATION

first "many-to-many" medium, as opposed to "one-to-many" media (such as TV) or "one-to-one" (such as the telephone). The focus of our book, to continue the movie analogy, is the "meeting" of these two colossal beasts. As they embrace, the stakes are enormous, and are ones in which we are all vested. Given the magnitude of the event, we have necessarily taken only a narrow perspective on the interaction of these two powerhouses: that of communication issues. This is not a "how-to" book nor a book offering practical advice to the public or physicians about finding and using medical information on the Internet. It is, instead, an attempt by academics for academics, students, researchers, and health professionals to explore the communication issues related to the consequences the Internet is having on health care as well as the way the exigencies of health care are affecting the use of the Internet. The gap we see in the literature on this area is an analysis of the way people actually use the Internet in pursuit of health care, with what consequences and implications. That is, few have probed what can be learned about human communication behavior when we look at the practice of medicine and the pursuit of health on the Internet. At this point, we know little about how traditional modes of human behavior are changing, and what might remain the same, as considerable personal and economic resources are shifted from traditional to electronic sites of health care delivery. Will the previously successful appeals to change behavior aimed at promoting good health be as effective in cyberspace? What new appeals will succeed? What policy questions need to be resolved so that the fruits of Internet health care can be harvested while avoiding further tragedies and abuse? How can social science theory enlighten us about communication issues in this brave new environment? By drawing on dispassionate and rigorous analysis experiences to date, we will better know what to expect for the future. A unique aspect of the book is that we look at a variety of settings and groups to grasp the diverse nature and impact of the Internet's use for health care. As a result, we believe our volume should help cross boundaries that have heretofore prevented greater progress both in understanding human communication behavior and in turning that understanding to the benefit of the public's health. This book is one of the first to apply systematic empirical tools to analyze the dramatic changes in health care associated with the Internet. Rather than being based primarily on ad hoc analyses or speculation, the authors have sought to ground the volume firmly in research. The book

Preface

xv

covers the gamut of activities from social interaction to e-commerce, and from regulatory regimes to Web site development and evaluation. Some courses will use this book as their primary text. These include college and graduate school courses that examine the Internet and electronic commerce, telecommunications, health communication, and communication campaigns. Nearly all research-caliber university libraries will want to acquire a copy. We would also expect strong interest from professional audiences, including health care and hospital administrators, federal and local policymakers, health care-oriented business people, and entrepreneurs interested in offering services. Finally, it seems quite likely that many physicians, social science researchers, and communication professionals will want to read this book to develop a broader understanding of the many relationships between the Internet and health communication. Thus what we have to say should be of both practical and conceptual significance to the provider community, the interested general reader, the health care entrepreneur and policymaker, the researcher, the teacher, and the student. We thank our colleagues at Rutgers, and most especially Brent D. Ruben, for their thoughtful encouragement of our devotion to research and writing. We at the School of Communication, Information and Library Studies (SCILS) seek to meld academic insight with pragmatic applications of communication and information science to advance important social goals. Our view is based on the fact that we are bringing together two important areas of communication: mediated technology (the Internet) and social action (experiences and expectations of communities, organizations, and individuals) to gain insight into an area of vital concern: health care delivery. Katz wishes to thank Robert K. Merton, Professor Emeritus at Columbia University, for the insights that he has so generously shared. Professor Merton has contributed invaluably to my understanding of telecommunication technologies by helping me apply an opportunity structure perspective. Irving L. Horowitz, PhD, as always, has been a constant intellectual guide and inspiration. Rice wishes to thank all the authors for their thoughtfulness, hard work, patience, and willingness to consider his suggestions, revisions, formatting guidelines, and page length restrictions. This has been a constant preoccupation for nearly the past year, and I have to say that it has been uniformly pleasant and insightful. It's been a great honor to be able to collaborate with so many wise and wonderful people.

xvi

THE INTERNET AND HEALTH COMMUNICATION

We thank Margaret Seawell, Leticia Gutierrez, and Claudia Hoffman from Sage for their encouragement and precise teamwork. We would also like to thank our families, whose sacrifices on behalf of this project have been so substantial. We will be making it up to them by finally being able to take them on the oft-postponed vacation they so richly deserve. We will be doing this immediately after we finish the follow-on projects that we will tackle immediately after we complete the research we are currently planning. —Ronald E. Rice —James E. Katz

Part I

An Overview of Experiences and Expectations

W

e take quite seriously the subtitle of our volume, "experiences and expectations." In order to be widely adopted, communication technology must fulfill a need; this is an obvious criterion. But it additionally must function in a way that is compatible with the predispositions, needs, and values of humans. This criterion is often overlooked by system designers. In turn, determining what these predispositions are must, at least initially, be predicated on experience. The Internet is a good example of this principle in action. Before the Internet existed, there were private prototype systems that incorporated all the elements of what we commonly think of as characteristics of the Internet; some of these even included elements that are beyond those that are currently available on the Internet. Depending on the system, these would include news and wire service feeds, video-on-demand, streaming conference presentations, virtual auditoriums, electronic mail, electronic reference books, text-to-voice applications, proto-Internet telephony, remote server access, online TV broadcasts, audio books, stereo music, and full-motion, high-fidelity videoconferencing. One such system was 1

2

THE INTERNET AND HEALTH COMMUNICATION

Telesophy, created at Bellcore in 1986 by Bruce Schatz and others (including coeditor Katz). (Telesophy is a neologism derived from ancient Greek meaning "knowledge at a distance.") What was missing from these early prototype systems, though, was a convenient graphical user interface. Only with the coming of the intuitive object-oriented interfaces of the 1990s did the Internet become a mass medium. Most readers will be familiar with the World Wide Web and the Apple computer-style operating systems (imitated by Microsoft Windows). It was enhancements of this caliber that made the Internet experience possible for tens of millions of people. But we are yet at the simplest of beginnings. The amazing and powerful applications are not only still to come but also are not even formed in the imagination. This brings us to the expectations that are held for the Internet fulfilling its potential for improving health care and well-being. For these expectations to be fully and equitably realized, the insights of social science research need to be brought to bear on the new systems as they are configured, made available, implemented, and used. To be successful, the health care communication systems of tomorrow, perhaps even more than those of other sectors such as e-commerce, need to be designed in ways that conform with the way humans like to work and interact. Moreover, the systems should be designed so that they incorporate the cultural and normative contexts of the particular subgroups of society to which they are trying to appeal. Given this situation, two tasks confront us. The first is to understand what has been attempted, both historically and in various contemporary cultural contexts. This information is valuable as we seek to determine what has worked—and what has failed—and the reasons for these divergent outcomes. The second is to look toward the future to identify what resources (including social-organizational ones) might be available and what problems need to be solved in what priority. By fitting the lessons of the past to ambitions for the future, we are in a better position to leverage our understanding and marshal our strength to help improve the public's health care. We can also better provide members of the public with the tools that will enable them to look after their own health. To lay the foundation for the understanding we seek to generate, we begin with two chapters, one assessing the current state of experiences with using the Internet for health information and communication, the other exploring influences on and forecasts of the Internet's role in health care. Chapter 1 provides an overview of the usage and content of Web sites and

Experiences and Expectations

3

the information for health care. Included in his examination are mailing lists, communities, and newsgroups as well as precursor and alternative medical and health information systems. Communication processes, certainly, are an important dimension to health care, and central to our concerns. So it is most appropriate that he explores important primary health care communication processes, including patient-physician communication and patient-patient support. He also probes what might be termed superstructural communication issues such as the credibility of online information, and risks of privacy invasion and tortious and criminal liability. Rice also examines the use of Internet-based health information systems as part of public communication campaigns. At the level of communication policy, he comments on issues of accessibility and barriers to access and use. Rounding out his investigation, Rice surveys issues of intellectual property and cross-provider boundaries. Finally, he suggests a summary model of the relationships among these topics and actors. In Chapter 2, Mittman and Cain provide an overview of the major drivers (developments, barriers, and advanced applications), trends, and issues for the future of the Internet and health care. This analysis, based in part on a commissioned study, discusses plausible near-term scenarios and consequences of the Internet on health care and health care on the Internet.

1 The Internet and Health Communication A Framework of Experiences RONALD E. RICE

H

ealth issues, especially research on health communication and health-related public communication campaigns, are becoming increasingly important for the public agenda. The same is true of new communication technologies, especially the Internet. The intersection of these two highly significant social trends is creating a rich, consequential, and challenging domain for users, health providers, researchers, and policymakers. The economic stakes are enormous. Health care consumes more than 13% of America's GNP; while much of this will remain in "bricks-andmortar" and "hands-on" sites such as hospitals and in-home delivery of services, much will also migrate to the Internet. Jupiter Communications reported that the online consumer health care market is expected to grow

AUTHOR'S NOTE: I thank James Katz for his help in developing this chapter. Preparation of this chapter was supported in part by a generous travel grant from Dr. Celia Romm, Central Queensland University, Australia, in connection with the Get Smart! Conference, November 1999.

5

6

EXPERIENCES AND EXPECTATIONS

to $1.7 billion by 2003 ("The Doctor Is Online," 1999), up from just $1 million in 1998, driven by prescription drugs, over-the-counter drugs, vitamins, and health-care products sold online. This technology promises to revolutionize the way health care is provided in at least as profound a way as did the movement of the site of health care delivery away from the barbershop. People will be empowered to gain expert insight into their problems and assess the options available to treat them. Diagnosis and cure will be dispensed interactively and electronically. At the same time, there are many opportunities for incorrect, misleading, fraudulent, and dangerous practices. Serious questions also arise about the ability of those with economic, intellectual, or physical limitations to participate effectively in this new environment. This chapter reviews the primary evidence and issues surrounding the use of the Internet for health communication. The following sections discuss: coverage of the topic in the print media, usage of Web sites, mailing/ discussion lists, online communities and newsgroups; medical computing, the health care industry, and community health information systems; patient-physician interaction, physician-physician interaction, patientpatient support, online medicine, online health information credibility, privacy and liability, policy and access; and the use of interactive media for health communication campaigns. The other chapters in this volume consider these issues in greater detail.

COVERAGE AND USAGE OF WEB SITES, MAILING LISTS, COMMUNITIES, AND NEWSGROUPS COVERAGE OF INTERNET AND HEALTH COMMUNICATION IN RECENT PUBLICATIONS Table 1.1 clearly shows the growing interest in this area, based on very specific keyword searches in five major commercial publication databases. The major growth has occurred in just the past 3 or 4 years.

General Usage Almost half of Internet users in 1997 reported looking for health information or support (FIND/SVP, 1997, cited in Eng, Maxfield, & Gustafson, 1998). A CyberDialogue/Internet Health Day survey of 2,000 Internet users estimated that more than 24.8 million people sought online health and

The Internet and Health Communication ν; ; V ;

7

Growth of Articles Concerned With Internet and Health Communication Retrieved From Business, Professional, Trade, and Technical Publication Databases, 1991 -1999, as of January 2 0 0 0 ABI/INFORM

Dowjones News & Business

3

Business Abstracts

b

Allied Health

Periodical Index

Medicine

c

Year

(%)

99

43

18

19

25

16

98

23

37

24

23

41

(%)

(%)

(%)

(%)

97

17

32

34

21

16

96

12

05

14

17

10

95

04

08

07

11

13

94

01

01

03

03

93

— —

00

92

— —

— — —

— —

Total

7,697

70

3,481

415

00

116

N O T E : Search query (on January 5, 2000): {Internet) or (WWW) and {health communication) or {medical information); 1999 figures are underestimates as many items are added to databases several months after their publication date. a. Dow Jones Interactive News and Business Headlines: 6,000 leading U.S. and international business newspapers, magazines, trade journals, newsletters, television and radio transcripts; searched fulltext. b. A B I / I N F O R M : 1,000 U.S. and international professional publications, academic journals, trade magazines; searched abstracts and fulltext. c. M E D L I N E : 3,600 journals covering medicine and health care, communication disorders, population biology, and reproductive biology; searched all fields.

medical content in 1998, up 44% from the prior year ("Double Mastectomy/ 1999). General use of e-health sites grew 176% in 1999, to 10.8 million in December (Media Meirix, 2000), much faster than the growth in general Internet usage. The CyberDialogue/Internet Health Day survey reported that about 52% of users sought out information on diseases; about one third sought out information on diet and nutrition, pharmaceuticals, and fitness; and about 15% sought out information on children's health. About one quarter of the disease-related searchers had joined an online support group (MSNBC, 1999). A recent Harris Poll found that nearly 70% of online users have researched a disease or medical condition, seeking information on, in decreasing frequency, depression (19%), allergies/sinus (16%), cancer 7

8

EXPERIENCES AND EXPECTATIONS

(15%), bipolar disorder (14%), arthritis/rheumatism (10%), high blood pressure (10%), migraine (9%), anxiety disorder (9%), heart disease (8%), and sleep disorders (8%). After the National Library of Medicine provided free Web access to MEDLINE in June 1997, usage jumped 1,000% to 75 million searches per year (Lindberg & Humphreys, 1998). The Health on the Net Foundation's third annual survey (HON, 1998) reported results based on 1,863 online responses. One of the biggest differences from the prior two surveys was that a majority of respondents were patients, and female. Health is one of the few areas of Internet usage where women outnumber men (53% to 47%) (Goldman-Sachs, 1999). Overall, 93% found Internet medical/health information useful, although 53% felt such information needs to be enhanced. Seventy-three percent of the respondents sought such information from their home. While 2 2 % sought such information for a patient, 29% sought it for themselves. Other beneficiaries included spouse (6%), child (9%), parent (8%), friend (17%), and relative (5%). Sixteen percent strongly agreed that the quality of the information needs to improve, with 37% agreeing, 37% neither agreeing nor disagreeing, 8% disagreeing, and 1% strongly disagreeing. Twentyfive percent strongly agreed that useful medical/health information was easy to find, with 43% agreeing, 17% neither agreeing nor disagreeing, 14% disagreeing, and 2% strongly disagreeing.

Web Sites A July 1999 study reported in Nature found that there were more than 15,000 health information Web sites categorized at the prime tier of both the Yahoo and Netscape directories. In addition, there were at least 44 professional-level Web sites that were dedicated solely to compiling and making available to their visitors other health-related Web sites (the figures reported in the following tables show even greater coverage). These numbers exclude ancillary topics related to health, such as health law and end-of-life processes. Many of these are commercially based, and represent a market capitalization of over $1 billion. In July 1999, drkoop.com was the most popular online health site, as evidenced by nearly 1.5 million hits per day, and retained this distinction until the end of 1999, when the OnHealth Network (http://www.onhealth.com/chl/index.asp) had more than 3.2 million unique users and nearly 16 million page views (Media Metrix, 2000), becoming the 82nd most frequently visited site on the Web.

The Internet and Health Communication

9

In 1998, the U.S. Department of Health and Human Services site, http:// www.healthfinder.gov, received about 4.7 million hits per month from 8,000 people per day. Links to comprehensive general, government, and commercial health sites, especially indexes, can be found at http://www. scils.rutgers.edu/~rrice/healsite.htm. See also this book's Appendix for the rapidly growing list of books that discuss and review online health information resources. Some sites offer particularly intriguing content, such as a live broadcast of a double-mastectomy (http://www.thehealthnetwork.com). The patient felt that potential breast-cancer patients might benefit from this experience, and others supported this increased sharing of health information and medical procedures, but others criticized the media hype associated with the event, as well as the possibilities for distractions and possible liabilities for the doctors involved ("Double Mastectomy," 1999). An early astounding site is the Visible Human Project, providing detailed views of a human body at 1-millimeter cross-sectional intervals (http:// www.nlm.nih.gov/research/visible/visible_human.html). The Web site of the Office of Alternative Medicine in the National Institutes of Health provides links to sites on acupuncture, aromatherapy, herbology, macrobiotics, naturopathic medicine, and yoga, among others (Slobodien, 1998). The National Library of Medicine has just opened an Internet-based registry of all ongoing clinical trials looking for participants (http://clinicaltrials. gov).

Mailing Lists There are many specialized online support groups, such as for cancer (see Landro, 1999). A search on Liszt.com provided 278 health-related mailing lists. Table 1.2 lists the major categories, with number of lists within each. Topics of these lists ranged from ethnicity-specific health issues, managed health care, and women athletes, to hypnosis and past lives, occupational health, distance health education, and holistic and natural health care. Liszt.com also found nearly 400 commercial lists (ads, press releases, announcements) in 28 categories ranging from allergy supplies, aromatherapy, and how to stop smoking, to natural healing, stress management, and vitamins and minerals. Meta-List.net, a search engine for nearly a quarter-million online newsletters and discussion lists, returned 501 results for the term "health com-

10

EXPERIENCES AND EXPECTATIONS Health-Related Categories, and Numbers of Online Mailing Lists Within Each, Found by Listz.com, as of October 1 9 9 9 ADD 5 AIDS 9

Homehealth 1 Hospice 2

Acupuncture 2 Addictions recovery 7 African American 1 Allergy 6 Alzheimer's 4 Arthritis 4 Autism 4 Blindness 7

Kids 4 Leukemia 3 Long-term care 2 Massage 1 Medicine 48 Men 1 Mental health 32 Midwifery 2

Cancer 18 Children 8 Chronic fatigue 6 Dentistry 11

Nursing 20 Occupational therapy 5 Organ transplants 2 Pain 6

Depression 7 Diabetes 4 Disabilities 23

Pregnancy 2 0 Prostate 4 Psychiatry 6

Diseases 49 Drug abuse 2 Eating disorders 5 Epilepsy 1 Exercise 7 Food 22 General 18 Hair 1 Health 5 Health professionals 21 Hearing loss 5

Safety 20 Skin 2 Stroke 1 Student 2 Support 74 Syndromes 18 Toxins 1 Twelve steps 9 Unclassifiable 1 Weight loss 2 9 Wellness 9

Holistic 6

Women 35

munication" (such as a list for health-promotion and disease prevention researchers) and 126 for the term "medical information" (such as the medicine and developmental disabilities list). The E-Zine-List (http://www. meer.net/~johnl/e-zine-list) shows that of the 80 most frequent keywords used to index and retrieve online newsletters and e-zines, "health" was 18th, with 180 e-zines. Of course, there are many other lists that may not be included in Listz.corn's or Meta-List.net's search processes. For example, askdrweil.com, which specializes in holistic and herbal medicine, lists more than 280 message boards.

The Internet and Health Communication

" ""

11

Online Health-Related DejaCommunities Found by DejaNews.com as of October 1 9 9 9

Australia's healthy investment, Baby boomers and aging, Bay Area older w o m e n , Bionutrition, Clubhouse (mental health), Darkaventure magnetic health products, Diagnosis: breast cancer, Esj nutritional products, For your good health, Geritarics 1 0 0 1 , Health and fitness, Health and lifestyle, Health bracelets, Health wealth and you, Healthcare IT careers, Healthshack.com, Herbalchest, HerbaLife, H I V the cure is there, Humans of the earth, Israel mental health, Jenny Craig support group, Life quest herbs, Living the holistic lifestyle, Managed care health professionals, M e d w e b India, Nursing homes—don't do it, Nutrition and immune support, Practice management options, R S D disability, Sexual health, Smokers survey, Spiritual, S u ko ming gung fu, Tai chi chuan, Toddler health, W h e e l council storytelling for prevention, Wolfsong herbs, w w w . a l l p e t s . c o m

Communities DejaCommunities is a new free service that lets people with common interests get together online. Interactions include ongoing discussions, chatrooms, Web site recommendations, group event promotion, polling, project collaboration, and announcements. The service also allows users to create their own public or private communities. Already, there are 50 communities relating to health, shown in Table 1.3. There were four communities identified by the term "health communication": Health care-human resources; mentally maimed families; Chinese medicine; and military medical student association. The 29 "medical information" communities included topics ranging from diving medicine, avoidant personality disorder, and a military medical student association to hyperbaric oxygenation therapy and AA alopecia areata.

Newsgroups In 1995, the last time individual newsgroup readership figures were provided (see news.lists), half of the top 10 Usenet newsgroups (each with from 8,000 to 22,000 readers monthly) were concerned with health topics. In decreasing order, they were: depression, diet, cancer, eating-disorder, and arthritis (Walther & Boyd, in press). More recently, DejaNews.com listed 17 main categories of "health & fitness" newsgroups, from Addiction to Mental to Surgery, with 107 subcategories, as shown in Table 1.4. A Listz.com search of newsgroups that had the word health in either the list

EXPERIENCES AND EXPECTATIONS

12 I

Categories of Health and Fitness NewsGroups Listed by DejaNews.com as of March 2 0 0 0

Addiction (Cigarettes); Alternative (Naturopathy); Industry (Education, Insurance, Networks); Media (Books, Health W e b Sites, Magazines, Newsletters, T V ) ; M e n ' s (Hair Loss, Impotence); Mental (Depression); Nutrition (Diet Aids, Diets, Supplements, Vitamins); Physical Conditions (Arthritis, Asthma, Cancer, Dental, Dermatological, Diabetes, Disability, Gastrointestinal, H I V and A I D S , Injuries); W o m e n ' s (Abortion, A n e m i a , Birth Control, Chronic Fatigue, Gynecology, Infertility, Maternity, Menopause, Menstrual, Osteoporosis)

name or its short description found 70 groups. Eight had FAQS (frequently asked questions sections) and 21 were moderated. Topics ranged from "really old medicine from India," biofeedback therapy, and chronic fatigue syndrome, to oxygen and ozone therapy, and Russian health documents.

Specialized Health Web Sites There are now so many Web sites about any given topic that specialist meta-search engines and gateways are emerging to help provide evaluations and guidance. As Table 1.5 shows, Invisible Web (http://www. invisibleweb.com), which is a meta-search engine that returns specialized rather than general lists or sites, provided 26 "health" subcategories pointing to 640 sites, while Yahoo! provided 43 "health" subcategories pointing to more than 19,000 sites. The Johns Hopkins reproductive health and population page has links to over 400 approved sites in 50 categories (jhuyccp.org/netlinks), and the Argus Clearinghouse provides detailed evaluations of its approved member health sites (clearinghouse.net). There are also many medical journals online (http://www.yahoo.com/ Health/Medicine/journals), with a growing number of them full-text and peer-reviewed.

PRECURSORS TO INTERNET HEALTH SYSTEMS: MEDICAL COMPUTING, HEALTH CARE INDUSTRY, CHINS, AND OTHER MEDIATED HEALTH INFORMATION SYSTEMS Recent Developments in Medical Computing The rise of the Internet for health communication is just one subcomponent of ongoing developments in medical computing and information

The Internet and Health Communication

13

Health Subcategories, and Number of Sites Within Each, Provided by Invisible Web (http://www. invisibleweb.com) and Yahoo! (http://www.yahoo.com), as of March 2 0 0 0 InvisibleWeb

Health

Alternative medicine 17 Children's health 12 Conferences 2 Diet and nutrition 23 Discussions 48 Diseases 90 Drugs 12 Find a doctor 48 Find a magazine or journal 1 Find answers to health questions 14 Fitness and exercise 8 Health care 5 Laws and regulations 7

Yahoo! Health Alternative medicine 564 Chats and forums 4 0 Children's health 150 Companies® Conferences 19 Death and dying® Dentistry® Disabilities® Diseases and conditions 7,265 Education 4 0 Emergency services 233 Employment 108 Environmental health 194 First aid 16 Fitness 161 General health 73 Health administration 63 Health care 3 6 0 Health sciences 2 6 Hospitals and medical centers 38 Institutes 33 Long-term care 114

Categories

Medical news 16 Medicine 174 Men's health 2 Mental health 3 Nursing 1 Parenting 13 Patient information 3 Publications 19 Reproductive health 3 Research 53 Sexual health 2 Statistics and demographics 56 Women's health 8

Categories Medicine 4,871 Men's health 2 6 Mental health 641 Midwifery 5 9 News and media 196 Nursing 424 Nutrition 2 0 6 Organizations 21 Pharmacy 1,054 Procedures and therapies 2 7 8 Public health and safety 179 Reference 92 Reproductive health 651 Senior health 78 Sexuality® Traditional medicine 178 Travel health and medicine 15 Web directories 4 7 Weight issues 75 Women's health 151 Workplace 69

@: These are pointers to categories that are cross-listed in other categories, pointing to more subcategories within them. Thus, this first-level listing underestimates the total number of Yahoo! sites that could be considered to be health related.

14

EXPERIENCES AND EXPECTATIONS

systems, large-scale changes in the health care industry (such as managed care/health maintenance organizations, and integrated delivery systems), and new kinds of health information users (such as medical suppliers, pharmaceutical companies, and health information system service providers; (Armoni, 1999; Eder, 2000; Committee, 1997, chap. 1; U.S. Congress, 1993, chap. 3; Winkler & Silberg, 1998). These new information and computing technologies are being considered, and implemented, to improve a wide range of health processes and outcomes, such as clinical information flow and care coordination among participants, ability to deliver care at the appropriate location of the client and provider, comprehensive capture and storage of clinical information, resource allocation and management, clinical management, patient management, clinical teaching and research, access by general practitioners, and productivity of clinicians and providers. Recent developments in medicine computing include online collections of data from controlled experiments, integration into the health curriculum, health history collection and diagnosis, Web-based clinical systems that access patient data from a variety of sources and display it in a variety of formats (what McDonald et al., 1998, call "canopy computing"), global monitoring of the epidemiology of various diseases (Flahault, Dias-Ferrao, & Lavanchy, 1998), and so on. These systems may involve the Internet for access and transmission structure, but the heart of their significance lies in connecting dispersed systems and databases (Forslund & Kilman, 2000). For example, the Australian "Doctor's Desktop" plan suggests bringing together four major application groupings: practice administration, practice scheduling (including capabilities to support preventative medicine), pharmaceutical services, and clinical services (More & Clarke, 1999), while other Webbased systems support the diffusion of practice guidelines (Anderson, Casebeer, Kristofoco, Carillo, & Smith, 2000). Coiera (1997) provides a comprehensive review and analysis of medical informatics, medical information systems, medical computing, the Internet, and telemedicine; Bashshur, Sanders, and Shannon (1997) provide a similarly broad coverage of developments in telemedicine systems; and Kissinger and Borchardt (1996) describe and analyze integrated health systems. Armoni (1999) and Eder (2000) consider other developments in health care information systems.

The Internet and Health Communication

15

A related development is the use of networked retrieval systems to diffuse medical information among doctors, clinicians, and researchers (Detmer & Shortliffe, 1997). However, in spite of the comprehensive National Library of Medicine's MEDLINE system, which provides access to 30 years of medical literature and 9 million records, many doctors are unaware of recent relevant information or have limited time and expertise to use such systems. More accessible and widely used Web-based online health information databases and search tools include: (a) Internet Grateful Med (http://www.igm.nlm.nih.gov), the free Web-based interface to the MEDLINE database, which has specialized features such as the ability to convert common query terms into specific NLM medical thesaurus terms, and general and detailed displays of results; (b) PubMed (http:// www.ncbi.nlm.nih.gov/pubmed), which provides a simpler interface to MEDLINE, though it does offer a "see Related Articles" (based on similarity of content) link, and a special specialized filter for clinical queries; and (c) the Medical World Search (http://www.mwsearch.com), which allows searching of thousands of medical sites using the NLM thesaurus (Rampil, 1998). The Johns Hopkins Center for Communication Programs offers online searchable databases on research, project, poster images, and full text publications (http://www.jhuccp.org). Detmer and Shortliffe (1997) describe several systems that have been designed using information science principles and techniques to improve the retrieval, credibility, use, and diagnostic decision support of medical information, including their MedWeaver system, designed as an interface that integrates a variety of specialized databases. One of the most sophisticated health-oriented search approaches is the MARVIN project (Baujard, Baujard, Aurel, Boyer, & Appel, 1998), which was first applied by the Health on the Net Foundation, to facilitate better access to valid medical information, in its MedHunt search engine, which uses a vector-based approach, translates among eight languages, and has both high recall and high precision. One Internet-based association, the Global Health Net (LaPorte, 1997; http://www.pitt.edu/HOME/GHNet/GHNet/html) integrates "health and telecommunication professionals [from academia, government, industry, and international agencies] who aim to develop an infrastructure to network everyone around the globe involved in public health and prevention efforts." Its underlying principle is that prevention investments and activities are far more cost-effective than medical treatment efforts.

16

EXPERIENCES AND EXPECTATIONS

The Health Care Industry Significant recent changes in the health care industry include pressure to reduce cost, penetration of managed care, rapid consolidation, intense competition, need to demonstrate improved quality, increased government influence, increasing demands from large employer groups and benefit consultants, differentiation of products, complexity of managing information as people integrate across services, and the ongoing need to gain a competitive advantage through growth (Ernst & Young, 1997). With respect to technology, managed care is intensifying information requirements: information must be used from a variety of sources, information must be delivered at the point of service, the health care model is shifting from an "illness" model to a "wellness" model, and decision making is being moved into real time rather than retrospective. Relevant legislation includes the 1996 Health Insurance Portability and Accountability Act, which provides standards for electronic transmission of health care information, the Fair Health Information Act of 1995-1996, and the Telecommunications Act of 1996 and its associated Universal Service Fund designed to provide some subsidies for public Internet connections.

Community Health Information Networks (CHINs) DEFINITION OF CHINs Community health information networks (CHINs) are a transaction system, a data repository, and an organization, as well as a social movement that began in the early 1990s and spread across the nation (Rubin & Aukema, 1997). The functions of CHINs include clinically oriented information systems development, shifting of economic risk, expanding managed care, and permitting of greater governmental involvement and monitoring (Dowling, 1997). Nearly 90% of CHINs have a single provider or provider associations as members, 85% have physicians, 55% have payer or medical service organizations as participants, while the government is a participant in less than 20% and purchasers in only 13%. Computer networking has fundamentally changed the way many cities have organized their drug abuse treatment systems (Moberg et a l , 1997). Where drug abuse treatment facilities once competed for patients and seldom shared data about patients' prior treatment, now their efforts are coordinated by a central intake facility. Since patient data are readily available through a com-

The Internet and Health Communication

17

puter network, patients can be quickly referred to the most appropriate treatments. Information networks have the potential to impact the community-wide approach to many other health issues. Pemble (1997) basically argues that CHINs are integration engines for the flow of health care information within a given community. She sketches out the endemic issues of a complex health care information system that includes topics of ownership, integrity, availability, utility, source control, errors/omissions, and leakage. Overcoming technical incompatibilities and lack of collaboration among systems and organizations will be a major challenge (Bysinger, 1997). As an example, the Wisconsin Health Information Network (WHIN; http://www.whin.net), established in 1992 as a joint venture between Ameritech and Aurora Healthcare of Milwaukee, includes a wide range of hospitals, health insurers, ancillary service providers, physicians, ambulance services, nursing homes, community clinics, and home health agencies. WHIN provides, from a single workstation, access to patient demographics and search, medical record abstract display, medication profiles, utilization review, outpatient scheduling, patient eligibility checking, insurance claim management, results reporting, patient referral, report transmission, e-mail, and so on. Imminent services include: transmittingX rays to the clinician, library search, transmitting prescriptions to the pharmacy, and electronic signatures. An external evaluation found that there was a faster response to requests, a decrease in the length of patient stays, and an increase in the recovery of previously lost charges (due to procedures missed in the billing process or because late submissions missed program filing deadlines; Pemble, 1997). Another example CHIN is the Community Health and Information Technology Alliance (http:// www.chita.org), a member-driven alliance of health care and technology organizations (including Microsoft, the Washington State Health Department, GTE Network Services, and University of Washington Medical Centre), operating as a not-for-profit foundation. It plans to implement a secure Intranet (CARENet) to link its care centers and primary care providers to optimize care coordination. Other integrated health services include the Massachusetts Health Data Consortium (http://www.mahealthdata.org), working with a range of technology partners including IBM, Hewlett Packard, CSC, and EDS; and the Minnesota Health Data Institute. Networks like CDC WONDER, INPHO, and SCARCNET have enabled public health workers to coordinate their responses to many problems, ranging from natural disasters to teenage smoking (Brownstein et al., 1997).

18

EXPERIENCES AND EXPECTATIONS SOCIAL DIMENSION OF CHINS

Moore (1997) and Puskin, Mintzer, and Wasem (1997) recommend that the design of CHINs arise from the information needs and practice styles of clinical care providers, and not from technical capabilities of computer networks. They envision a telecommunications village within which telecommunication health services (health promotion community wellness programs), as well as telemedicine practices, improve the lives of people everywhere, including remote rural areas. The ultimate value of a CHIN, Kahn (1997) argues, is less in the management of health care of an individual than in its ability to exploit databases for analysis and systemic improvement. Based on her experiences as a clinical nurse moderator for a special computer network service, the ComputerLink, Moore (1997) discusses challenges such as relationship development and the establishment of group norms (moderator leadership style, managing confidentiality, acceptable topics, response to dangerous or inaccurate advice, acceptable level of typos, and messages warning of decreased participation near the end of projects). However, recommendations for successful CHINs are sometimes impossible to put into practice, due to problems such as the inability of a CHIN to know even what is care-/cost-effective, let alone what provider does it best (Rubin & Aukema, 1997). Kahn (1997) provides a case study of the consolidation of entities including the Barnes-Jewish and Christian Health Service located in Missouri and southern Illinois. The mission of this CHIN was to acquire the capacity to measure the current health and changes in the health of the people and communities it serves as well as to acquire the capacity to relate these changes to a particular intervention of the CHIN. He specifies the various problems that were encountered; most notable was the difficulty the organizations faced in making various data sets compatible.

Other Mediated Health Information Systems Health information systems do not necessarily have to be implemented on a large community-based computer network. Boberg et al. (1997) and Hawkins et al. (1997) describe CHESS (the Comprehensive Health Enhancement Support System), initially a DOS-based system but now available at chess.chsra.wisc.edu, that allows a variety of modules to be networked to the advantage of individual users. Among the modules are "question and answer" sections to address a range of issues people may

The Internet and Health Communication

19

have in dealing with a health crisis, a tutorial on getting help and support, comprehensive referral directories, a variety of personal stories (written by professional journalists), a comprehensive dictionary, and finally, a lifestyle assessment inventory. There is also a module that helps guide people through difficult decision making. Another section provides an action plan to deal with health issues. It also includes an "ask an expert" section, allowing questions to be asked anonymously of physicians. Finally, there are discussion groups allowing participation at any time. Particular modules include topics of sexual assault, breast cancer, and an AIDS module. Evaluations show that CHESS is widely accepted, heavily used, and can have a significant positive impact on health service utilization and lifestyle. Alemi and Stephens (1997) advocate using the telephone as a comfortable, convenient way to receive health care information and alter behavior. In analyzing a phone-talk bulletin board, they found that the system was used by an undereducated, poor, and drug-using group of patients. This, of course, is a population that is extremely hard to reach by traditional or computer-based means. They also positively assess the prospects for remote management of patients via the telephone.

MAJOR COMMUNICATION ISSUES Health care requires accurate, accessible information (education, self-care, treatment, medical office management, cost and quality of care management) and timely and effective communication (patients with health care providers, physicians with other health care providers and technicians and insurers, patients with patients and significant others, payers with providers, manufacturers and distributors with providers, and health promoters and campaign designers with at-risk or otherwise salient populations; Goldman-Sachs, 1999). The following sections summarize some of the major issues, and describe current usage, with respect to these flows of information and communication.

Patient-Physician Communication Online information could have potentially substantial benefits in reduced patient anxiety and decreased time and cost in obtaining health information, in better evaluations of potential doctors, and in better-informed patients who can then engage in deeper discussions with their physicians and be more likely to follow recommended treatments. Online

20

EXPERIENCES AND EXPECTATIONS

communication may improve interaction among other health care providers, increase interaction and intimacy between doctors and patients, enable the discussion of sensitive topics, provide easy access to quick opinions, avoid telephone tag, improve patient compliance with treatments, and increase patient health knowledge (Hodge, Gostin, & Jacobson, 1999; Lemaire, 2000; Spielberg, 1998; Turner, 2000; Weinberg et a l , 2000). Certainly the current status of patient-physician communication has room to improve. Harris and Associates (1997) reported a large-scale survey of patients (n = 1,008) and physicians (n = 230), exploring the principle that open and honest physician-patient communication has a "tremendous impact on the quality of health care" (p. 1). Ninety-three percent of the physicians believed that many serious problems could be avoided with more discussion, yet 6 1 % felt that they didn't receive adequate communication training in their medical schools. Patients say that physicians don't listen well and provide difficult-to-understand explanations. Consequences of these problems include patients' avoiding seeing their doctors. Thirty-one percent of patients recently postponed seeking medical advice; 26% of patients were reluctant to talk to or seek advice from doctors, due to embarrassment (25%), feeling they would be wasting the doctor's time (8%), feeling their symptoms weren't important enough (11%), fearing bad news (8%), and being scared of treatment options (7%). On the other hand, physicians say that patients withhold important information, don't follow medical directions, and so on. Forty-eight percent of the physicians thought that not having enough time to go to a doctor's office was a "very" or "somewhat" major reason that patients don't ask questions or talk about health problems, while 67% said that not having enough time to spend with patients was a "very" or "somewhat" serious problem in their practices. Indeed, reasons that patients turned to other sources for health care information included easier accessibility/quicker (52%), not wanting to "waste" the doctor's time (14%), and feeling more comfortable (11%). These differed according to age and gender. A 1995 survey of users of AOL's Better Health & Medical Network reported that 6% said they were able to avoid a visit to the emergency room because of information from the site, and 26% said it saved them from at least one doctor visit (Lowes, 1997). The Internet is beginning to challenge, redefine, and even increase these communication problems. Davis and Miller (1999) reported that more and more patients are showing up with Internet printouts, becoming more informed about a particular topic than their doctors. While some doctors re-

The Internet and Health Communication

21

sist patients who come in with Internet research (Kahin & Keller, 1995), which may be incomplete, misinterpreted, or outright quackery, other physicians know that well-informed patients are better patients. Thus recommendations include providing good review and summary sites for patients, providing them accurate search terms, requesting copies of materials be sent to them before the scheduled appointment, and indicating how much extra charge for consultation will be involved (Lowes, 1997). In a 1998 survey of more than 1,000 California residents, those with Internet access were less likely than were those without access to report consulting their physicians and health care providers for information about medical conditions (Pennbridge, Moya, & Rodrigues, 1999). While more than 75% of all respondents ranked physicians and health care providers as the most trusted sources, those with Internet access ranked the Internet as the third most trusted source (those without access ranked it fifth). Intriguingly, regardless of access, the respondents also ranked the Internet as the fourth (out of eight) most distrusted source, behind newspaper/magazine/journal articles, family member/friend, and telephone advice line. However, because all physician-patient communication must be guaranteed to be confidential, and online communication is not considered secure, most doctors (as well as therapists, lawyers, and financial advisors) will usually not communicate any possibly confidential information with clients via e-mail, cordless or cellular phones, or fax. On the other hand, doctors who do facilitate such interaction are likely to take patients away from those who do not, over and above the business advantages of interconnecting one's practice through the Internet with suppliers, insurance companies, and banks (Eder & Darter, 1998). And online clinics are providing information and diagnosis to thousands of medical consumers (Stroh, 1999). They find that online patients are less passive, although in some cases the virtual waiting room has a backlog of online patients lasting 2 or more hours! Such clinics attempt to avoid liability issues by warning that the online doctors will not "enter into a physician-patient relationship" or "engage in any conduct that involves the practice of medicine" (p. 2), facilitated by enforced anonymity. However, it is possible that physicians may be held legally responsible for not becoming aware of relevant medical information that would be easily obtained through the Internet or other medical information systems (Hodge et a l , 1999). Borowitz and Wyatt (1998) report on the development, content, and effort associated with providing e-mail physician-patient consultation. They reviewed nearly 3 years of e-mail consultation requests ( N = 1,239) sent to

22

EXPERIENCES AND EXPECTATIONS

one division at the University of Virginia Children's Medical Center. Nearly 40 were received each month, with 8 1 % sent by parents, relatives, or guardians; 10% by physicians; and 9% by other health care professionals. Nine percent of the requests were seeking a second opinion, implying one interesting use of online resources—avoiding questioning doctors face to face—while 22% requested general information. Reading and responding took about 4 minutes per message. The authors conclude that "e-mail provides a means for parents, guardians, and health care professionals to obtain patient and disease-specific information from selected medical consultants in a timely manner" (p. 1321). Hodge et al. (1999) review other studies of physician-patient e-mail interaction. Considering the perspective of the physician, Eysenbach and Diepgen (1998) surveyed 58 physicians and Web masters (found on Web sites for dermatological information retrieved through search engines) concerning their attitudes toward unsolicited patient e-mail requests for medical advice, stimulated by a fictitious acute skin problem described in an e-mail sent out by the researchers. Half of the physicians responded to the request. Although 3 1 % would not provide advice without seeing the patient's skin, 93% recommended that the person see a physician, and 5 9 % actually referred to the correct "diagnosis." The study concluded that approaches to handling unsolicited e-mail, as well as the kinds of responses to medical requests, vary widely, representing "a striking lack of consensus . . . on the theoretical and practical handling of unsolicited patient e-mail messages and their judgment of this topic" (p. 1334). Thus standards are needed to protect both physicians and patients. The American Medical Informatics Association (AMIA) has recently developed guidelines for patient-physician e-mail that emphasize privacy concerns (Kane & Sands, 1998). Spielberg (1998) notes that most physicians use e-mail for reference and physician-to-physician communication rather than communication with patients. However, early physician e-mail users note a variety of advantages, such as a better medical record (indeed, to avoid liability, such messages must be stored as part of the patient's record, rather than treated as casual, ephemeral conversation), more considered response, reduction of media costs, increased service provision through Web site e-mail addresses, the use of online forms for patient information collection, and hypertext links to online resources. Nevertheless, there are several legal and ethical issues still under debate. Encryption will increase a sense of confidentiality but can never

The Internet and Health Communication

23

guarantee it, which is still a physician's responsibility, subject to legal statues and regulations, including the Electronic Communications Privacy Act, which protects against eavesdropping on most phone and digital communications. Even encrypted messages, if sent to a patient at the employer's e-mail address, are legally susceptible to monitoring and search by the employer. E-mail privacy against governmental searches is not protected unless both the sender and receiver can show they had exclusive access to their messages, which is typically unlikely. A related issue is the need for informed consent associated with e-mail communication, as supported by the American Medical Informatics Association (Spielberg, 1998, p. 1356). And the actual identity of either patient or physician is currently difficult to guarantee via e-mail. As with standard medical practice, practice norms will develop around e-mail communication, such as response time and completeness of response. Two related thorny issues are jurisdiction in cross-state e-mail communication, and whether medical software and hardware are subject to FDA approval. Links to other Web sites maybe problematic, as they may be interpreted as endorsement or as antitrust practices among a limited set of providers. Spielberg's analysis identifies a range of new expectations, practice standards, and potential liabilities that are associated with new communication technologies for health communication. Good practice will involve awareness that e-mail communication may become part of the patient's medical record, discussion with the patient about implications of electronic communication, and mechanisms to ensure confidentiality.

Physician-Physician Communication There is great potential for improving the communication and service provision among the 650,000 physicians, 2 million nurses, tens of thousands of medical researchers, and 1 million administrative health care professionals in the United States (Goldman-Sachs, 1999). The Healtheon Corporation survey noted above reported that 85% of the 10,000 surveyed physicians were using the Internet (Stroh, 1999). Eder and Darter's (1998) survey of 86 members of a northeast chapter of the ΑΜΑ found that although half were using the Internet to obtain professional medical information, only 14% indicated they would recommend it as a medical information resource for patients. About one quarter of the respondents said they had, or were planning to have within the year,

24

EXPERIENCES AND EXPECTATIONS

a Web site. A study of more than 500 physicians with and without Internet access found that while there were no differences in access or usage between obstetricians and family physicians, or even between urban and rural physicians, urban physicians participated more frequently in user groups and exchanged information with other medical professionals (Eitel, Yankowitz, & Ely, 1998). Physicians with Internet access, and younger physicians, were more likely to consider the Internet a valuable clinical resource. A recent survey by the ΑΜΑ of over 1,000 office-based U.S. physicians found that usage of the WWW (World Wide Web) by doctors rose from 20% in 1997 to 37% in 1999, with 58% of physicians having a computer but not currently having WWW access planning on doing so within the next half year (ΑΜΑ, 1999). As of October 1999, up to 80% of all U.S. physicians were online (Credit Suisse First Boston, 1999), though Goldman-Sachs (1999) suggests the use of the Internet and the Web is more likely 30% to 40%, with the rest being regular e-mail. As a Goldman-Sachs (1999) report concludes, the "paper-intensive health care industry is ripe for electronic data interchange through connectivity vendors" (p. 1). Many sites, such as Australia's http://www. medeserv.com.au and http://www.hcn.net.au, provide extensive online services and resources for health professionals, such as evidence-based research databases, tutorial, textbooks, news, Web site indexes, pharmaceutical databases, full-text, world health sites, discussion forums, and so on. Many companies are beginning to offer full-service Web-based support for enrollment, eligibility, and claims processes, such as MedOffice. Medscape.com, Claimsnet.com, and Healtheon/WebMd (for coverage of such e-health issues, see http://www.wrhco.com/research/coverage/ ehealth/index.html). The American Medical Association has developed a 30-city physician continuing medical education course on advanced Internet applications for health care, an extension of its more basic Physicians Accessing Internet program (ΑΜΑ-Internet Health Road Show, 1999). One study of a physician/researcher discussion list (the anesthesiology discussion group; Worth & Patrick, 1997), based on a content analysis of 635 consecutive messages in 1 month and surveys from 28 participants, found that the practitioners rated the overall quality of responses to clinical questions at 7.1 (from 1, poor, to 10, superb), and their likelihood of using the list again for such questions at 8.7. Practitioners in solo or small group practices were more likely to use the discussion list for consultation and rated the quality of information higher than those in large practices.

The Internet and Health Communication

25

A survey of 83 academic and 43 industry drug information centers found that their professionals used the Internet daily, though as of the end of 1997, fewer than 25% of the centers had a homepage (Johnson & Wordell, 1998). Forty percent of industry, and 24% of academic, centers accepted inquiries about drug information via e-mail (though these were less than 3% of all their inquiries), while more than 75% of both types of centers used email to communicate with other drug information professionals. More than 65% of the centers use the WWW as part of their practice. Other than inquiries about current medical news (reported by around 80% of the centers), the most frequent types of information sought by drug information specialists on the Web were, for academics, drug availability, product information, regulatory, and therapeutic uses, and, for industry, regulatory, therapeutic, news groups, and recreation. Academic sites provided links to other WWW sites, a newsletter, and formulary, while industry sites also provided package inserts and product advertisements. Both academic (82%) and industry (69%) representatives agreed that the FDA should regulate industry-based drug information Internet sites.

Patient-Patient Support Communication One of the most discussed uses of the Internet for patient communication is online support groups. A recent study by Walther and Boyd (in press) identified four primary dimensions of reasons for, and benefits from, online support groups: The social distance provided greater expertise, management of potential stigma, and candor; anonymity helped protect their personal identity concerning sensitive topics; characteristics of the medium allowed users to manage their interactions with respect to expressiveness, stigma, and obligations; and access was available any time and any place. As of 1998, there were over 50 anesthesia-related discussion groups alone (http://www.eur.nl/cgi-bin/wgt4.pl; Rampil, 1998)! Goggins et al. (1998) summarize an evaluation of the Johns Hopkins Pancreatic Cancer Web site (http://www.pathology.jhu.edu/pancreas). They report very large numbers of accesses to the site, and messages posted to the discussion board, far exceeding what might be expected from the number of patients diagnosed with pancreatic cancer annually, indicating considerable unmet medical information and care needs. More than three quarters of the discussion board posters were female, and nearly 90% were relatives of patients, indicating a powerful emotional and cognitive support resource. In

26

EXPERIENCES AND EXPECTATIONS

February 1998,14.5% of the messages requested information, 12.8% requested or acknowledged support, 34.1% provided information, and 33.6% provided emotional support. The site also generated a number of referrals to the Johns Hopkins hospital. The authors propose that sites dedicated to diseases for which the providers have specialist expertise, such as from university health centers, help guarantee quality information and stimulate usage. Brennan and Fink (1997) review research showing that computer networks can help to provide "key social support necessary to increase individuals engagement in health promoting lifestyles" (p. 180). They studied one week's use of ComputerLink, a system providing information, decision-making utilities, and communication and online support for persons with Alzheimer's disease. Participants posted messages to share information, express their feelings, and offer support. In particular, messages provided evidence of social support in the form of affect (liking and admiration), affirmation (acknowledging appropriateness), advice, and information. Social support is positively associated in general with overall health lifestyle patterns (p. 167). Scheerhorn (1997) reports on the HIGHnet messaging system for hemophiliacs. Hemophiliacs, like groups of people with other medical conditions, have particular needs, categorized by good medicine, good physician, good information, good education, peer support, and economic support. Their analysis of 2,259 messages over 18 weeks among 58 members indicated that training was a potent influence on system usage. Messages concerned six themes: education, outreach, advocacy, improved psychological and physical health, business, and cost savings. Smyth, Feinstein, and Kacerek (1997) report on a Cleveland-based computer mediated Alzheimer's support group, noting factors that influence adoption and diffusion, including issues of perceived social presence of mediated interactions. 7

Preece and Ghozati (Chapter 11, this volume; see also Preece, 1998) report on their analysis of 100 textual communities, indicating that empathy is present to some extent in most Internet communities. Preece (2000) expands this concern with online support communities to describe comprehensively how to design "thriving" online communities (emphasizing usability, sociability, and purpose). Note that online support community members apparently do not have to be active "posters" to gain benefits. Nonnecke and Preece (1999) found, through a rigorous analysis of health and software support distribution lists (DLs) and other very large DLs, that

The Internet and Health Communication

27

lurkers (subscribed members who posted no messages over a 12-week period) ranged from 0 to 99% on 77 health support groups (with a mean of 45.5% of the members lurking), compared to 50% to 98% (mean, 82% of the members) of nonhealth groups. Lurking was not associated with number of DL subscribers, while lurking was negatively correlated with number of messages posted per day for DLs with more than 500 members, but essentially uncorrected with smaller DLs, indicating that lurkers tend to avoid high-traffic sites. But even when actual people are on the "other end" and participate, such as in an online support group, there's no guarantee the person has the illness they are discussing or even is the person he or she claims to be, as some people use support groups to play out "sick roles" and exhibit "factitious disorders" (Stephenson, 1998). Culver, Gerr, and Frumkin (1997) analyzed 1,658 consecutive messages from 313 authors during 5 months in 1994 from an online discussion group concerning painful hand and arm conditions (http://www.sorehand@ ucsfvm.ucsf.edu). Seventy-two percent of the messages were posted by an affected person, and onlyO.3% bya physician, with 67.6% of the messages providing information and 21.4% requesting information. Of the 55.9% messages addressing a medical topic, 79% provided medical information, with 89.3% of those by users without professional medical training! Nearly one third of this information was coded as being inconsistent with generally agreed medical practice. Sources of evidence were as follows, for nonprofessionals and then professionals: personal experience (61% vs. 13.5%), no source (29.8%, 67.6%), and published source (9.2%, 18.9%). There was very little challenging of advice, information, or conclusions drawn from experiences, though 10% of the messages expressed frustration or complaints about medical care. Patterson, Shaw, and Masys (1997) warn that "online self-help groups are a fragile meritocracy of written fact and opinion, in which each reader must decide what is true and what is merely unsubstantiated belief" (p. 238). Culver et al. (1997) conclude that medical discussion groups are a mixture of "snake oil" and "self help"; "appropriate diagnosis and treatment may elude participants in such electronic bulletin board discussions, on a scale that is unprecedented" (p. 470).

Interactive Media and Communication Campaigns The aspect of communication and information technology with the greatest implications for health promotion is interactivity, related both to

28

EXPERIENCES AND EXPECTATIONS

interacting with a system or data and to interacting with other people (Brennan & Fink, 1997; Rimal & Flora, 1997; Street & Rimal, 1997). Street and Rimal (1997) argue that the defining dimensions of interactivity are user control (the extent to which a user can modify the content or form of the mediation) and responsiveness (or extent to which a response takes into account a prior action). Interactive media can improve health promotion because of increased learning, information seeking, information processing, and individualized knowledge by current or potential patients or interested parties. Computer networks increase the potential of interactive systems by making available a wide variety of resources, participants, and applications through one system. Street and Rimal's review of past research on interactive health systems reveals considerable confounding among treatments, media features, and content, so that it is hard to draw definitive conclusions as to benefits of interactivity. They integrate this research with the concept of interactivity by proposing a three-stage model of health promotion using interactive technology: implementation and use (influenced by institutional, technological, and user factors), the user-media-message interaction, and health outcomes (intermediate attitude and cognitive outcomes, and longer-term health and prevention outcomes) influenced by psychosocial factors. Rimal and Flora (1997) consider the distinctions among media features that may be best associated with specific health domain attributes (such as addictiveness, skills, and heredity) and individual attributes (such as demographics, psychographics, and sociocultural aspects). These distinctions include (a) multimodality, (b) networkability, (c) temporal flexibility (asynchronicity), (d) segmentation capability, (e) interactivity, (f) sensory vividness, (g) modifiability, (h) availability, (i) cost, and (j) ease of use. For example, temporal flexibility, modifiability, and interactivity foster cognitive rehearsal of the desired behavior, increasing the likelihood that the behavior will be applied in actual situations. Or, individuals with low involvement may well benefit from less-interactive media, such as videos, than interactive Web pages. Segmentability of the content, such as through using Web "cookies" (short lines of data stored in a user's browser directory, depending on the user's acceptance of them, that indicate the prior use of a specific Web page or service) to tailor Web content to prior searching behavior, would increase the matching of information to one's readiness for change. Skinner and Kreuter (1997) summarize theoretical foundations for identifying the kind of health behaviors (asymptomatic screening, lifestyle

The Internet and Health Communication

29

modifications, cessation of addictive behaviors, medical regimen compliance, precaution adoption) that would respond to particular interactive media interventions. Reviewing the health belief model, efficacy theory, attribution theory, the theory of reasoned action, and the transtheoretical model, they show what specific behaviors and interventions would be motivated by each theory, using different kinds of new media to enable, reinforce, and predispose. For example, an interactive program can identify salient health beliefs and then provide simulations or information geared to those assumptions. Alternatively, a system that has diagnosed the efficacy level of a patient can help focus the person on internal or external causes for a particular unhealthy behavior. The theory of reasoned behavior would propose that individuals be connected to influential or knowledgeable others, say in a support group, to increase the persuasability of suggested health behaviors. Many communication campaigns are beginning to integrate interactive Web sites into overall communication strategies, such as the REACT Web site for reducing delay in responding to heart attacks (Finnegan et al., Chapter 7, this volume; Simons-Morton et al., 1998). Indeed, information systems can be used to help develop and tailor health messages and campaigns based on the needs, interests, and concerns of specific groups or individuals (Kreuter, Farrell, Olevitch, Brennan, & Rimer, 2000). Other developments include interactive CD-ROMs for campaign planning and country/city analysis, such as the SCOPE CD by the Johns Hopkins Center for Communication Programs (Piotrow & Kincaid, 2001) and the CDCynergy CD by the Centers for Disease Control (Parvanta & Freimuth, 2000). Both of these combine comprehensive census, media usage, and epidemiology databases with simulations of stage-by-stage campaign development and implementation.

MAJOR INDUSTRY AND POLICY ISSUES Online Medicine Online medical advertising is growing rapidly, from $58 million in 1999 to a projected $265 million in 2002, still small compared to the total medical advertising and promotion spent in the United States of $8 billion in 1998 (Goldman-Sachs, 1999). Online pharmaceutical sites are generating considerable revenue. In May 1999 they took in nearly $1 billion for prescription drugs, $300 million for over-the-counter drugs, $434 million for

30

EXPERIENCES AND EXPECTATIONS

vitamins and supplements, and $700 million for personal care products. Users are still wary of commercial health Web sites, however, indicating they are more likely to go to nonprofit, hospital, patient advocacy, other, and pharmaceutical Web sites first (Goldman-Sachs, 1999). Online sales of medications have generated considerable controversy. Cross-border commerce via the Internet can escape regulatory evaluation, so some products are promoted for symptoms that do not correspond to their intended use; also, quality, manufacturing, purity, safety, efficacy, and enforcement cannot be guaranteed. The purchase of medical products via the Internet allows patients to avoid medical advice and treatment, including an understanding of drug interactions and side effects ("General Policy Issues," 1998). Online pharmacies per se are "just another channel of distribution" as long as a physician-patient relationship exists (Cohen, 1999). While it is not strictly illegal to do so, providing prescriptions to patients that a doctor has not physically examined is considered unethical by the ΑΜΑ, unacceptable by the National Association of Boards of Pharmacy, and likely to be fined by medical quality assurance commissions (Stolberg, 1999a, 1999b). A recent study, however, found that of 77 Web sites offering online purchase and direct delivery of Viagra, 52 required users to release the company from liability, only 42 required users to complete an online medical history questionnaire that could have provided information on whether Viagra would be safe and effective for the patient, and only 27 of those indicated that a doctor would review the information (Armstrong, Schwartz, & Asch, 1999). One of the problems is that jurisdiction is widely and partially scattered among federal regulators, FDA, FTC, and state agencies (Cohen, 1999; see also Chapter 17, this volume). However, the National Association of Boards of Pharmacy has begun to develop a process to certify legitimate Internet drugstores (Cohen, 1999). This association requires online pharmacies to be licensed in all the states to which they ship drugs; upon verification, they are allowed to display the Verified Internet Pharmacy Practice Sites (VIPPS) icon. Some services, such as Cyberdocs, avoid many of these problems by being insured for malpractice, having doctors provide advice only in their areas of expertise, and dealing only with patients from states where the doctors are licensed. The World Health Organization has proposed a range of recommendations for discussion, including review legislation of cross-border activities, establish informational Web sites, develop voluntary codes of conduct, consider cooperative licensing, report problem cases to WHO and Member States, and so on ("General Policy Issues," 1998). President Clinton's

The Internet and Health Communication

31

2001 federal budget will request authorization for the Federal Drug Administration to verify the quality of Internet-based pharmacies that fill online prescriptions, and to expand their investigation of Web sites (New York Times, 1999). The growth of this practice is not only fueled by the obvious e-commerce benefits, but also social and cultural trends, such as the rapid adoption of "lifestyle drugs" (such as impotence, hair loss, and diet pills), the attraction of anonymity when requesting certain kinds of medications, the ability to obtain drugs not approved in the United States from foreign sites, and online drug interaction analyses. As an indication of this anticipated usage, one survey found that almost half of the respondents (18% often, and 3 1 % most likely) indicated that they would use the Internet to order prescription refills and make doctor's appointments (Pennbridge et al., 1999). An obvious negative consequence is the ability of consumers to selfprescribe drugs and to obtain drugs that have not been approved by the FDA.

Credibility of Online Health Information As with some of the studies of online support groups, analysis of Web pages raises significant questions about the relevance, coverage, and legitimacy of much Internet health information. Hersh, Gorman, and Sacherek (1998) analyzed 629 pages retrieved from Metacrawler based on 50 questions selected from a database of clinical questions. They were judged to be "neither clinically applicable nor of high quality" (p. 1307); 89% were not even applicable to the search question, and most pages had no more than one "quality" measure (such as site affiliation). Rose, Bruce, and Maffulli (1998) identified the 25 most frequently used terms obtained from 100 orthopedic patients as best describing their medical condition, and entered them as search terms into five search engines. (As an aside, 2 0 % of the terms had been misspelled by patients, implying that regular users may not find relevant Internet results for a good proportion of their searches because of unfamiliarity with the appropriate terms!) Of the 5,947 retrieved pages, 20% contained patient information, 19% professional information, 9% commercial information, 2% sports news, 8% unavailable, and 4 1 % other (non-English pages, chat rooms, etc.). The primary result was that only 7% of the pages they retrieved (in this case about fever in young children) were actually relevant to the patients' conditions. Impicciatore, Pandolfini, Casella, and Bonati's (1997) study similarly found that rela-

32

EXPERIENCES AND EXPECTATIONS

tively few of the pages provided complete and accurate information. Biermann, Golladay, and Baker (1999) searched for Web sites concerning a specific form of cancer, Swing's sarcoma. Based on 400 of the 27,000 resulting sites using four search engines, only half actually contained information on this topic, and, of those, only 60% had peer-reviewed information. The authors noted that in addition to the time-consuming and cumbersome search process, users may end up with incorrect and harmful information. McClung, Murray, and Heitlinger (1998) arrived at a similar conclusion from their study of 60 articles from the first 300 hits on a search for information about childhood diarrhea. They reported that only 2 0 % conformed to the current diagnosis and treatment guidelines of the American Academy of Pediatrics; so 80% had inaccurate or out-of-date information. They concluded, "Patients must be warned about the voluminous misinformation available on medical subjects on the Net." Many have argued that professional medical organizations need to inform and educate not only patients but also doctors about reliable Internet health sites (Henson, 1999). Because of the consequentiality of health information, there are several initiatives to develop credibility and accuracy standards for online information. These include the Health On the Net Foundation's HON label (HON, 1998); the Australian Department of Health and Aged Care's site (http://www.healthinsite.gov.au, which emphasizes reliable and relevant health information, based on a contracted "information Partnership" between the site and accredited/evaluated organizations, which then receive the Healthlnsite Web page icon); the New York Online Access to Health (NOAH, http://www.noah. cuny.edu; Voge, 1998); Rees's consumer health information source book (Rees, 1998); the Health Summit Working Group (1998); the Argus Clearinghouse (http:// www.clearinghouse.net); the Johns Hopkins reproductive health and population Web page (http://jhuccp.org/netlinks); and other approaches (Jadad & Gagliardi, 1998). In June of 1999, the Federal Trade Commission initiated a program to fight false and deceptive health claims for products advertised or sold via the Internet, including applying existing legislation. The Ε-Health Ethics Summit, sponsored by the In- ternet Healthcare Coalition (including actors such as America Online, drkoop.com, Healthwise, Medscape, Healtheon, and WebMD.com), has recently proposed a voluntary code of conduct for Web sites' collection and use of personal information (Woody, 1999). Standards include informed consent, editorial control over accepting advertising content, and disclosure of all sponsorships or financial incentives (http.V/www.ihealthcoalition. org/community/

The Internet and Health Communication

33

draft.html). Another ethics code push comes from an organization called the Internet Healthcare Coalition (IHC). In addition to involvement from the American Medical Association, the IHC also represents the views of profit-making companies such as those affiliated with the HI-Ethics committee, as well as nonprofits, academic journals, advocacy groups, and other Web-based medical information services.

Privacy and Liability Issues of privacy, confidentiality, and security in both well-recognized and enterprise-specific medical information systems (Lindberg & Humphreys, 1998) are compounded by the interorganizational linkages characteristic of CHINs and networked health systems (Barrows & Clayton, 1997; Donaldson & Lohr, 1994). The three primary information security goals in health care include "prevention of unauthorized disclosure of information . . . prevention of unauthorized modification of information . . . and prevention of unauthorized or unintended withholding of information or resources" (Barrows & Clayton, 1997, p. 300), leading to 14 areas needing security policy development for computer health networks, ranging from user authentication, data protection, hard copy materials security, legal and liability issues, and system reliability and backups, to audit trails and informed consent. Pemble (1997) also emphasizes the crucial areas of security and authorization for data access. Other aspects of privacy and confidentiality noted in other sources include: authentication of and security profiles for individuals; validating individuals' access to systems and databases; levels of priority and associated access means; levels of security associated with different information, policies, and procedures for those who have the responsibility of protecting and maintaining the confidentiality of information they can access; ensuring integrity and protection of transferred patient-client information; applying privacy policies and measures to interdependencies among various databases; ensuring protection during transmission and conversion from manual to computer systems; access and integrity audits; validating and maintaining passwords; firewalls; access to patient information by third parties; disaster recovery; protection of remote access interfaces; encryption; electronic authentication; and so on. The U.S. Institute of Medicine and the National Academy Press published a comprehensive book on this issue (Committee, 1997). The Office of Technology Assessment, before it was dissolved, also released at least three related reports on this

34

EXPERIENCES AND EXPECTATIONS

topic (http://www.princeton.edu/~ota; U.S. Congress, 1993). Another set of issues that overlaps with privacy and confidentiality is "standards," such as unique and uniform client and service provider identifiers; data definitions of encounters, diagnoses, and treatments; and data and system security. The National Library of Medicine has two large funding programs in these areas, both emphasizing health data confidentiality and standards. One supports increased Internet access for education, research, clinical care, and administration by health professionals, while the other fosters the development and use of the National Information Infrastructure (Nil) for public health and health care research (Federal Telemedicine Directory, 1998). Schneider (1997) acknowledges that there are fears that partner notification programs, such as those used to fight AIDS and sexually transmitted diseases, could suppress participation in programs. However, he argues that a computer-mediated partner notification program that will respect confidentiality of patients and partners will actually increase participation. Slack (1997) also believes using computers properly can empower patients and give them greater control over their own destiny, enhancing their health. Unfortunately, although the author makes powerful assertions about the value of this approach, little supporting evidence is provided. Regardless of the likely value and security, most respondents have major concerns about the privacy and confidentiality of medical information, and about making their medical records available via the Internet (Pennbridge et al., 1999). Credibility and privacy of health information are directly related to legal issues (Hodge et al., 1999). Personally identifiable health information is now more available through intra- and interorganizational networks, providing benefits such as greater patient autonomy, improved treatment, faster diagnoses, reduction of adverse drug interactions and reactions, broader dissemination of medical research and practice, improvements in research, better monitoring of morbidity and mortality, and increased security and audit trails. However, such systems also generate legal challenges associated with privacy of identifiable health information, data reliability and quality, and tort-based liability (p. 1466). There is no comprehensive legal protection of personally identifiable health information; instead, there is a wide variety of existing federal and state privacy laws and several proposed federal privacy laws, such as Health Insurance Portability and Accountability Act and regulations by the Department of Health and

The Internet and Health Communication

35

Human Services. The DHHS has provided Congress a set of recommendations covering limitations on disclosure of health care information, required patient authorization for health information distribution, access to and correction of health information by patients, accountability for use and distribution of health information, and the priority of national and public interests over individual privacy interests (Hodge et al., 1999). The authors argue that these areas are interdependent, such that increased privacy improves reliability and quality of information, which in turn reduces the likelihood and need for tort-based liabilities. They provide seven recommendations for health information privacy legal reform.

Policy and Access In spite of the promise of computer-based and computer-mediated medical information and communication, it is still true that people with preventable health problems and those who have little or no health insurance are also those least likely to have access to the necessary technologies (Eng et al., 1998). A wide variety of barriers exist, such as cost, location, illiteracy, physical ability, and capacity. Public and governmental efforts are needed to reduce the gap between health information "haves" and "have nots." Enget al. discuss a number of these, from providing public and residential access, to diverse applications, improving access through research, increasing health and technology literacy, and integrating universal access into health planning, at the levels of private-sector health services, federal and state funded health programs, federal grants, charitable organizations, corporate marketing, and private and public long-term community investment. They argue that such activities have both moral and pragmatic motivations: Most health information is produced by public funding, so should be made public; improved public health benefits society at large; and improved access reduces health costs. Harris (1995) proposed "three clear indicators of progress toward our goals of improved health for all at affordable costs": increased access, improved quality, managed costs. She also described three primary characteristics of new media: interactivity, interconnectedness (due to digitization and infrastructure), and user-driven (both producers and consumers can customize information and systems). This leads to a 3 χ 3 matrix of improved health care goals by new media characteristics: "customized information breaks down educational barriers, universal access reduces financial barriers, participatory learning challenges motivations barriers;

36

EXPERIENCES AND EXPECTATIONS

simulation enhances education, distributed collaboration coordinates services, all decision makers are participants; and service options are linked to individual health status, multiple service providers share network costs, all participants are accountable" (p. 14). McGinnis, Deering, and Patrick (1995) argue that new media provide the potential for improving disease prevention and health promotion, including "online information resources, interactive multimedia educational materials and computer-assisted decision-support programs" (p. 127). However, they note two major challenges for health in America: health inequality, and prevention. There are vast disparities in health in the United States: Even though we spend more than other nations, costs are rising faster than inflation, and health levels are lower than in some other nations. Information-based activities can foster prevention activities, which influence nearly half of all deaths in the United States. However, most health costs are currently allocated to the treatment of "leading causes of death" instead of to the (preventable) "real causes" such as tobacco, diet, alcohol, firearms, sexual behavior, motor vehicles, and drug use. Healthy People 2000 (National Disease Prevention and Health Promotion Objectives, 1990) identifies and explains those activities that should have the greatest influence on improving health status, with 300 objectives in 21 priority areas. Central to such efforts is patient education and counseling, both information-based efforts. The U.S. Public Health Service intends to take advantage of the recent national initiative to wire schools, libraries, clinics, and hospitals to create a "seamless, interactive, information-based health support system" (p. 136). Perhaps the most accessible of these public health information services is Internet Grateful Med, noted above. Fisher (1995) argues that free market forces may not be enough to produce either the national information infrastructure or the health information products necessary to provide sufficient health prevention and treatment. The necessary components of the M I should be provided for every home as well as the current goal of classrooms, libraries, clinics, and hospitals. These include broadband capacity (e.g., for Xrays), open/switched/interactive networks (allowing content choice and individual production as well as consumption), upstream video (for distant diagnosis, video publishing), upstream interaction (for peer support via discussion groups), search tools (directories), geographic and economic universal access (from central business districts to rural communities), and content (medical records, knowledge, and health information). The National Library of Medi-

The Internet and Health Communication

37

cine is a good initial model for such a service, as it not only provides health information and online analysis but also helps ensure the credibility of that information. Similarly the New York Online Access to Health site, http://www.noah.cuny.edu, aims to provide "quality health information . . . that is accurate, timely, relevant, and unbiased" (Voge, 1998, p. 326) by means of many links to and from a wide variety of partners, as well as fulltext articles, in both English and Spanish. This site, designed by library professionals, has extensive and well-researched search mechanisms and approaches, including using a controlled list of subject terms. Further, differences in intellectual property issues between traditional print and new online sources could reduce access and fair use of electronic publications, data, or facts (Lindberg & Humphreys, 1998). Gorry, Harris, Silva, and Eaglin (1995) analyze the problem of boundaries and separations between providers and patients due to specialization, access, and institutions. Solutions involve interactions among strategies, people, structures, and technologies, across organizations. They discuss a case of providing multimedia, networked health services involving technology companies, a social services consulting firm, universities, social services agencies, and governmental agencies. Of course, major health agencies are moving toward providing many services through Web-based forms. Neu, Anderson, and Bikson (1999), for example, provide a case study of current and potential uses of the Internet to provide Medicare services. Vickery (1995) discusses how new media might improve the management of health services demand through selfcare, as the management of supply is insufficient. Four components of demand are considered: morbidity (illness, prevention, health habits, accidents, secondary prevention through screening), perceived need (knowledge of risks and benefits, assessing medical problem, severity of problem, ability to self-treat, self-efficacy—all influenced by knowledge, education, culture, social support, attitudes of health providers—these explain more than 40% of variance in decisions to seek care), patient preference (economic appropriateness, informed choice, risk aversion, care for terminal illnesses), and nonhealth motives (sick leave, disability, compensation benefits). Self-care interventions involving lifestyle, immunization, safety, self-management, informed choice self-help, life management, screening, and provider information are associated with a variety of medical benefits and cost savings. While mediated self-care would provide immediately available, personalized, and multimedia information, it would be less successful at providing emotional and social support. Thus a fully integrated

38

EXPERIENCES AND EXPECTATIONS

infrastructure would "support interactions among individuals, the demand management system, and the medical care system" (Vickery, 1995, p. 58). As another example, Zallen (1995) described the Harvard Community Health Plan's increasing emphasis on providing services that allow members to become more involved in their own health care, as part of the managed health care movement. Services include online initial diagnosis, health assessment and health risk appraisals, and online health support groups. Zallen emphasizes the shift from information being provided by the clinician to the patient, to the patient seeking information from comprehensive systems, including public "health booths" providing interactive video and online information.

SUMMARY Some of the relationships among some of the major issues discussed so far can be summarized by the overall model shown in Figure 1.1. Current practices and uses of the Internet for health communication and information, and the outcomes associated with them, are highly socially situated and not just outcomes of specific technology implementations. They are influenced by (or are interdependent with) Stakeholders, Systems and Infrastructure, available Applications, and Policy, Legislation, Regulation, and Economic forces. Applications include education, continuing medical education, testing, consultation, support, clinical trials, diagnoses, analyses, interaction, collaboration, information flows, online support, online communication campaigns, e-commerce, search and retrieval, records, and images. Stakeholders include patients, physicians, health care professionals, insurance companies, resources, social networks, administrative and technical support, customers, referrals, supplier and purchaser commercial organizations, and so on. Relevant aspects of Stakeholders include their characteristics, information-seeking behaviors and needs, social agenda, and personal needs. For example, recently Hoffman and Novak (1998) have argued that there is a digital divide among U.S. Internet users, with African Americans being dramatically less likely to use the Internet than other ethnic/racial groups. If this assertion is correct, it has dramatic implications for the quality of health care that this portion of the citizenry receives. Systems and Infrastructure includes the Web, National Information Initiative, Internet II, telemedicine, Community Health Information Net-

The Internet and Health Communication

ASSESSMENT, EVALUATION *

STAKEHOLDERS

SYSTEMS & INFRASTRUCTURE

APPLICATIONS

39

CURRENT, CHANGING PRACTICES & USES

OUTCOMES AND CONSEQUENCES AT VARIOUS LEVELS FOR VAROUS STAKEHOLDERS

POLICY, LEGISLATION, REGULATION, ECONOMICS

Figure 1.1. General Model of Components of Internet and Health Communication System

works, and characteristics of media sources and channels. Relevant Policy, Regulations, and Issues include privacy, authentication, intellectual property, access, quality, equity, efficacy, educational practices, personalization, barriers, and trust. General categories of Outcomes and Consequences include rewards and peer evaluation, health outcomes, institutions, knowledge communication, and economic structures and markets. These may vary highly across stakeholders and levels of analysis, such as the individual, family, community, organization, and industry. The outcomes, as well as the various applications, practices, and uses, may be more or less extensively Assessed or Evaluated. Examples include health care and access models, information, doctors, institutions, quality, efficacy, and credibility/seal of approval. Such assessments and evaluations may (should) influence redesign and change in the prior factors. Clearly, the intersection of the Internet and health communication is a socially significant, ethically and politically consequential, dynamic and innovative, intriguing and interesting, and methodologically challenging arena.

REFERENCES Alemi, R, & Stephens, R. (1997). Electronic communities of patients: Computer services through telephones. In P. F. Brennan, S. J. Schneider, & E. Tornquist

40

EXPERIENCES AND EXPECTATIONS

(Eds.), Information networks for community health (pp. 247-270). New York: Springer. American Medical Association [ΑΜΑ]. (1999). 2000ΑΜΑ study on physicians' use of the World Wide Weh Washington, DC: Author. ΑΜΑ-Internet health road show. (1999). [Online]. Available: http://www.amaassn. org/about/roadshow/index.htm Anderson, J., Casebeer, L., Kristofoco, R., Carillo, Α., & Smith, M. (2000). Using Web-enabled technology to promote the adoption of practice guidelines. In L. Eder (Ed.), Managing healthcare: Information systems with Web-enabled technologies (pp. 172-185). Hershey, PA: Idea Group Publishing. Armoni, A. (1999). Healthcare information systems: Challenges of the new millennium. Hershey, PA: Idea Group Publishing. Armstrong, K., Schwartz, J., & Asch, D. (1999, October 28). The direct sale of Sildenafil (Viagra) to consumers over the Internet. New England Journal of Medicine, 341(18), 1389-1393. Barrows, R., Jr., & Clayton, P. (1997). Privacy, confidentiality, and CHINS. In P. F. Brennan, S. J. Schneider, & E. Tomquist (Eds.), Information networks for community health (pp. 299-316). New York: Springer. Bashshur, J., Sanders, J., & Shannon, G. (Eds.). (1997). Telemedicine theory and practice. Springfield, IL: Charles C Thomas. Baujard, O., Baujard, V., Aurel, S., Boyer, C, & Appel, R. (1998). Trends in medical information retrieval on Internet. Computers in Biology and Medicine, 28, 589601. Biermann, J. S., Golladay, G., & Baker, J. (1999). Evaluation of cancer information on the Internet. Cancer, 86(3), 381-390. Böberg, Ε., Gustafson, D., Hawkins, R., Bricker, E., Pingree, S., McTavish, R, Wise, M., Owens, B., & Botta, R. (1997). CHESS: The Comprehensive Health Enhancement Support System. In P. F. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 171-188). New York: Springer. Borowitz, S., & Wyatt, J. (1998). The origin, content, and workload of e-mail consultations. Journal of theAmencan Medical Association, 280(15), 1321-1324. Brennan, P., & Fink, S. (1997). Health promotion, social support, and computer networks. In R. L. Street, Jr., W. Gold, & T. Manning (Eds.), Health promotion and interactive technology: Theoretical applications and future directions (pp. 157-169). Mahwah, NJ: Lawrence Erlbaum. Brownstein, J. N., Oberle, M., Miner, K., Alperin, M., Howze, E., & Patrick, K. (1997). CHINs: A public health perspective. In P. F. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 117-134). New York: Springer. Bysinger, B. (1997, January). Using the Internet to support provider-patient collaboration. Health Management Technology, pp. 44-45. Cohen, R. (1999, July 12). Internet drugs: Authorities unsure of jurisdictions as shady e-merchants flourish. Princeton/Metro Times, pp. 1-2. Coiera, E. (1997). Guide to medical informatics, the Internet and telemedicine. New York: Oxford University Press.

The Internet and Health Communication

41

Committee on Maintaining Privacy and Security in Health Care Applications of the National Information Infrastructure; Computer Science and Telecommunications Board Commission on Physical Sciences, Mathematics, and Applications, National Research Council. (1997). For the record: Protecting electronic health information. Washington, DC: National Academy Press. Credit Suisse First Boston. (1999, October). Ε-health II: Beyond the business plan. Report released at Intel's Internet Health Day II, New York. Culver, J., Gerr, F., & Frumkin, H. (1997). Medical information on the Internet: A study of an electronic bulletin board. Journal of General Internal Medicine, 22(8), 466-471. Davis, R., & Miller, L. (1999, July 14). Net empowering patients. USA Today, pp. 1A2A. Detmer, W., & Shortliffe, E. (1997). Using the Internet to improve knowledge diffusion in medicine. Communications of the ACM, 40(8), 101-108. The doctor is online: The prognosis is excellent for health Web sites. (1999, September 21). PC Magazine, p. 28. Donaldson, M., & Lohr, Κ. (Eds.). (1994). Health data in the information age: Use, disclosure and privacy. Washington, DC: National Academy Press. Double mastectomy being shown live on Internet. (1999, October 20). Trenton Times, p. A8. Dowling, A. (1997). CHINs—The current state. In P. F. Brennan, S. J. Schneider, & E. Tomquist (Eds.), Information networks for community health (pp. 15-40). New York: Springer. Eder, L. (2000). Managing healthcare: Information systems with Web-enabled technologies. Hershey, PA: Idea Group Publishing. Eder, L., & Darter, M. (1998). Physicians in cyberspace: A growing number of doctors are discovering the business benefits of a Web presence. Communications of the ACM, 41(3), 52-54. Eitel, D., Yankowitz, J., & Ely, J. (1998). Use of Internet technology by obstetricians and family physicians. Journal of the American Medical Association, 280(15), 13061307. Eng, Τ. R., Maxfield, Α., & Gustafson, D. (1998). Access to health information and support: A public highway or a private road? Journal of the American Medical Association, 250(15), 1371-1375. Ernst & Young. (1997). The role of the Internet in health care: Current state. Ernst & Young, along with 3Com Corporation, ConnectedHealth.Net, Sprint Healthcare Systems, Inc., Sun Microsystems, and Sybase, Inc. [Online]. Available: http://www.ey.com/health Eysenbach, G., & Diepgen, Τ. (1998). Responses to unsolicited patient e-mail requests for medical advice on the World Wide Web. Journal of the American Medical Association, 280(15), 1333-1346. Federal Telemedicine Directory. (1998). Washington, DC: Office for the Advancement of Telehealth. FIND/SVP Inc. [commercial online database]. (1997). The 1997 American Internet user survey: Top content choices, May 1997. [Online]. Available: http:// www.findsvp.com

42

EXPERIENCES AND EXPECTATIONS

Fisher, F. D. (1995). But will the new health media be forthcoming? In L. Harris (Ed.), Health and the new media: Technologies transforming personal and public health (pp. 209-227). Mahwah, NJ: Lawrence Erlbaum. Hahault, Α., Dias-Ferrao, V., & Lavanchy, D. (1998). FluNet as a tool for global monitoring of influenze on the Web. Journal of the Amencan Medical Association, 280(15), 1330-1332. Forslund, D., & Kliman, D. (2000). The impact of the global, extensible electronic health record. In L. Eder (Ed.), Managing healthcare: Information systems with Web-enabled technologies (pp. 3-13). Hershey, PA: Idea Group Publishing. General policy issues: Dangers of purchasing medical products through the Internet. (1998). WHO Drug Information, 12(1), 1-2. Goggins, M., et al. (1998). Use and benefits of a Web site for pancreatic cancer. Journal of the Amencan Medical Association, 280(15), 1309-1310. Goldman-Sachs. (1999, November 11). Health-e opportunities in ehealth? Technology: Internet/Ehealth. New York: Author. Gorry, G., Harris, L., Silva, J., & Eaglin, J. (1995). Health care as teamwork: The Internet collaboratory. In L. Harris (Ed.), Health and the new media: Technologies transforming personal and public health (pp. 87-105). Mahwah, NJ: Lawrence Erlbaum. Harris, L. (1995). Differences that make a difference. In L. Harris (Ed.), Health and the new media: Technologies transforming personal and public health (pp. 3-18). Mahwah, NJ: Lawrence Erlbaum. Harris & Associates. (1997). Take time to talk: A survey of primary care physicians and the public (Study No. 728320/1). New York: Author. Hawkins, R., Pingree, S., Gustafson, D., Böberg, Ε., Bricker, Ε., McTavish, F., Wise, Μ., & Owens, Β. (1997). Aiding those facing health crises: The experience of the CHESS project. In R. L. Street, Jr., W. Gold, & T. Manning (Eds.), Health promotion and interactive technology: Theoretical applications and future directions (pp. 79-102). Mahwah, NJ: Lawrence Erlbaum. Health Summit Working Group. (1998). Cnteria for assessing the quality of health information on the Internet: Policy paper. Available: http://hitiweb.mitretek.org/ docs/policy.pdf Henson, D. (1999). Cancer and the Internet. Cancer, 86(3), 373-374. Hersh, W., Gorman, P., & Sacherek, L. (1998). Applicability and quality of information for answering clinical questions on the Web. Journal of the American Medical Association, 280(15), 1307-1308. Hodge, J., Jr., Gostin, L., & Jacobson, P. (1999, October 20). Legal issues concerning electronic health information. Journal of the American Medical Association, 282(15), 1466-1471. Hoffman, D., & Novak, T. (1998). Bridging the racial divide on the Internet. Science, 280(5362), 390-391. HON. (1998). HON's third survey on the usage of the Internet for medical and health Internet purposes. Health On the Net Foundation. MEDNET 98—The Third Annual World Congress on the Internet in Medicine, Electronic Poster EP8. [Online]. Available: http://www.hon.ch/cgi-bin/quest/quest Jun98

The Internet and Health Communication

43

Impicciatore, P., Pandolrmi, C, Casella, N., & Bonati, M. (1997). Reliability of health information for the public on the World Wide Web: Systematic survey of advice on managing fever in children at home. British Medical Journal 314,1875-1879. Jadad, Α., & Gagliardi, A. (1998). Rating health information on the Internet. Journal of the American Medical Association, 279, 611-614. Johnson, S., & Wordell, C. (1998). Internet utilization among medical information specialists in the pharmaceutical industry and academia. Drug Information Journal, 32, 547-554. Kahin, B., & Keller, J. (1995). Public access to the Internet. Cambridge: MIT Press. Kahn, Μ. (1997). Enterprise-wide clinical data integration. In P. R Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 41-54). New York: Springer. Kane, B., & Sands, D. (1998). Guidelines for the clinical use of electronic mail with patients. Journal of the American Medical Informatics Association, 5,104-111. Kissinger, K., & Borchardt, S. (1996). Information technology for integrated health systems. New York: John Wiley. Kreuter, Μ., Farrell, Ό., Olevitch, L., Brennan, L., & Rimer, B. (2000). Tailoring health messages: Customizing communication with technology. Mahwah, NJ: Lawrence Erlbaum. Landro, L. (1999, October 19). Alone together: Cancer patients and survivors find treatment—and support—online. Wall Street Journal, pp. R12, Fl 9. LaPorte, R. (1997). Improving public health via the information superhighway. [Online] . Available: http://www.the-scientist.library.upenn.edu/yrl997/august/ opin_970818.html (The Scientist, 11 [16], p. 10). Lemaire, M. (2000). The Web-enabled patient. In L. Eder (Ed.), Managing healthcare: Information systems with Web-enabled technologies (pp. 222-238). Hershey, PA: Idea Group Publishing. Lindberg, D., & Humphreys, B. (1998). Medicine and health on the Internet: The good, the bad, and the ugly. Journal of the American Medical Association, 280(15), 1303-1304. Lowes, R. (1997). Here come patients who've "studied" medicine online. Medical Economics, 74(2), 175-184. McClung, H. J., Murray, R., & Heitlinger, L. (1998). The Internet as a source for current patient information. Pediatrics, 101(6), E2. McDonald, C, et al. (1998). Canopy computing: Using the Web in clinical practice. Journal of the American Medical Association, 280(15), 1325-1329. McGinnis, J. M., Deering, M. J., & Patrick, K. (1995). Public health information and the new media: A view from the public health service. In L. Harris (Ed.), Health and the new media: Technologies transforming personal and public health (pp. 127141). Mahwah, NJ: Lawrence Erlbaum. Media Metrix. (2000). Internet traffic data. New York, January. [Online]. Available: http://www.wrhco.eom/research/eletters/wrh/curTent.html#table Moberg, D., Fuller, D., Gossage, J. P., Littman, P., Mulvey, K., Shwartz, M.,& Vetter, J. (1997). The target cities program: Management information systems for drug abuse treatment. In P. R Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 135-158). New York: Springer.

44

EXPERIENCES AND EXPECTATIONS

Moore, S. (1997). Computer networks as environment for care: Dynamics of the clinical encounter. In P. F. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 287-298). New York: Springer. More, D., & Clarke, P. (1999). The general practice computer system project: A doctor's desktop for Australia. International Journal ofMedical Informatics, 55,65-75. MSNBC. (1999). Web health searches increasing [Online]. Available: http:// www.msnbc.com/news/209252.asp National disease prevention and health promotion objectives. (1990). Washington, DC: U.S. Department of Health and Human Services, Public Health and Science Department. Neu, C, Anderson, R, & Bikson, T. (1999). Case study: The Health Care Financing Administration and the Medicare program. In Sending your government a message: E-mail communication between citizens and government (pp. 31-58). Santa Monica, CA: RAND. New York Times. (1999). [Online]. Available: http://www.nytimes.com/99/12/28/ news/washpol/clinton-pharmacy.html Nonnecke, Β., & Preece, J. (1999). Lurker demographics: Counting the silent. Unpublished paper, University of Maryland, Baltimore County, Department of Information Systems. Parvanta, C, & Freimuth, V. (2000). Health communication at the Centers for Disease Control & Prevention. American Journal of Health Behavior, 24(1), 18-25. Patterson, T, Shaw, W., & Masys, D. (1997). Improving health through computer self-help programs: Theory and practice. In P. F. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 219-246). New York: Springer. Pemble, K. (1997). Information infrastructure for health communities: The Wisconsin Health Information Network (WHIN). In P. F. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 73-100). New York: Springer. Pennbridge, J., Moya, R., & Rodrigues, L. (1999). Questionnaire survey of California consumers' use and rating of sources of health care information including the Internet. Western Journal of Medicine, 171, 302-305. Piotrow, P., & Kincaid, D. L. (2001). A strategic approach to international health programs. In R. E. Rice & C. Atkin (Eds.), Public communication campaigns (3rd ed.), pp. 249-266. Thousand Oaks, CA: Sage. Preece, J. (1998). Empathic communities: Reaching out across the Web. Interactions, 2, 32-43. Preece, J. (2000). Thnving online communities: Usability and sociability. New York: John Wiley. Preece, J., & Ghozati, K. (1998). Empathic communities: Exploration ofl 00 online communities. Baltimore, MD: University of Maryland Information Systems Department. Puskin, D., Mintzer, C, & Wasem, C. (1997). Telemedicine: Building a world system for today and tomorrow. In P. F. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 271-286). New York: Springer.

The Internet and Health Communication

45

Rampil, I. (1998). Medical information on the Internet. Anesthesiology, 89, 12331245. Rees, A. M. (1998). The consumer health information source book (5th ed.). Phoenix, AZ: Oryx. Rimal, R., & Flora, J. (1997). Interactive technology attributes in health promotion: Practical and theoretical issues. In R. L. Street, Jr., W. Gold, & T. Manning (Eds.), Health promotion and interactive technology: Theoretical applications and future directions (pp. 19-38). Mahwah, NJ: Lawrence Erlbaum. Rose, S., Bruce, J., & Maffulli, N. (1998). Access the Internet for patient information about orthopedics. Journal of the American Medical Association, 280(15), 1309. Rubin, R., & Aukema, M. (1997). CHMIS evolution: The metamorphosis of a community health information movement. In P. R. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 101-116). New York: Springer. Scheerhorn, D. (1997). Creating illness-related communities in cyberspace. In R. L. Street, Jr., W. Gold, & T. Manning (Eds.), Health promotion and interactive technology: Theoretical applications and future directions (pp. 171-185). Mahwah, NJ: Lawrence Erlbaum. Schneider, S. (1997). Postscript for section 2. In P. F. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 159-170). New York: Springer. Simons-Morton, D. G., Goff, D. C, Osganian, S., Goldberg, R. J., Raczynski, J. M., Finnegan, J. R., Zapka J., Eisenberg, Μ. S., Proschan, Μ. Α., Feldman, Η. Α., Hedges, J. R., & Luepker R. V. (1998, July). Rapid Early Action for Coronary Treatment: Rationale, design, and baseline characteristics. REACT Research Group. Academic Emergency Medicine, 5(7), 726-738. Skinner, C. S., & Kreuter, Μ. (1997). Using theories in planning interactive computer programs. In R. L. Street, Jr., W. Gold, & T. Manning (Eds.), Health promotion and interactive technology: Theoretical applications and future directions (pp. 39-65). Mahwah, NJ: Lawrence Erlbaum. Slack, W. (1997). Patient-computer dialogue and patients' right to decide. In P. F. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 55-72). New York: Springer. Slobodien, H. (1998, August). It's more than herbs, Herb. New Jersey Medicine, pp. 15-16. Smyth, K., Feinstein, S., & Kacerek, S. (1997). The Alzheimer's disease support center: Information and support for family caregivers through computer mediated communication. In P. F. Brennan, S. J. Schneider, & E. Tornquist (Eds.), Information networks for community health (pp. 189-204). New York: Springer. Spielberg, A. (1998). On call and online: Sociohistorical, legal, and ethical implications of e-mail for the patient-physician relationship. Journal of the American Medical Association, 280(15), 1353-1359. Stephenson, J. (1998). Patient pretenders weave tangled "web" of deceit. Journal of the American Medical Association, 280(15), 1297. Stolberg, S. G. (1999a, June 27). Internet prescriptions boom in the "wild west" of the Web. New York Times, pp. 1,14.

46

EXPERIENCES AND EXPECTATIONS

Stolberg, S.G. (1999b, July 4). From M.D. to I.P.O., chasing virtual fortunes. New York Times, p. 3. Street, R. L., Jr., & Rimal, R. (1997). Health promotion and interactive technology: A conceptual foundation. In R. L. Street, Jr., W. Gold, & T. Manning (Eds.), Health promotion and interactive technology: Theoretical applications and future directions (pp. 1-18). Mahwah, NJ: Lawrence Erlbaum. Stroh, Μ. (1999, June 29). The doctor is . . . online: "Clinics" making virtual house calls are altering the traditional physician-patient relationship. Star-Ledger, p. 2. Turner, J. (2000). Telemedicine: Generating the virtual office visit. In L. Eder (Ed.), Managing healthcare: Information systems with Web-enabled technologies (pp. 5968). Hershey, PA: Idea Group Publishing. U.S. Congress, Office of Technology Assessment. (1993). Protecting pnvacy in computenzed medical information (OTA-TCT-576). Washington, DC: Government Printing Office. Vickery, D. (1995). Demand management, self-care, and the new media. In L. Harris (Ed.), Health and the new media: Technologies transforming personal and public health (pp. 45-63). Mahwah, NJ: Lawrence Erlbaum. Voge, S. (1998). NOAH—New York Online Access to Health: Library collaboration for bilingual consumer health information on the Internet. Bulletin of the Medical Library Association, 86(3), 326-334. Walther, J., & Boyd, S. (in press). Attraction to computer-mediated social support. In C. Lin & David Atkin (Eds.), Communication technology and society: Audience adoption and use of new media. Cresskill, NJ: Hampton Press. Weinberg, J., Klein, S., Waxman, B., Anderson, D., Banerjee, P., & Klepper, R. (2000). Patient-doctor interconnectivity: Improving health care management and patient compliance with Web technology. In L. Eder (Ed.), Managing healthcare: Information systems with Web-enabled technologies (pp. 188-204). Hershey, PA: Idea Group Publishing. Winkler, M., & Silberg, W. (Eds.). (1998). Computers in medicine [Special issue]. Journal of the American Medical Association, 280(15). Woody, T. (1999). Ε-health ethics summit releases code. YahoolNews [Online]. Available: http://www.dailynews.yahoo.com/htx/is/200000218/bs/20000218006. html Worth, E. R., & Patrick, Τ. B. (1997). Do electronic mail discussion lists act as virtual colleagues? Journal of the Amencan Medical Informatics Association, 4, 325-329. Zallen, B. (1995). Member-centered managed care and the new media. In L. Harris (Ed.), Health and the new media: Technologies transforming personal and public health (pp. 21-43). Mahwah, NJ: Lawrence Erlbaum.

The Future of the Internet in Health Care A Five-Year Forecast ROBERT MITTMAN MARY CAIN

T

he use of the Internet in health care has attracted a lot of attention lately. Numerous publications from the New York Times to the Journal of the American Medical Association have featured stories on how consumers are finding medical information on the Web and changing the balance of power in the practice of medicine. Health care practitioners are also using the Internet—to keep up in their fields, to communicate with patients, and to consult with each other. But there is a dark side to Internet medicine. Inspiring stories of lives saved through the Internet get equal billing with hair-raising tales of Web-based quackery. What are we to believe? This forecast begins by describing the driving forces behind some of the high expectations—the good reasons people are excited about the Web. It then presents some of the barriers to the implementation and diffusion of health care applications on the Internet—some sobering analyses to temper the enthusiasm. Finally, it presents forecasts of six leading-edge appli47

EXPERIENCES AND EXPECTATIONS

48

Driving Forces, Barriers, and Leading-Edge Applications Associated With the Future of the Internet in Health Care Driving

Forces

21 st-century health consumers Consumer experiences with other industries: Internet shopping and e-mail

Barriers

Leading-Edge

Security concerns

Consumer health information services

Characteristics of the Internet

Applications

Online support groups for patients and caregivers

The mixed quality of information on the Internet

Health care provider information services

Characteristics of the Internet

Physician ambivalence

Provider-patient e-mail

Market forces in health care

The disarray of health care information systems

Communications infrastructure and transaction services

Lack of resources for Web development

Electronic medical records

Too many standards

cations for health care on the Internet. Table 2.1 summarizes each of these three topics.

DRIVING FORCES The Internet has already made substantial inroads into the delivery, administration, and reimbursement of health care services, as well as into consumer health information. Several factors are driving this growth, including (a) consumer demands, (b) consumer experience with Internet shopping and e-mail through other industries, (c) the characteristics of the Internet, and (d) market pressures in the health care industry.

21st-century Health Care Consumers A growing share of health care consumers are what the Institute for the Future has come to call "new consumers"—people who are actively involved in making choices about the health care they receive (Kyrouz, Holt,

Future of the Internet in Health Care

49

Mittman, & Everett, 1998). New consumers have three characteristics that distinguish them from more traditional consumers: cash, college, and computers. The Institute for the Future (IFTF) has projected from U.S. Labor Department statistics that in 2005, 48% of U.S. households will have an annual income of $50,000 or more, more than 54% of Americans will have attended at least one year of college, and 65% of U.S. households will own a personal computer (Kyrouz et a l , 1998). 1. New consumers have more disposable income, giving them discretion to make choices about what they purchase. New consumers can choose between saving for their children's education or saving for a vacation. They can choose between purchasing a more expensive health plan or spending out-of-pocket for alternative medicine services. The point is, new consumers have money to spend, and a multitude of businesses and services are vying for those dollars. 2. New consumers have the analytical sophistication that comes from attending at least a year of college. New consumers apply that analytical approach to many of their purchases. They are more likely to seek information about health and health care choices than traditional consumers. 3. New consumers have access to computers, either at home or work, or both. They have experience in using computers and the Internet to help them seek and analyze information about their health decisions. They are much more likely than other consumers to turn to online health information sources and published medical references. New consumers are a growing share of the population, rising from 2 5 % in 1978 to 45% in 1998 to an estimated 52% in 2005. This means that all health care organizations will have to attend to their needs, while continuing to meet the needs of traditional consumers. New consumers' expectations about the level of service and interaction they receive from health care service providers have been shaped by their experiences with retailers and financial services firms. They expect free choice of doctors, control over the type of treatments they receive, access to good information about their care, and extremely high levels of customer service. Many of them also expect to be able to interact with their health care providers and payers by Internet and e-mail as part of that service. They go online to get health information and expect their health care providers to be as well versed in what's on the Web as they are.

50

EXPERIENCES AND EXPECTATIONS

Consumer Experiences With Other Industries: Internet Shopping and E-Mail Experience with other industries has helped shape the expectations that new consumers bring to health care. Part of that recent experience includes the use of the Internet for transactions—particularly for shopping. During the 1998 and 1999 holiday shopping seasons, consumer-oriented Web commerce exploded. Amazon.com, Dell Computers, eBay, REI, CDnow, and many others did a booming retail business on the Internet. Retail transactions on the Net in 1999 exceeded $10 billion—only 0.4% of total retail sales, but still a significant figure. The media covered Internet shopping extensively. And any stock with ".com" in its name traded at incredible multiples of earnings (assuming they made any profit at all). At the same time, there are now more than 100 million e-mail users in the United States (Harris Interactive, 1999). Many older Americans—about 25%—now use e-mail (Corey 1999), often starting as a way to communicate with their grandchildren. Many e-mail users go on to use the Web for other purposes. However, not everyone will be a new consumer. If about half of the American population will have access to a PC and the Internet at home or work by the year 2005, about half of the population won't. An ever-widening disparity between the "haves" and "have nots" with regard to technology is correlated with income, education, and ethnicity. Inner-city residents and new immigrants, in particular, have been left out of the information revolution. Such traditionally underserved populations are less likely to have electronic technologies in their homes and are more likely to get information through talking rather than reading (Kyrouz et al., 1998; National Telecommunications and Information Administration [NTIA], 1999). A significant sector of consumers will not have access to information and services on the Internet. Even though the cost of PCs has dropped to below $500, some will not be able to afford to purchase the hardware and software and pay for Internet access. Others will be able to afford the computers but will face barriers of language, literacy, and education. Still others will choose not to go online.

Future of the Internet in Health Care

51

Characteristics of the Internet The Internet is particularly well suited to meet the needs of new consumers and health care organizations, for several reasons. 1. Inexpensive. Access to the Internet costs less than $20 per month, while the cost of an Internet-capable personal computer is below $500. For the 50% of consumers who don't have the resources of new consumers, Internet access is increasingly common in libraries and schools. In December 1999, President Clinton said that 50% of America's schools and more than 80% of its classrooms are wired for the Internet, and he set a national goal to have all connected by the end of 2000 (Hunt, 1999). 2. Easy to Use. Browser technology has made the act of navigating through complex information spaces much simpler through point-and-click interfaces. Although the sheer quantity of health information on the Web maybe overwhelming, most Web sites are easy to use. 3. Democratic. Before the development of the Web, only large institutions could afford to disseminate their point of view to a wide audience. The Web has dramatically lowered the barriers to entry for people and organizations that want to spread their message. From the point of view of information seekers, it is possible to get a broad diversity of viewpoints on any health issue. 4. Long Distance. The delivery of health care services has always been an inherently local phenomenon. The Internet is an inherently global phenomenon. This means that it is possible to get medical information, and even medical advice, from geographically dispersed sources. This, too, contributes to the diversity of information available. 5. Increasing Functionality. The Web platform is gaining in functionality and sophistication. In addition to its core browser functions, it now includes client- and server-side Java and will soon include XML (extensible Markup Language). These provide a much richer set of capabilities for deployment of interfaces and for integration across health care information systems.

52

EXPERIENCES AND EXPECTATIONS

Market Forces in Health Care For most of this century, American health care was isolated from the kinds of market forces that pushed other industries to become more efficient and provide excellent customer service. But market forces have been at work in health care for almost a decade now, in the form of managed care, employer purchasing coalitions, assertive government payers and regulators, and consumer organizations. The system has adapted. Many of these adaptations favor the use of information technology and the Internet. Physicians, hospitals, and health plans all have recognized that the larger they are, the more market power they can command. As a result, a wave of consolidation has swept the industry in the past 3 years, with, for example, about 170 hospital mergers per year and the consolidation of 18 large health plans across the country into 6 (Institute for the Future [IFTF], 2000). At the same time, provider organizations around the country have created vertically integrated enterprises, spanning physicians, hospitals, home care, labs, imaging centers, and even health plans. Web technologies—intranets, extranets, and the Internet—can serve as a low-cost, rapidly deployable platform for disseminating information across vertically and horizontally integrated health care organizations. Managed care has increased the flow and diversity of information about health care. One result of the cost-containing pressure of managed care is that the simple indemnity insurance world of charges, claims, and payments is nearly gone (see Figure 2.1). It has been replaced with the "hassle factor" of managed care—discounts, provider networks, eligibility verification, precertification, referral authorization, drug formularies, chart reviews, and so on. Large employers, purchasing coalitions such as the Pacific Business Group on Health, accreditors such as the National Committee on Quality Assurance, and the government are all demanding that health plans and providers show they deliver high quality and good value. These factors all increase the information coming in and out of health care organizations. The Internet will be a channel for an increasing share of those transactions. Perhaps the most significant indicator of market forces in health care is the growth of brands. Along with the competition and consolidation has come an increase in advertising of health plans, hospitals, medical groups, specialty clinics, and pharmaceutical products. Hoping to drive consumers

Future of the Internet in Health Care

53

Performance-based

Episodic prepayment, APGs Global capitation Partial capitation Discounted FFS (with risk pool withheld) Discounted FFS

FFS (including Medicare FFS) 1997

2002

2007

2010

Figure 2.1. The Decline and Fall of Fee-for-Service (FFS) Medicine S O U R C E : Institute for the Future (2000).

toward their products, pharmaceutical companies spend increasing sums on direct-to-consumer advertising, from $600 million in 1996 to an expected $1.8 billion in 1999. Like other advertisers, health care advertisers have promoted their Web sites on television, on billboards, and in print ads. Advertising has raised the profile of health care Web sites. Advertising online is the subject of many discussions about revenue streams. In health care, online marketing is expected to be a boon to medical marketing. According to researchers at Cyber Dialogue, pharmaceutical companies spent an estimated $915 million on direct-to-consumer advertising in the first half of 1999 to influence an estimated 34 million U.S. adults who requested specific prescription medications from doctors. The vast majority of this spending, $530 million, ended up on television, while print came in second with expenditures of $370 million. Internet ad spending amounted to just 1%, $10 million (Cyber Dialogue, 1999). Yet Forrester Research projected in September 1999 that Internet advertising will reach $33 billion by the year 2004.

54

EXPERIENCES AND EXPECTATIONS

BARRIERS Although the scene is set for the Internet to move rapidly into the health care mainstream, a number of barriers will inhibit its deployment. None of the barriers is fatal, but together they will slow adoption of Web technologies substantially. The main barriers to Internet adoption are security concerns, characteristics of the Internet, distrust of the quality of Internet information, physician ambivalence, the disarray of legacy information systems, the lack of resources for Web development in health care organizations, and a confusing array of "nonstandard standards."

Security Concerns If there's one thing people are more guarded about than their financial information, it's private information about their health. Moving to electronic commerce in health care and using the Internet will send large quantities of private health information zinging across open networks. The perception of a lack of security (as much by providers and insurers concerned about liability and bad publicity as by consumers) will inhibit the use of the Internet for personal clinical information in the near term (Cyber Dialogue, 2000). The "security challenge" breaks down into six components: 1. Protecting servers and databases from unauthorized intrusion or modification 2. Authenticating the identity of senders and recipients 3. Protecting the integrity of the message itself 4. Ensuring nonrepudiation (i.e., making sure senders cannot falsely deny they sent a given message) 5. Establishing audit trails 6. Ensuring the confidentiality of messages Various technologies and procedures are being developed to tackle these security problems, including the following:

Future of the Internet in Health Care

55

1. Data encryption to protect data integrity and confidentiality, such as DES and RSA (private key or public key encryption technologies) 2. Digital signatures, the extra data appended to a message that identifies and authenticates the sender and message data using public-key encryption, hardware tokens, and biometric solutions such as fingerprint readers to authenticate users (Free Online Dictionary of Computing [FOLDOC], 1999) 3. Firewalls, virus protection software, smart cards (that store key information on a portable card that cannot be intercepted), and other hardware and software, all designed to protect network integrity 4. Once the technology is in place, ongoing procedures must be implemented to ensure continued security. These procedures, such as the use of passwords, multitiered access to different levels of information, and controls over physical access to information, must be rigorous but efficient; they mustn't hold up the core processes of the organization. Perhaps more significant than the interception of health care information or unauthorized release by an inside party is the possibility of institutional breaches of confidentiality; for example, the sale of individually identifiable information to marketers or the use of private data for medical underwriting by an insurer. These breaches don't lend themselves to technical solutions, but rather to laws, detection of violations, enforcement, and punishment (Goldman, Hudson, & Smith, 2000; see also Katz & Aspden, Chapter 18, this volume). The European Union already is enforcing strict medical data security standards. At some point in the next 10 years, the American market, government, courts, or some combination of the three, will demand improved security and confidentiality. The U.S. health care industry needs to get involved in this legislative process or risk the imposition of burdensome regulations. The industry must develop an understanding of the capital and operating costs of implementing system security techniques and factor this into the business equation. Three competing tendencies will shape the perception of security on the Internet. First, health care organizations will develop, deploy, and publicize secure information systems—confidence will legitimately increase. Second, there will inevitably be several well-publicized incidents of people

56

EXPERIENCES AND EXPECTATIONS

being harmed by public releases of their health care information—those exceptional cases will shape the debate. Third, in the end, people and organizations will learn to live with a less-than-perfect combination of technologies and policies, just as the policy of limiting the fraud liability of credit card holders to $50 has largely allayed the fears of Internet shoppers.

The Characteristics of the Internet Just as there are characteristics of the Internet that will drive its use in health care, there also are several characteristics that will impede its diffusion. 1. Instability and Technology Churn. The underlying technology of the Web is undergoing rapid mutation—"Web-years" are measured in human weeks. Most health care organizations traditionally are bureaucratic and slow to change, underinvest in information technology, and have to devote scarce resources to keep up with the pace of change. 2. Browser Technology Is Weak. Despite the additions of XML and Java, Web client software (browsers) is not very capable. It does a poor job, for example, of handling linking from on-screen forms to underlying databases. This will impede the linkage of the Web to the databases of health care legacy systems. 3. Search Engines Can't See Everything. Dynamic Web pages make up an increasing share of all content on the Web. Because of how dynamic pages are created, the Web crawlers on search engines cannot see them. This means that a person searching for health care content will miss much of what is available. Of more than 13 million registered domain names, less than 20% are thought to be indexed and found by search engines (Kaiser, 1999). 4. Low Bandwidth Links to Most Homes. Health care information lends itself well to rich content such as images, animations, and video. Most users reach the Web through modem connections at data rates that do not give responsive performance for that type of content.

Future of the Internet in Health Care

57

The Mixed Quality of Information on the Internet The Internet provides ready access to a vast body of health information for consumers, information that at its best can equip consumers to lead healthier lifestyles, detect potential medical problems early, work more collaboratively with physicians to treat illness, and learn of effective treatments to which a local provider may not have access. At its worst, however, the information can undermine health objectives. Consumers can be misled into, for example, self-destructive beliefs and behavior changes, ineffective or harmful treatments, and false medical understandings that undermine relationships with their physicians. Concerns about the quality of health information on the Internet are in some respects no different from quality issues faced in other communication channels. Indeed, advice from friends and family, never the most accurate, is perhaps the most common source of health information for consumers (Kyrouz et al., 1998; see Napoli, Chapter 3, this volume). Some characteristics of information on the Internet, however, exacerbate problems of quality. These characteristics include: 1. Inexpensive and easy publishing. The Internet allows thousands of times more health information publishers to gain access to a national and global audience. 2. Anonymity. It is difficult to verify, without checking outside the Internet, who is behind a particular Web site or Internet message. 3. Pace. Internet sites can be readily added and changed. Furthermore, breaking news breaks more quickly over the Internet, pressuring publishers to be less rigorous with their fact checking. 4. Transcending regulators. With physical location of little relevance in cyberspace, fraudulent operations can readily access the American market from countries with limited regulation covering fraud and other potential infractions. Several approaches will address the quality of health information on the Web. A number of rating services, such as Health On the Net (http://

58

EXPERIENCES AND EXPECTATIONS

www.hon.ch), which measures sites against its code of conduct focusing on credibility; and Health Summit Working Group (http://www.hswg. com), which measures credibility, content, disclosure, links, design, interactivity, and caveats, have been put in place. These likely will proliferate, leading to confusion about which provide reliable ratings. Consumers will sort through the cacophony of health care sites the way they do any crowded product category—they will gravitate toward brands. Large health care organizations with established brands, such as the Mayo Clinic, will continue to extend them onto the Internet. Some Internetbased health care businesses will establish strong brands of their own. Finally, some physicians, consumers, or others who are concerned about the quality of health care information on the Web will seek to regulate its content. Because the Internet, as a huge social communication experiment, is growing rapidly beyond any governmental purview, those attempts will likely be futile. Nonetheless, both Internet taxation and online prescribing are the subject of legislation pending before Congress.

Physician Ambivalence Medical culture is extremely conservative and cautious, especially when it comes to technologies that could alter the doctor-patient relationship. The Internet is one such technology. Physicians (or at least physician offices) have moved significantly to adopt information technology. Computers in medical offices are commonly used for routine administrative functions, such as scheduling, electronic claims, and checking eligibility. They also are used in some clinical applications. A 1997 survey of physicians by the Institute for the Future and the Lou Harris organization found that 46% of physicians and/or their staff had accessed background information like Medline via the Internet or an online service; for doctors only, the figure was 3 8 % . For accessing clinical patient information, the percentages were 42% and 22%, respectively; for recording patients' clinical information like histories, 30% and 15%; for transmitting clinical information to other doctors electronically, 28% and 13%; and for receiving computerized treatment records, 8% and 6%. Doctors' experience of the quality of information on the Internet is mixed. Hersh, Gorman, and Sacherek (1998), analyzing 639 pages retrieved in 50 searches seeking to answer common clinical questions, found that 99% did not disclose possible conflicts of interest, 89% were not appli-

Future of the Internet in Health Care

59

cable to the question, 82% didn't provide the date the page was posted or updated, 80% did not provide the author's credentials, 69% did not indicate the author, and 60% were not oriented toward health care professionals. O n l y l % provided any evidence-based resources (see also, Rice, Chapter 1, this volume). Physicians are both threatened and fascinated by the Web. Many understand the value of having well-informed patients and the role the Internet can play in educating their patients. At the same time, they are concerned about losing control over the interaction with their patients. In a 10-minute office visit, they have trouble answering all the questions of a patient armed with 400 pages printed from the Internet. And they fear being overwhelmed with messages if they provide an e-mail address. On balance, their ambivalence will keep many of them from enthusiastically embracing the Internet (see also Katz & Rice, Chapter 19, this volume).

The Disarray of Health Care Information Systems The legacy information systems of most health care players—insurers, hospitals, and physicians—are not ready for prime time. Clinical information systems in labs, pharmacies, and hospitals are fragmented and don't communicate well with each other. The enrollment and eligibility databases of most insurers are weeks to months out of date. The essence of electronic commerce is that trading partners can perform transactions (i.e., view and change information in databases) in real time. Customers and partners expect that the information in these databases is correct and complete. Most health care customer service departments give the illusion that the information is good by placing clever people on the telephone between the customer and the database. Once customers gain direct access (preferably before they get direct access), the information must be brought and kept up to date. This "data-cleaning" process will take time and money, and will delay the onset of direct transactions through the Internet. Most transaction-intensive health care players, notably health insurers, have a mix of database and transaction systems, ranging from old flat-file, batch systems to real-time, relational databases. Some even are experimenting with object-oriented databases. To provide integrated access to information via the Internet, information from all of those databases must

60

EXPERIENCES AND EXPECTATIONS

eventually be available in real time. At some point, those reliable old batch systems must be replaced. This is not an insurmountable problem, but, again, it will take time and money to resolve.

Lack of Resources for Web Development Information systems departments in most health care organizations are not Web oriented (see Eder & Wise, Chapter 14, this volume). They have resisted making the investments needed to build a Web presence. Many are underfunded—health care as a whole spends much less on information technology, only 2% to 4% of the average health care company's operating budget, than other information-intensive industries, which spend between 7% and 10% (IFTF, 2000). Many of their resources were being diverted to resolving the Year 2000 problem. Although the direst projections about the chaos it would spread were avoided, a lot of information systems' departmental resources were heavily invested in fixing the date fields in mission-critical applications. While this didn't preclude the development of Internet applications, it surely delayed them for months or years, as the attention of health care information professionals was focused on this urgent task. Health care organizations must compete with a range of other industries to attract the best Web designers and developers, who are in extremely short supply.

Too Many Standards Health care has plenty of standards-setting bodies. Overall, however, standards have been unevenly adopted. Unlike the uniform bar codes for grocery items or automotive parts, there are differing, overlapping, and competing sets of standards in many areas of health care information technology. Health care has a data interchange standard, HL7 (Health Level 7), that is widely used, but not universally so, and is imperfect in many areas. Medical equipment suppliers, pharmacists, and health insurers all have their own electronic data interchange (EDI) standards and distinct electronic commerce efforts. A critical missing piece is a single semantic and terminology standard for diagnoses, procedures, and outcomes. SNOMED, a nomenclature classification system created for the indexing of the entire medical record, including signs and symptoms, diagnoses,

Future of the Internet in Health Care

61

and procedures, for example, is used by many organizations, but it is proprietary and not complete. Standards setting in any industry is a slow, political process. The parochial and vertically differentiated nature of standards in health care will inhibit the near-term development of electronic commerce in the industry and its move to the Internet. That diversity will limit the rate of diffusion of comprehensive electronic commerce in health care, including electronic commerce on the Internet.

LEADING-EDGE APPLICATIONS A number of applications will be at the leading edge of Internet use in health care. This section presents six of these applications, including a description of the type of application with some of the leading examples or precursors, and a forecast of how the application will evolve in the next 5 years.

Consumer Health Information Services As Internet usage increases, so does the public's search for health information. In 1999, Harris Interactive reported that almost 70% of adult Internet users searched for health information online. The applications used to search the Web for health information comprise a wide range of specialized search engines and databases. People searching for health information also vary a great deal in level of interest and expertise. The same search engine may handle requests from a physician looking for the latest medical journal article on an obscure condition and those from a lay person who wants to know whether canker sores are contagious. Health information online can take the form of anything from clinical research results to advice columns written by people who aren't even health professionals. The search for health information online often leads to information overload and misinformation. Newsgroups, chat rooms, and listservs are another way for individuals to pull information from the Internet by reading and posting comments online (see Rice, Chapter 1, this volume). With tens of thousands of newsgroups to choose from, an Internet user can post to the list (i.e., can

62

EXPERIENCES AND EXPECTATIONS

choose to participate by responding to elements of the "discussion" or by starting a new thread) or just "lurk," reading what others have written. Newsgroups with a high degree of interaction around a particular medical issue essentially become an online support group.

FORECAST Several developments will influence consumer access to information on the Internet in the next 5 years. Plans and providers will react to the demands of "new consumers" to be involved in their care and will provide an increasing amount of information online. Because much diverse, inconsistent, and incoherent information is now available on the Web, consumers need help in sorting through the extensive "noise" on health care Web sites. Search engines with a stronger editorial voice that compile and index health information rather than simply accumulate information will win loyal users. Figure 2.2 shows the number of health-related pages indexed on the World Wide Web by the search engine AltaVista since 1996, using specific search terms. A small number of trusted services that rate the quality of information on health care sites will emerge from efforts such as Health On the Net (http://www.hon.ch). Mainstream sites will comply with the raters' guidelines to get seals of approval. At the same time, the multitude of sites that don't care about ratings will continue to thrive. New forms of health care information will proliferate online. With the growing availability of inexpensive data and storage, more images, animation, video clips, and interactive learning systems will be put in place. Few consumers, however, will have the bandwidth to be able to take advantage of that rich content in the near future. Online purchases of prescription drugs and of over-the-counter items like vitamins and nutraceuticals (vitamin-fortified food) will increase substantially over the forecast period (see Bayer, Chapter 17, this volume). Health e-commerce will be used particularly for routine refills of prescriptions (often done by mail order now), items that are difficult to locate, and potentially embarrassing items (such as incontinence diapers and Viagra). This growth will result in some regulation. Issues like the interstate practice of medicine and privacy of consumers' information will dominate.

Future of the Internet in Health Care

63

Thousands of Pages 20 "Health"

15

10

5 "Medicine" "Pharmaceutical" 0 Jan 96

Dec 96

Mar 98

Oct

Figure 2.2. Thousands of Pages Indexed From the World Wide Web by the Search Engine AltaVista for the T e r m s Health, Mediane, and Pharmaceutical S O U R C E : Institute for the Future (1998).

Many health care report cards, consumer reports, and consumerfriendly ratings of health plans, hospitals, and physicians, such as those provided by healthgrades.com, will be available online. They will be substantially more usable than current report cards. Patients, especially those with chronic diseases, will go beyond information services and begin to use personalized health records services to track their health status. These systems will allow patients to perform health risk assessments online and to track their adherence to diet, exercise, or medication regimens.

64

EXPERIENCES AND EXPECTATIONS

Online Support Groups for Patients and Caregivers Thousands of online support groups give individuals with a specific health concern a place to "congregate" on the Internet for support and information. Frequently, peer support groups are set up by individuals as alternatives to professionally led support groups (see Preece & Ghozati, Chapter 11, this volume). Online support groups are convenient for busy or housebound people, providing support at all times of the day and night. One of the greatest boons to family caregivers has been the online selfhelp and support group where they can meet with others performing similar functions and get expert help. Support group members go online, reaching out for information and commiseration, and, in so doing, create an information commons. This commons is filled with anecdotes about how to deal with a new diagnosis, especially of a rare disease where there is a dearth of research. The powerful ability of the online community to transcend geography, time, and culture is a great comfort to people isolated by disease. Support groups vary by level of interaction, by level of clinical monitoring and availability, and by level of public access. Most groups are either self-moderated by a patient or caregiver or by an advice nurse, who is either compensated minimally or not at all. Support groups sponsored by health care organizations are often either a marketing or revenue-generating tool. Some commercial groups mine data and resell the information collected by the group, looking for patterns in the data. With an increased focus on wellness, some health care organizations believe patient empowerment—brought about by support groups and increased patient education—results in reduced costs and healthier patients. People who understand and believe in their treatment will be more compliant with treatment, and, ultimately, the quality of care will improve and costs will go down. Studies have shown that support group users have fewer hospital admissions and reduced lengths of stay, translating into up to a 30% savings in hospital costs (Maunsell, Brisson, & Deschenes, 1995; Spiegel et a l , 1989).

FORECAST Many of the concerns about the source and quality of information for consumer health information services apply to online support groups as well. In addition:

Future of the Internet in Health Care

65

1. Organizations will be cautious about offering specific medical advice because of fears of liability regarding advising patients and perhaps "practicing medicine" online. Individuals will feel much less constrained about offering medical advice. 2. Some physician organizations will resist the growth of online support groups, since the lack of reimbursement provides little incentive for their involvement. Online support groups will become a flash point between patient groups and physicians. 3. Physicians will feel an incursion onto their clinical turf and a loss of power as their patients gain the support of their online community. Growth in online support for patients will expand in two major directions—more bandwidth leading to new media and a much richer set of support services. 4. Patient education, compliance, and support programs will merge on the Internet. Psychosocial support groups; disease-specific research; physician and provider assessment; information about alternative medicine; educational video and animation programs; and patient self-care support tools, such as software that tracks diet and issues reminders to take medications, will all be available in integrated, disease-specific forums. Some will be sponsored by health plans or providers, while others will be led by patient groups. 5. Beyond the 5-year time frame, as bandwidth to the home and workplace increases, online support groups may begin to add new media choices beyond the simple text messaging (chat) done now. Internet telephony will allow multipoint voice chats, for example. Eventually, people will be able to do video-mediated support groups, though they may stick to chat and use the video channels to share prerecorded information. Most groups treasure and will maintain anonymity until a community of trust is developed, at which point video or face-to-face meetings will occur.

Health Care Provider Information Services Health care professionals have to keep up with an increasing volume of medical information—one study estimates that in general medicine, a physician would have to read 19 articles a day, 365 days a year to stay current (Wayne-Doppke, 1997). Doctors and other health care professionals currently use a range of sources, including journals, textbooks, drug compendia, other physicians, and pharmacists. Most of those information sources fall short—they are not available when needed, are out of date, or

EXPERIENCES AND EXPECTATIONS

66

are hard to search. New information tools that provide valid, timely information would be a boon to harried physicians. The Internet, to a limited extent, meets many of these information needs. Use of online information by providers has become increasingly common—72% of libraries in academic medical centers had Internet access in 1998, up from 24% in 1993. As more sources have become available, demand has jumped. MEDLINE, with 9.2 million references to peer-reviewed articles, saw its usage jump tenfold to more than 75 million searches per year when it introduced free Web access in June 1997 (WayneDoppke, 1997). Nonetheless, technology proxies such as nurses, office staff, or research librarians often search the Web on behalf of busy physicians, and when physicians self-report that they use the Internet, they are often referring to such proxies. Health care professionals account for 34% of MEDLINE's searches.

FORECAST The increase in the use of the Internet for information by physicians will be largely incremental. No breakthrough applications are likely, nor are any discontinuities. Usage of the Web for enhancing clinical knowledge will increase as newly minted doctors who have been trained to do online searching come into private practice. Already, new graduates are making their internship and residency decisions based partly on the sophistication of the information systems at a particular hospital. Continuing medical education, while a natural fit for the Internet because it is based on disseminating information, faces some resistance by CME accreditors and will grow only slowly (see Witherspoon, Eastin, & Cook, Chapter 9, this volume). Physician-oriented sites that filter out most of the random content from the Internet and concentrate clinical information sources such as peer-reviewed journals will develop and increase, particularly by physicians who don't have the information-searching skills to sort out the "noise" on the Internet themselves. Faster growth will occur among nursing and other health care professional sites where there is a stronger culture of collaboration. Most medical journals will have a Web site that reprints current and back issues for subscribers. Some journals are prereleasing important arti-

Future of the Internet in Health Care

67

cles before full peer review has been completed. NIH's EBIOMED, a online journal service, is leading the way and creating controversy by releasing articles before they are published in a journal and allowing for public and peer review to take place simultaneously. New functionality will increase the usefulness of the Web for practicing physicians—evidence-based medicine guidelines that can be linked to electronic medical records and medical atlases with animation and video images. With increases in broadband and integrated software systems, the complexity of the health care record can finally be brought together in one place.

Provider-Patient E-Mail Physicians have been reluctant to embrace e-mail with their patients. Harris Interactive reports that only 13% of physicians use e-mail, a smaller number than report using the Internet (Harris, 1999). Physicians have expressed fears similar to those concerning the telephone in its early days—a barrage of inappropriate e-mails, an inability to assess the priority of email communications, and potential breaches in security and privacy of doctor-patient communications. E-mail, meanwhile, has been enthusiastically embraced by the general public, with approximately 100 million subscribers in 1999. Like the telephone, e-mail will, in time, become an important channel for patient-physician communications (but see Katz & Rice, Chapter 19, this volume).

FORECAST Electronic mail between providers and patients is going to increase rapidly, both in volume and in the types of communications. Communications by e-mail between physicians and their patients will increase in certain communities, with patients driving the shift. As more patients put demands on providers for e-mail, it will become more common for physicians to put their e-mail addresses on their business cards. Physicians will resist the increased level of direct communication. Initial uses of health-related e-mail will be for administrative functions, such as coordination and scheduling. In some areas, for some doctors and con-

68

EXPERIENCES AND EXPECTATIONS

sumers, patient-provider e-mail will be a competitive advantage, generating word-of-mouth referrals and a patient-friendly reputation. Communications between consumers and physicians who are not their doctors will increase much more rapidly. Patients will seek out Web sites that offer information from and consultations with physicians. (Many of these sites will be offered as a free service of organizations seeking to attract patients, while others will be offered for a fee; see Rice, Peterson, & Christine, Chapter 10, and Witherspoon, Chapter 9, both this volume). While practicing medicine over the Web is illegal, many physicians now provide some guidance and informal second opinions to consumers. Consumers today can use the Internet to interact with physicians they have never seen to get prescriptions for some medications. Besides professional and ethical problems, there are jurisdictional problems with interstate or even intercountry transactions. At a minimum, patients are increasingly printing out health information from Web sites and discussing it with their physicians during office visits. Disease state management (DSM) communications will increase as a number of health plans, specialized DSM companies, and pharmaceutical DSM companies use e-mail for the protocol-based management of chronically ill patients (Runde et a l , 1999). We forecast an increase in employerdriven demonstrations and trials in chronic diseases such as congestive heart failure, asthma, and diabetes. Among other things, e-mail will be used to track a patient and collect information on outcomes, health status, and quality of life. Leading-edge provider organizations will put in place a range of applications to make patient-physician e-mail more effective: (a) Form-based e-mail that helps a patient characterize the priority and type of request; (b) alerts to use the telephone or make an office visit for urgent conditions; (c) flow tracking and control to ensure that all messages are followed up; (d) automated responses for simple communications and frequently asked questions (FAQs); (e) linkages to online health risk assessments; and (f) explicit guidelines about how any message will be used, including whether it will be placed in the medical record, to whom it will be forwarded, and how long it will be kept.

Future of the Internet in Health Care

69

Communications Infrastructure and Transaction Services All of the applications described up to this point capitalize on the unique characteristics of the Internet; they represent new approaches to communicating about health care. A range of other applications use the Internet as a replacement for or a complement to existing communications infrastructure or transaction services. The justification for using the Internet will be largely economic—it will replace other communications technologies that are more costly or less capable (see Eder & Wise, Chapter 14, and More & McGrath, Chapter 15, both in this volume). These include: 1. Electronic claims transmitted via the Internet versus dial-up or leased phone lines 2. Remote telemedicine consultations using the Internet to transmit data, voice, video, and images 3. Transmission of clinical information in clinical trials 4. Submission of FDA filings by pharmaceutical and medical equipment manufacturers We look at the first two of these applications. A number of electronic claims clearinghouses, such as claimsnet.com and MedEAmerica, are now accepting claims via the Internet. The largest claims processors, such as Envoy/NEIC, NDC, and HBOC, have not yet embraced the Internet as a medium, but all are studying it. Telemedicine is the use of communications and video technology to bring specialized clinical expertise to remote sites for aid in diagnosis and treatment, and for educational purposes. Telemedicine over the Internet is already happening at hundreds of sites around the globe as clinicians "store and forward" images to one another.

FORECAST The Internet is likely to play a valuable role in health care's move to electronic commerce, especially for claims processing, but also for other paper transactions that can be done much faster and less expensively electronically (see Figure 2.3). Specific developments are likely to include the following scenarios.

70

EXPERIENCES AND EXPECTATIONS

Percent 100

80

Hospital c l a i m s ^ - - — ^

60 Pharmacy claims 40

-•"Äffhealth claims 20 Physician claims 0

'— 1991

1996

1994

Figure 2.3. Electronic Claims Are Dominant in Hospitals and Pharmacies (Percentage of all claims) S O U R C E : Faulkner and Gray (1997), Directory

of Annotated

Medical

Payments.

The large claims clearinghouses and their customers have taken a waitand-see approach to claims on the Web because of security concerns. Assuming that the Health Care Financing Administration (HCFA) sticks to its revised ruling supporting encryption for electronic claims and that the Health Insurance Portability and Accessibility Act (HIPAA) rules, when issued, acknowledge some form of encryption as sufficient to safeguard security, we will see a shift toward the Web (IFTF, 2000; see Baur, Deering, & Hsu, Chapter 16, this volume). The Internet will not replace the clearinghouses. They will have a continuing role tracking and auditing claims submissions, providing customer service, cleaning and editing claims (especially without real data standards or sophisticated practice management

Future of the Internet in Health Care

71

systems), and consolidating and dispatching claims. Claims and encounters, especially for hospitals and pharmacies, and materials and inventory management already are moving rapidly to electronic commerce and the Internet. Other transactions, including eligibility and enrollment, coordination of benefits, utilization review, precertification, referral authorization, and lab and radiology will begin to move to the Internet in the next 2 to 5 years. Telemedicine applications will develop more slowly. It is unlikely that any significant proportion of health care will be delivered remotely in the next decade. The lack of reimbursement for most telemedicine consultations and interstate medical licensing conflicts will limit the deployment of telemedicine services. Also, bandwidth limitations on the current Internet will hinder the use of live video. Most telemedicine developers will have to wait for access to expanded bandwidth. Nonetheless, several factors, such as dropping communication costs and provision in Medicare funding for telemedicine in underserved rural areas, will provide the push telemedicine providers need to get remote care out of the pilot phase for at least traditional telemedicine applications such as video-based consultations, remote radiology, and remote education (see Whitten, Eastin, & Cook, Chapter 12, this volume).

Electronic Medical Records Health care providers are at the beginning of a slow transition toward electronic medical records (EMRs). Most applications of computers in medical offices and hospitals are for administrative rather than clinical functions. Clinical functions of computers have typically been in departmental "islands of automation," usually in the pharmacy, lab, and radiology departments. Most physicians would not consider using a computer during an office visit and would not take the time to transcribe their notes on a keyboard after the visit. Many of these systems are largely homegrown and proprietary. We estimate that fewer than 5% of physicians are now using a comprehensive electronic patient record. A number of institutions have developed their own Web-based front-ends for their electronic medical records, such as Wishard Memorial Hospital and Clarian Health Partners (Indianapolis), University of California (San Diego), Columbia Presbyterian Medical Center (New York), University of Washington (Seattle), Care Group (Boston),

72

EXPERIENCES AND EXPECTATIONS

the Mayo Clinics, and Centre Hospitalier (Rennes, France; McDonald, 1998; see Eder & Wise, Chapter 14, this volume). Most of these systems get information from legacy information systems, localized database server systems that reside onsite, and present it on a browser front-end. They do not integrate data from across legacy systems, but simply present it in a consistent format.

FORECAST Web-based front-ends to electronic medical records will attract a lot of attention and development effort in the next 5 years. 1. Just as hospitals and clinics were not able to push vendors of clinical information systems to create open architectures in the 1980s and 1990s, they will have difficulty in getting the vendors to build interfaces to the Web. Vendors will resist installing a technology that makes it easy to substitute competitors' systems. 2. Standards work, which always proceeds slowly, will have to be done to get vendors to agree on how applications should link to the Web. Health Level Seven (HL7), a standard for health care information, will likely incorporate XML (extensible Markup Language) to create the same general look as conventional (HTML) Web documents, but better able to represent structured records. 3. The efforts to build tools that agglomerate and represent information from disparate systems' electronic medical records will not result in systems that integrate information well enough to provide effective, real-time decision support for physicians. Integration at the display and interface level will not be sufficient. 4. Although there will be a lot of activity in electronic medical records for the next 5 years, it will not provide solutions that are sufficiently compelling to drive widespread adoption, and fragmentation will continue.

REFERENCES Corey, A. (1999). Older Americans among fastest-growing group of Internet users. Columbia, MO: Columbia Daily Tribune. Cyber Dialogue. (1999). Press Release, New York, NY.

Future of the Internet in Health Care

73

Cyber Dialogue. (2000). Ethics survey of consumer attitudes about health Web sites. Oakland: California HealthCare Foundation. Forrester Research. (1999). Press Release, Cambridge, MA. Free online dictionary of computing. (1999). Available: http://www.instantweb.com/ Goldman, J., Hudson, Z., & Smith, R. (2000). Report on the privacy policies and practices of health Web sites. Oakland: California HealthCare Foundation. Harris Interactive Press Release. (1999). September. Hersh, W., Gorman, P., & Sacherek, L. (1998). Applicability and quality of information for answering clinical questions on the Web. Journal of the American Medical Association, 250,1307-1308. Hunt, T. (1999). Clinton wants every American connected to the Internet [Associated Press wire story]. Institute for the Future. (2000). Health and health care 2010: The forecast, the challenge. Trenton, NJ: Robert Wood Johnson Foundation. Kaiser, J. (1999). NET NEWS: Search engines fall short. Science, 285, 295. Kyrouz, Ε. M., Holt, M., Mittman, R, & Everett, W. (1998). Twenty-first century health care consumers. Menlo Park, CA: Institute for the Future. Maunsell, E., Brisson, J., & Deschenes, L. (1995). Social support and survival among women with breast cancer. Cancer, 76, 631-637. McDonald, C. J., Overhage, J. M., Dexter, P. R, Blevins, L., Meeks-Johnson, J., Suico, J. G., Tucker, M. C, & Schadow, G. (1998). Canopy computing: Using the Web in clinical practice. Journal of the American Medical Association, 250,1325-1329. National Telecommunications and Information Administration (NTIA). (1999). Falling through the Net: Defining the digital divide. Washington, DC: U.S. Department of Commerce. Runde, D., Falcon, R., Mittman, R., Cain, M., Everett, W., Bernard, S., & SarasohnKahn, J. (1999). Weaving disease management into the fabric of patient care. Menlo Park, CA: Institute for the Future. Spiegel, D., et al. (1989). Effect of psychosocial treatment on survival of patients with metastatic breast cancer. Lancet, 2, 888-890. Wayne-Doppke, J. C. (1997). Medicine on the Net. Santa Barbara, CA: COR Healthcare Resources.

PART II

Sources of and Experiences With Online Medical Information

n this section, we seek to define and portray the user experience. By the term user, we mean the people who access the Internet for health care purposes. These people include the consumer as well as the physician and other health care professionals. By access, we mean not only seeking information about medical conditions per se (often called searching), but also information about resources, problem solving, the valuations, and sharing social experiences (also often called browsing or grazing). The gamut runs from instrumental to affüiative and expressive purposes. This section offers an examination of the array of uses to which the traditional mass and database media have been put for accessing medical information in order to have a baseline from which to compare the Internet. This examination includes a literature review, a national survey of users, and a state-wide survey of physicians. We also take an in-depth look at the experiences of two users. Clearly, there is no "typical" user, and a case study included here is no exception. However, we believe that to understand best the context within which medical information is gathered it is useful to develop both a detailed and nuanced view via qualitative materi-

I

75

76

THE INTERNET AND HEALTH COMMUNICATION

als and a high-level view via quantitative materials. We make no claims that the groups included here for analysis are truly representative of the current situation. However, we do think there is merit in making a first approximation. Any snapshot of what users are doing in terms of the Internet must be out of date the instant it is taken. This having been said, we still see merit in investigating the situation as it was at the turn of the millennium. In some sense, human behavior changes little; hence, the experiences and observations that we are able to gain today can yield important dividends for gaining insight into the emerging situation. Napoli reviews the research on the health information-seeking process. The chapter discusses models of health information seeking and draws upon these models to suggest an analytical framework for research on the use of the Internet for health information. Aspden and Katz's chapter reports on a nationwide telephone survey of Internet users. Nearly half reported that they had looked for medical information on the Internet, and a majority of those had done so more than three times in the prior year. The results show that the Internet is not just a health information repository: It also facilitates interactions among doctors, health care professionals, and patients; and among members of support groups of those with common health interests. While these users felt that the Internet health care information was valuable, nearly a third also indicated concerns about reliability, misleading information, and commercial information. In this early stage of Internet use, many of those populations that especially need health care and medical information—the elderly and low-income families—are least likely to have ready access to online health information. They argue that health professionals and product manufacturers need to provide more advice about choosing health care Web sites; evaluating health information; and online confidentiality, privacy, and misrepresentation. Aspden, Katz, and Bemis summarize New Jersey physicians' use of and attitudes toward the Internet and their patients' use of the Internet. They identify differences between physician users and nonusers on attitudes toward patient use of the Internet, and discuss issues surrounding the accuracy of Internet health care information. Anigbogu and Rice's chapter has two parts. The first is a narrative summary of the experience of a woman with no prior computer experience as she seeks information on infertility and local infertility clinics, as the case researcher helps her with using the Internet and finding substantive information. The second is a comparison of information on infertility clinics in

Online Medical Information

77

the area found on the Internet to that found in the local telephone book. Although a chapter of this nature is somewhat novel in an academic volume, we found the contribution intriguing. Moreover, it serves as a useful antidote to some of the more glib assertions about the way computers and the Internet are being used by many segments of the population.

3 Consumer Use of Medical Information From Electronic and Paper Media A Literature Review PHILIP M. NAPOLI

M

edical practitioners and researchers recently have begun to realize the enormous potential of the Internet to drastically alter how medical information is disseminated and obtained (e.g., Chamberlain, 1994; Sonnenberg, 1997). Already, the Internet is being used to provide a variety of medical information services, including online psychological counseling (LeBourdais, 1997), health information for college students and adolescents (Cheiten & Waters, 1995; Fulop & Varzandeh, 1996), and doctorpatient interactions designed for expanding and improving managed care (Zallen, 1995). A recent count estimates that more than 800 hospitals have information available on the World Wide Web and that more than 25,000 Web sites are devoted to some aspect of health information (Izenberg & Lieberman, 1998a, p. 215). Online users appear to be taking advantage of this enormous body of information. According to Ferguson (1997), approximately 25 % of search engine queries on the Web involve people looking for information or advice on health-related matters (p. 252). This situa79

80

ONLINE MEDICAL INFORMATION

tion has led some within the medical community to pronounce that the Internet has introduced new paradigms in health care and health communication (Kassirer, 1995; Widman & Tong, 1997). These explosive developments in the ways that health information is disseminated and accessed have influenced the development of new areas of specialty in health care. At the broad level, "public health informatics" is devoted to "the science of applying Information Age technology to serve the specialized needs of public health" (Fried, Blum, & McDonald, 1995, p. 239; see also Bader & Braude's 1998 discussion of "patient informatics"). A newer, more audience-centered subspecialty, called consumer health informatics (CHI), focuses on the study of consumer interfaces in health care systems (Ferguson, 1997). CHI has recently been officially recognized by the American Medical Informatics Association as a subspecialty of medical informatics (Ferguson, 1997), indicating that the medical community recently has begun to recognize the importance of consumer-centered research in the areas of health information and new technologies. As Bader and Braude (1998) point out, "With WWW technology becoming ever more pervasive, physicians are at a disadvantage if they do not know the extent of their patients' information-seeking habits" (p. 409).

IMPORTANCE OF HEALTH INFORMATION SEEKING Unfortunately, at this point much more research attention has been paid to the capabilities and potential of these new health communication services than to how consumers actually seek health information on the Internet. As Sonnenberg (1997) has noted, "More studies are needed to define what information patients are seeking, what they find, and how it affects their health care" (p. 151). This neglect of the information-seeking process within the specific context of the Internet parallels broader patterns in health communication research. According to Johnson and Meischke (1993), "little research has focused on the receiver as an active information seeker and processor" (p. 344). Similarly, decision-making research has focused on how salient information affects judgment, to the neglect of the process of information acquisition (Pezza, 1990, p. 34). A greater understanding of the health information-seeking process on the Internet is essential for a number of reasons. First, it is important to gain a clearer understanding of how the Internet fits among the more tra-

Consumer Use of Medical Information

81

ditional sources of health information. It is always a mistake to consider a new form of communication independently of the preexisting forms of communication. Often, the various forms of communication function interdependently. A clearer understanding of these interdependencies can come from detailed analyses of how consumers use and evaluate Internetbased health information, in relation to how they use and evaluate health information obtained from more traditional sources, such as the mass media and interpersonal contact. Along related lines, it is essential to consider Internet use within the context of established theory and research on health information seeking. At this early stage in the Internet's development, it is still unclear as to the degree to which Internet use is adhering to or undermining established models or processes of information-seeking behavior. Those within the health care field have come to realize the value of accurate and easily accessible information for improving and maintaining public health (e.g., Bluming & Mittelman, 1996). However, the availability and accuracy of information alone does not guarantee that the public will make educated health care decisions. A greater understanding of the behavior of the receivers of health information, and the processes by which they receive such information, is essential for maximizing the effectiveness of information exchanges between health care providers and the general public. Thus "information-seeking, the process through which a patient actively learns about his or her disease and treatment, plays a pivotal role in how individuals successfully manage their own health care" (Turk-Charles, Meyerowitz, &Gatz, 1997, p. 85).

RELEVANCE OF THE INTERNET TO HEALTH INFORMATION SEEKING The importance of understanding the process of seeking health information becomes magnified when we consider important developments within the health care held, such as the ongoing trend toward decentralization in health information sources. Increasingly, the responsibility for health-related matters is passing to the individual, who must choose from a variety of information sources and then use the information acquired to select options for health, prevention, and treatment. The social norms that cast doctors and public health officials as the bro-

82

ONLINE MEDICAL INFORMATION kers of medical information are yielding as individuals actively seek information. (Johnson & Meischke, 1991, p. 39)

The development of the Internet certainly has the potential to accelerate this trend toward decentralization of health information (Izenberg & Lieberman, 1998b), so much so that some have speculated that the Internet may eventually supplant more traditional community health networks (Simpson, 1996), which have long been considered central to the diffusion of health information (Milio, 1996). If the benefits of such a societal transition are to outweigh the costs, it is essential that the sources of health information operate from a thorough understanding of the dynamics of the health information-seeking process. Failure to do so raises the possibility that individuals will increasingly make health care decisions using incomplete or inaccurate information and that much vital health information will be neglected or ignored. The remainder of this chapter reviews the literature on the health informationseeking process, with an emphasis on the Internet's potential role and position within this process. This chapter represents an effort to link the theory-based research that has been conducted primarily within the context of traditional media and information sources with the early research on Internet use that has, to this point, been more descriptive in nature. The result is a preliminary analytical framework (presented in the final section) that is intended to guide future research into how the Internet will fit into the array of information sources currently available and into the factors related to Internet use as a source of health information.

THE HEALTH INFORMATION-SEEKING PROCESS In assessing the potential role and function of the Internet in the realm of health communication, it is important to begin with a foundation in the dynamics of the health information-seeking process. Though the research in this area is sparse, there have been a few important efforts to develop and test theoretical models of the health information-seeking process. Before reviewing these models and their implications for research on Internet use, it is first necessary to delineate exactly what constitutes information seeking. Information seekinghas been defined as the "purposive acquisition of information from selected information carriers" (Johnson &

Consumer Use of Medical Information

83

Meischke, 1993, p. 350). Traditionally, the information-seeking process has been conceptualized according to three dimensions. These are (a) method, (b) scope, and (c) depth (see Lenz, 1984). Method refers to the information channel selected. Scope refers to the number of different elements within an information channel that are consulted. Depth refers to the extent of exposure to the specific information channel elements consulted. Thus analyses of the health information-seeking process need to explore multiple dimensions within individuals' choices and uses of particular sources of information. What, however, leads to the process of seeking health information? At the most general level, uncertainty is an important factor. Researchers have demonstrated that the degree of uncertainty about the outcome of one's activity is an important motivating factor for seeking information (see Pezza, 1990). However, not all individuals are motivated to action by such uncertainty. Those individuals with an "uncertainty orientation" (e.g., those who desire to know more and to attain clarity about themselves or their environment) will tend to seek out new information. In contrast, those individuals with a "certainty orientation" tend to "maintain clarity" by rejecting unfamiliar thoughts and ideas. Consequently, they are not moved by uncertainty to seek new information; rather, they limit their searches for information to sources that will be supportive of preexisting beliefs (Pezza, 1990). Within the context of health communication, these findings are particularly important in that they suggest that not all individuals possess an unqualified receptivity to health information, as health practitioners have frequently assumed in the past. Thus there maybe those individuals who, regardless of the nature of their information environment, or their level of knowledge about their health condition, will not engage in active searching for new health information. Consequently, failure to seek new information may not be a function of simply demographic or technological factors, which have received most of the emphasis in health communication research (e.g., Chamberlain, 1994; Pingree et al., 1996). Psychological factors may play an important part as well. Consequently, research on the use of a new technology such as the Internet should simultaneously consider demographic, technological, and psychological factors as potential predictors of Internet use. The remainder of this section outlines three distinct models of health information seeking: (a) the synergy model, (b) the comprehensive model of information seeking, and (c) the sense-making approach. As this section will illustrate, each of these models raises important issues regarding the

84

ONLINE MEDICAL INFORMATION

role and function of the Internet in the health information-seeking process.

The Synergy Model The general process of health information-seeking has been modeled in a manner that owes much to the "Two-Step Flow" research of Katz and Lazarsfeld (1955) and its subsequent applications within the context of the diffusion of innovations (Rogers, 1995). According to the synergy model developed by Schooler, Flora, and Farquahar (1993), individuals are first alerted to particular health topics by the mass media. At this stage, awareness is created and individuals are motivated to make initial behavior change attempts. However, the hurdles encountered in making such behavior changes then prompt individuals to seek further information and support. To fulfill these needs, individuals turn to their interpersonal network, where it is possible to achieve greater levels of situation specificity and immediate feedback (Johnson & Meischke, 1991, p. 38). In addition, they simultaneously consult media that provide more detail and depth regarding the particular health issue (see Schooler et a l , 1993, pp. 604-606). It is these later interactions that are generally assumed to be the cause of any ensuing behavior change. As this model suggests, certain types of mediated communication have different characteristics. Some function primarily to provide general information to large audiences, while others function along lines similar to interpersonal communication, providing more detailed information (see Hawkins, Gustafson, Chewning, Bosworth, & Day, 1987). For this reason, it is important to acknowledge a key point made by Johnson, Meischke, Grau, and Johnson (1992), that "in the field of communication, there is an institutionally supported dichotomy between interpersonal and mass communication that hinders the development of health communication theory and research" (p. 192). Clearly, certain types of media can serve very different functions. In applying this theoretical perspective to Internet use, the key question obviously revolves around determining the exact nature of audience perceptions of the characteristics of the Internet as a communication medium. These perceptions will likely determine exactly where in the chronological sequence of the synergy model the Internet is likely to fit. The unique capacity of the Internet to function as both an interpersonal and a mass medium further accentuates the need to approach the inter-

Consumer Use of Medical Information

85

personal-mass distinction as a continuum with multiple dimensions, as opposed to as a strict dichotomy.

The Comprehensive Model of Information Seeking More detailed explorations of the relevant characteristics of health information sources can be found in research by Johnson and Meischke (1993), who have developed perhaps the most extensive model of the health information-seeking process. Their comprehensive model of information seeking (CMIS) is the result of a synthesis of three theoretical perspectives pertaining to health behaviors and media use. These are (a) the health belief model; (b) uses and gratifications research; and (c) a model of media exposure and appraisal. Unlike the synergy model, which focuses only on the chronology of information seeking, the CMIS addresses the causal factors related to health information seeking and source selection. The model postulates that four health-related factors (demographics, direct experience, salience, and beliefs) determine two information carrier factors (perceptions of information carrier characteristics and utility), that, in turn, determine information-seeking actions (Johnson & Meischke, 1993). Looking first at the health-related factors, the CMIS draws heavily upon the health belief model (see Rosenstock, 1974a, 1974b) in outlining those factors likely to influence the process of selecting sources of health information. Demographic factors have frequently been associated with various types of health behavior, including information seeking. For instance, it has frequently been reported that older individuals engage in less health information seeking than younger individuals. This may be due to the fact that older individuals adhere to the more traditional perspective that physicians are the single authoritative source of medical information and health decisions (Turk-Charles et al., 1997). However, a number of studies have found that higher education levels also correlate with more information seeking (Turk-Charles et al., 1997, p. 87). The interaction between age and education (higher levels of education in younger vs. older adults) raises the possibility that age-based differentiations in levels of information seeking are actually a function of differing educational levels (TurkCharles et al., 1997). The second key health-related factor in the CMIS is direct experience. Not surprisingly, an individual's level of direct experience with a particular health issue frequently has been found to be a significant factor affecting

86

ONLINE MEDICAL INFORMATION

information seeking (Lenz, 1984). Thus individuals suffering from a particular ailment, or who are close to somebody suffering from the ailment, are much more likely to seek information about the ailment than individuals who are simply curious or concerned about contracting the ailment. The third factor, salience, refers to the "personal significance of health information to an individual and is related to the degree of perceived health threat an individual feels" (Johnson & Meischke, 1993, p. 347). Clearly, there is a significant likelihood of overlap between direct experience and salience, as direct experience with a health issue is likely to increase the salience of that issue. The final key health-related factor that affects information seeking is beliefs. According to Johnson and Meischke (1993), beliefs in this case refer to the extent to which individuals perceive they control the future or perceive that there are effective methods of prevention and control. Defined this way, the term beliefs may be a bit broad. Indeed, efficacy may be a better label for this motivational factor, as the specific belief in question focuses exclusively on the degree to which the individual is capable of improving his or her health situation. Taken together, these four factors are expected to determine the perceptions of the utility of various information sources for providing health information. Thus, according to Johnson and Meischke (1993), "in the CMIS, information seeking is driven by factors traditionally incorporated in models of health behaviors, but its exact form is shaped by information carrier factors" (p. 348). These information factors can be organized into two broad categories: (a) information carrier characteristics and (b) utility. Evaluations of information sources according to these evaluative categories are presumed to be directly linked with exposure to these sources (Johnson & Meischke, 1993). Specific information carrier characteristics investigated include editorial tone, understandability, and communication potential (which pertains to issues of style and comprehension). These factors provide a useful starting point for investigating potential factors related to Internet use, though the evaluative criteria should be expanded to include important characteristics of the Internet. Early research on the reasons for the use of online health information sources could be useful in identifying the most appropriate criteria. In a study of the use of Cleveland's Free-Net (the nation's first free, open-access computer health information system), the most commonly mentioned reason for use was that the individual did not want to bother his or her physician. Other common reasons included the desire for a speedy response, the fact that the individuals did not under-

Consumer Use of Medical Information

87

stand their physicians' answers or were too embarrassed to ask a physician in person (Hekelman, Kelly, & Grundner, 1990). Similar research by Ferguson (1997) found that individuals highly value the degree to which Internet information can be highly tailored to the individual's particular concerns and can often be obtained in a very understandable form. These results suggest that anonymity, interactivity, timeliness, and clarity could be important evaluative criteria driving Internet use for health information. Whereas information carrier characteristics involve a direct evaluation by an individual of a particular medium, utility refers to whether the information contained within the information source is important and relevant to the individual's specific needs. Thus exposure to information sources can be conceived of as the product of a combination of the needs of the receiver and the attributes of the message (Johnson & Meischke, 1993, p. 349). In their test of the CMIS, Johnson and Meischke (1993) found that the health-related factors contributed only minimally to information seeking, while the information factors had much greater explanatory power. It's important to note, however, that these results were obtained within the very specific context of the use of magazines among cancer patients. In discussing the broader implications of their results, the authors conclude that "at least for the general population sample not directly confronted by the disease, factors that account for media exposure generally may be much better explanatory variables for exposure to health-related information than more purely health-oriented variables" (pp. 358-359). This conclusion has important implications for research on the use of the Internet for health-related information. Specifically, it suggests that future research needs a strong background in—and emphasis on—the factors that affect media exposure and the choices made among alternative media technologies.

The Sense-Making Approach The sense-making approach differs from both the synergy model and the comprehensive model of information seeking in two fundamental respects. First, it is less concerned with predicting information-seeking behavior than with dissecting and evaluating the information-seeking process. Second, it advocates a departure from the traditional source-receiver conceptualization of the health information-seeking process that charac-

88

ONLINE MEDICAL INFORMATION

terizes both the synergy model and the CMIS. Developed primarily by Dervin (see 1989 for a review), the sense-making approach emphasizes that information is socially constructed. That is, one cannot assume that information (health information, in this case) has truth value. Rather, one must assume that "information is created by human observers, is inherently a product of human self-interest, and can never be separated from the observers who created it" (Dervin, 1989, p. 72). Consequently, the communication of information should be conceptualized not in terms of the transmission of information from a sender to a receiver, but in terms of a dialogue between both parties involved in the information exchange, with the institution (in this case, the health information provider) and the audience conceptualized as equal partners (Dervin & Nilan, 1986). As Dervin (1989) notes, "The sense people make of the media messages is never limited to what sources intend and is always enriched by the realities people bring to bear" (p. 72). Sense-making research has typically focused on how individuals deal with "gaps" or discontinuities in their experiences or knowledge (Dervin, 1989). Research in this vein has consistently found that information seekers generally first reach out to the closest or most convenient information sources. Issues of credibility or expertise do not factor into these initial information-seeking efforts except to the extent that they come into play if or when the individual tries to understand why the information source was not helpful (Dervin, 1989). Research by De Pietro and Clark (1984), which used the sense-making approach to study adolescents' selection of health information sources, emphasized the importance of individuals accessing a variety of different types of information sources (e.g., mass media, peers, family). Diverse "contact profiles" (as opposed to "single source" contact profiles) were found to be positively related to improved decision-making capacity and awareness of new information sources (De Pietro & Clark, 1984, pp. 426-427). These findings should curb any tendencies toward considering the Internet as an effective replacement for more traditional health information sources, as they suggest that expanding the range of types of sources that individuals consult is central to improving health decision making and knowledge. The sense-making approach's focus on how individuals cope with gaps seems particularly relevant to the role of the Internet in the health information-seeking process, given the little we know at this point about how individuals make sense of and utilize this new, complex, and unique information tool. The traditional sense-making methodology of asking respon-

Consumer Use of Medical Information

89

dents to recall and reflect upon their information-seeking process step by step has the potential to reveal important insights into the detailed dynamics of how individuals use the Internet as a health information source, and more important, the difficulties, challenges, and uncertainties that consistently affect users. Obviously, such insights can lead to improvements in how Internet-based health information sources are designed. This last point is particularly important in that it illustrates a key aspect of the sense-making approach—that it is as much about the institution providing information as it is about the audience (Dervin, 1989, p. 84). As was mentioned above, the sense-making approach conceptualizes the audience and the institution as equal partners in a dialogue. Consequently, sense-making research in health communication contexts has frequently produced recommendations for how health institutions can reconfigure their structures and processes for information provision (e.g., Dervin, Nilan, & Jacobson, 1981). In these early, essentially experimental, phases of designing Internet-based health information services, the sense-making approach could provide a useful tool for evaluating the effectiveness of—and suggesting changes to—these information services.

SOURCE SELECTION FOR HEALTH INFORMATION Each of the models discussed above focuses to some degree on the process of selecting sources of information. The next step, then, is to examine how various sources of health information are evaluated, which in turn provides an indication of how the different sources fit into the health information-seeking process. Health information sources traditionally have been divided into three categories. These are (a) formal sources, such as family doctors and other health professionals; (b) informal sources, such as friends and relatives; and (c) commercial and media sources, such as television, newspapers, and magazines. At the point of general exposure, it is perhaps not surprising that television is one of the most frequently cited sources of health information (Izenberg & Lieberman, 1998b, p. 276). Research conducted by Hofstetter, Schultze, and Mulvihill (1992) found that individuals identified television as their most common source of exposure to health information, followed closely by newspapers. Interpersonal communication placed third. Radio placed last among those media included in the questionnaire. Unfortunately, magazines were inexplicably

90

ONLINE MEDICAL INFORMATION

excluded from the questionnaire, despite their centrality as a source of health information (Wade & Schramm, 1969; Wright, 1975). Similarly, the Internet was neglected as a potential source, no doubt a reflection of the recency with which the Internet has begun to be recognized as a potentially important source of health information. When magazines are included as a potential source of health information, they generally rate higher than either television or newspapers in terms of frequency of exposure (Wade & Schramm, 1969; Wright, 1975). The key point here is that, when researchers examine health information sources at the passive level of general exposure, the traditional mass media (television and magazines, in particular) rate among the most prominent sources. However, this pattern changes significantly in contexts in which specific information is being sought. For example, among individuals seeking information regarding where to receive medical care, interpersonal sources such as physicians, friends, relatives, and neighbors were much more prominent than mass media sources (Reagan & Collins, 1987). In fact, within this specific context, television was mentioned by fewer than 1% of the respondents. However, specialized media such as flyers and newsletters figured prominently in this more directed information seeking (Reagan & Collins, 1987, p. 562), reflecting the degree to which different types of media can perform different functions. Similar research by Booth and Babchuk (1972) produced comparable results, with the authors concluding that, "Mass media appeared to be wholly inconsequential in selecting health services" (p. 97). It's also worth noting that this research found that only one in four respondents consulted a physician in his or her decision-making process, reflecting the degree to which the trend toward decentralization in health decision making has been prominent for decades (Booth & Babchuk, 1972, p. 94). Of course, tests of the information-seeking models described above also require investigation into the evaluative dimensions of various sources of health information, given that the evaluative criteria are presumed to guide source selection decisions. Research in this vein has uncovered a number of important differences between information sources. For instance, in terms of perceived accuracy, understandability, clarity of presentation, and novelty of information, magazines have rated significantly higher than either television or newspapers, with television generally following a close second and newspapers placing third (Johnson & Meischke, 1992b, p. 153). Thus, within the context of mass mediated sources of information, magazines appear to rate the highest among con-

Consumer Use of Medical Information

91

sumers of health information. Nonmediated forms of communication such as contact with physicians and health organizations generally rate higher than mass media sources across a number of key dimensions, including credibility and accuracy. However, mass mediated sources rated equally well in terms of understandability (Johnson & Meischke, 1992a). This result undermines the common assumption that interpersonal channels are generally more effective in communicating complex information than mediated channels. As Johnson and Meischke (1992a) point out, there is significant evidence that many patients have a difficult time communicating with their physicians and often do not receive the information they desire during consultation (p. 440). Other interpersonal channels, such as friends and family, were generally evaluated least positively across all six major dimensions (accuracy, credibility, understandability, clarity, utility, novelty). Clearly, each source of health information contains a unique set of strengths and weaknesses. Consequently, it is not surprising that those seeking health information generally consult multiple sources (De Pietro & Clark, 1984; Finnegan, Viswanath, Kahn, & Hannan, 1993). Related research found that individuals rate mediated sources of information as least preferable across a variety of functions (general health information, specific health information, preventative information, coping, etc.), yet these sources remained among the most prominent sources of health information (Johnson & Meischke, 1991). As Johnson and Meischke (1991) state, "the media are the source of most of the public's health information in spite of the clear preference for information from other channels" (p. 42).

Source Selection and Internet Use These results suggest the existence of an unfulfilled need in terms of available sources of health information. It is possible that the Internet may possess the unique collection of attributes necessary to fill this need. Certainly, a key question involves exactly where the Internet might fit according to these evaluative dimensions. Its hybrid nature makes it difficult to determine whether it would most closely approximate mass media, authoritative interpersonal sources (e.g., physicians), health organizations, or informal interpersonal sources (e.g., friends and family) in terms of evaluations of attributes such as credibility, clarity, and accuracy. It is also possible that the Internet would display evaluative characteristics based upon some combination of these categories. Regardless, it is safe to say

92

ONLINE MEDICAL INFORMATION

that we know relatively little at this point about its perceived strengths and weaknesses as a source of health information and how these perceived strengths and weaknesses are affecting its use. Research at this point suggests that the Internet will function primarily within more specialized contexts, serving as a tool for individuals seeking specific health information within a quasi-interpersonal setting. Research on the use of an interactive computer system designed specifically for HIVinfected people found that discussion groups were by far the most frequently used aspect of the service (Pingree et al., 1996, p. 341). These results suggest that the social support dimension of interpersonal health communication identified in Schooler et al/s (1993) model maybe central to Internet use. This emphasis on creating interpersonal networks reflects the notion of community building that has frequently been associated with sufferers of specific diseases (see Rogers et al., 1995; Scheerhorn, Warisse, & McNeills, 1995). The Internet may provide a type of intimate, supportive communication that health care providers have traditionally been unable to provide (Johnson et a l , 1992, p. 184). At the same time, the Internet can provide a level of anonymity that cannot be achieved in traditional interpersonal contexts. This combination of attributes may represent an important gap among previous modes of health communication that the Internet is now capable of filling. The vital next step, then, is to determine whether these attributes of the Internet (community, anonymity) are indeed primary motivators for its use in health contexts. Along related lines, it is important to determine the types of health information that Internet users seek and obtain. The fact that most of the little research that has been conducted to date on the use of the Internet for health information has focused on specific health contexts (e.g., HIV sufferers, hemophiliacs) makes it difficult to determine whether the average user is relying upon it for highly specific and detailed versus more general information, or to fulfill social support versus informational needs. However, within these more particularistic health contexts, the educational function of the technology has been quite prominent, if not as prominent as the social support function (Hawkins et al., 1987; Pingree et al., 1996, p. 341; Scheerhorn et a l , 1995, p. 319). In terms of evaluative dimensions, perceived credibility and accuracy of Internet information are key areas of concern. The accuracy of much of the health information available on the Internet is certainly questionable. Both the speed with which information becomes available and the absence of any centralized control mechanism increase the likelihood of in-

Consumer Use of Medical Information

93

accurate or unreliable information being made available on the Internet (Jadad & Gagliardi, 1998). A recent study concluded that only 48% of a sample of online health and medical resources was developed by credible authorities (Izenberg & Lieberman, 1998b, p. 277). As Sonnenberg (1997) points out, "Literally anyone with control of a Web site can claim to be a medical expert" (p. 151). Consequently, the dangers of medical disinformation are significant. Health information is, after all, an "experience good." That is, evaluations of its quality cannot be made until after the information has been acted upon (acting upon poor-quality health information, and consequently learning of its poor quality, can of course be harmful or fatal). Those sites that claim to provide objective evaluations of the credibility and accuracy of other health information sites generally lack any systematic evaluation criteria (Jadad & Gagliardi, 1998). Thus the health site ratings services essentially suffer from the same credibility problem as many of the sites they claim to evaluate. Within this environment of enormous amounts of available information, with few mechanisms for verifying its accuracy, it becomes increasingly important to determine the degree to which consumers of health information available on the Internet are cognizant of such credibility issues, and the degree to which evaluations of credibility and accuracy factor into their decision to seek information via the Internet and their processing of the information retrieved. Unfortunately, there has been little research to this point that has investigated how consumers evaluate the credibility of health information on the Internet. Some researchers have suggested that individuals are capable of accurately choosing credible online sources of health information (Widman & Tong, 1997), though the empirical basis for these conclusions is tenuous, at best (Sonnenberg, 1997). Future research needs to provide a clearer picture of how the Internet compares to more traditional sources of health information, not only in terms of credibility, but across other evaluative dimensions as well.

TOWARD AN ANALYTICAL FRAMEWORK FOR THE STUDY OF INTERNET USE The descriptive data gathered thus far, along with the theoretical models that have been developed, provide a useful starting point for developing an analytical framework for investigating the process by which individuals

94

ONLINE MEDICAL INFORMATION

consult the Internet for health information. Specifically, future analyses must investigate the chronology of Internet use and the psychology of Internet use. Chronology, in this case, refers to the point in the health information-seeking process at which Internet use occurs. That is, does Internet use typically occur after exposure to health information from other sources (mass media, friends, physicians, etc.), or does it precede, and perhaps promote, exposure to other sources? Psychology, in this case, refers to the user factors that promote Internet use. Are there particular perceptions of the Internet that seem to promote its use? Are there particular personality or demographic types that exhibit a much greater reliance on the Internet? It seems likely that particular user characteristics will correlate with preferences for particular information source characteristics, which, in turn, will lead to choices of particular information channels. In terms of user characteristics, the key factors that need to be explored include technological sophistication; uncertainty orientation; demographic factors such as age, gender, education, and ethnicity; and health-related factors such as current health status, range and intensity of health concerns, knowledge and interest in health issues, and exposure to health information from other sources. Recurrent perceptual attributes that likely factor into the decision of whether or not to use the Internet include its level of convenience, anonymity, reliability, interactivity, and asynchronicity. Certainly, it is important to determine whether information-related factors are more powerful predictors of Internet use than health-related factors, as they were in Johnson and Meischke's (1993) study of cancer patients' use of magazines. Such insights will be useful in determining whether the current structure and characteristics of the Internet are serving as impediments to its use as a health information source or whether simply expanding the availability of the Internet will have substantial effects on the levels of health information individuals obtain. In many ways, this analytical framework owes much to Webster and Phalen's (1997) audience exposure model, in which audience exposure is seen as determined by a combination of media factors (factors pertaining to the characteristics of the media available) and audience factors (factors pertaining to the characteristics of the audience members). The Webster and Phalen model has been developed and applied primarily within the context of television viewing, though the authors suggest its potential utility in other media contexts. Of course, differentiating audience and media factors becomes a bit difficult within the context of Internet use for health

Consumer Use of Medical Information information, given that audience perceptions of the characteristics of medium are presumed to be an important factor. These perceptions media factors in that they relate to the characteristics of the medium, they are also audience factors in that their perceptual nature raises possibility that they can vary across individual audience members.

95 the are yet the

CONCLUSION This chapter has reviewed the research on the health information-seeking process and has used this research to outline a preliminary analytical framework for investigating the use of the Internet for health information. The descriptive data gathered thus far regarding the extent of Internet use for health information and the reasons for such use, along with models of information seeking and audience exposure that have been developed within other media contexts, provide a useful starting point for theoretically grounded investigations into the nature and dynamics of the use of the Internet for health information.

TRENDS AND IMPLICATIONS FOR THE NEAR-TERM FUTURE The Internet's role in the provision of health care continues to evolve. Perhaps the most important step in its recent evolution is its expansion from a source of health information to an increasingly comprehensive source for purchasing health-related products and services. Analysts of the " e health" industry have generally utilized separate "Content" and "Commerce" labels when classifying health-related Web sites; however, such distinctions are beginning to blur (Bickert, 1999), particularly as commerce becomes the engine driving Web-based content providers across a variety of fields. Future research on the role of the Internet in health information seeking therefore needs to become integrated with research on the process of online purchasing of health products and services. Such an integration will allow for a more complete understanding of the Internet's unique functionality in the health media mix.

96

ONLINE MEDICAL INFORMATION

REFERENCES Bader, S.A., & Braude, R. (1998)." Patient Informatics": Creating new partnerships in medical decision making. Academic Medicine, 73(4), 408-411. Bickert, M. (1999, July). The impact ofecommerce on legacy health-care companies [Online]. Available: http://www.cyberdialogue.com Bluming, Α., & Mittelman, P. S. (1996). Los Angeles Free-Net: An experiment in interactive telecommunication between lay members of the Los Angeles community and health care experts. Bulletin of the Medical Library Association, 84(2), 217-222. Booth, Α., & Babchuk, N. (1972). Seeking health care from new resources. Journal of Health and Social Behavior, 13, 90-99. Chamberlain, M. A. (1994). New technologies in health communication: Progress or panacea? American Behavioral Scientist, 38(2), 271-284. Cheiten, S., & Waters, M. (1995). Comprehensive school health education and interactive media. In L. M. Harris (Ed.), Health and the new media: Technologies transforming personal and public health (pp. 145-162). Mahwah, NJ: Lawrence Erlbaum. De Pietro, R., & Clark, Ν. M. (1984). A sense-making approach to understanding adolescents' selection of health information sources. Health Education Quarterly, 11(4), 419-430. Dervin, B. (1989). Audience as listener and learner, teacher and confidante: The sense-making approach. In R. E. Rice &: C. K. Atkin (Eds.), Public communication campaigns (2nd ed., pp. 67-86). Newbury Park, CA: Sage. Dervin, B., & Nilan, M. (1986). Information needs and uses: A conceptual and methodological review. Annual review of information science and technology (Vol. 21, pp. 3-33). White Plains, NY: Knowledge Industries. Dervin, B., Nilan, M., & Jacobson, Τ. (1981). Improving predictions of information use: A comparison of predictor types in a health communication setting. In M. Burgoon (Ed.), Communication yearbook, 5. New Brunswick, NJ: Transaction. Ferguson, T. (1997). Health online and the empowered medical consumer. Jot η t Committee Journal on Quality Improvement, 23(5), 251-257. Finnegan, J. R., Jr., Viswanath, K., Kahn, Ε., & Hannan, P. (1993). Exposure to sources of heart disease prevention information: Community type and social group differences. Journalism Quarterly, 70(3), 569-584. Fried, Α., Blum, H. L., & McDonald, M. (1995). Public health informatics: How information-age technology can strengthen public health. American Review of Public Health, 16, 239-252. Fulop, M. P., & Varzandeh, Ν. N. (1996). The role of computer-based resources in health promotion and disease prevention: Implications for college health. Journal of American College Health, 45(1), 11-17. Hawkins, R. P., Gustafson, D. H., Chewning, B., Bosworth, K, & Day, P. M. (1987). Reaching hard-to-reach populations: Interactive computer programs as public information campaigns for adolescents. Journal of Communication, 37(2), 8-28. Hekelman, F. P., Kelly, R., & Grundner, Τ. Μ. (1990). Computerized health information networks: House calls of the future? Family Medicine, 22(5), 392-395.

Consumer Use of Medical Information

97

Hofstetten C. R., Schultze, W. Α., & Mulvihill, Μ. M. (1992). Communications media, public health, and public affairs: Exposure in a multimedia community. Health Communication, 4(4), 259-271. Izenberg, N., & Lieberman, D. A. (1998a). The Web, communication trends, and children's health—Fart 2: The Web and the practice of pediatrics. Clinical Pediatrics, 37(4), 215-221. Izenberg, N., & Lieberman, D. A. (1998b). The Web, communication trends, and children's health—Part 3: The Web and health consumers. Clinical Pediatrics, 37(5), 275-285. Jadad, A. R , & Gagliardi, A. (1998). Rating health infonnation on the Internet: Navigating to knowledge or to Babel? Journal of the American Medical Association, 279(8), 611-614. Johnson, J. D., & Meischke, H. (1991). Cancer infonnation: Women's sources and content preferences. Journal of Health Care Marketing, 11(1), 37-44. Johnson, J. D., & Meischke, H. (1992a). Differences in evaluations of communication channels for cancer-related information. Journal of Behavioral Medicine, 15(5), 429-445. Johnson, J. D., & Meischke, H. (1992b). Mass media channels: Women's evaluations for cancer-related infonnation. Newspaper Research Journal, 23(1-2), 146159. Johnson, J. D., & Meischke, H. (1993). A comprehensive model of cancer-related infonnation seeking applied to magazines. Human Communication Research, 19(3), 343-367. Johnson, J. D., Meischke, H., Grau, J., & Johnson, S. (1992). Cancer-related channel selection. Health Communication, 4(3), 183-196. Kassirer, J. P. (1995). The next transformation in the delivery of health care. New England Journal of Medicine, 332(1), 52-54. Katz, E., & Lazarsfeld, P. (1955). Personal influence: The part played by people in the flow of mass communication. New York: Free Press. LeBourdais, E. (1997). When medicine moves to the Internet, its legal issues tag along. Canadian Medical Association Journal, 257(10), 1431-1433. Lenz, Ε. R. (1984). Information seeking: A component of client decisions and health behavior. Advances in Nursing Science, 6, 59-72. Milio, N. (1996). Electronic networks, community intermediaries, and the public's health. Bulletin of the Medical Library Association, 84(2), 223-228. Pezza, P. E. (1990). Orientation to uncertainty and infonnation seeking about personal health. Health Education, 21(2), 34-36, 49. Pingree, S., Hawkins, R. P., Gustafson, E., Böberg, Ε. Β., Wise, Μ., Berhe, Η., & Hsu, Ε. (1996). Will the disadvantaged ride the infonnation highway? Hopeful answers from a computer-based health crisis system. Journal of Broadcasting & Electronic Media, 40(3), 331-353. Reagan, J., & Collins, J. (1987). Sources for health care infonnation in two small communities. Journalism Quarterly, 64(2-3), 560-563, 676. Rogers, Ε. M. (1995). The diffusion of innovations (4th ed.). New York: Free Press.

98

ONLINE MEDICAL INFORMATION

Rogers, Ε. Μ., Dearing, J. W., Rao, Ν., Campo, S., Meyer, G., Betts, J. R, & Casey, M. K. (1995). Communication and community in a city under siege: The AIDS epidemic in San Francisco. Communication Research, 22(6), 664-678. Rosenstock, I. M. (1974a). The health belief model and preventive health behavior. In Μ. H. Becker (Ed.), The health belief model and personal health behavior (pp. 2759). Thorofare, NJ: Charles B. Slack. Rosenstock, I. M. (1974b). Historical origins of the health belief model. In Μ. H. Becker (Ed.), The health belief model and personal health behavior (pp. 1-8). Thorofare, NJ: Charles B. Slack. Scheerhorn, D., Warisse, J., & McNeilis, K. S. (1995). Computer-based telecommunication among an illness-related community: Design, delivery, early use, and the functions of HIGHnet. Health Communication, 7(4), 301-325. Schooler, C, Flora, J. Α., & Farquahar, J. W. (1993). Moving toward synergy: Media supplementation in the Stanford Five-City Project. Communication Research, 20(4), 587-610. Simpson, R. L. (1996). Will the Internet supplant community health networks? Nursing Management, 27(2), 20, 23. Sonnenberg, F. Α. (1997). Health information on the Internet: Opportunities and pitfalls. Archives of Internal Medicine, 157(2), 151-152. Turk-Charles, S., Meyerowitz, Β. E., & Gatz, M. (1997). Age differences in information-seeking among cancer patients. International Journal of Aging and Human Development, 45(2), 85-98. Wade, S., & Schramm, W. (1969). The mass media as sources of public affairs, science, and health knowledge. Public Opinion Quarterly, 33(2), 197-209. Webster, J. G., & Phalen, P. F. (1997). The mass audience: Rediscovering the dominant model. Mahwah, NJ: Lawrence Erlbaum. Widman, L. E., Tong, D. A. (1997). Requests for medical advice from patients and families to health care providers who publish on the World Wide Web. Archives of Internal Medicine, 157(2), 209-212. Wright, W. R. (1975). Mass media as sources of medical information. Journal of Communication, 25(3), 171-173. Zallen, B. G. (1995). Member-centered managed care and the new media. In L. M. Harris (Ed.), Health and the new media: Technologies transforming personal and public health (pp. 21-43). Hillsdale, NJ: Lawrence Erlbaum.

4 Assessments of Quality of Health Care Information and Referrals to Physicians Λ Nationwide Survey PHILIP ASPDEN JAMES Ε. KATZ With research assistance from Ann Bemis

T

he Internet has made massive amounts of information and advice available to the more than 50% of the U.S. population that uses the technology. Much of these vast data are concerned with health care and medical information. Anecdotal and longitudinal survey evidence suggest that people are increasingly turning to this store of health care information (Cyber Dialogue, 1999). Indeed, for some medical practitioners, it is now almost a daily occurrence for a patient to bring to a consultation information derived from the Internet ("Health Information on the Internet," 1999).

7

AUTHORS NOTE: The preparation of this chapter was partially funded by the Robert Wood Johnson Foundation. Thanks are also due to Philip Napoli for helping with the literature search.

99

100

ONLINE MEDICAL INFORMATION

This emerging source of health care information could have a transformative effect on the delivery of health care. As patients become better informed via the Internet, medical professionals cease to be the main source of health care information. As a result, some doctors already believe that the way doctors practice medicine may have to change to respond to these changing circumstances (George, 1997). But physicians worry about the quality of information available from the Internet. Many doctors believe the information is often misleading and sometimes harmful. Many Web sites are not subject to any editorial control, a traditional source of quality assurance. Further, many Web sites are provided by commercial organizations that naturally want to show their products and services in the best light. These concerns about quality have motivated some doctors to call for rating systems to help consumers distinguish the good advice from the unsound (Payton, 1997). Academic research on Internet use for accessing medical information has, so far, been largely descriptive. Researchers have devoted particular attention to describing the sources of Internet health information available (e.g., Ferguson, 1996) and the processes by which this information is being made available to the public (e.g., Hersh, Brown, Donohoe, Campbell, & Horacek, 1996; Redman et al., 1997). Those studies that have focused on users have been largely descriptive as well (e.g., Marshall, 1989; Wood, Wallingford, & Siegel, 1997). For example, Ferguson (1997) found that approximately 25% of search engine queries on the Web involve people looking for information or advice on health-related matters. Against this background of anecdotal evidence and descriptive research, we conducted a national telephone survey to gain some empirical understanding of the extent to which Americans are obtaining and using health care information from the Internet. Specifically, the survey sought to answer the following questions:

1. To what extent are Internet users accessing medical information via the Internet? What are the characteristics of users who do this most frequently? 2. To what extent is the health care information derived from the Internet discussed with others, including medical professionals? 3. What do people think about the quality of medical information derived from the Internet?

Quality of Health Care Information

101

METHOD Data for our study were taken from a November 1997 national random telephone sample, surveyed by a commercial firm working under our direction. The survey yielded 2,148 respondents, of whom 30% reported being Internet users, 10% reported being former Internet users, 5 0 % reported being aware of the Internet but never being users, and, finally, 10% reported not having heard of the Internet. The sample of Internet users was augmented by a second national sample of 153 Internet users. In total, we surveyed 800 Internet users on their use of and attitudes toward health care information derived from the Internet. In addition, all respondents were asked about social and demographic characteristics.

RESULTS Substantial Minority of Internet Users Access Health Care Information In answer to the question, "Have you ever looked for medical information on the Internet?" a substantial minority, 4 1 % of our 800-strong sample, reported that they had. We sought to characterize those more likely to access health care information using a multiple linear regression model with propensity to access medical information as the dependent variable and several demographic and Internet usage measures as independent variables: age, race (white or nonwhite), gender, highest educational level achieved, and reported household income. The Internet usage variables were self-assessed Internet skill level (novice, average, above average, excellent), year started using the Internet, and hours online in the week previous to survey. Table 4.1 provides the results of our regression analyses. Those likely to access health care information were (in decreasing order of importance) more likely to have higher Internet skills, be female, be white, be older, and be greater Internet users.

Those Who Access Health Care Information Generally Do so Several Times For those who reported accessing health care information via the Internet, we asked them how many times they did this in the year prior to

102

ONLINE MEDICAL INFORMATION

Predicting Propensity to Access Health Care Information, Frequency of Access of Health Care Information, Propensity to Discuss Health Care Information with Others, and Skepticism Predictor

Access

Frequency

Discuss

Skepticism

Demographics Age

-.04*

.07

-.02

-.08

Race

-.12*

.34

-.08

-.34

Gender

-.12**

.40**

-.12*

-.16

Education

-.01

.06

-.01

-.16***

Income

-.04

-.04

-.09**

-.15

-.19*

Usage Internet skill level

.20*

-.08*

.01

.01

.02

.03

-.04*

.00

-.05

-.06

Year started Internet hours online R F Ν

.13

2

13.6*** 674

.04

.08 3 g***

2.3* 274

277

.10 4 7*** 277

* p < .05; * * p < .01; * * * ρ < .001 N O T E : Values are standardized beta coefficients.

the interview. A majority of such respondents reported accessing health care information three or more times. Specifically, 39% reported doing so once or twice, 22% three or four times, 16% five to nine times, and 23% 10 or more times. Those with higher frequencies of access were more likely to be female and to have higher Internet skills.

Most Discussed Health Care Information With Others; Many Discussed With Medics Of those who reported accessing Internet health care information, 73% (242 out of 329) reported discussing the information with others. Of these, 50% reported discussing the information with doctors or other health professionals. Thus, of those who accessed health care information via the Internet, 37% (122 out of 329) reported discussing what they had read with health care professionals. Those more likely to discuss health care information with others were more likely to be more affluent, have higher Internet skills, and be female. Strangely, we could find no demographic or

Quality of Health Care Information

103

Internet usage predictors of the propensity to discuss health care information with health care professionals.

Most Believe Internet Information to Be Valuable, but Many Are Wary In order to measure respondents' belief in the value of information gained from the Internet, we asked them, "Thinking only of the last time you read medical information on the Internet, how would you rate the information?" Thirty-three percent reported it was highly valuable, 6 1 % somewhat valuable, and 5% no value at all. Thus the vast majority believed the information was somewhat valuable, and a third believed it was highly valuable. Here again, we were not able to identify any demographic or Internet usage factors as predictors of belief about the value of the information. We found that this belief in the value of health care information derived from the Internet is tempered by considerable wariness on the part of respondents. Of those who had accessed information, 24% reported reading unreliable material, 27% reported reading misleading information, and 3 1 % reported reading information they thought was too commercial. The results of these three questions were highly correlated—those who had read unreliable material were more likely to have read misleading information and to have found the information too commercial. From the answers (yes = 1, no = 2) to these three questions, we created a skepticism scale by adding for each respondent the numerical scores for all three questions (Cronbach's alpha = .74). Thus a respondent with a score of 3 would be highly skeptical, and a respondent with a score of 6 would not be skeptical at all. Those who reported being more skeptical were likely to be better educated and to have higher Internet skills. We also examined whether skepticism related to some of the other Internet health information variables we have been discussing. We found that those with greater skepticism accessed health care information more frequently and were more likely to discuss health care information gathered online with other people. However, skepticism levels did not appear to be statistically related to the propensity to discuss Internet information with health care professionals or with beliefs about the value of health care information obtained during the most recent occasion health care information was accessed.

104

ONLINE MEDICAL INFORMATION

DISCUSSION In summary, we found that a large proportion of Internet users access health care information via the Internet. Many did this several times in the year prior to the survey. The Internet is clearly now an important source of health information, competing with other media sources, such as television, magazines, and newspapers, and formal (medical professionals) and informal (friends and relatives) sources. But the Internet is not just a source of health information. It is a medium that facilitates interactions between doctors and patients, for example, via Web sites and e-mail, and also interactions between members of groups with common health interests (e.g., diabetics and family members responsible for caring for diabetics) via chat rooms or bulletin boards. The health impacts of these interactive activities need to be calibrated and evaluated. Propensity to access health care information via the Internet and frequency of access appear related to self-assessed Internet skills. Those with higher Internet skills are presumably better able to navigate the Internet and seek out health care information when needed. Further, women are more likely than men, controlling for Internet skill level, to access health care information on the Internet, and to do so more frequently. No doubt, this reflects women's greater tendency to receive health care and their greater involvement in the health care of close relatives (children, parents). In addition, older people are more likely to access health care information, reflecting the greater need for health care as one gets older. It is perhaps not surprising that Internet skill levels appear to be an important indicator of accessing health care information. The Internet is a much more complicated and less familiar medium to navigate than traditional media such as magazines, newspapers, and television. Internet users need to master the technology, including search engines, to identify Web sites, to find the useful material within Web sites, to navigate links between Web sites, and to retrace steps to pursue other avenues of information. Regarding the extent Internet users share health care information with other people, we found that almost three out of four of those who accessed health care information discussed the information with others. More than a third of those who accessed health care information discussed this information with a doctor or other health care professional. It will be interesting to investigate, through future studies, how the Internet compares with

Quality of Health Care Information

105

other media in the way health information is shared with others and used to influence behavior. Our results show also that although respondents generally appear to find Internet health care information valuable, this belief is tempered by wariness about misleading, unreliable, and too commercially focused information. This skepticism could well increase as Internet surfing expertise improves. To counteract this skepticism, health care professionals and product manufacturers should be ready to advise the public on recommended health care Web sites and to caution the public about acting on information derived from the Internet without taking appropriate professional advice. In addition, health care professionals and product manufacturers should seek to protect themselves from the consequences of misrepresentation on the Internet. Lastly, it should be recognized that Internet users are still a minority of the general public, with certain groups overrepresented as compared to the general population. For instance, users are more likely to be younger, more affluent, better educated, and white (Katz & Aspden, 1997). From the health care perspective, many of the most vulnerable—the elderly and low-income families—are underrepresented among the population of Internet users. Ways need to be found to extend Internet access to these needy groups.

REFERENCES Cyber Dialogue. (1999). Impacts of the Internet on the doctor-patient relationship: The use of the Internet health consumer (Survey report, 1999 based on the American Internet User Survey; New York) [Online]. Available: http://cyberdialogue.com/ pdf s/wp/wp - cch -1999- doctors. ρ df Ferguson, T. (1996). Health online. Reading, MA: Addison-Wesley. Ferguson, T. (1997). Health online and the empowered medical consumer. Joint Committee Journal on Quality Improvement, 23(5), 251-257. George, J. (1997, December 15). Web med: Doctors adapt as more patients use the Net to learn about illnesses, but some worry about misdiagnoses. Philadelphia Business Journal [Online]. Available: http://www.amcity.com:80/philadelphia/ stories/121597/f ocusl .html Health infonnation on the Internet. (1999, February 16). Washington Post, p. Z17. Hersh, W. R., Brown, K. E., Donohoe, L. C, Campbell, E., M, & Horacek, A. E. (1996). CliniWeb: Managing clinical information on the World Wide Web. Journal of the American Medical Informatics Association, 3(4), 273-280.

106

ONLINE MEDICAL INFORMATION

Katz, J., & Aspden, P. (1997). Barriens to and motivations for using the Internet: Results of a national opinion survey. Internet Research Journal: Technology, Policy & Applications, 7(3), 170-188. Marshall, J. G. (1989). Characteristics of early adopters of end-user online searching in the health professions. Bulletin of the Medical Library Association, 77(1), 4855. Payton, D. (1997, May 1). Take medical sites with a dose of salt. Philadelphia Inquirer. Redman, P. M., Kelly, J. Α., Albright, E. D., Anderson, P. F., Mulder, C, & Schnell, Ε. Η. (1997). Common ground: The HealthWeb project as a model for Internet collaboration. Bulletin of the Medical Library Association, 55(4), 325-330. Wood, F. B., Wallingford, Κ. T, & Siegel, Ε. R. (1997). Transitioning to the Internet: Results of a National Library of Medicine user survey. Bulletin of the Medical Library Association, 85(4), 331-340.

5 Use of the Internet for Professional Purposes A Survey of New Jersey Physicians PHILIP ASPDEN JAMES Ε. KATZ ANN BEMIS

E-mail your doctor any questions before your appointment. "The Smart List: How You Can Get Wiser in 20 Minutes." Ladies Home Journal, March 2000

T

he above captioned advice might be light-years from the characteristic tips concerning canning fruit and getting stains out of blouses for which in the 1910s this enduring women's magazine was best known. But the advice proffered at the turn of this millennium might be as useless to many women (or men) today as it would have been to the Journals readers a century earlier. Of course we can expect that the situation will change. 7

AUTHORS NOTE: The preparation of this chapter was partially funded by the Robert Wood Johnson Foundation. Thanks are also due to Mary J. Deering (Department of Health and Human Services) and the New Jersey Medical Society for contributions to the questionnaire design and for supplying the survey sample.

107

108

ONLINE MEDICAL INFORMATION

But currently, despite its substantial and growing importance for millions of Americans as a health care information source (Aspden & Katz, Chapter 4, this volume), the Internet remains largely noninteractive when it comes to communication between patient and doctor (Eysenbach & Diepgen, 1999). This is one area where there is great disparity between what patients seek from their doctors and what the doctors seem willing to make available to them. Indeed, a major national study of Internet users, conducted in the summer of 1999, showed that a vanishingly small proportion—just 4%—of the Internet health care-using public had ever had email contact with their physicians (Cyber Dialogue, 1999). It has often been said that doctors lag behind many other professional groups in using e-mail and adopting computer technologies, and more specifically, the Internet (American Medical Association, 1998; Susman, 1997). However, this situation clearly seems to have changed. For although in 1996 only 15% of physicians were using the Internet for clinical reasons, by early 1999 this number had surged to 70%, according to a corporatesponsored study (Health Management Technology, 1999). These physicians also seem to be healthy users of the Internet for e-mail. Apparently there is substantial disparity between the way physicians might use the Internet and the way their patients, both existing and potential, do. Hence we thought it would be useful to understand in a systematic way what the usage patterns and interest levels are of physicians. This is the genesis of the study reported here. It had two immediate sources. As part of our research program on the socioeconomic impact of the Internet, we have investigated the extent Internet users turned to the Internet for health care information (Aspden & Katz, Chapter 4, this volume). A late1997 survey showed that a large minority of Internet users (41%) had turned to the Internet for health care information. Of those who had accessed health care information, more than a third had discussed the information with health care professionals. However, of those who had accessed information, at least a quarter were wary about the information they had accessed, with 24% reporting unreliable health care information, 27% reporting misleading health care information, and 3 1 % reporting information they thought was too commercial. These results stimulated us to examine the situation from the physician's perspective. The second impetus for the work reported in this chapter came from Ann Bemis, a doctoral student at Rutgers University in fall 1997, who carried out a survey of physicians' use of the Internet for professional purposes. The intriguing results from a small sample of physicians from a New Jersey teaching hospital encouraged the three authors of this chapter to extend the survey objectives in both scope and size, and attempt to:

The Internet for Professional Purposes

109

1. Assess the level of physician usage of the Internet for professional purposes 2. Determine physicians' attitudes toward the Internet and its impact on their work 3. Identify differences between users and nonusers in their attitudes toward patient use of Internet 4. Focus on issues surrounding the accuracy of Internet health care information

METHOD In late 1998 we contacted by mail a sample of New Jersey physicians drawn from the membership of New Jersey Medical Society. We enclosed a fourpage survey instrument and a stamped self-addressed envelope for returning the completed survey. The survey instrument was reviewed by and contained questions suggested by the New Jersey Medical Society and the Department of Health and Human Services, Washington, D.C. We received an acceptable response: 220 completed questionnaires out of 801, a 27.5% response rate. Our questionnaire was short and thus contained few questions to characterize the survey respondents. We asked for the respondent's age, gender, and main area of specialization from a list of 32 different specialties. Taking the respondents as a group: 52% were under 51 years of age, 88% were male, and the most frequently reported specialty was internal medicine (15% of respondents), followed by obstetrics/ gynecology (8%), ophthalmology (8%), surgery (8%), pediatrics (7%), and family practice/general practice (7%). These results seemed to be an accurate reflection of the characteristics of the overall membership of the New Jersey Medical Society.

RESULTS More Than Three Out of Five Use the Internet for Professional Purposes Sixty-two percent of respondents used the Internet for professional purposes. Users were significantly younger than nonusers. Sixty-two percent of users were aged up to 50 years old, while only 36% of nonusers

110

ONLINE MEDICAL INFORMATION

Table 5.1 Age

Number of Internet Users by Age User

Nonuser

Total

< = 50

84

30

114

>51

51

54

105

Total

135

84

219

N O T E : Users younger than nonusers ( χ = 15.2, 1 df, ρ < .001) 2

were aged up to 50 (see Table 5.1). We found no difference in usage rates based on gender.

Physician Internet Usage Is Multisourced and Multipurposed For Internet users, we provided respondents with a list of types of Internet information and encouraged them to check all the types they used. On average, respondents checked almost three sources. Bibliographic databases (83%) and medical journals (57%) were the most frequently accessed sources, followed by e-mail (55 % ) , government information (36%), information from other providers (21%), the Web site of the Medical Society of New Jersey (15%), discussion groups (12%), and other sources (11%) (n = 133). Regarding the professional purposes for using the Internet, we provided respondents with a list of purposes and again encouraged them to check all the purposes that applied to them. On average, respondents checked more than 2.5 purposes. Clinical care (73%) and current awareness browsing (70%) were the most frequently cited purposes, followed by research purposes (46%), teaching purposes (26%), administration purposes (22%), reading for a course of study (14%), and other purposes (10%) (n = 132).

For Physician Users, the Internet Is Having an Important Impact on Work In terms of the impact of the Internet on work, it would appear that the availability of the Internet has altered communication patterns with both colleagues and patients. Thirty-one percent of users reported that the Internet had altered communication patterns with other professionals, while 18% had used e-mail to communicate with patients.

The Internet for Professional Purposes

111

Most Recent Patient With Internet Information Last l-Net

Patient

User

NonUser

Total

Within week

47

Week-month

33

23

56

Month-1998

34

28

62

Before 1998/never Total

13

60

9

14

23

123

78

201

N O T E : Users more recent than nonusers ( χ = 12.8, 3 df, ρ < .005). 2

To determine the overall impact of the Internet, we asked users to answer the question, "What has been the impact of the Internet on your work?" on the basis of a 5-point scale. One fifth reported that the Internet had a significant (15%) or major (6%) impact, half reported that the Internet had a moderate (29%) or better impact, while half (48%) reported a slight effect, and 2% indicated no effect (n = 134).

Professional Interaction With Patients With Internet Information Is More Productive We next explored the impact on patient-physician interactions of Internet information gathered by the patient. Almost three out of five physicians in our survey (both users and nonusers) reported seeing a patient within the past month who discussed with the physician information gathered from the Internet. As Table 5.2 shows, for users (65%), the occurrence was more frequent than for nonusers (46%). In terms of the impact of Internet information on the patient-physician interactions, 33% of physicians believed that the interactions were more effective, while only 14% believed them to be less productive. Interestingly, there was no significant difference between the opinions of users and nonusers.

Physicians With Major Concerns About Accuracy of Internet Health Information Our earlier survey indicated that at least a quarter of the public had concerns about the quality of Internet health care information. As expected, we found that physicians also had major concerns about the trustworthy-

112

ONLINE MEDICAL INFORMATION

ness of Internet health care information. In response to the statement, "Too much information from the Internet is untrustworthy," 5 8 % of respondents agreed or strongly agreed, while only 8% disagreed or strongly disagreed. There was no statistical difference between the views of users and nonusers. We then asked users to identify the way they checked the accuracy of Internet health care information. In the questionnaire we provided a list of ways and invited respondents to check all that applied. On average, users checked two primary approaches: using one's own judgment (78%) and reading peer-reviewed sites (46%). Of the remaining approaches, 3 6 % discussed the content with colleagues, 2 1 % checked with a professional organization, only 8% reported calling or e-mailing the source of the information, and 5% indicated other ways (n = 132). We next asked both users and nonusers how Web site quality should be promoted. We again provided a list of options. More than two thirds (69%) of the respondents believed that Web site quality should be promoted through accreditation by a professional entity (no significant difference between users and nonusers). Two fifths (41%) believed that quality should be promoted through a clearinghouse providing expert opinion (again, no significant difference between users and nonusers). These were followed by 2 1 % supporting a rating system provided by a consumer watch organization, 9% through federal regulation, 8% by other means, and 6% by federal nonregulatory oversight (n = 202). Taking the two most frequently cited approaches together—accreditation by a professional entity and the availability of a clearinghouse providing expert opinion—23% of respondents indicated the use of both, 6 5 % one or the other approach, and 12% neither approach.

Physicians Believe They Should Be Proactive About Patient Use of Internet Information We next asked a series of questions about physician attitudes toward patients' use of Internet health care information. Overall, the responses to these questions indicated that physicians (both users and nonusers) believe that they should be proactive about advising patients about their use of Internet health care information. In some situations, physician users want to be significantly more proactive than nonusers.

The Internet for Professional Purposes

>';*f^/^ S;?:^| ;

113

Physicians' Opinions About Responses to Patient's Use of the Internet

What Is Your View About the Following Statements? "Physicians should . . ." ". . . recommend relevant Internet sites that they know are accurate." Recommend

Sites

Agree/Strongly agree Neutral

User

Ν on user

Total

102

49

151

27

22

49

3

6

9

132

77

209

Disagree/Strongly disagree Total

Users agree more strongly (χ = 4 . 3 , 1 df ρ < .05). 2

". . . discourage the patient from using the Internet as a source of medical information." Discourage

User

Patient

Agree/Strongly agree Neutral Disagree/Strongly disagree Total

Ν on user

Total

8

10

18

27

32

59

98

39

137

133

81

214

Users disagree more strongly ( χ = 13.8, 2 df, ρ < .001). 2

". . . remain neutral about information the patient obtains from the Internet." Remain

Neutral

User

Ν on user

Total

Agree/Strongly agree

16

14

30

Neutral

27

33

60

90

33

123

133

80

213

Disagree/Strongly disagree Total

Users disagree more strongly ( χ = 15.4, 2 df, ρ < .001). 2

The format of this series of questions was the same. Respondents were given a statement and asked to indicate which of "strongly agree/ "agree," "neutral/' "disagree" or "strongly disagree" best described their opinion regarding the statement. As Table 5.3 shows, about three quarters of respondents agreed or strongly agreed with the statement, "Physician should recommend relevant Internet sites that he/she knows are accurate." Users agreed more strongly with this statement than nonusers. Sev7

114

ONLINE MEDICAL INFORMATION

enty-seven percent of users agreed or strongly agreed with the statement as compared to 64% of nonusers. We posed the above statement in a negative way: "Physicians should discourage the patient from using the Internet as a source of medical information." Overall, about 6 0 % of respondents disagreed or strongly disagreed with this statement (see Table 5.3), but, as before, there was a major difference between the views of users and nonusers. Seventy-four percent of users disagreed or strongly disagreed with this statement, but only 48% of nonusers did so. Forty percent of nonusers were neutral about the statement. Finally, we posed a statement similar to the above in a neutral way: "Physicians should remain neutral about information the patient obtains from the Internet." About 60% of respondents disagreed or strongly disagreed with this statement (see Table 5.3). There was, again, a major difference between the views of users and nonusers. Sixty-eight percent of users disagreed or strongly disagreed with this statement. Of nonusers, 4 1 % disagreed or strongly disagreed with the statement and another 4 1 % were neutral about the statement. Taking the responses from these three questions together, the results of our survey indicate that users are strongly supportive of patients' using Internet health care information, with about three quarters making supportive statements about the use of Internet information by patients. Nonusers were also supportive of patients using Internet health care information, but much less so than users. About half of the nonusers in our sample were strongly supportive of use of Internet information by patients, and a third were neutral. Next we asked respondents for their opinions about various aspects of Internet health care information. There was overwhelming support for the statement, "Physicians should advise the patient to check with a physician before acting on information from the Internet." Ninety-four percent of respondents agreed or strongly agreed with this statement. There was no statistical difference between the views of users and nonusers. There was also very strong support for warning patients about inaccuracies in medical information on the Internet. Eighty percent of respondents agreed or strongly agreed with the statement, "Physicians should warn patient of inaccuracies in medical information on the Internet." Again, there was no statistical difference between the views of users and nonusers.

The Internet for Professional Purposes

115

Finally, there was mixed support for physicians checking Web sites used by patients. Thirty-seven percent agreed or strongly agreed with the statement, "Physicians should check the site the patient used to assure accuracy of the information," while 28% disagreed or strongly disagreed with this statement. Once again, there was no statistical difference between the views of users and nonusers.

Physicians Believe They Should Be Offered Internet Training Against the background of strong expressions of the need to be proactive about patient use of the Internet, we asked respondents whether physicians should be offered training in Internet use. A very large majority, 87% of the respondents, believed physicians should be offered training in Internet use. There was no statistical difference between the views of users and nonusers. This result suggests that as a prerequisite for physicians to be proactive about patient use of the Internet, physicians themselves must be given training in Internet use.

PUBLIC POLICY DISCUSSION The Internet Is Now a Major Source of Health Care Information Combining the results of our surveys of Internet users and physicians, we can postulate that the Internet is now a major source of health care information for Internet users and physicians. In late 1997, we found that 4 1 % of Internet users had used the Internet for health care information, and in late 1998 we found that 62% of our sample of New Jersey physicians had used the Internet for professional purposes. The above percentages are now (year 2000) almost certainly significantly higher (Cyber Dialogue, 1999). Indeed, for some the Internet is probably the main source of health care information. In a matter of a few years, the Internet has become a major source of health care information. The time has come to start investigating whether the use of Internet health care information has changed consumer behavior patterns; for example, are consumers interacting in different ways with the health care system? Are they consuming more or less health care resources? Certainly this is a

116

ONLINE MEDICAL INFORMATION

worthwhile question in light of the fact that physicians are writing prescriptions at a growth rate of 6% compared to the underlying population growth rate of less than 1%.

Two Models of the Use of Internet Health Care Information Could Emerge How will consumers respond to the availability of health care information on the Internet? One model—the adjunct model—assumes that consumers will use Internet health care information while continuing to participate in the existing health care system; for example, patients continuing to see their doctors on a regular basis. The other model—the self-service model—assumes consumers will take greater charge of their own health care and reduce their contact with the health care system. In extreme cases, the consumer will use information from the Internet and other sources for self-diagnosis and self-treatment (e.g., lifestyle changes and self-medication). Clearly, the self-service approach has considerable implications for health care professionals and public health administrators. Further, it might lead to pressure for certain prescription medications to be made available over the counter at supermarkets and pharmacies.

More Attention Needs to Be Given to the Quality of Internet Health Care Information For Internet health care information to be effective in improving health care, the issue of its quality must continue to be addressed. In our late 1997 survey (Aspden & Katz, Chapter 4, this volume), about a quarter of Internet users had concerns about the quality of Internet health care information. In our late 1998 survey, more than half the physicians in the survey expressed concerns about the quality of Internet health care information. In addressing this issue, we first need to understand better how to reassure physicians and consumers about the quality of the health care information provided. Some will be reassured if the information is from the government; some will be reassured if the appropriate medical professionals have reviewed the information; while others will be reassured if they can ask questions of the information author. We must also learn how best to package the information so that the user can quickly judge the credibility of the information provided. Further, we must help consumers and physicians navigate the Web to sites that provide the types of reassurance

The Internet for Professional Purposes

117

they are seeking. For example, specialized search engines could be developed that search only sites that provide the required type of reassurance, for instance, search only sites that have been reviewed by medical professionals. Second, we need to develop more formalized approaches to providing reassurance. One approach is the "seal of approval" method whereby a third-party authority ascertains on a regular basis whether the quality of the information on a Web site is acceptable. A variant on this is when a third-party authority recommends Web sites for the public to use. For example, a team of researchers at the University of Michigan (Biermann, Golladay, & Baker, 1999) reviewed references to Ewing's sarcoma on 400 Web sites. On the basis of the review, the researchers recommended only a handful of sites. Another way to provide reassurance is via a self-administered code of good practice. Today, the credibility of many health care Web sites rests on the credibility of the source of the information—a national association of medical professionals, a prestigious hospital. Nevertheless, consumers can legitimately ask for an explicit statement on quality assurance. Some sites already provide this, for example, http://www.healthfinder.gov (see Baur, Deering, & Hsu, Chapter 16, this volume). Such a statement, at a minimum, should cover the source of the information provided, how it was reviewed, how often it is to be reviewed, how the Web site is financed, the business and funding of the owner of the Web site, the criteria for adverts placed on the Web site, and how visitors can query information on the Web site.

Expanded Internet Training for Health Care Professionals The results of our survey of New Jersey physicians indicate two clear messages. Physicians wish to be proactive regarding patient use of the Internet, and to do this, physicians expressed a strong need for training in Internet skills. No doubt other groups of health care professionals (had they been included in the survey) would also have expressed a desire for Internet training. For new health care professionals, training in Internet skills can be added to the standard professional curriculum. Special programs, however, need to be developed to reach out to those health care professionals who qualified before the Internet became a consumer service. One option

118

ONLINE MEDICAL INFORMATION

would be to use the Internet itself to provide training and guidance on the use of the Internet for health care purposes.

What About Those Without Access to the Internet? Our two surveys focused on Internet usage by physicians and consumers. But, what about those physicians and consumers without access to the Internet? Considering the case of physicians first, and presuming that their lack of access is not due to lack of resources, we envisage that the proportion of physicians using the Internet for professional purposes will steadily rise for the following reasons. First, the likely increasing availability of Internet training and support will encourage more physicians to become Internet users. Second, pressure from their patients who have accessed Internet health care information will lead to more physicians becoming Internet users. Third, older physicians, predominantly nonusers, will retire, raising the proportion of active physicians using the Internet for professional purposes. With regard to use of the Internet by the public, ever since the Internet became a consumer service there have been concerns about a Digital Divide between users and nonusers on a number of demographic dimensions. For example, Hoffman and Novak (1998) examined the Digital Divide from the perspective of race. Our own research (Katz & Aspden, 1997) has indicated that the Digital Divide on each of the dimensions of gender, age, educational attainment, household income, and race has been narrowing. According to our data, the gender gap appeared to have closed for users starting in 1997, and the education gap is closing rapidly. However, for the dimensions based on household income, age, and race, the gap in 1997 was still wide. The elderly are particularly underrepresented among Internet users. In our data set, 3% (5 out of 184) of those starting in 1997 were aged 65 and over—well below the population proportion of 17% for those 65 and over. The elderly, low-income families, and African Americans are in greater need of health care resources, and, therefore, health care information, yet they are less likely to have access to the Internet. This lack of access could be partly remedied by making available Internet access in public places and physicians' offices while at the same time providing guides on how to access health care information via the Internet.

The Internet for Professional Purposes

119

REFERENCES American Medical Association. (1998, October 21). Access to health information and support. Journal of the Amencan Medical Association, 280,1371-1375. Biermann, J., Golladay, G., & Baker, L. (1999). Evaluation of cancer information on the Internet. Cancer, 86(3), 381-390. Cyber Dialogue. (1999). Impacts of the Internet on the doctor-patient relationship: The use of the Internet health consumer. Survey report based on the American Internet User Survey. New York. Eysenbach, G., & Diepgen, Τ. (1999). Patients looking for information on the Internet and seeking teleadvice: Motivations, expectations, and misconceptions as expressed in e-mails sent to physicians. Archives of Dermatology, 135,151-156. Health Management Technology. (1999, March). Physician use of Internet explodes. Health Management Technology, pp. 2, 20. Hoffman, L., & Novak, T. (1998). Bridging the racial divide on the Internet. Science, 280, 390-393. Katz, J., & Aspden, P. (1997). Internet usage—The Digital Divide will not disappear soon: The results of three nationwide surveys. Unpublished manuscript, Rutgers University. Susman, E. (1997). Doctors lag behind executives in using e-mail and the Internet. Telemedicine and Virtual Reality, 21,1-2.

6 Expectations and Experiences of Seeldng Infertility Information via the Internet and the Telephone Directory JUNE ANIGBOGU RONALD E. RICE

T

his chapter provides two perspectives on how a person new to the Internet might experience the accessibility and usability of health information. The first part follows a woman as she tries to understand what appears to be a problem of infertility in her marriage. With our help, she turned to the Internet to find resources outside of her regular doctor and her family. As she was a complete novice at computing and Internet usage, this case study provides interesting insights into how "ordinary" people might actually first experience this new environment, rather than as Utopian advertisements portray, and how this experience might change people's perceptions of their problems, their health providers, and even their own relationships. The second part considers a related dimension of this case: To what extent do traditional information resources, here the local telephone directory and hospital referrals, and new information resources, here specialized Internet lists and search engines, reflect a common physi121

122

ONLINE MEDICAL INFORMATION

cal health environment? Is it accurate to say that the Internet is "just" a faster and easier way to find the same information? Finally, both of these cases show that the process of using the Internet for health information is far more complex, and involves many more tacit assumptions and prior knowledge, than is typically portrayed in the media or conveyed in training materials.

PART I: RESEARCHING INFERTILITY ON THE NET: A CASE STUDY Mr. and Ms. Doe wish to remain anonymous. June met Ms. Doe through another friend, and considers her an acquaintance. Their initial phone conversation was held on May 27, 1999. Situation: A couple married for 7 years, in their early and late forties, residing in Union, New Jersey, have been trying to have children for 4 years. The wife confides that she has been to an OB-GYN and undergone a number of female-oriented tests that proved that she would not have any difficulties in the area of reproduction. She then confided that it was her husband who eventually was examined and found to have a low to nonexistent male reproductive count. "Male infertility is a new frontier that had not been dealt with until recent times" (http:// maleinfertility. com/contact.html, Center for Male Reproductive Medicine and Microsurgery). The following pages are the dialogue between Ms. Doe and the first author, from the beginning computer sessions to her personal investigation of the doctor's inquiries and her consideration of possible treatments for herself and her husband.

GETTING ONLINE: ISSUES OF PHYSICAL, EXPERIENTIAL, AND COGNITIVE ACCESS Phone Conversation on June 9, 1999, with Ms. Doe I called Ms. Doe on June 9 to verify whether she was still interested in being shown how to get online. She said she definitely was still interested in the session, so we decided to meet on Friday, June 11, at 3 p.m. at my place in New Brunswick. I asked her if it would it be okay if I taped her. She

Information via Internet and Telephone

123

said, "Definitely not, are you trying to Linda-Tripp-me-up?" I said, "No, I would never do that to anyone. Could I take notes?" Ms. Doe agreed.

First Meeting, June 11 at My House Our first meeting was an eye-opener for me. My definition of computer illiteracy was definitely inaccurate. I always thought that to be computer literate, one had to know how to design computer programs and use other computer program designs, as well as have the ability to set computers up on a network. So I always said that I was computer illiterate. After beinginvolved in a training session with Ms. Doe, I realized that I am computer literate according to some standards. When we sat down, I kept in mind that she wanted to learn how to get on and off the Internet to investigate fertility issues and available clinics. I thought it would be a quick and easy process of clicking on an icon and waiting for the connection to take place over the phone line. I thought that the majority of individuals in the United States had some knowledge of computers. As Ms. Doe got adjusted in front of the monitor, she said anxiously, "OK, what do I do?" I spoke in a casual, matter-of-fact voice, so as not to talk down to her or underestimate any computer skills I thought she might have. I responded, while pulling out the keyboard: "Click your mouse on the [browser] icon, and just wait a few minutes and you're online." Ms. Doe looked at the keyboard tray, which included the mouse, and then at me and said, "What mouse? Where is the mouse?" At that point I knew this would take some time! I picked up the mouse, which was on the left side of the keyboard, and placed it on the right side, since she was right-handed, at the same time telling her, "This is your mouse." I said this with a straight face and trying not to show my feelings of surprise and disbelief. Ms. Doe: It does look like a mouse, its tail goes to the computer, (we laughed) I repeated the same instructions, apologizing for the unseen keyboard, thinking that maybe she was nervous about the fact that this friend of a friend (me) had knowledge of her private problem. June:

OK, now click your mouse on the Internet icon.

124

ONLINE MEDICAL INFORMATION

Ms. Doe: What or where is the icon? I casually explained the convenience of an icon, and then I explained how she could also go to the start button, and then to any computer program. I thought, how can she not know this? My curiosity got the best of me, so I asked, "Do you have a computer at home?" Ms. Doe: Yes, but my husband only uses it on special occasions, and I am not allowed to touch it at all until I learn how to use it. Sounds crazy, huh? I ain't computer literate... never was interested in computers until now. This is a special occasion, looking for a clinic or hospital online for my husband. I got the idea from my girlfriend. June:

What Microsoft program do you have at home, '95 or '97?

Ms. Doe: I think my husband said Microsoft Word 4.6 or 7.1 really am not sure. What did I do, the computer stopped!? I was taking notes and had not noticed what took place to make the computer freeze. June:

What exactly did you do?

Ms. Doe: I pushed both buttons on the mouse at the same time, I think. June:

Push Alt, Control, and Delete. This is a way of rebooting or restarting the computer. (I pushed the buttons in one smooth move for her, and the computer came back up.) Now push the left button on the mouse.

I watched as she clicked the button and found that she held it down too long. The computer froze again. This happened three times; after the third time the computer came back up, I took her hand along with the mouse so she could feel how my finger bounced or doubled clicked the mouse. Ms. Doe:

Oh, I get it!

When we finally were dialing onto the Internet, Ms. Doe asked what were the little whirling sounds going on. I explained that we were being hooked up online through the telephone wire to my Internet service provider.

Information via Internet and Telephone

125

Ms. Doe: I don't know what my husband's server is, do you have to have it to get online? June:

Yes. (we connected) Now we are hooked up to the outside world.

I showed her the two little green computers at the bottom of the screen that also told us we were online. Ms. Doe: Whoa! Isn't that something! So, I can contact anything or anyone online? I tried to explain the difference between e-mail, being person to person, and the Internet, which was one person communicating with the world; that was hard, as I considered myself a novice. June:

Now, let's do our research. Once you're online you use different search engines to look up anything you're interested in. In your case, you want a health issue, infertility. So, let's go to a Web page that has many search engines to choose from, my professor, Dr. Rice's . . .

Ms. Doe:

(interrupting) Wait! You're going too fast, what is a Web page, and I think I understand a search engine, it sounds self-explanatory. A computer engine that searches things out, like researches for you . . . right, like a car engine drives you around?

June:

Yeah, something like that.

As long as it made sense to her! Dr. Rice's Web page came up. Ms. Doe: Look, that is so neat, is that his name moving? June, what does www stand for? I hear it all the time on TV. I know the com stands for computer, right? I printed Dr. Rice's Web page and explained what a Web page was, and she began to read what the computer abbreviations stood for. Ms. Doe: Oh wow, stupid me, com stands for commercial, www is World Wide Web, here are all the meanings, like a computer dictionary, neat-o. I wondered what that http meant, it says here that it stands for hypertext transfer protocol. I bet my

126

ONLINE MEDICAL INFORMATION husband don't know this. Oh yes, wait 'til I show him. I'll be on that computer in no time. Thank you so much for this.

June:

You're welcome, no problem. Now, let's do some research on the health issue.

I instructed her to click on the meta-search engine and investigate fertility for men. Ms. Doe: Is that like Yahoo? Can we go to the Yahoo on line . . . I want just to check out a music group... u h u m . . . the Spice Girls. She clicked Yahoo! slowly, like a little baby that just had learned how to walk, being cautious with each new step, which hung the computer up again. I showed her how to push Refresh and continue and not to be afraid to push/click the mouse fast. I showed her where to click on music, and then type "spice girl" as the group she wished to search. Ms. Doe: That little finger pointing, what does that mean? June:

You can go into that site, just double click.

She double clicked after she was in the Spice Girl site, on a song. It began to play. Ms. Doe: My god . . . this is like TV or computer radio with pictures. I think that she is avoiding the health issue, or maybe she is really into the excitement of the new computer experience. She also wanted to know if she could transfer information from online to her computer. I showed her how to hold the mouse button down and highlight the info, then press Control-letter C, then bring up the computer word processing program, and then push Control-letter V. I did it the first time so she could see how it was done. Then she tried on her own. Ms. Doe: It worked... I did it, I can't believe it, I can copy from the online world! I wonder if my husband can do that?! I think there are other issues she is dealing with in her home with hubby besides the infertility one, because she keeps referring to her husband like they are in competition or she knows something that he might not know. June:

(trying again) Now let's check health issues online.

Ms. Doe: Wait, look, I want to click on shopping.

Information via Internet and Telephone

127

She clicks on shopping, then types in J. Crew. J. Crew's colorful site pops up. Ms. Doe: WOW! This is great! I can't believe this, it is really space age, that's modern technology for ya. Have you ever shopped on the World Wide Web? June:

No.

I went on to explain that shopping online is fine but, once your info is out there a lot of companies can contact you, and I did not want that. It made sense to her, and she agreed. Ms. Doe: It's getting late—I have a ways to go to get home. I really appreciate meeting with you, and your help. I think this is enough to get me going with health research. June:

When next week do you wish to meet again?

Ms. Doe: Maybe Monday. She called on Monday, but could not make it because of sickness. I told her I got info for her via mail regarding a clinic, and I gave her a number to call.

A NEW WORLD OF INFORMATION: CHANGES IN PERSONAL, FAMILY, AND PHYSICIAN RELATIONS Phone Conversation on July 9 I called Ms. Doe regarding her progress on the Internet. We had not met because of illnesses. She informed me that she had been searching the Net at home using the search engines from Dr. Rice's Web site that I had printed out for her. June:

Well, how's the research on the Internet going?

Ms. Doe:

I am doing OK, I move slow and sometimes I freeze the mouse when I don't click fast enough, so . . . I do that little Alt-Control-Delete trick. But now I'm becoming more familiar with the computer and the Internet thing. I feel I understand how it works, I just can't get off the computer, I mean I really can't stop . . . it's like I'm going into other

128

ONLINE MEDICAL INFORMATION worlds inside my home. June, you know what I m e a n . . . I am going some place, without actually going some place. It's mind baffling! Now my husband has to cook his own dinner, because I'm on the Internet and I haven't cooked. I think he's happy for me. I think I have talked my husband into getting the latest computer package or program put on the computer. I told him that we have to move into the new millennium.

June:

I know exactly what you mean, the world of computers is enlightening, right? I'm glad that you're getting better on the computer, and you're getting it upgraded. So, how about the health information, how much progress have you made regarding your fertility research?

Ms. Doe

You know, I found out anything that I type into those little long squares and push the search button, after waiting forever the information that I requested comes up, usually more than I ask for or want. There are so many things about fertility for women, and I saw some of the tests that they did on m e . . . in the hospital to make sure I could have kids. It's a hysterosalpingogram. All the doctor said was everything looks OK, but he never went into detail about what he saw, or what he actually did. All I know is that I hurt like everything. But in that fertility site for women I saw in color the procedure that was done to me, and why they did it. It was to see if I had any fibroids. Now I'm going to call my gynecologist to see if my uterus was a normal shape and . . . I have so many questions to ask my doctor about myself.

I explained to Ms. Doe that that is the process, and having options is good. Also, it is up to her to narrow her search, or choose the Web site that is of interest to her or appropriate for her. She agreed, replying, "Uhum ahumm." June:

Well, that's good. What about your husband?

Ms. Doe: There's not much information about male infertility on the Internet. But that urologist, we have made an appointment for July 14. We know that his male count is low, b u t . . . we'll just see what happens.

Information via Internet and Telephone

129

June:

There is a lot of info on male infertility. Go into http:// www.medicine.uiowa.edu/urology or just input "male infertility" in any search engine, or long search squares [which is how Ms. Doe identifies the input boxes] and a number of Web sites will be shown so you can research them the next time you come over.

Ms. Doe:

OK. You know we can talk on the phone if you don't mind.

June:

That's fine with me, that is quite funny because I was going to ask you if the conversations on the phone were more convenient for you. Can I call you Friday, July 16, just to touch base with you on the research?

Ms. Doe:

I guess so I thought I gave you enough information, June, but since you have really helped me with this infertility research and learning the computer and the Internet, I will call you next Friday on the 16th, OK? In the meantime, I will check out the male infertility info on the Net. Now, my husband is private so whatever the outcome, please keep it private.

Phone Conversation July 16, After Mr. Doe's Appointment With the Urologist I called Ms. Doe around 6 p.m., knowing that she promised to call me back, and after not hearing from her that afternoon. I hoped that I was not intruding. Mr. Doe answered. June:

Hi, Mr. Doe, I am June Anigbogu, an acquaintance of your wife. How are you?

Mr. Doe:

Hello, yes uhum I'm fine and you? Uhum, hold the line a minute . . . Hon!! Come to the phone! [he tries in vain to cover the phone] It's that one research lady.... June with the last name. [He whispers, and I hear it] Honey, you don't gotta stay long with her . . . O K ? "

Ms. Doe came to the phone. I felt that this conversation was not going to be an easy undertaking. From the sound of her husband, she might not be able to talk freely. Maybe that was why she didn't call me back as she had promised . . . my speculation.

130

ONLINE MEDICAL INFORMATION

Ms. Doe: Hello June, I'm really busy right now, b u t . . . I can give you a few minutes. June:

A h h . . . OK. [she was curt but I understood why] I know that you and your husband have just returned from the urologist, I hope the appointment was favorable. I do not wish to disturb the both of you, but was just wondering if you were able to talk to your GYN regarding the information that you found on the Internet?

Ms. Doe: W e l l . . . . I did. I'll get back to you on that, because it's like I said, I am busy. Maybe I'll call you tomorrow. OK? 'Bye now. June:

OK, I understand. 'Bye.

Her husband was there, so she could not talk freely. I'll call her back if need be before next class, to reach closure for the case study.

Phone Conversation on August 2 Ms. Doe did not call me back (maybe she was going to, but I was too impatient) so I called her at midday on Monday, hoping that it was a better time for her to speak freely. June:

Hi. How are you? I called because you might have missed me. I have been out and in, so I thought I would touch bases with you. To see if you called. Is this a good time to talk?

I felt like I was harassing her, like in the movie Cable Guy. Ms. Doe: Hi June I hadn't called, but I guess this is as good a time as any. You have been awfully helpful to me so [She immediately jumped into what she wished to discuss with me.] I did call my GYN and asked him the details of my hysterosalpingogram. I tell you, I think he was shocked that I asked, because he asked me who have I been talking with, and did I get a second opinion with another GYN. I said no, I just went to the Internet and I found what a normal uterus looks like and the minor or major complication that can be found during a hysterosalpingogram. June:

I hate to interrupt, but could you spell that for me?

Information via Internet and Telephone

131

Ms. Doe: Sure, I got this spelling from the Internet... H-y-s-t-e-r-os-a-l-p-i-n-g-o-g-r-a-m. Anyway, I couldn't believe what he told me. He said that the hysterosalpingogram showed that there was a cyst on one of my fallopian tubes. And according to the pelvic sonogram I have a small fibroid. He said my uterus was not enlarged but it did show a slightly bulky contour. I asked him why didn't he tell me before now? He said "because you can still have children, so does it matter?" I was upset because it's my body, and I let him know it. My silly GYN said are you going to let me take care of your body or a computer? I told him "you, but you have to tell me everything about me!" Well, anyway, June, I guess I am still upset and I am venting. I am glad you called, I will call you this Friday without fail regarding what the urologist said regarding my husband, whether it's good as well as bad, after that I think I was helped by you and used by you enough. Don't you think? June:

Well, I hope things turn out well, and please don't think I was using you. I hope that we have helped each other.

Ms. Doe: Yes, w e l l . . . maybe s o . . . . Sometimes it's good to not know so m u c h . . . about things. Oh what am I saying?! June, it was good talking with you, I've gotta go. I'll call you Friday, OK? June:

OK. Have a good one.

THE INTERNET AS EMPOWERMENT Phone Conversation on August 6 Ms. Doe finally called me on her own—I was shocked! Ms. Doe: Hello, June? I bet you're surprised that I called, huh? I thought the least I could do is initiate the last call, right? She was letting me know in a nice way that this would be her last call. June:

Yes, this is a surprise.

After exchanging 2 minutes of pleasantries, we finally began the conversation about her situation.

ONLINE MEDICAL INFORMATION

132

Ms. Doe: Well, first I would like to thank you for the referral to the urologist, it proved to be quite positive. I think that I had told you that my husband had a low male reproduction count. We took his lab results with us so the doctor could analyze the report. In a nutshell... the doctor said that yes the count is low which makes for the abnormal analyses. But the report also shows good sperm morphology and good swimming velocity, this is the good news. He said the best way for us to have children is IVF, that is in-vitro fertilization, where the sperm and egg are retrieved and fertilized, and then placed into the uterus for potential pregnancy. I had already read about this procedure from the Web site that you gave me—www.inciid.org. There is so much positive information in that program about infertility. I corrected her and told her it is a Web site and not a program. Ms. Doe: I can't believe I am talking so free and unrestricted about this. I found out that everyone talks freely on the computer about everything, you know? June, life is not as restrained as we make it. Although it will cost a lot of money for the procedure, my husband is really depressed about it. I guess it's a man thing. He doesn't want me to say anything to our friends about the procedure. We'll start this right away as soon as we see what our insurance covers. We'll hopefully have children. I guess I won't be talking to you any more, my husband has signed me up with a computer class at a junior college. June, my husband is a very private person, and I am sorry for any inconveniences that might have occurred. But, if you need any more help please feel free to call me. One thing I found out about that Internet: It's the information key that unlocks any door to the outside world, and I have become a part of that world. It's more informative than TV, I just keep comparing it to TV.

She finally pauses. June:

Well, I am glad everything worked out for you and your husband. I believe you have a point about the Internet because

Information via Internet and Telephone

133

the urologist and some of the other information that you received was from the Internet. Ms. Doe: You know, although I needed the fertility info, I mainly needed to get out of the house. The computer got me out of the house, I got help with our fertility problem, and I made my own GYN know not to take me or my body for granted. At first I was going to change my GYN after I found out he was holding back information about my body. But now that he knows that I have my own personal resource—the computer—to check things out, I know he'll be honest. So, I don't feel like breaking in another GYN. I feel so changed, so much in control. I have the edge on just about everything, you know what I mean? June:

Not really. Edge? What is your edge?

I had an idea what she meant, but I wanted her to clarify her "edge" in detail, because this might be the last time we talk. Ms. Doe: My edge is the computer, June. Don't you see it? I don't care what it is concerning my health, my house, or my husband, I can find it on the Internet. Speaking of which, that is my next step, I am sure there are resources on the Internet about husbands who are control freaks. He thinks that I don't know what he is doing. My edge is the computer. I have a way out, while I am inside my home. June, I think I hear the hubby, he was not supposed to be home this early. I guess I better go, thank you again for everything. June:

OK, I thank you for your time, and once again I am glad everything turned out fine for you. Take care! 'Bye.

Ms. Doe:

'Bye, now.

What might have grown into a good friendship remained only a business agreement and case study. Maybe because I knew too much about her personal life, or maybe because her husband did not want her to get too close to anyone who knew too much about their fertility condition. It would be very wrong for me to perceive her husband as a domineering person on a power trip since I do not know him or have any background about his experiences and situation.

134

ONLINE MEDICAL INFORMATION

Phone Conversation on August 9 I called Ms. Doe to see if she would be interested in reading what I had written, and adding anything to the case study. Ms. Doe: No, I feel that I have given enough of myself and time as a guinea pig, I don't mean any harm. In addition, just like you have received something, so have I. If it weren't for our meeting, I would not have inquired with a computer about my health. Through the Internet experience I got nerve or courage to confront my doctor. My doctor was someone that I always trusted and never questioned, I almost reverenced him, it was a Southern thing. Because my Mom felt the same way about her doctor. I now know he is a man who can make mistakes or hold back information. I will continue interacting within the computer world. I am a part if it and it has become a part of me now. It was great meeting you, and maybe I'll see you around. 'Bye. As I told her 'bye, I thought about the impact the computer and the Internet had on Ms. Doe. I wondered what would eventually happen to her. I gave Ms. Doe the www.rba-online.com/male.html site, which I felt would be very informative for her husband regarding infertility. She thanked me and we said our goodbyes. It seemed Ms. Doe had other issues with her husband that she was dealing with besides infertility. Or maybe it was the infertility issue that was predominant in the home and that made other conflicts or issues occur. Nevertheless, her experience with the Internet opened a number of possibilities and raised a number of issues for her and her husband.

PART II: HOSPITAL AND CLINIC REFERRALS FROM THE TELEPHONE DIRECTORY COMPARED TO THOSE LISTED ON THE INTERNET, IN THE NEW JERSEY COUNTIES OF MIDDLESEX, SOMERSET, AND UNION This section analyzes fertility-related hospitals and clinics in three counties in New Jersey (Middlesex, Somerset, and Union) identified from the telephone directories and subsequent referrals, and from the Internet, in an attempt to compare the "health resources environment" represented

Information via Internet and Telephone

135

through these two media. Two questions served as the primary motivations for this study: How well do the two sources reflect a shared physical reality of health resource providers? And what kind of processes and search activities might be required to become aware of this health resources environment? Using the prior case on infertility information as a backdrop, we attempted to document all the efforts and results associated with this experience. As with the case study, we hope this analysis emphasizes the many difficulties, circuitous paths, implicit knowledge, and inconsistent information involved in finding appropriate health resources. Vendors, implementors, health service provides, Web site designers, researchers, educators, and novice and expert users should become more realistic about just what is involved in such activities. This is also written from June's perspective.

Results From the Telephone Directory I live in New Brunswick, New Jersey, which is located in Middlesex County. I knew that it would be feasible to begin my research at home with my phone book, which I thought would be part of the New Brunswick and Edison areas of Middlesex. I went directly to the yellow pages and wrote down all the hospitals and clinics that I thought would cover these geographic areas. I knew this portion of my research might be a little difficult given that I am from Des Moines, Iowa, and have been a resident of New Brunswick for only 3 years; I am still familiarizing myself with the city and county regions. In view of that, I would need to make sure that the hospitals and clinics for Union, Middlesex, and Somerset counties that I had written down were correct. To check this I went to the Alexander library on the Rutgers University campus to check other phone books for these specific counties to make sure I did not omit any of the hospitals or clinics in the chosen areas. I thought there would be specific or separate phone books for the Union, Middlesex, and Somerset counties. Once inside Alexander library I spoke with the librarian at the reference desk, asking her where the phone books were. First she laughed, and then said that the library does not carry phone books anymore, and for me to use the yellow pages on the computers. She began to tell me that she was actually one of the librarians on Busch Campus and was waiting behind the desk to go to lunch with a fellow librarian. The Busch Campus librarian began to clarify how they had gotten rid of their phone books because they were taking up space, replacing them with specialized science directories.

136

ONLINE MEDICAL INFORMATION

She explained that, in place of the phone books, they provide total use of the Internet, the Iris online catalog, and the online yellow pages. I thanked her and explained that my research still needed a hard-copy phone book. The Busch Campus librarian referred me to the New Brunswick Public Library. I later found out that the Alexander Library did in fact have phone books! That day I went to the New Brunswick Public Library reference department. The librarian was a recent graduate of the Rutgers Master of Library Science (MLS) program. After I explained my research need, she directed me to the phone books, and she made copies of postal zip codes so I could verify the county areas where the hospitals and clinics are located. The New Brunswick librarian also gave me the Web site for the New Jersey Hospital Association, http://www.njha.com. This Web site provides a listing of New Jersey hospitals alphabetically or by county, and was useful when I began the research for hospitals and clinics on the Internet. After accumulating all the required information, I went back to the Alexander library to double-check my phone book findings, to make sure I didn't exclude any institutions. The librarian explained that each phone book has a map in the front, showing the county areas that are covered inside. I made a copy of the map and then grouped the health institutions by county areas. I was happy to find out, after looking in a number of different phone books, that all the hospitals and clinics would be contained in one phone book. The librarian informed me that I should have this exact phone book at home. From the looks of the front cover, I indeed had a copy at home! I had not realized that Middlesex, Somerset, and Union counties were all contained in my home phone book, because I am new to the state. The results of the phone book research showed Middlesex with a listing of seven hospitals, Somerset with two hospitals, and Union with four hospitals. The phone book also had a listing for 16 clinics in the tri-county area: 7 were abortion clinics, while the other 9 clinics were a mixture of dialysis care and mental health care facilities. Out of the 16 clinics, I researched only the 5 that corresponded with the infertility issue. I attempted to contact each health institution by telephone and requested to be referred to a fertility clinic. I thought the phone investigation would be a 1-day event, but instead it ended up being a 3-day ordeal because I had to wait for the callbacks. I took on the role of a desperate older woman who needed a fertility referral. The operator at the first hospital responded immediately in a very empathic voice, "We don't have a fertility clinic, but I believe XXX Medical Center does; here is the address and

Information via Internet and Telephone

137

phone number." I used this speech for the remaining health institutions. Either the operator gave me a clinic name and address, or she referred me to the hospital social worker who, after asking a host of questions, gave me a referral to a fertility clinic. The majority of the operators or social workers were very sympathetic. There were two hospital operators that I considered rude. One talked to another operator while talking to me and then got upset with me because she did not understand what I was saying, so I had to repeat my problem twice to her. The other operator started out asking about my problem; as I related my story to her she was repeating it to her co-workers and joking about it. I could only imagine the anguish that would be felt by a person really in need who was treated in such a way. However, other than these incidents, the overall phone contacts with the operators, social workers, and physicians' referral services were positive, with rewarding responses. I even received a follow-up letter from one of the hospitals regarding the referral they had given to me. One of the hospitals was closing, and I just happened to call them on their last day. They were sending all their patients and inquiries to two other hospitals. In this particular instance an Internet listing could be updated immediately, while the phone book would not reflect those changes until the next issue came out, so a phone book would be inaccurate for a while.

Results From the Internet My research experience with the Internet while searching for hospitals and fertility clinics was just the opposite of my experience with the phone books at the libraries. With the Internet, the investigation for hospitals and fertility clinics was less confusing and involved no stress or inconvenience. Everything was at my fingertips via the keyboard and Web links through search engines. I started my Internet research by using search engines. The first was the bigbook.com yellow pages. After getting online, I input gte.superpages. com into the computer. The search phrase that I used was "hospitals in New Jersey," which resulted in 266 hospitals in New Jersey. By continuing to use different search engines, I received a broader scope of what was available on the Internet. For example, through the Switchboard search engine I found a relevant health service. This site had a few hospitals that are located in New Jersey that the health service uses as hospital referrals

138

ONLINE MEDICAL INFORMATION

for New Jersey residents. I called the health service, asked for a urologist, and was referred to one. I then went to http://www.inciid.org (The InterNational Council on Infertility Information Dissemination), which was a start in narrowing my geographical search. This Web site was very rewarding. It has a geographic directory of physicians treating infertility. I was able to zero in and obtain the names of physicians and a few clinics in the tri-county area of New Jersey: Middlesex, Union, and Somerset. This particular Web site would be beneficial to any couple doing an in-depth investigation on their fertility problem and wanting immediate answers. The inciid.org site index includes roundtable chat transcripts on various fertility topics, and chat rooms. The site also has Fertility Industry News, which includes infertility treatment and technology news reports, and fact sheets, which are repositories of essays, articles, commentaries, and facts. There is also a glossary of infertility terms and acronyms, another method this site uses to educate infertile couples. Inciid's Web site is dedicated to the enlightenment of the infertile couple, even down to a humor page to make them laugh. The next search engine I used was Yahoo! I entered "infertility" as the search word, although I was to search hospitals and fertility clinics first. I thought I would also acquire a few interesting sites for my case study on Ms. Doe at the same time. The search resulted in 460 sites. I pulled a few down for Ms. Doe that pertained to male infertility. Next, one of the Web links that appeared was http://www.hospitallink.com; this site contained some of the major hospitals in the wider New Jersey area. Using a meta-search engine allowed me to try out other search engines. These search engines included MetaCrawler, where I found IVF of North Jersey: http://www.ivfofnorthjersey.com. Also, Infoseek gave me http:/ members.aol.com/fertilmd; this site was very useful. It was called Dr. XXX's Infertility Home Page/Infertility Info Center, and contained causes of infertility and information on reproductive surgery, although not as many Web pages and links as inciid.org. On the Lycos search engine, the homepage stated, "Ask Jeeves any question." Jeeves was a smiling male cartoon figure. I did not ask Jeeves anything, although I thought the computer suggestion was cute. I went straight to the search box and input "New Jersey infertility clinics." I did not receive a list of clinics per se, but instead a host of Web pages that could connect with clinics: for example, http://www.infertility.to/clinics.html. Also, the site included Web pages that gave information on the "Partial Birth Abortion Ban act of 1997" (http://www.senate.gov/~rpc/release/

Information via Internet and Telephone

139

1997/PARBLN.704.htm), and stories behind the story (e.g., the story, "Who are the egg donors," http://www.wnet.org/archive/innovation/showl/ html/3sb-egg.html, about women who were selling their eggs). I changed my search to "health clinics in New Jersey" and received more Web pages—some of them were on New Jersey Health Initiative 1998 (http:// www.rwjf.org/librarynjhl-98.html), which is a grant-giving foundation. Using the Lycos search engine resulted in 17 articles, with 9 of them having Web sites. The majority of the articles were related to the issue of infertility. Some articles were unrelated, however, such as an article on doctors standing up to HMOs, "AIDS Care October 1998," and "Intelihealth— Home to Johns Hopkins Health Information: Mental Health." One relevant site was South Jersey Online-Women's Health (http://www.southjersey. com/promos/women/health/pregnancy/cfmsout hjersey.com.html). Lycos also had many IVF counseling centers listed in New Jersey. I chose the (Cornell) Center for Reproductive Medicine and Infertility (http://www.ivf.org/ptcood/direct/nj.htm), though it was not in New Jersey, to examine next in the Lycos search engine. I wanted to investigate what the Center's Web page offered couples. It was a nice Web page; the school seal added to the credibility of the page. The organization specialized in male reproductive medicine and microsurgery. As stated before, the Lycos search engine had a number of helpful links, mainly Web page articles from the news media and Web sites. But I also noticed at the end of page 2 that there was a Web page for an IC Yellow Page Web that was unrelated to fertility clinics. This site was a business clearinghouse, stating that individuals and businesses could find financial solutions. It seemed that the Lycos search engine allowed unrelated Web site topics to appear in a specified search. I briefly visited the AltaVista search engine, entering "New Jersey/infertility clinics": 2,495,115 web pages were available! Like Lycos, AltaVista had a mixture of other topics that were not related to my search. The homepage for the Galaxy search engine was very impressive. The title was, "Galaxy, The Professional's Guide to a World of Information." This statement might put a thought in the Web user's head, such as, "This information must be the best if professionals use this search engine." As a novice, I believed that some professionals must use this search engine because Galaxy was well set up. I liked its sophisticated Web site layout, with the Web sites in numerical order and each one followed by an outline. I entered "male infertility" in the Galaxy search engine, and 53 documents came up, 1 of which was http://www.fertilitext.org. The resulting Web

140

ONLINE MEDICAL INFORMATION

pages were all focused on my topic of male fertility, unlike on Lycos and AltaVista, where unrelated topics appeared.

Discussion: Hospital Referral Research Versus Computer Research Table 6.1 summarizes the results for each of the three counties for entities listed both in the telephone directory and on the Internet, the telephone directory only, and the Internet only. The research on the Internet and the verbal research on the phone with the hospitals were two contrasting experiences. The Internet research was personal; I could enter into a search box anything I wanted that pertained to anyone or to myself. In other words, I could personalize my search to suit my own needs and research goals. With the Internet's research I can go beyond the boundaries of my own environment and in a short time investigate other realms of social and cultural society. I can research ad infinitum without sharing my personal needs or problems with anyone else. I felt more in control with the Internet research. It was so wonderfully overwhelming, I had almost to pull myself away. It was easy to look up and find I had been on the computer almost 5 hours a day with no breaks. Some of my time was spent reading the available material, other time going from Web site to link to Web site and from one search engine to another. Sometimes I just wanted to see how far I could go on the Internet. The research with the hospitals via word of mouth referrals as well as the library research was personable because I received live, not necessarily face-to-face, but real-time verbal telephone contact. However, responses from this part of the investigation were not all rewarding or positive. B e cause of human nature, a response was occasionally inaccurate (such as the original advice from the library, leading to unnecessary trips elsewhere) or became unpleasant (as in several of the telephone calls). On the phone, people could treat me anyway they desired, either rudely or courteously; they had the upper hand as information providers. So, what actually made the call easy or hard was determined by the response from the personal contact I had with the operators. For instance, if the hospital operator was positive or sympathetic, the call went smoothly. Even if the operator was impersonal, I considered it a positive call if I got the information that I wanted without a host of negative responses. The overlap between the telephone directory referrals and the Internet resources showed that many of the hospitals that are on the Web are also in

Information via Internet and Telephone

141

Summary of Comparative Results of Search for Infertility Resources in Three New Jersey Counties via the Telephone Directory and Hospital Referrals, and the Internet MIDDLESEX C O U N T Y Both telephone directory and Internet: O n e hospital, listed on both njha.com (included "university" in name) and gte.superpages.com (listed it as a medical center). The operator said they didn't have a fertility referral but gave the phone number of the next entity (university hospital). O n e university, listed on both njha.com (different phone number) and gte.superpages.com. The operator transferred the researcher to the hospital social worker, w h o provided a fertility clinic and w a s very polite. O n e medical center, listed on both gte.superpages.com (with a slightly different name) and njha.com. The operator transferred the researcher to a resources specialist w h o w a s polite and very kind, and who found a convenient doctor in the area. O n e medical center, listed on njha.com only (with a slightly different name). T h e operator w a s rude and asked about the researcher's problem, and at the same time joked about the call with co-workers during the phone conversation, and said simply to " c a l l your local hospital." O n e medical center, listed on njha.com only (with a different phone number). T h e center w a s actually closing permanently, and asked to have their number removed from the researcher's personal phone directory. O n e center, listed only on gte.superpages.com (with the same phone but a different address). T h e operator w a s too busy to handle the call, but took the researcher's number and called back with a fertility specialist; was very professional. And one health service, listed only on njha.com (with a slightly different name). Telephone directory only: Three centers and 1 " c h o i c e . " The latter operator asked the reason for the c a l l , as they w e r e a therapy c l i n i c . The operator explained that they were not listed that w a y in the telephone directory. She said they had no fertility listings, to call a gynecologist for a referral, and seemed very annoyed with the c a l l . Internet only: Ten listings, 4 on inciid.org/new-jersey, 6 on gte.superpages.com, and Ί on njha.com. There were 5 centers: 1 doctor, 1 outpatient clinic, 1 dentist (!), 1 hospital, and 1 surgical. UNION COUNTY Both telephone directory and Internet: O n e hospital, listed on both gte.superpages.com and njha.com. The operator wondered w h y the researcher called, the researcher explained that they were listed as a hospital; they referred her to another hospital. O n e specialized hospital, listed only on njha.com. The operator w a s very kind and transferred the researcher to their physicians' referral service, w h i c h provided a referral and followed up with a letter! O n e medical center, listed on both gte.superpages.com and njha.com. T h e operator w a s very sympathetic and referred the researcher to a fertility c l i n i c .

(continued)

142

ONLINE MEDICAL INFORMATION

Continued O n e hospital, listed on both gte.superpages.com and njha.com. T h e operator wanted details about the fertility condition and gave a referral w h i l e talking with other operators in the background. Telephone directory only: O n e medical group; no phone contact made. Internet only: T w e l v e entities: 4 hospitals, 3 medical centers, 1 specialized hospital, 1 health system, Ί development center, 1 institution for rehabilitation, and 1 dialysis center (!). T w o hospitals were listed on both gte.superpages.com and njha.com, and one of these had different phone numbers and no street number on the two sites. O n e specialized hospital w a s listed twice on gte.superpages.com, with different street spellings. SOMERSET C O U N T Y Both telephone directory and Internet: O n e medical center, listed only on njha.com. T h e operator referred the researcher to a hospital social worker w h o asked a host of questions and then provided a referral. Telephone directory No entities.

only:

Internet only: Six entities: 1 center w a s listed on two sites (ihr.com/infertility and ivfnj.com). Four (2 medical centers, foundation, hospital) were listed only on njha.com. O n e fertility c l i n i c w a s listed only on inciid.org/new-jersey.

the phone book, though the Internet had more hospital and clinic options available to an individual than did the phone book. The overlap in the resulting "health environments" varied widely, even across the three counties. How current or reliable that information was also was an important issue to consider. When I tried to assess both research situations (the telephone and the Internet), I saw myself retaining a certain amount of personal and cognitive power on the computer. Although I consider myself a novice, I saw myself confidently driving around the information superhighway and aggressively investigating new sites and search engines. I felt powerful with what little I knew, and unthreatened: No one was there but me and my friend the computer to interact with as I progressed online. I felt more comfortable giving out personal information via computer than having impersonal contact over the phone.

Information via Internet and Telephone

143

The research over the phone was different—I was in need of information that was at others' disposal. They were in the position to say what they wished to me, and I felt I had minimal power. Although I felt the computer investigation was the least complicated and the phone book and phone referral investigation full of anxiety and confrontation, another individual might feel differently.

CONCLUSION Several results are quite clear from this experience and analysis. First, the health resources in one's physical environment can appear radically different between the traditional telephone directory and the specialized Internet listings for the same locale. There were many more Internet listings than telephone directory listings. Further, there was considerable divergence, even among the specialized Web pages and search engines. Also, sometimes specific information (entity name, phone number, address) varied between (and sometimes within!) the different sources. It is simply not the case that the Internet reflects the same health environment (at least with respect to infertility) that the traditional telephone directory and referral process does; the Internet seems considerably more expansive. Relying solely on the Internet would not omit many providers or referrals in the local area. As seen from the summary above (and in Table 6.1), telephone interaction is usually quite personally supportive and courteous, a kind of interaction missing from the Internet search attempts. Note, however, that several of the telephone interactions were themselves quite unsatisfying, discourteous, and unprofessional. So we must avoid overromanticizing the "personal" nature of the telephone (and, by implication, direct faceto-face interaction, such as the initial library advice, even ignoring the vastly greater time and effort required for such contacts). Some interactions with personal representatives of health entities may be less satisfying and less emotionally supporting than using the Internet. Finally, as with the case study of the woman seeking infertility information, the experience of searching the Internet can be daunting, requiring considerable experience and persistence, and accumulated tacit knowledge. Yet it can also be personally rewarding, opening up new vistas of awareness and providing access to a wide, never fully knowable, range of health resources.

Part III

Experiences Developing and Evaluating Health Information Sites

I

n this section, we present studies that apply the formal methods of social science to gain insight into communication issues in the context of Internet-based services. The topics covered include evaluations of health intervention strategies and governmental Web sites. They also cover evaluations of Web sites that are designed to serve various publics but that are also constrained by a variety of forces. These forces include—for the private sector—the need to seek profits and demonstrate an effective business model while at the same time satisficing if not necessarily satisfying the customer. On the other hand, the demands on the public sector are quite different. Their need is to maximize public service while not violating laws requiring governmental neutrality or norms that require they be noncompetitive with the private sector. We also believe that an inspection of the formal methods presented here shows the communication discipline to good advantage. We expect that as the services available via the Internet continue to expand, so too will the understanding of what services and packaging will best serve the interests of particular user segments.

145

146

THE INTERNET AND HEALTH COMMUNICATION

Finnegan, Alexander, Rightmyer, Estabrook, Gloeb, Voss, Leviton, and Luepker address a continuing issue in public health: How can communities incorporate, replicate, or customize intervention programs and strategies developed through formal research projects? Their chapter describes the use of a Web site to disseminate health intervention "technology/ or intervention planning processes, strategy development, materials, and process-based experience in intervention implementation, specifically addressing the problem of patient delay in seeking care for heart attack symptoms. Schneider, Frechtling, Edgar, Crawley, and Goldstein discuss and analyze the purpose, goals, development, and multi-methods evaluation of the medicare.gov Web site for Medicare, the largest health insurance program in America. The evaluation covers a wide range of domains, such as user characteristics, reactions and uses, attainment of site standards, usability, methodological considerations in Web site evaluation, and the role of assistive technology. The evaluation included online surveys of Internet users and specifically of visitors to the site, focus groups, "think aloud" interviews, expert assessments, and assessments by visually impaired users. The results led to a set of recommendations for revisions to the site. Witherspoon analyzes the Web sites of leading HMOs to measure how much they exploit the potential of this technology to meet their simultaneous goals of health maintenance of their members at the lowest possible cost. Using the Mayo Clinic's 0 @ s i s Web site as a benchmark and the transtheoretical model of health communication as a guide, she examines the presence and comprehensiveness of certain types of information and levels of interactivity on the selected Web sites. Rice, Peterson, and Christine develop a typology of Web site features (text entries, e-commerce, multimedia content, navigation assistance, interactivity, search methods, and other) based on prior research and pilot studies. They then use that coding typology to evaluate a set of commercial, and a set of governmental, health database sites. 7

7 Using the Web to Assist Commimities in Public Health Campaign Planning: A Case Study of the REACT Project JOHN R. FINNEGAN, JR. DEBORAH ALEXANDER JASON RIGHTMYER BARBARA ESTABROOK BERNADETTE GLOEB MELI Ν DA VOSS LAURA LEVITON RUSSELL V. LUEPKER

A

continuing issue for disseminating public health campaign research is how communities may incorporate, replicate, or customize intervention programs and strategies developed and tested through research projects (Backer, Liberman, & Kuehnel, 1986; Butler-Jones, 1996;

AUTHORS' NOTE: This research was supported by grants of the National Heart, Lung, and Blood Institute, Bethesda, MD (U01-HL53211) and an education materials grant from Genentech, Inc., San Francisco, CA.

147

148

HEALTH INFORMATION SITES

Cameron, Brown, & Best, 1996; Goodman & Steckler, 1990; Green & Johnson, 1996; McLeroy, 1994; Mittelmark, 1996; Murray, 1986). Although Standard dissemination methods, including published journal articles, conferences, presentations, and consulting, convey useful information about health intervention planning and strategies, they seldom capture the richness and detail of community-based public health interventions. Communities interested in developing interventions to address specific health problems do not always have access to such detail, and it is seldom available on demand. The emergence of new information technology, including especially the Internet and the World Wide Web (WWW), offer innovative means to address this issue (Friede, Blum, & McDonald, 1995; Gradstein, Hofman, & Reuben, 1995; Jennett & Premkumar, 1996; Mrad & Lenert, 1998). This case study chronicles the use of new information media to disseminate health intervention "technology/' which we define here as intervention planning processes, strategy development, materials, and process-based experience in intervention implementation and evaluation. Specifically, this case study illustrates the use of the World Wide Web to make available to communities intervention technology addressing the problem of patient delay in seeking care for heart attack symptoms. Funded by the National Heart, Lung & Blood Institute (NHLBI), the REACT Project (Rapid Early Action for Coronary Treatment) was a community trial to target this important public health problem (Simons-Morton et a l , 1998). Although the specific application emphasizes the problem of patient delay in heart attack, much of the intervention planning process and attendant education strategies are universally applicable to public health problems generally.

BACKGROUND Despite advances in cardiac treatment, patient delay in seeking care for heart attack symptoms remains a major public health problem in the United States and other countries (Cobb, Weaver, Fahrenbruch, Hallstrom, & Copass, 1992; Luepker, Raczynski, Simons-Morton, & REACT Study Group, 2000; Weaver, 1995). The 1980s witnessed the emergence of thrombolytic agents and other reperfusion therapies designed to open clogged arteries. Studies of these advances have shown that heart attack mortality and heart muscle damage could be reduced further if treat-

Public Health Campaign Planning

149

ment were delivered to patients more rapidly (Conti, 1990; Hennekens, O'Donnell, Ridker, & Marder, 1995). Improvements in emergency medical services and emergency medical department treatment protocols (Gibler, Walsh, Levy, & Runyon, 1992; Sharkey, Brunette, Ruiz, et a l , 1989) have reduced organizational sources of treatment delay; patients themselves remain the most significant source of delay. In the United States, persons experiencing heart attack symptoms delay a median of some 3 hours before deciding to seek emergency medical care. This comprises about 65 % of total delay time in administering treatment and hasn't changed much since the 1980s (Goldberg, Brady, Müller, et al., 1990; Goldberg, Gurwitz, Yarzebski,et a l , 1992). To test the hypothesis that community-based interventions may reduce patient delay in seeking care for heart attack symptoms, the National Heart, Lung and Blood Institute (NHLBI) funded the REACT Study, a community trial, in 1994 (Feldman et al., 1998; Raczynski et a l , 1999; Simons-Morton et al., 1998). The project was a randomized community trial composed of 10 pairs of communities (N = 20; population range = 70,000-240,000) in each of five U.S. regions (Midwest, Northeast, Northwest, South, and Southwest). (Six university centers were involved in the study: the University of Alabama at Birmingham; the University of Massachusetts, Worcester; the University of Minnesota, Minneapolis; Oregon Health Sciences University, Portland; University of Texas, Houston; and the University of Washington, Seattle. In addition, the New England Research Institutes [NERI], Watertown, Massachusetts, served as the study coordinating center, and NHLBI investigators were directly involved in the research.) Matched-pair communities were selected in part to minimize overlap of media markets and thus forestall potential contamination of the design. Half the communities in each region were randomly assigned to receive an 18-month intervention (April 1996-September 1997) to reduce patient delay (treatment condition); the other half were assigned to a control condition and therefore received no intervention during the study period (Feldman et al., 1998). The intervention implemented in the treatment communities included four components: (a) community organization (e.g., organization of community and health leaders into coalitions and advisory groups); (b) public education consisting of mass media (e.g., newspaper and television stories and public service advertising), targeted media (e.g., direct mail, video tapes), and speaker presentations to groups and organizations; (c) professional education (e.g., training presentations to physicians, nurses, and

150

HEALTH INFORMATION SITES

Emergency Medical Services technicians); and (d) high-risk patient education (e.g., presentations to patients admitted to hospitals with a heart attack diagnosis). Although final results of the REACT Study are in review at this writing (Luepker et al., in press), investigators decided to initiate an intervention dissemination project for several reasons. First, control communities in the randomized design were promised access to REACT campaign-planning methods and strategies post intervention to secure their cooperation as research design controls. Second, investigators wished to provide an ongoing source of information and training for treatment communities that indicated a desire to continue intervention efforts. Third, investigators wished to capture "frontline" experiences of community intervention specialists prior to the close of the study. Fourth, the National Heart, Lung and Blood Institute's National Heart Attack Alert Program (NHAAP) viewed the REACT Study as an important source of information in the future for communities to tackle the problem of reducing patient delay.

PLANNING AND IMPLEMENTING THE WEB SITE Effective use of new media technologies requires systematic planning similar to processes used to plan public health interventions themselves. To begin with, the REACT project developed a core team composed of seven members. These included two REACT investigators with experience in campaign development and evaluation; intervention coordinators who had developed and implemented the REACT intervention; and other staff with skills in communication and information packaging. The team also included an individual with special skills in health informatics, including Web site development, programming, and related technical expertise. The team approach provided a mix of skills: direct experience with the REACT intervention; technical expertise in communication, writing, graphic layout, and information packaging; and skills in Web programming. As the team proceeded to plan, it identified, obtained, and coordinated needed expertise, feedback, and advice from REACT investigators and staff from around the country. The team followed a 10-step planning and implementation process proposed by Rightmyer (1998), similar to the Social Marketing process used in many public health campaigns. Social Marketing is a planning scheme that adapts commercial marketing approaches to the dissemination of

Public Health Campaign Planning

151

Steps in Planning the REACT Web Site 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

Preliminary questions, data gathering Define scope and mission Detailed outline of Web site sections Visual representation of site organization Site layout, page design decisions, and user interface Collection of materials Technical training Technical development (implementation) of the design Development and implementation of a maintenance plan Development and implementation of an evaluation plan

ideas and behavior (Kotier & Roberto, 1989). First articulated in the early 1970s, it requires setting measurable objectives, gathering formative data to segment target audiences in communities, developing and implementing an appropriate campaign to achieve objectives, and measuring impact and effects. The steps adapting this planning approach to the development of the REACT Web site are summarized in Table 7.1.

Preliminary Questions and Formative Data Gathering At this stage, the team sought first to clarify the purpose of the dissemination activity and the clear advantages of using the Web compared with, or in combination with, other more standard dissemination strategies. Simultaneously, the team identified potential users to create a profile of needs and their likely level of Web familiarity and skills. In addition, the team discussed technologies required for meeting user needs, and whether these were feasible technically and economically. The team decided with consultation from REACT investigators and staff that the primary purpose of the dissemination activity was to provide lay community and health professional leaders in REACT control communities with postresearch access to all phases of the intervention: planning, materials, experts, and the richness of intervention experience by staff who carried it out. However, following consultation with field intervention staff, the team also recognized that such activity could be useful to treatment communities planning to carry on patient-delay campaigns following the close of the formal study. After consulting with NHAAP staff, the team also recognized that the REACT experience, whatever the final re-

152

HEALTH INFORMATION SITES

suits of the study, may also provide insights for any community wishing to address the problem of patient delay. The team considered that while dissemination activity was directed at the problem of patient delay per se, much planning information and experiential insight might also prove useful to communities planning to address other kinds of public health problems. In so doing, the team emphasized the feasibility, practicality, and experiential usefulness of the intervention components necessary for program success, rather than summative treatment of whether the campaign theory worked per se. In analyzing the potential usefulness of the Web, the team recognized that a mix of active organizing strategies and access to training and reference materials would be required. The dissemination process envisioned would combine group presentations to leaders in REACT control communities, with continuing on-demand access to the REACT intervention. The team chose the Web as best suited to provide on-demand access and updated material with the added characteristics of multimedia and interactive formats. The Web is also best suited to link communities with experts to share experiences with others involved in similar activities around the country. The team also assessed likely end-user Web access and skills. Because the primary audience was judged to be community and health professional leaders (high socioeconomic status), access to the Web and basic user skills were likely to be good, based on profiles of Web users (Mechanic, 1999; Weston, 1997). Although some leaders were likely to have high-speed institutional access to the Web, the team judged that others would have slower modem access. This assessment was an important consideration in designing the Web site to serve slower modem users as well, and suggested simplicity of design. The team also acknowledged that while the site would be aimed at community and health professional leaders primarily, other persons were likely to locate the site by informal "surfing" or through formal Web searches. Many of these persons were likely to be seeking personal health information about heart attacks rather than information about organizing community interventions, the primary purpose of the Web site (Mechanic, 1999). Because of this, the team decided that a statement of the site purpose was needed early upon access. The statement urged individuals seeking personal information about heart attacks or heart disease to link to other sites better conveying such information. Appropriate links would be provided (e.g., the American Heart Association).

Public Health Campaign Planning

153

Definition of the Scope and Goals of the Project With preliminary questions asked and information gathered, the team made final decisions about the specific mission and objectives of the Web sije. In general, the team defined the Web site's scope using the REACT intervention research as a model Thus the main portion of the site would provide training in the general planning process followed in the intervention (Social Marketing) and in each of the four principal strategy areas: community organization, community education, health professional education, and high-risk patient education. The intervention planning section would provide step-by-step detail on REACT formative evaluation strategies; identification of target audiences; program and message elements; intervention delivery channels; materials development; intervention implementation; evaluation; and supporting reports, documentation, and resources. Each of the specific strategy areas would include detailed steps, rationales, and supporting materials and resources. The team also identified several features to enhance the training modules. These included summaries of field intervention experiences in each strategy area and a means of identifying and interacting with REACT investigators and field staff via e-mail. The team defined an initial development time line calling for the site to be functioning reasonably within a period of about 60 days, at which time dissemination meetings with REACT reference communities were scheduled to begin (autumn 1998).

Detailed Outline of the Project's Components (Blueprint) At this stage, the team allotted responsibility to various members for creating a detailed outline of each Web site section's content. These were presented to the team as a whole, to other REACT investigators and staff, and then discussed and revised. Outlines were a necessary and useful step in clarifying the concepts and content of each section and in defining the materials and multimedia and interactive formats needed.

Visual or Graphic Representation of the Site Organization The team also produced multiple draft schematics of the Web site's organization. Displayed in Figure 7.1, the final version divided the site into

154

HEALTH INFORMATION SITES

Links to browsers, helper applications

SPLASH PAGE WELCOME

Links to other heart sites

REACT background

What is a heart attack?

Introduction Site overview

Problem of delay

I Registration, log-in MAIN MENU INTERVENTION PROCESS

1

1

Findings from the field 1 COMMUNITY ORGANIZATION

Target groups

"I

The Inessage

COMMUNITY EDUCATION

1

"""

ι

PROFESSIONAL EDUCATION

Talk lo the experts

ι

PATIENT EDUCATION

Download materials

Figure 7.1. Organization of the REACT Web Site

two large sections with several subdivisions in each section. The top section was designed as an introduction to the Main Menu, which contained the principal purpose, content, and site design. The site's point of entrywas a Splash Page, a simple graphic page that appears when the site URL is entered into a Web browser (http://www.epi.umn.edu/REACT). This linked to a Welcome page that provided a preliminary description of the site's purpose and described the software requirements to use the site most efficiently. Software requirements were relatively simple, keeping in mind the needs of end users using slower modem links. A Web browser, of course, was the basic required software. However, the site was designed to work best with recent versions of browsers that were identified and linked to free download sites (Netscape™, Internet Explorer™). In addition, links connected users to a portable document format (.pdf) reader (Adobe Acrobat™) that permitted downloading designed text documents, and to a video extension (Quicktime™) that permitted use of video files. The welcome page was linked to an introduction and site overview that more specifically described the site's purpose and organization. This subsection also encouraged persons looking mainly for personal health infor-

Public Health Campaign Planning

155

mation about heart attacks or heart disease to link to sites with more appropriate content (e.g., the American Heart Association). However, links were also provided to pages describing the REACT Study, how to recognize heart attack symptoms, and the problem of patient delay. An important site page was devoted to some 65 links to public health sites, including meta-directories, directories, search engines, health statistics, and community health planning resources. These resources also linked individually to sections in the main menu. The main menu (inside the large box of Figure 7.1) contained the site's central training and materials sections. The section begins with an overview of the intervention planning process utilizing a Social Marketing approach described earlier. It provides training in the general intervention planning process and connects each step to the specifics of the REACT intervention. From this section, users link to the specific strategy development areas of community organization, community education, health professional education, and high-risk patient education. Each of these sections links to a description of the study's target groups, message elements, and intervention experience. This last section, "Findings from the Field," summarized experiential results from focus groups of field intervention staff who were interviewed by a REACT investigator following completion of about two thirds of the intervention. "Talk to the experts," a section providing e-mail links to REACT investigators and staff, may be sorted online by area of expertise or by region of the country. Each of the intervention methods sections included links to each of the others and to sample educational materials, some of which could be downloaded or viewed on site.

Site Layout, Page Design Decisions, and User Interface While the content organization of a Web site is central to its purpose, considerations of layout, graphic design, and user interface are vital to its ease of navigation and effectiveness in communication and presentation format. The REACT Web team spent quite a bit of time "surfing" other Web sites, especially those concerned with Web design. With lower-tech end users in mind, the team opted for a general page design that would be readable on small as well as large computer monitors. The team avoided complex, heavily layered page designs that often take longer to download.

156

HEALTH INFORMATION SITES

Instead, the team opted for a two-column page design with a shaded left border anchoring an index and other links to each section and the main menu. The team also opted for color-coded banners identifying the various sections to permit end users to navigate the site more easily. Color coding and graphic design of the Web site were adapted from the REACT intervention itself. Working with the same professional graphic design group that created the REACT intervention, the team created a "look" for the Web site that included a color palette, type styling, and graphics. These helped to coordinate the site's visual look with previously produced REACT video and print materials. The page design included many cross-links and anchors to permit users to navigate quickly and easily. Along similar lines, the Web team developed a "fast track" access path for physicians and other health professionals with limited time to surf the whole site. This approach provided health professionals with direct links to access quickly those parts of the site with immediate relevance to their work (e.g., professional training materials).

Collection of Materials During this stage, the Web team gathered various materials that would be needed to implement the site. Materials collected included print, audio, and video intervention materials; pictures; charts; documents; and links. An important aspect of this stage was the transfer of materials into Webbased formats that could be integrated with the page design. Because nearly all REACT materials were originally written and produced by project staff, few copyright issues were encountered. However, some video materials utilized brief commercial film clips to illustrate Hollywood depictions of heart attacks in the movies. If film clips were produced in the past 75 years, they required copyright clearance from the producers, directors, and often the actors themselves, and also the payment of a royalty. Royalties vary based on a film's age (cheaper if produced before 1970; more expensive after) and the conditions of use. Use of a film clip for a commercial purpose may run $3,000 or more for a minute or less. Use of film clips in "educational" settings (i.e., no money exchanged in distributing a product) permitted a discount. Because REACT qualified as an educational program, the most expensive royalty it paid was about $1,500 for a 30-second film clip. Securing clearance was time-consuming, even for a staff member familiar with the commercial film industry (about 3 to 6 months of diligent phone calling and draft legal agreements).

Public Health Campaign Planning

157

Technical Training Although team members were familiar with the Web as end users, only one possessed deep technical knowledge about Web page implementation and more sophisticated programming. With the emergence of newer Hypertext Mark-up Language (HTML) software, however, Web page development no longer requires deep knowledge of programming. Nevertheless, even basic Web page construction requires a certain amount of training in the idiosyncrasies of Web protocols, file formats, and hierarchical structures. The health informatics member of the team was designated as Webmaster and provided other team members with essential training. He also evaluated several Web page development programs for creating templates that team members could easily use in implementing their portions of the site and selected one for sophistication yet ease of use by nontechnical team members. Training included basic use of these programs to transfer existing REACT materials into Web-based formats.

Technical Development (Implementation) of the Design The team developed a page template for each section and began to construct the Web site based on the design and content plan. The implementation of the basic design from start to initial operation required about 60 working days (60 x 5 persons x 8 hours per day = 2,400 hours).

Development and Implementation of a Maintenance Plan Unfortunately, many Web sites are initially developed and implemented without regard to maintenance. Many Web sites rapidly become outdated and less useful, especially when links to other Web sites change or disappear altogether. In the case of the REACT Web site, the team regarded maintenance as an important issue from the start. The team currently maintains the site, checking and monitoring links and users and revising and updating pages as needed. Newer generations of Web management software make the task of maintenance somewhat easier by automatically checking and flagging links that become outdated.

158

HEALTH INFORMATION SITES

Development and Implementation of an Evaluation Plan Relatively little study has been made of the role and effects of the World Wide Web in public health settings. This is not surprising, given its relative novelty as a tool of health promotion and health education. Both formal and informal evaluation methods are available, though as yet little applied (Fernsler & Manchester, 1997). These include programs that automatically track Web site traffic to evaluate usage, and formal online surveys that can be incorporated as part of the site itself and linked to a database for easy analysis of responses. Surveys also may be conducted of users by e-mail (assuming that the site has included a method of obtaining users' e-mail addresses). There are also less formal methods available, including online or in-person user discussion groups, review by peers, usability studies (focusing on ease of site use), and anecdotal evidence. All are important to continued formative evaluation of Web sites and are similar to ongoing "message testing" strategies used in planning campaigns. In the case of the REACT Web site, the team used other REACT investigators and staff initially to evaluate its usability. This was helpful in refining the site to meet the needs of distant users with modest equipment. The team also implemented a Web tracking system to record information about the site's use and collected some anecdotal information. Using the Web Analog™ program, the team measured the site's traffic during two periods of implementation: for a period of 6 months after its activation (December 1997-May 1998) and for a period of 6 months about 1 year later (May-October 1999). Of particular interest to investigators were the site's traffic patterns, access by specific domains and referrals, and use of specific content areas. In the 12 months examined, the site received 23,896 successful requests, with 22,124 of these for specific pages (see Table 7.2). The vast majority of requests (80.4%) occurred during the second period, with October 1999 the busiest month (the most recent month surveyed). Requests during the first period averaged 25.7 per day, with a fourfold increase averaging 106.2 per day during the second period. In all, these included requests for 768 distinct Web site files accessed from 1,331 identifiable, distinct hosts. This dramatic increase occurred despite relatively little publicity about the site. From an examination of the host report, it is clear that this was due partly to Web search engines picking up and displaying the site in response to specific searches for information. Although the REACT site was not inde-

Public Health Campaign Planning

:

;V* )ÜtyZ*

Month/Year

5

Days

159

REACT Web Site Traffic During Two Six-Month Periods Cumulative Days

Number of Requests

Cumulative Requests

Number of Pages

Cumulative Pages

Period 1: December 1997—May 1998 Dec 1997

31

31

285

285

275

275

Jan 1998

31

62

1,426

1,711

1,258

1,533

Feb 1998

28

90

736

2,447

636

2,169

Mar 1998

31

121

824

3,271

735

2,904

Apr 1998

30

151

990

4,261

878

3,782

May 1998

31

182

420

4,681

346

4,128

1,349

1,349

1,307

1,307

Period 2: May 1999-—October 1999 May 1999

31

31

Jun 1999

30

61

3,665

5,014

3,390

4,697

Jul 1999

31

92

3,334

8,348

3,131

7,828

Aug 1999

31

123

3,467

11,815

3,302

11,130

Sep 1999

30

150

3,432

15,247

3,201

14,331

Oct 1999

31

181

3,968

19,215

3,665

17,996

—

363

—

23,896

—

22,124

Totals NOTES:

Average requests per day, Period 1:

25.7

Average requests per day, Period 2 :

106.2

Distinct files requested, total: Distinct hosts served, total:

768.0 1,331.0

pendently registered with search engines, its parent site (http:// www.epi.umn.edu) was registered. In this process, Web search engines typically scour each page of a registered site to collect and check all of its links to other sites. A daily traffic summary showed that the site has been used (in rank order) mainly on Fridays, Wednesdays, Thursdays, and Tuesdays. The majority of requests, more than 1,000 per hour, occurred between 11 a.m. and 7 p.m. Central time, with a similar spike from midnight to 1 a.m. Table 7.3 reports trends in site use according to domain of the end user. The overwhelming majority of requests came from commercial (48.2%) and educational sites (26.3%) from within the United States. Nonprofit

160

HEALTH INFORMATION SITES ;

REACT Web Site Traffic by Accessing Domain

Domain Commercial (.com)

Number of Requests

Percentage of Total

11,527

48.2

Educational, U.S. (.edu)

6,286

26.3

Unresolved addresses

3,139

13.1

Network (.net)

1,361

5.7

Nonprofit (.org)

632

2.6

Government, U.S. (.gov)

473

Other, U.S. (.us, .mil) International Asia (Japan, Malaysia, Mongolia) Australia-New Zealand Europe (Total) Central Europe (France, Hungary, Switzerland) Northern Europe (Denmark, Ireland, Netherlands, United Kingdom) Southern Europe (Albania, Bulgaria, Greece, Italy, Spain) Near East (Israel, Saudi Arabia) North America (Canada, Mexico) South America (Argentina, Chile, Columbia, Peru, Uruguay) Total

51 427 14 38 141 3 88 50 3 220 8 23,896

2.0 Ο c U Ä Q.SJ Ο c ~σ ο U Ο) Μ- u CD ο J- w ω tjjT)

ο

Cl _Q _o

c

δ

Β ν

Ο

Ο. CU

~D

ο

^_

Ε cu u

ε §

1/5

03 " u U • ^

2 «> / Ο 03

ί=! °-

"Ο Φ ι0λ3

ω (υ

> ts cu cu

CD

t/T ^

j —•{

API

/

\aPI Hbätfeare

Purepac(US) Russeis (NZ) Com. Drug Andrews Lab Sigma Pharma

339

Soul P. (Man)

PeCC Health Care Supply Chain

\\\

Gosford

Cn^n^«!a4\^^^ Tet

/

\

Healthsense \ '—— PNSSynergy

\ SIGMA CO.

AMCAL ^

W S PATTISON

\

" -~ Guardian

\

Pharmacy Care PPS (NZ) SP Chemists

\\

Wovwo

y

A

Wyong Η

Long Jetty

^s^\\ Extend-A-Care \.

Ν. \\

\

RHobart

\^jy^^» RAdelaide

^^^^^^hVlvTedShpppe χ

\ F Η FAULDING

Faudling Η.

HcoA

^y^ ^^l^^^^^^^^^^^^^\

/ \\

Austin

^^Ballarat

P^jacist Advice

^^^^

J^^£^~^^—~--~-

^^^^—/~

Abbott

Figure 15.2.

Pharma

/JpfMayne ^^^^>^Oßsford PharTna^-J^^^^^^

. HSA

PharmaPack

SB Sterling

DOCTORS

y

Amed

AstraZeneca

Rallarat

^W/l^^~^ \y££ry$^^^

"

SB

GROCERIES

^ ^^^2^^^—"'

HMR Amrad

\.

St Vincents \

Essendon \ \

Westmead Pmacq Η RNS Alfred

340

ONLINE HEALTH COMMUNITIES AND E-COMMERCE

lian agent for a number of overseas manufacturers of health care products and equipment. On the other hand, Australian Pharmaceutical Industries (API) serves as wholesaler for many manufacturers but is the exclusive distributor for its hospital supply manufacturing subsidiary, Amed Supplies.

MAN to RET/HOS/EU Links Global pharmaceutical manufacturers often have direct links to several retail outlets and most hospitals, including those already being serviced by the five wholesalers. Global manufacturing giant Glaxo Wellcome, in particular, has begun distributing directly to pharmacists through a single agent. In contrast, Baxter Healthcare P/L, while maintaining links with wholesalers and distributors, also supplies products (like its sterile fluids) directly to retailers, hospitals, and even to end users (home-based patients).

Links With Distributors Wholesalers sometimes use the expertise and facilities of distributors. These also distribute the products of some manufacturers directly to hospitals. Likewise, other distributors can also pick, pack, and ship orders for manufacturers to retailers (like pharmacies).

WHO to RET Links The major drug wholesale distributors in Australia have formed alliances with pharmaceutical retail groups in order to increase their market share. Some wholesalers sell only to pharmacies while others also sell consumer health care products through grocery stores and specialty shops.

MAN/WHO to HOS Links While most public and private hospitals get their supplies directly from manufacturers, the wholesalers are also vying to supply drugs to the hospital market. Manufacturers and wholesalers also market their products (particularly drug samples) directly to health practitioners or doctors through their medical representatives.

Competitive Collaboration

341

RET to HOS to End-User Links Retail outlets have direct links to public hospitals. Similarly, almost all end users (such as hospital patients and ordinary consumers) get pharmaceutical products directly from retailer outlets, doctors, and/or hospital dispensaries. However, manufacturer Baxter Healthcare also supplies sterile fluids directly to home-based patients all over Australia. If direct manufacturer-patient consumer links were further developed, this would obviously pose a major threat not only to PEG/PeCC projects but to the key stakeholders. Yet PEG software would appear to be precisely what is required to allow an implementation of this type.

PECC ADVANTAGES Clearly, if the central focus of PeCC, to reduce waste in the health industry by improving SCM, is achieved, this will be its major advantage. One recent study (DIST/PW, 1998) suggests that up to $340 million a year from an annual budget of $1.4 billion could be saved if the health care SCM is reworked and the Internet is used for ordering barcoded consumables (Head, 1998). PeCCs project director estimates the savings as being far greater, linking a current "culture of waste" to more stringent expectations, such as those in the retail sector. Indeed, savings of 24% are predicted, as are major reductions in transaction times from order placement to delivery and payment, error-free receipt of orders, improved services to customers, and broader health sector stakeholders. PeCCs work in ΕΑΝ numbering/barcoding will help to optimize such improvements. Moreover, the study also found that the PeCC process could enhance benefits more broadly than just health care consumables, to include medical records, cost evaluation, and patient billing systems. Some specific benefits include: • Reduction in transition time from order placement to delivery and payment • Reduction in costs of overall procurement (order costs reduced from $75 to $5) • Comprehensive information on exact stock movements • Just-in-time ordering system can be established • Better matching of demand and supply by manufacturers and suppliers

342

ONLINE HEALTH COMMUNITIES AND E-COMMERCE

• Error-free receipt of orders and improved integration with order entry systems by eliminating rekeying time and errors • Rationalizing of other trading documents, such as turnover orders and possibly electronic invoicing and payment instructions to financial institutions • Improved service to customers (leading to faster payment for suppliers) • Improved inventory management and accountability within the hospital sector • Increased efficiencies in hospitals and reduced shrinkage, allowing funds to be better focused on actual patient care • Less reliance on proprietary IT systems • Major move toward standardization, opening the way for much greater interoperability • Ability to realize electronic commerce benefits without a massive investment in IT • Capability to make much greater use of customer usage and ordering patterns information This alliance is characterized more by a common search for new skills, information, knowledge, and, ultimately a cost-saving new process. In fact, PeCC may be unique, unshackled by traditional bureaucracy and yet supported by substantial government support (personal communication, interviews, 1999). More broadly, the process PeCC has been nurturing in terms of e-commerce will provide the sort of confidence in the health system that has been lacking for a long time. The private system, focused on bottom-line outcomes and advantages, will lead the public sector, and the pressure exerted on suppliers from the private sector will flow on to its public counterpart. Suppliers themselves will then pressure the public system in order to maximize the effort so that all key players are on the bandwagon together.

PECC DIFFICULTIES From inception, PeCC has had to contend with difficulties relating to the broad issue of change management in a fairly traditional and conservative industry. A new system that makes individuals more accountable; challenges current power bases; threatens jobs and vested interests in proprietary systems and threatens territorial boundaries; and challenges the pre-

Competitive Collaboration

343

vious system that hid traces of theft and/or waste is bound to be resisted by many stakeholders. Changing cultures focused on patient care without other considerations to one that also encompasses more cost consideration is also a major task. The broader problem of increased visibility and accountability is also a politically sensitive issue at all levels, especially given a focus on hospital waste and any criticism of hospital performance or government knowledge. Consequently, it is not surprising that government relations have also been a difficulty—both between Commonwealth and State governments, and also between government and industry—in terms of responsibility, accountability, culture, and politics. For example, PeCC has been successful in working with the supply side of the health care sector. The demand side of PeCC/EANnet involves many influential government agencies. Public hospital management and state government officers have different expectations in terms of e-business applications to broader hospital supply and to other aspects of acute care. Even the need to integrate the various hospital information systems is fraught with difficulty. That is because political issues of electoral sensitivity are involved, health care costs are a major social policy issue, and government performance in this area is always on show. The numbers game has also proved difficult for PeCC, with the complete cooperation of all those involved in the system, including 1,100 hospitals and about 1,000 suppliers, still a long way off in 2000. Furthermore, for some, the progression of ΕΑΝ numbering has not proceeded as quickly as had been earlier hoped, and there is evidence of a lack of total commitment by all organizations in the supply chain (personal communication, interviews, 1999). In this sense the change is perceived by many as more evolutionary than the radical and rapid change desired by the few passionate change leaders. Another ongoing challenge is dealing with the potential negatives of fully implementing supply chain management incorporating e-commerce and ΕΑΝ numbering and barcoding of products, particularly in the hospital sector. These include cost concerns for manufacturers/suppliers, wholesalers, transport companies, and hospitals, possibly initially totalling $170 million, with annual maintenance and upgrading costs running at 10% to 20% of this figure. Staff redundancy costs and possible associated conflict, covering approximately 3,000 jobs (lost mainly from hospitals, wholesalers, and transport companies), may amount to a once-off cost of $30 million (Price Waterhouse, 1998).

344

ONLINE HEALTH COMMUNITIES AND E-COMMERCE

For success to occur, there is an ongoing need to manage alliance relationships over time, and appropriate leadership within organizations involved (not only at the senior level) is critical, as is the need to move the agenda away from the technical domain owned by CIOs and into the realm of strategy and ownership by business managers. The IT industry itself needs to take the issue more seriously and to become more client focused rather than stubbornly continuing to market proprietary systems.

PECC AS COMPETITIVE COLLABORATION Areas of Collaboration There are many levels of collaboration evident in the PeCC Project, involving federal and state governments and their departments and affiliated bodies (e.g., CSIRO), universities, industry bodies and industry associations, commercial organizations, plus international entities. While there is ongoing competition for health IT resources, there is, however, a growing recognition of the need to collaborate and cooperate between the states, and between the federal government and the states. Alternatively, it has been suggested that while the private sector is increasingly prepared to do business through strategic alliances, the public sector has far less capacity to understand such relationships and little competence in managing them effectively. Furthermore, there is some suggestion that a system that works overseas, or in the Australian private health sector, may not be adopted by the public because the Australian system is considered different and the private hospital system is "treated with disdain" (personal communication, interviews, 1999). In Australia the public health sector dominates and is far more complex than the private. Perhaps the heart of the difference is, as one interviewee indicated, "In Australia health is seen as a cost; in America it is regarded as an industry." Within such diverse perceptual frameworks it is hardly surprising that attitudes toward technological and strategic developments also differ. In terms of competitive collaboration, PeCC's facilitation of agreement among five major pharmaceutical wholesalers in the PEG project stands out. It is a competitive alliance producing a new process-a new way of doing business-that can be learned and distributed as both a private and a common good among the alliance participants who are simultaneously PEG collaborators and competitors in the pharmaceutical markets. Leader firms are typically not eager to cooperate with other leaders in the same in-

Competitive Collaboration

345

dustry, partly because of antitrust concerns and partly because of intense rivalry. In general, they cooperate only at the instigation of governments, and then in carefully structured and monitored schemes and in specialized "precompetitive" areas (Doz & Hamel, 1998, p. 99).

Alliance Motivation This study supports the contention of an eclectic theory of alliance motivation based on changes in the external trading environment, revealing the need to correct an internal resource inadequacy in order to maintain competitive advantage (Child & Faulkner, 1998). The partners in the PEG project, for example, certainly fit this category, finding themselves with complementary resources and perceived inadequacies in new requisite skills and the need to learn new ways of organizing and managing. Deficiencies of EDI and the need to get a new system off the ground with reduced individual risk, and the need to convince suppliers of its efficacy, were cited as powerful motivators (personal communication, interviews, 1999). There is also "sincerity in terms of a united objective—a common system for all to benefit with all participants having the same responsibility for implementing the system internally" (personal communication, interviews, 1999). Stakeholders had diverse motivations. A major motivation for suppliers to become involved is that e-commerce will lead to more reliable and predictable payment cycles, important particularly to SME suppliers. It is also likely to have a ripple effect throughout the health sector and a flow back to insurers and hospitals. Public hospitals are spurred on by the growing need to understand the cost of health consumables fully and to avoid waste. Private hospitals, in order to stay in business and function effectively, need to find ways of saving money and to reestablish a profit margin in an environment where revenue is stagnant and costs increasing. The health funds are also searching for ways to reduce operating costs and look to e-commerce, using industry standards, as a better way of dealing with claims. Special bodies, such as the Department of Veterans Affairs, a large spender within both the private and public health industry, needs to have better data that would enable them to identify costs at each episode of care, and to ensure cost-effective outcomes for more efficient ageing services. Similarly, the Department of Defense is also a large spender in the industry and requires better information, choice, and more control over supply and reduced costs (ΑΡΗ, 1999).

346

ONLINE HEALTH COMMUNITIES AND E-COMMERCE

Pivotal to the whole process is the role and motivation of a key stakeholder group: the nursing profession, one that can constrain or enable the changes required. This group, as important if not more so than pharmacists, has to be convinced of benefits to the profession, the organization, and, critically, to their patients.

Learning Certainly, there were political motivation and agendas in the broad background to the PeCC alliances. Yet in many ways this alliance was clearly also both a learning and a business experience—learning from each other in partnership about new ways of organizing and doing business, and also establishing a more competitive position in the marketplace; and motivated by exploring innovation, learning, and internalization of new skills (Doz & Hamel, 1998; Koza & Lewin, 1998). Clearly, also, the strong positions of the participants within the industry, accompanied by high status, credibility, and reputation, were a useful magnet to draw them together. Learning occurred at various levels, with participants acquiring and learning new technological capabilities and innovative ways of doing business in the competence acquisition process (Lei, 1997). Both individual and collective learning seems to have occurred in this alliance (Larsson, Bengtsson, Henriksson, & Sparks, 1998). Moreover, a key feature seems to have been a heavy emphasis on learning with, rather than learning from, partners, plus mutual rather than competitive learning (Child & Faulkner, 1998). One interviewee suggested that positive collaboration had evolved in a situation where partners were comfortable; there were few reservations and no threats.

Trust Trust, therefore, is a key element, developed and nurtured by those playing boundary roles and by alliance managers through the ongoing meetings and formal and information communication mechanisms. In this sense, "trust may be reasonably viewed as an emergent and epiphenomenal property of successful alliances" (Koza & Lewin, 1998, p. 259). The basis for trust in this alliance lay in an informal agreement at the outset that all participants would achieve the same level together and then it would be up to the individual organizations to use their new skills inhouse in the most efficient way possible to gain competitive advantage

Competitive Collaboration

347

(personal communication, interviews, 1999). This agreement, not to move ahead until all partners are ready, defeats the "race to win" syndrome discussed in much of the collaboration literature. Nevertheless, there still is a pragmatic guardedness as to just how much one exposes to competitors. Competition is alive and well among the original gang of five, with ongoing jostling for greater market share, in spite of collaboration in the PEG project. Upholding strategic consistency is complex in an alliance where participants are both collaborators and competitors.

Coordination and Control Coordination and control have been made easier by the very pragmatic motivation and commitment of the partners. Coordination has generally been of an intensive interdependent type with participants applying complementary resources to the common activity in an integrated way to produce a new "service" or business process (Grandori, 1998). Control has very much relied on behavior and process controls rather than some more rigid mechanisms or an emphasis purely on output controls (Koza & Lewin, 1998). Moreover, a successful balance has been achieved between collaboration and competition. A sensible compromise between rigidity and flexibility has also been maintained. Partners have recognized and acted on the basis of understanding the alliance "as an evolving relationship punctuated by a series of commitments, steps, and 'bargains' explicitly negotiated or implicitly accepted over time" (Doz & Hamel, 1998, p. 20). Governance and coordination, therefore, have built on the foundation of partner consensus agreement and little formal authority structure. Effective communication has been critical. Face-to-face (and teleconferencing) regular meetings among participants have been complemented mostly through the liaison or boundary-spanning roles occupied by the key representatives of the partners and major stakeholders involved at the frequent PEG meetings, for example, or the less frequent and more formal Council meetings. There is extensive communication at and between PEG regular meetings, and a ready exchange of shared experiences in implementing systems.

NEAR-TERM TRENDS AND IMPLICATIONS It is increasingly clear that the technology involved plays a minor role in what is essentially a major change across the health sector. This was recog-

348

ONLINE HEALTH COMMUNITIES AND E-COMMERCE

nized in one recent workshop organized by PeCC with Australian private hospital representatives, wherein five code words were decided as a mission statement for the group—communication, cooperation, compromise, critical mass, common standards (ΑΡΗ, 1999). Only the last is really technology-related per se; the others are all about change management and collaboration. Moreover, given the need to track collaboration across the PeCC projects over time, longitudinal research over the next few years seems vital in order to see the real evolution and management of collaboration past the formation and implementation (and perhaps more halcyon) days of partnership. It is also important to contextualize better such relationships within industry changes that are likely to occur within the next decade. This is especially true given that the health sector still faces major challenges, including the actual, widespread implementation and acceptance of e-commerce-based business solutions. Longer-term research would also permit better evaluation of collaboration and performance than is at present possible—not only evaluation of the alliance but also individual organizational performance; financial, operational, and organizational effectiveness; and tangible and intangible benefits. Some countries have mandated a PEG-like system, and this is a possibility in Australia. The majority of players believe it better for now that industry continues to work collaboratively and manage the change process at such a pace that ensures all stakeholders can attain mutual benefits from a new open system. There seems to be no diminishing of the competitive positioning of those involved in the alliance, but the collective interest has won out through tacit regulation by cooperative norms and a form of trust or civic culture promoting the good of the whole (Carney, 1998). Certainly even in the broader umbrella of PeCC we are witnessing a range of legitimate alliances, demonstrating excellent fit and based on complementary strengths and united goals. What this represents, as Gomes-Casseres (1996) suggests, is an alliance and e-commerce revolution that will not only change health industry players but the very way in which that industry itself is organized. Analysts suggest that the Australian pharmaceutical industry is "on the edge of a precipice" (Ferguson, 1998, p. 48) because of the rapid changes affecting this sector today. Each of the major players is currently undergoing alteration in positioning to make way for changing trends, such as longer trading hours, price benefits at supermarkets, and consumer demand for convenience. Many believe that the pharmaceutical industry will transform from one that is predominantly product-based to one that is infor-

Competitive Collaboration

349

mation-based, and that service and location will become of overwhelming importance to the consumer. This is almost certainly an overstatement. For close to two decades now (see, e.g., Barrett, 1986-1987), much has been made of ways in which superior information (facilitated by novel uses of IT) can be employed to provide organizations with a competitive edge. However, a closer analysis of most of the oft-quoted, classic examples (e.g., American Hospital Supplies, American Airlines, Thomson's Holidays, and Merrill Lynch) reveals that these are mostly instances of clever applications of operational support systems, rather than strategic, information systems. PEG, too, at least in its current form, could be classified as an operational support system. Nevertheless, we see enormous potential for the development of exciting, genuinely strategic information systems built upon the current PEG foundation. If this is to occur, much will depend on the attitudes of the partners and, in particular, their risk-taking propensity.

REFERENCES ΑΡΗ. (1999). Minutes of the Australian Private Hospital Supply Chain Workshop, June 29,1999, Sydney. Barcett, S. (1986-1987). Strategic alternatives and inter-organizational system implementations: An overview. Journal of Management Information Systems, 3(3), 516. Bryan, M. (1999, May 7). Australia running third in the race for Internet commerce. Australian Financial Review, p. 60. Buono, A. (1997). Enhancing strategic partnerships: Intervening in network organizations. Journal of Organizational Change, 10, 251-266. Carney, M. (1998). The competitiveness of networked production: The role of trust and asset specificity. Journal of General Management, 21(1), 1-23. Child, J., & Faulkner, D. (1998). Strategies of co-operation. Oxford: Oxford University Press. Department of Industry, Science & Tourism (DIST)/Price Waterhouse. (1998). Getting business online. Canberra, Australia: DIST. Doz, Y., & Hamel, G. (1998). Alliance advantage. Boston: Harvard Business School. Eisenhardt, Κ. (1989). Building theories from case study research. Academy of Management Review, 14, 532-550. Ferguson, A. (1998, August 3). Trouble in the drug trade. Business Review Weekly, pp. 48-50. Gallagher, P. (1999). Turnover order project. Unpublished manuscript, Canberra. (Contact Patrick Gallagher, Project Manager, PeCC, [email protected]) Gallagher, P., McCrea, P., & McGrath, C. (1977, April). PeCC project mid-term report. Canberra: Department of Industry, Science and Tourism.

350

ONLINE HEALTH COMMUNITIES AND E-COMMERCE

Gomes-Casseres, Β. (1996). The alliance revolution. The new shape of business rivalry. Cambridge, MA: Harvard University Press. Grandori, A. (1998). [Editorial]. Organization Studies, 19(4), v-xiii. Gulati, R. (1998). Alliances and networks. Strategic Management Journal, 19, 293317. Head, B. (1998, June 5). A net cure for hospital waste. Australian Financial Review, p. 45. Inkpen, A. (1998). Learning and knowledge acquisition through international strategic alliances. Academy of Management Executive, 12(4), 69-80. Inkpen, Α., & Beamish, P. (1997). Knowledge, bargaining power, and the instability of international joint ventures. Academy of Management Review, 22(1), 177-202. Joint Statement from Australia and the United States on Electronic Commerce. (1998, November 30). [White House press release]. Kilbane, D. (1999, January). Australia pilots pharmaceutical EDI systems to save $340 million. Automatic I.D. News, 1-2. Koza, M., & Lewin, A. (1998). The co-evolution of strategic alliances. Organization Science, 9(3), 255-264. Larsson, R., Bengtsson, L., Henriksson, K., & Sparks, J. (1998). The interorganizational learning dilemma: Collective knowledge development in strategic alliances. Organization Science, 9(3), 285-305. Lei, D. (1997). Competence-building, technology fusion and competitive advantage: The key roles of organizational learning and strategic alliances. International Journal of Technology Management, 14, 208-237. Lutz, S. (1997). Learning through intermediaries: The case of inter-firm research collaborations. In M. Ebers (Ed.), The formation of inter-organizational networks (pp. 220-237). Oxford: Oxford University Press. McCrea, P. (1999). Reducing the cost of supply chains. Unpublished manuscript, Canberra, Australia. (Contact Patrick Gallagher, Project Manager, PeCC, casprel@ ibm.net) Mcintosh, T. (1998, September 1). Hard drive. The Age and Sunday Age, p. 14. Mintzberg, H., Dougherty, D., Jorgensen, J., & Westley, F. (1996). Some surprising things about collaboration—Knowing how people connect makes it work better. Organizational Dynamics, 25(1), 60-70. Nalebutt, B., & Brandenburger, A. (1997). Co-opetition: Competitive and cooperative business strategies for the digital economy. Strategy & Leadership, 25(6), 2935. PeCC briefing document. (1998, November 26). Canberra, Australia: Department of Communication, Information Technology, and the Arts. PeCC & Royal College of Nurses Association. (1999). Canberra, Australia: National Office of Information Economy. PEG manifesto. (1999). Unpublished paper. Canberra, Australia. Price Waterhouse. (1998). PECC—The way forward. Canberra, Australia: Department of Communication, Information Technology, and the Arts. Sachs, P. (1995). Transforming work: Collaboration, learning and design. Communications of the ACM, 38(9), 36-44. Yin, R. (1994). Case study research: Design and methods (2nd ed.). Newbury Park, CA: Sage.

PART V

Public Policy Experiences and Expectations

lublic policy profoundly affects the direction and deployment of new JL technologies in society, certainly in terms of the Internet and health communication. Policy connects technology (i.e., the Internet) with users (e.g., patients and their families, scientists, lawyers, accountants, doctors). Policy can retard or prevent the creation of technologies that might otherwise be enthusiastically adopted by the public (and the reverse!). A case in point is cellular mobile telephony, which was delayed for more than a decade as the U.S. Federal Communications Commission (FCC) sought to determine what would be the optimal policy for the equitable deployment of the scarce resource. Meanwhile, the Europeans were not only enjoying this convenience, but in some cases were having their lives saved by it. Hence, the policy dimension is one that must be considered at the early stages of a new technology, the Internet being no exception. When considering policy issues, every program is viewed through at least two lenses. These lenses are efficacy and equity. In terms of the Internet and health care, the federal government is seeking to achieve progress in both of these areas. As demonstrated by the Baur, Deering, and Hsu chapter, the U.S. government has committed itself to a vigorous interagency effort aimed at offering useful Web sites that can provide guidance 351

352

THE INTERNET AND HEALTH INFORMATION

concerning health care and health programs to the general public. A more controversial policy of the federal government, though, is to underwrite the deployment of the Internet to hospitals (and schools). What is unclear about this program is, given the expenditure of billions of dollars, who precisely will benefit and to what extent. These questions are all the more pertinent since the money to pay for this program is raised from those who use the telephone for, in many cases, access to the Internet itself. This regressive tax makes it more difficult for these very same people to afford Internet service. Another dimension along which policy can be considered is the question of novelty. Some hold that the Internet merely requires the extension of already well-established principles to govern the new intellectual terrain that it has made available. This view holds that while clarifications and extensions may be required, there is no need to fundamentally rethink issues. A contrasting view is that the Internet is, sui generis, something unique unto itself. Health care, medicine, and medical advice will be dispensed in special new ways that transcend and transgress traditional approaches. A consequence of this view is that new policies must be considered and constructed. In particular, new legal regimes must be formulated and implemented to address these unprecedented problems and issues. The approach of Bayer, who addresses issues of online prescriptions in this section, would suggest the former view. The approach that Katz and Aspden take toward medical record privacy and security would suggest the latter view. Baur, Deering, and Hsu discuss the federal role in development, delivery, and oversight of online health information and interactive health technologies. Specific topics include quality of information; fraud and abuse in online advertising and sales; privacy; equal access to Internet and technology. They summarize specific recent initiatives such as healthfinder.gov, the federal gateway to reliable consumer health and human services information; the Science Panel on Interactive Communication and Health (www.SciPICH.org); oversight efforts of the Federal Trade Commission and the Food and Drug Administration; and the "National Health Information Infrastructure," which is an evolving concept emphasizing linkages among content and technologies to serve clinical care, public health, and consumer empowerment. These exciting concepts should do much to foster the delivery of health information to the public. Bayer assesses the laws and regulations governing traditional pharmacies, compares those to coverage of online pharmacies, and analyzes the

Public Policy Experiences and Expectations

353

advantages and disadvantages of both approaches. He essentially argues that current legal and regulatory approaches are adequate to the situation. Katz and Aspden argue that the technology, regulations, controls, and social contexts motivating traditional security and privacy practices, such as in the defense and financial industries, are fundamentally different from those motivating security and privacy practices and needs in the health care community. This leads both to inappropriate approaches as well as wider and more unaccountable consequences. In light of the inadequacy of current regulations, and the impending additional ones requiring universal health identifiers and all-electronic record keeping, they believe a fundamental shift in rights and responsibilities is appropriate.

16 ehealth Federal Issues and Approaches CYNTHIA BAUR MARY JO DEERING LESLIE HSU

T

he U.S. government has several broad mandates to guide its activities in the health arena. These mandates include providing leadership, ensuring public health and safety, creating and disseminating public information, and educating the public about health. The federal government has a particular interest in the health care sector because it is one of the major providers of health care services directly through the Department of Defense, the Department of Veterans' Affairs, and the Indian Health Service, and through grants to clinics. The federal government is also a major insurer through Medicare and, in partnerships with states, Medicaid, and the State Children's Health Insurance Program (SCHIP). As a result, the federal government has become increasingly involved in efforts to measure and improve health in the United States. The primary framework for these efforts is Healthy People, the national agenda for health improvement. Healthy People 2010—the third in a series of decade-long initiatives—identifies health priorities and sets specific goals and objectives to address the nation's health needs (U.S. Department of Health and Human Services [DHHS], 2000). 355

356

PUBLIC POLICY

Healthy People 2010 is a national effort that involves state and local health departments, health professional and voluntary organizations, educational and research institutions, schools and businesses, faith-based organizations, and community-based programs. It is led by the Department of Health and Human Services (DHHS) and coordinated by the Office of Disease Prevention and Health Promotion (ODPHP) within that department. Its development involved stakeholders from the public and private sectors, with input from individuals around the country. Healthy People 2010 has two overarching goals: to increase the quality and years of healthy life and to eliminate health disparities. It includes 28 focus areas, ranging from specific diseases, conditions, and risk factors such as arthritis, injury, and tobacco use, to cross-cutting issues such as access to quality care, health communication, medical product safety, and public health infrastructure. These focus areas encompass more than 460 separate objectives that support the two main goals. The goals and objectives are grounded in science, reflect public consensus, and measure progress and achievements. Healthy People 2010 is the disease prevention and health promotion agenda for the United States in the first decade of the new century. The breadth of the health challenges encompassed by Healthy People 2010 calls for an equally expansive set of tools to address them. The dynamic rise of new information and communication technologies offers innovative possibilities for health enhancement, ehealth technologies create numerous opportunities to improve health. Interactive technologies can help individuals find health information, track their health concerns over time, compare treatment and medication options, and receive regular reminders to use preventive services and adhere to treatment and health maintenance plans (Eng & Gustafson, 1999), all of which are recognized by public health leadership as the basis for good health and the accomplishment of Healthy People objectives. Other benefits from the use of ehealth applications include making information, services, and peer support available to people when and where they need them (Ferguson, 1996; Harris, 1995; U.S. General Accounting Office, 1996); supporting behavior change, such as quitting smoking or avoiding risky sexual practices (Strecher et al., 1994); and linking clinical, public health, and personal health information with appropriate safeguards to maintain confidentiality (National Committee on Vital and Health Statistics, 2000).

ehealth

357

The public sector is actively exploring ways to use interactive technologies to support disease prevention and health promotion goals and objectives; contribute to longer, healthier lives; and help eliminate health disparities. The following chapter discusses federal roles and responsibilities in the area of ehealth and presents a federal perspective on trends, opportunities, and potential problems.

DEFINITIONS There are several ways to conceptualize the use of the Internet and other information and communication technologies in the health sector. The most broad term is ehealth, which refers to the use of electronic technologies in health, health care, and public health. Table 16.1 describes the various functions of ehealth and provides examples of each function. Interactive health communication (IHC), the term used by the Science Panel on Interactive Communication and Health, emphasizes the "interaction of an individual with or through an electronic device or communication technology to access or transmit health information, or to receive or provide guidance and support on a health-related issue" (Eng & Gustafson, 1999, p. 8). Examples of IHC activities include searching the Internet for health information, engaging in online risk assessment and risk behavior change programs, utilizing medical decision-support tools, participating in online chat or support groups, sending electronic mail to health care providers, and managing personal health records online. Telemedicine is more narrow in scope than telehealth, referring primarily to clinical care applications. An Institute of Medicine report defines telemedicine as "the use of electronic information and communications technologies to provide and support health care when distance separates the participants" (Field, 1996, p. 1). Telemedicine functions can include patient consultations when the main health care provider is in one location and the patient is in another; consultations and mentoring between health care providers at a distance; medical procedures at a distance, for example, remote surgery; the use of peripheral equipment to collect patient information such as blood pressure and body temperature; the forwarding of medical images and other data; and distance continuing medical education. A fourth term is informatics, a field of study concerned with the broad range of issues in the man-

358

PUBLIC POLICY

ii^m^

h e a l t h Continuum Examples

Function Reference

Electronic publishing, catalogues, databases

Self-help/self-care

Online health information, support groups, health risk assessments, personal health records

Plan/provider convenience services

Online scheduling, test and lab results, benefit summaries

Consultation and referral

Doctor-patient or doctor-doctor consultations via telemedicine systems, remote readings of digital images and pathology samples

E-health-commerce

Sales of health-related products and services

Public health services

Automated data collection, data warehouses, online access to population survey data and registries, advance detection and warning systems for public health threats

agement and use of biomedical, consumer health, public health, and nursing information (Shortliffe & Perreault, 1990). Medical informatics specialists are concerned with, for example, the development of standardized vocabularies to facilitate electronic storage and retrieval of information, electronic medical record systems, and data mining and warehousing systems. This chapter uses the term ehealth to refer to the broadest possible range of interactive technologies applied to health and health care.

FEDERAL ROLES AND RESPONSIBILITIES Federal roles and responsibilities in ehealth are defined in relation to the broad mandates described above, as well as by more specific regulatory and legislative mandates. Some federal ehealth activities respond to the need to provide regulatory oversight in particular areas; others occur because they are part of an agency's or department's general mission. For example, ODPHP was created in 1976 as the Office of Health Information and Health Promotion and, as part of its mission, engages in the creation and dissemination of consumer health information, whether the communication medium used is the telephone, the postal service, or the Internet. The Office, however, does not have any regulatory authority over other en-

ehealth

359

tities that also create and disseminate health information and cannot sanction those who distribute poor-quality information. The Food and Drug Administration (FDA), which is a DHHS agency, and the Federal Trade Commission (FTC), on the other hand, are regulatory bodies that have specific responsibilities to investigate and sanction those who engage in illegal activities, such as deceptive or unfair acts or practices, making false or misleading claims about health products and services, and selling or dispensing a prescription drug without a valid prescription. Federal mandates in ehealth are to regulate health claims and the promotion of medical products; pursue fraud; create and disseminate reliable and valid health information; collaborate with the private sector in standards-setting efforts; promote quality and access for both health information and services; and engage in public and professional education. In addition to regulatory boundaries, other limits to federal action include First Amendment protections of speech, which include, for example, the protection of personal testimonials about the benefits or harms of particular health practices; the mostly private nature of the U.S. health care system in which clinicians work for themselves or a health care organization rather than publicly operated health facilities; the reliance on professional and private organizations, such as medical societies and specialty groups, and internal reviews at hospitals and managed care organizations to police malpractice; and the responsibility of state boards, not federal agencies, to license health professionals and pharmacies. Additional factors related to the Internet and described in more detail below are the commitments of parts of the Executive, Legislative, and Judicial branches to promote use of the Internet in, for example, commerce, education, and health; the development of electronic commerce (e-commerce), and industry self-regulation of ethical business practices (U.S. Government Working Group on Electronic Commerce [U.S. Government Working Group], 1999; see also Clinton, 1999). As a result, the federal government has not taken a strongly interventionist position in many parts of the ehealth sector, and instead fulfills an important function as a leader and convener of stakeholders. The Clinton administration has strongly supported the development of the Internet and the application of information technologies to address issues in different sectors of society. In December 1999, President Clinton issued a memorandum that indicated that the federal government should be more involved in promoting the use of the Internet, and he directed heads of executive departments and agencies to take actions to adopt poli-

360

PUBLIC POLICY

cies that will remove barriers to private sector investment in Internet applications; explore partnerships with companies, state, local, and tribal governments, and other entities, such as nonprofit organizations and universities; explore innovative mechanisms for fostering a national discussion on the potential of the electronic society; consider policies to promote the electronic society; and review the recommendations of the President's Information Technology Advisory Committee (Clinton, 1999). In the health arena, Clinton directed the Secretary of Health and Human Services to "identify additional steps that can be taken to promote expanded access to higher quality, cost effective health care to underserved rural communities and inner city clinics, and other health-care applications of information technology" (Clinton, 1999). President Clinton also made the closing of the "Digital Divide," the gap between those with ready access to the Internet and other emerging technologies and those without access, one of the primary issues of his second term. The issue of the Digital Divide is discussed in greater detail below. Well before these presidential directives, federal agencies had created and maintained programs and activities in several areas of ehealth (see Table 16.2). Federal agencies have established Web sites that provide information about their programs, services, and subjects within their areas of responsibility. There are electronic federal gateways for many types of consumer and professional information. ODPHP manages healthfinder.gov, the official federal gateway for consumer health information. The National Library of Medicine (NLM) provides free access to PubMed and MedlinePlus, the library's electronic catalogues of the peer-reviewed health and medical literatures. Federal agencies also provide online selfhelp and self-care tools. For example, as part of the National Cholesterol Education Program, the National Heart, Lung and Blood Institute (NHLBI) of the National Institutes of Health (NIH) provides online tools for users to test their knowledge of cholesterol, heart disease, and appropriate food portion sizes, and to create diets for healthful eating. The National Cancer Institute (NCI), also part of NIH, has an online Breast Cancer Risk Assessment Model Tool, and healthfinder® plans to include a personal health guide for users to track the preventive health services they receive. In the area of services, the Departments of Defense and Veterans' Affairs have incorporated telemedicine into their health services delivery systems. In addition, the NLM and the DHHS Office of Rural Health Policy and Office for the Advancement of Telehealth have provided funding for many of the active telemedicine programs in the United States.

ehealth

361

rült

Agency

or

Examples of Federal Agencies and Their Involvement in ehealth

Department

ehealth

Activities

Relevant

URL

Department of Defense

Telemedicine services; computer-based patient records; minimally invasive surgery; interactive medical manuals; distance learning; haptic feedback devices; medical modeling and simulation; virtual reality

www.matmo.org/ www.tatrc.org

Veterans Affairs

Telemedicine services; computer-based patient records

w w w. va. go v/te I e m ed/ default.htm

Department of Health and Human Services

www.hhs.gov

Agency for Healthcare Quality and Research

National Guideline Clearinghouse, information on quality of care

www.ahrq.gov

Office for the Advancement of Telehealth (Health Resources and Services Administration)

Funding of telemedicine projects

http://telehealth.hrsa.gov/

Office of Disease Prevention and Health Promotion

Health information provider; ehealth evaluation; policy analysis of online health activities

www.healthfinder.gov www.scipich.org www.health.gov/healthypeople

Food and Drug Administration

Online purchasing

www.fda.gov

Health Care Financing Administration

Health insurance information provider

www.medicare.gov

Centers for Disease Control and Prevention

Public health information provider

www.cdc.gov

Indian Health Service

Telemedicine services, computer-based patient record

www.ihs.gov

National Institutes of Health

Health information provider through its various institutes, such as the National Heart, Lung and Blood Institute and National Cancer Institute

www.nih.gov/

National Library of Medicine (National Institutes of Health)

Funding of telemedicine projects; Internet^; health information provider

www.nlm.nih.gov/

362

PUBLIC POLICY

POLICY ISSUES Although there is great enthusiasm about the potential benefits of ehealth, the rapid proliferation of technologies is also generating concern and raising policy issues, both old and new. The policy issues raised by the use of ehealth fall into six general areas: one, the lack of mechanisms to ensure the quality of health information that is being widely disseminated through electronic media to the public; two, the lack of evaluation research to determine the impact of ehealth on health care utilization, access, continuity, quality, and health status; three, the use of the Internet to engage in potentially fraudulent or deceptive electronic health commerce (e-health commerce); four, the online collection of personal health information and the potential abuses of privacy and confidentiality; five, the inefficiencies, errors, and missed opportunities to improve health and well-being that occur because of the lack of a fully integrated national health information infrastructure; and six, inequities in access to the Internet and appropriate health information. Increasingly, ehealth applications are being developed according to a commercial model in contrast to the previous nonprofit model of health information and education. Previously, public and medical libraries, government agencies, nonprofit groups, health care professionals and organizations, public health officials, and health educators were the primary sources of health information and education because of their training and commitment to public service. Increasingly, many individuals who do not have medical, public health, or social science training are repackaging health information, some of which is already publicly available, in proprietary formats, and disseminating it to the public in the form of commercial health information services, health assessment and management tools, and systems for managing personal health records. The organizing principle of e-commerce is not public service, but attracting "eyeballs" and creating "sticky" content that keeps individuals within a single Web site. Even nonprofit entities are creating for-profit arms to get access to the advertising monies and investment capital that are needed to develop and operate a commercially competitive Web site. Little is known about the impact of these applications on users' health and the health care system because this type of evaluation is rarely part of the business model (Eng & Gustafson, 1999). One group of health and technology experts warned that without greater attention to outcomes, ehealth "may result in inappropriate treatment, delays in care, damage to patient-provider relationships, violations

ehealth

363

of privacy and confidentiality, wasted resources, delayed innovation, unintended errors, and an increased technology and health gap" (Eng & Gustafson, 1999, p. 2). The federal government's response to these policy issues is framed by the constraints discussed above as well as by the constantly evolving nature of the technologies themselves and consumer demand for health information and ehealth services. Federal activities to address these policy issues and public concerns are discussed in turn below.

Quality of Information Numerous departments and agencies in the federal government are engaged in efforts to define and promote the quality of online health information resources. Recognizing the importance of the quality issue, Objective 11-4 in Healthy People 2010 is to, "Increase the proportion of healthrelated World Wide Web sites that disclose information that can be used to assess the quality of the site" (U.S. Department of Health and Human Services, 2000, pp. 11-16). The specific quality elements are drawn from the recommendations of the Science Panel on Interactive Communication and Health (see next section) and are framed as a matter of disclosure (Eng & Gustafson, 1999). The objective states that, at a minimum, Web sites should disclose (a) the identity of the developers and sponsors of the site (and how to contact them) and information about any potential conflicts of interest or biases; (b) the explicit purpose of the site, including any commercial purposes and advertising; (c) the original sources of the content on the site; (d) how the confidentiality of any personal information collected from users is protected; and (e) how the site is evaluated and (f) how the content is updated. To help consumers and health professionals find reliable health information on the Internet, ODPHP, in collaboration with other federal agencies, operates http://www.healthfinder.gov, a free health information and referral Web site developed by the U.S. Department of Health and Human Services. The site was launched in 1997 by Vice President Al Gore and DHHS Secretary Donna Shalala to make available to the public a reliable, easy to use, commercial-free online health information resource, healthfinder® is a gateway site with links to more than 5,600 online resources, carefully selected from more than 300 federal organizations and 1,400 nonfederal organizations, including major nonprofits, universities, state health departments, and private health organizations. It covers more

364

PUBLIC POLICY

than 1,000 topics and includes many resources in Spanish. In 1999, healthfinder averaged about 400,000 user sessions per month and approximately 4.5 million user sessions for the year. The site has received 40 national awards. Other federal agencies, such as the Food and Drug Administration (FDA) and the Federal Trade Commission (FTC), are engaging in public education activities to make consumers aware of the need to protect themselves when they browse and buy on the Internet (http://www.fda.gov; Federal Trade Commission, 1999a). The FTC is using its authority as a consumer protection agency to make consumers aware of possible fraud, privacy violations, and other consumer protection issues on the Internet. The FTC operates two Web sites, http://www.ftc.gov and http://www.consumer, gov, that have consumer information about online fraud, deception, and privacy. The consumer.gov site has a section specifically on health misinformation and scams. The FTC operates several "sting" Web pages that are designed to highlight for consumers the tell-tale signs of a scam. The FTC, in conjunction with the U.S. Department of Commerce, has also sponsored public workshops on issues related to online commerce, particularly the protection of users privacy. Other FTC activities are discussed in a following section. 7

Evaluation of ehealth's Impact DHHS also has initiatives to educate the public, health professionals, and the developers of ehealth applications about the need to assess the quality and impact of ehealth and online health information. In 1997, ODPHP convened the Science Panel on Interactive Communication and Health (SciPICH), a 14-member nonfederal panel to study and report on the field of interactive health communication (IHC). The message of Wired for Health and Well-Being, the Panel's final report, is that the Internet and interactive technologies hold promise to communicate health information and to influence attitudes and behavior, but they could cause harm without more rigorous evaluation to determine quality, safety, and effectiveness (Eng & Gustafson, 1999). In its review of the literature, the Science Panel found evidence that interactive tools can relay individualized health information on demand; facilitate individuals' decision making; improve provider-patient communication; promote the adoption and maintenance of positive health behaviors; promote exchange of information and emotional support among individuals with similar health concerns; help indi-

ehealth

365

viduals manage their own health problems; and promote appropriate utilization of health care services (Eng & Gustafson, 1999, pp. 13-14). The report contains a set of recommendations to improve the quality and effectiveness of IHC as well as evaluation tools that can be used to assess and report on IHC applications. These tools include an evaluation template that developers can use to assess the quality and impact of their application, and a sample disclosure label to provide information to the public about the application. The Science Panel's report identifies four strategies that will help to realize the promise of IHC: (a) strengthen evaluation and quality of IHC; (b) improve basic knowledge and understanding of IHC; (c) enhance capacity of stakeholders to develop and use IHC; and (d) improve access to IHC for all populations. The panel concludes that each of the stakeholder groups—developers, purchasers, health and information professionals, researchers, consumers, and policymakers—has a responsibility to promote and participate in the evaluation of IHC applications so that the overall quality of IHC improves. The Institute of Medicine (IOM) has issued a separate report on telemedicine systems that also calls for rigorous evaluation to determine the impact of telemedicine on a number of factors, including the clinical care process, patient status or health outcomes, cost of providing telemedicine services, and users' satisfaction (Field, 1996). The IOM recommends that telemedicine systems be evaluated using the same controlled trial methods that are applied to other biomedical activities, which means that evaluators should determine telemedicine's impact in relation to the best available alternatives. From this perspective, the goal of the evaluation should be to provide decision makers with enough information to determine whether telemedicine is at least as good as, if not better than, other possible ways of delivering health information and services. This approach, argues the IOM report, is the best way to promote informed decisions about the best uses of telemedicine technologies.

ehealth-Commerce Current U.S. government policy simultaneously encourages private sector leadership in the development of the Internet and reminds online entrepreneurs that agencies will use existing laws to fulfill traditional government functions to protect consumer health and safety (President's Working Group on Unlawful Conduct on the Internet, 2000; U.S. Government Working Group, 1999). The Clinton administration's approach to the

366

PUBLIC POLICY

rapid growth of the Internet and electronic commerce (e-commerce) in particular is summarized in five principles (U.S. Government Working Group, 1999): 1. The private sector should lead. 2. Government should avoid undue restrictions on electronic commerce. 3. Where governmental involvement is needed, its aim should be to support and enforce a predictable, minimalist, consistent, and simple legal environment for commerce. 4. Government should recognize the unique qualities of the Internet. 5. Electronic commerce over the Internet should be facilitated on a global basis.

The Internet industry has been very vocal in its opposition to new regulation or the expansion of regulatory authority of existing agencies. Within this context, government agencies, such as the FTC, are working with online businesses to develop a model of self-regulation for the industry that relies on private organizations, such as the Better Business Bureau Online and the National Association of Boards of Pharmacy, to develop online codes of conduct and seal-of-approval programs, such as the Verified Internet Pharmacy Practice Sites Program (VIPPS), and enforce compliance. Because of individuals' concerns about privacy and protecting the confidentiality of personal health information, e-health-commerce has been generating a lot of public debate about acceptable online business practices. The proliferation of Web sites that offer health-related information, goods, and services is testament to consumers' interest in these categories and in new tools to manage their own health. In particular, there has been rapid growth in the online drugstore category with the appearance of virtual drugstores and pharmacies that now compete with "brick-and-mortar" drugstores and pharmacies and pharmacy benefits managers. Online drugstores and pharmacies sell everything from cotton balls to prescription drugs, and consumer demand for health items seems insatiable. The anticipated benefits of Web sites that sell health products include greater availability and choice, speed and ease of choosing and ordering products, lower prices, the opportunity to consult on demand with health profes-

ehealth

367

sionals who are associated with the site, and improved tracking of health conditions and medications. However, for the full public health benefits to be realized, these online vendors would need to be integrated into systems that would permit monitoring and reporting of problems with drugs or other products and to provide immediate warnings of potential adverse interactions among medications dispensed by different sources to individual patients. The Medical Products Safety focus area of Healthy People 2010 includes two separate objectives, 17-1 and 17-2, about these issues (U.S. Department of Health and Human Services, 2000). At the same time, online drugstores and pharmacies represent an expanded marketplace for the illegal sales of unapproved drugs, prescription drugs without valid prescriptions, and products marketed with fraudulent health claims. Media investigations of online drug sales have spurred public attention. Investigative reports of Viagra, a prescription drug for sexual dysfunction in men, have been especially popular, and there have been several occasions when reporters used the names and information of pets, children, or deceased persons to order Viagra over the Internet (U.S. House of Representatives, 1999). The three federal agencies with the most clear regulatory authority in the area of online sales are the Department of Justice (DOJ), the Federal Trade Commission (FTC), and the Food and Drug Administration (FDA). All three have testified before Congress on the matter of illegal online drug sales and have been members of a highranking interagency Working Group on Unlawful Conduct on the Internet, chaired by the Attorney General. The current position of the Working Group is that if additional resources and training were available to pursue online illegal activities, existing authority and regulations would be sufficient to fulfill their responsibilities (President's Working Group on Unlawful Conduct on the Internet, 2000). Although federal agencies are currently following the policy that laws and regulations from the offline world are sufficient to regulate the online world, the Internet poses many challenges to existing legal and regulatory frameworks, especially when online transactions regularly cross state and national boundaries. States have traditionally been responsible for the licensing of health professionals and pharmacies to protect local citizens. Some states, for example Kansas, have filed lawsuits against companies, doctors, and other individuals for practicing medicine or pharmacy in the state without legal authority and violating the Consumer Protection Act. The Clinton administration recently proposed expanding FDA authority to certify pharmaceutical Web sites and to levy fines of up to $500,000 for

368

PUBLIC POLICY

selling drugs without valid prescriptions or operating without an FDA certification (Office of the White House Press Secretary, 1999). Since 1996, the FTC has used the Internet to investigate deceptive practices (Federal Trade Commission, 1997, 1999a; 1999c). It conducts "Surf Days" during which staff identify deceptive Internet practices and then search the Internet to find violations. Since 1997, the FTC has conducted two Health Claims Surf Days during which they identified approximately 800 Web sites and Usenet groups that contained fraudulent health claims. In June 1999, they launched "Operation Cure.All," a consumer health education campaign and law enforcement action against false or misleading health claims on the Internet, and announced the prosecution of four companies as a result of the 1998 Surf Day (Federal Trade Commission, 1999b). The development of sales of online health care services, such as online consultation with a health professional for a fee, has been much slower. The diffusion of telemedicine, or long distance consultation between doctor and patient or doctor to doctor, has been inhibited by the lack of national licensing and the reluctance of insurance companies to reimburse for telemedicine visits. The sanctity of the doctor-patient relationship is also a factor. Doctors who are online and dispensing advice in chat rooms or other ask-a-doc services frame their activities as patient education rather than doctor-patient consultations. One online service that is based on real-time chats with physicians explicitly states that the exchange of electronic mail between the user and the physician does not constitute the basis for a doctor-patient relationship and will not be accorded the same protections. Gradually, barriers are being breached. A health Web site already offers an auction service for elective, nonemergency surgery and will likely be followed by imitators (MedicineOnline.com, 2000; Rustad, 1999). The service will allow consumers to post their requests for a desired surgical procedure and give physicians registered with the service the chance to bid.

Online Privacy and the Protection of Personal Health Information The rapid development of the Internet and the ability of electronic media to record and store every keystroke have motivated widespread debate about individuals' privacy rights, including the right to be anonymous on the Internet and the right to have personal information safeguarded on-

ehealth

369

and offline. A survey of Internet users found that 66% of those who seek health information online are concerned or very concerned about their personal privacy and 49% of online health information seekers are even more concerned about their privacy when they are on the Internet than off (California HealthCare Foundation & Internet Healthcare Coalition, 2000, p. 2). The protection of personal health information in an electronic environment is a matter of particular concern. The basis of concern is the ability of electronic media to collect—often unbeknownst to the user—and store large amounts of detailed, personal information that can be processed, altered, accessed, copied, and distributed to an unlimited number of people or network nodes. Personal information includes not only data, such as name and address, explicitly provided by individuals while they are online, but also information generated as a result of being online, such as an email address, an online alias, or an IP address, that may be connected to other personal information. A recent study of 21 health-related Web sites found that "visitors to health Web sites are not anonymous, even if they think they are. Through mechanisms such as cookies, profiling, banner ads, and clickstreams, sites are collecting information about individuals, often without their knowledge or consent" (Goldman, Hudson, & Smith, 2000, p. 4). Cookies, for example, are pieces of code that are transferred from Web sites to users' computers so that the site can track a user's online movements, such as which advertisements are read or how long the user stays on a particular page. Users do not have to give consent for cookies to be employed by a Web site. Many commercial Web site operators claim cookies are a customer benefit that allow the site to recognize return visitors and to provide personalized service based on information the system collected during a user's previous visits. From the businesses' perspective, cookies provide information that can be exhaustively mined to tailor product pitches and other advertising and create ever more specific user profiles. One of the premises of the e-commerce model is that online businesses have an advantage over their brick-and-mortar counterparts because the former can use Web technologies to compile highly detailed profiles of individuals' interests and habits, both viewing and purchasing, and can target more precisely potential customers. In surveys, Internet users report that they are willing to provide some personal information in exchange for a very personalized experience and protections against unauthorized sharing of data with others (California HealthCare Foundation & Internet Healthcare Coalition, 2000, p. 3; Mabley, 1999).

370

PUBLIC POLICY

Online information collection practices raise new concerns about privacy and confidentiality beyond those surrounding traditional medical records. Personal health information obtained in delivering care has enjoyed special status as sensitive and to be disclosed only on a need-to-know basis, even if its legal protection has not been fully comprehensive. This special status has been rapidly eroding even in the case of paper-based records, however, because of the release of information to multiple parties to obtain reimbursement. The potential harm that can occur from the disclosure of personal health information includes embarrassment, social stigma, and discrimination (Hodge, Gostin, & Jacobson, 1999). Searching the Internet for health information; using interactive health communication tools, such as online health risk assessments; and exchanging electronic mail with physicians in chat rooms add a new dimension to privacy and confidentiality concerns. Individuals are now generating a lot of personal health information outside clinical consultations and health care organizations. Protection of this type of personal health information represents uncharted domain for public policy. As of Fall, 2000, there was no comprehensive, national legislation in the United States that protected the confidentiality of health records in all settings. Existing federal and state laws provide selective privacy protections (Hodge et al., 1999; Pritts, Goldman, Hudson, Berenson, & Hadley, 1999). The Privacy Act of 1974 covers health information held by federal agencies but only in the same way it protects all other personal information, without reference to the special character of health information. The debate in the United States about the protection of electronically stored and transmitted personal information is being propelled not only by domestic considerations to reassure consumers about the trustworthiness of online transactions, but also by international pressure from countries with more stringent privacy protections. The European Union (EU) and the United States have been in negotiations to determine the basis for U.S.-based companies' compliance with the EU Directive on Data Protection (European Parliament, 1995). The Directive requires that member states not transfer data (including health data) to countries outside the EU unless the recipient country "ensures an adequate level of protection," which, from the EU perspective, appears to be lacking in the United States. The U.S. Department of Commerce has led the negotiations with the EU and concluded a "safe harbor" privacy agreement in July, 2000 (U.S. Department of Commerce, 2000). In the health sector, groups in the United States have proposed "Best Principles for Health Privacy" (Health Privacy Working

ehealth

371

Group, 1999); Health Internet Ethics: Ethical Principles for Offering Internet Health Services to Consumers (Hi-Ethics, 2000); an e-Health Code of Ethics (Health Ethics Initiative, 2000) and legislative frameworks (Hodge et al., 1999) to fill the gaps in existing privacy laws and regulations. In the fall of 1997, DHHS Secretary Donna Shalala recommended to Congress that it enact national legislation to provide fundamental privacy rights for patients and to define the responsibilities of those who provide and pay for health care (U.S. Department of Health and Human Services, 1997). She offered five principles on which to base legislation to protect individually identifiable health care information. These principles are (a) boundaries: health care information should be disclosed for health purposes only, with limited exceptions; (b) security: health information should not be distributed without patient authorization absent a clear basis and those who receive such information must safeguard it; (c) consumer control: persons are entitled to access and amend their health records and to be informed of the purposes for which the health records are used or disclosed; (d) accountability: those who improperly handle health information should be criminally punished and subject to civil recourse; and (e) public responsibility: individual privacy interests must not override national priorities of public health, medical research, preventing health care fraud, and law enforcement in general. Many medical records confidentiality bills have been and are being debated in Congress, although to date all have failed to progress. Moreover, Congress missed the August 1999 deadline it set for itself to pass health privacy legislation under the Administrative Simplification provisions of the Health Insurance Portability and Accountability Act (HIPAA) of 1996. The provisions are intended to reduce the costs and administrative burdens of health care by making possible the standardized, electronic transmission of many administrative and financial transactions that are currently carried out on paper. Once the legislative deadline passed, HIPAA required the Secretary of Health and Human Services to issue regulations by February 2000. DHHS issued draft privacy regulations in the fall of 1999. The public comment period ended February 17,2000, and DHHS received more than 50,000 public comments. Final regulations are scheduled for Fall, 2000. Even as the regulatory process progresses, Congress can act at any time and pass legislation that would supersede the regulations. The proposed regulation would apply to health care providers who engage in electronic health transactions, health plans, and health care clearinghouses (U.S. Department of Health and Human Services, 1999). They

372

PUBLIC POLICY

would be obliged to permit individuals to see their records and offer corrections, and, with some exceptions, to permit individuals to see who has seen their records. They could not disclose information unless the patient authorized it, the disclosure is required by other law, or the disclosure is specifically permitted in the regulation. Information could be disclosed without consent for treatment, payment, or other operation of the health care system, but individuals could request that information not be disclosed in specific instances. Information could be disclosed for certain other specified purposes, such as public health, oversight of the health care system, and research, often after stated conditions. Covered entities would have to give patients an explanation of their information practices and would have to have internal procedures to ensure that information is handled in accordance with the regulation. There would be criminal penalties for improper disclosure of information, and civil money penalties. In the area of e-commerce, which includes e-health-commerce, the FTC, often in partnership with the Department of Commerce, has been in the lead in the investigation of online consumer privacy issues. Since 1994 the Commission has held a series of public workshops, conducted surveys of online privacy practices, issued reports, and testified before Congress (Federal Trade Commission, 1999a). FTC involvement in e-commerce is grounded in its mandate to protect consumers from deceptive and fraudulent practices and in the promotion of consumer confidence in an emerging marketplace. As of mid-2000, the Commission encouraged self-regulation, which includes online privacy seal programs, in which businesses adhere to a code of conduct in exchange for the right to display a seal of approval from an organization such as the Better Business Bureau Online or TRUSTe; legislation to mandate adoption of fair information practices; and emerging technologies to allow consumers to protect their privacy online (Federal Trade Commission, 1999c; 2000b). Fair information practice has four elements: notice, consent, access and correction, and security. Individuals must be notified of the uses of their personal information, give their consent to its use, have access to their information and be able to request corrections, and be assured that their information will be secure. The FTC is conducting research on fair information practices on the Internet. It has completed a current survey of privacy notices on Web sites and issued a report to Congress (Federal Trade Commission, 2000b). In addition to selfregulation and legislation, the FTC created an Advisory Committee on Online Access and Security to advise the Commission on implementation of access and security measures (Federal Trade Commission, 2000a). The

ehealth

373

committee has issued its final report outlining the FTCs options and the full range of actions that the Commission should consider (Federal Trade Commission Advisory Committee on Online Access and Security, 2000).

NATIONAL HEALTH INFORMATION INFRASTRUCTURE (NHII) The National Committee on Vital and Health Statistics (NCVHS) is the public advisory body to the Secretary of Health and Human Services and Congress in the areas of health data, statistics, and national health information policy. Part of the Committee's charge is to encourage the evolution of a shared, public/private national health information infrastructure that will promote the availability of valid, credible, timely, and comparable health information (National Health Information Infrastructure Work Group, 2000). The National Health Information Infrastructure (NHII) Work Group defines the NHII as the sum total of the information technologies, standards, applications, systems, values, and laws that support all facets of individual health and health care. The Work Group has been studying the NHII from three separate but overlapping perspectives, called dimensions by the Work Group. The three are the Personal, Health Care Provider, and Community Health dimensions. The Work Group is using the dimension concept as a means to identify and address the current and anticipated health information infrastructure needs of consumers, patients, health professionals, and other stakeholders in the national health system, including the people and organizations who will be performing the work of attaining the Healthy People 2010 goals. The Personal Health Dimension (PHD) addresses the elements of the NHII designed for the improved management of personal health and wellness information and more informed health care decision making. The PHD will consist of health and wellness information supplied by individuals for personal use and for use by their health care providers, and summaries of health care system interactions (e.g., hospital discharge summaries) supplied by providers, to create a longitudinal and cross-institutional record of health care and outcomes. The Personal Health interface will ensure the privacy and confidentiality of personal health information while empowering consumers to manage their health better. Individuals will be able to designate levels of access to different parts of the PHD depending upon the sensitivity of information. Potential PHD technologies,

374

PUBLIC POLICY

including smart cards and the Internet, will allow for convenient access to personal health information in the various settings in which the information is utilized, such as at home, work, or school; while traveling; and during medical emergencies or when changing health care providers. The PHD will also include links to high-quality and reliable information to help individuals and clinicians make better health and wellness decisions. The Health Care Provider Dimension (CHD) is the electronic record that is the interface between the health care system and the NHII. The HCPD will be the electronic record of clinical care, similar to, and evolved from, computerized patient records now being developed. The HCPD will allow convenient, real-time, and 'round-the-clock accessibility to accurate and stable clinical information for clinicians, allied health care professionals, payers, and other authorized parties, and will incorporate clinical decision support technologies for improved quality of care. While the Work Group recognizes that different technologies and applications will support the HCPD at different institutions, it believes that data and format standards currently under development will allow for the electronic integration of health records throughout the health care system and with other components of the NHII, including the Personal Health Dimension and the public health information infrastructure. As a result, patient outcomes will improve because of improved flow of accurate clinical information for health care decision making, reducing the likelihood of medical errors and reducing health care costs, while maintaining the strictest safeguards to ensure the confidentiality of personal health information. The Community Health Dimension is the component of the NHII that will support communication and information to improve public health. This dimension will provide public health professionals timely access to data for the identification of immediate health threats to the public and for ongoing assessment of public health. The information will be multifaceted and permit the identification of longitudinal changes in the population's health, intergroup disparities, emerging health problems, and health emergencies. Some of the benefits will be facilitating the design of programs and polices for improvements in population health, designing interventions to respond to emerging health problems and health emergencies, targeting and evaluating health intervention programs and policies, and providing data to researchers for basic and applied studies. Much of the data in this dimension are currently collected under existing federal and state public health laws and regulations. Confidential public health

ehealth

375

data will remain so in this record through appropriate and state-of-the-art confidentiality rules and security systems.

Equal Access to Emerging Technologies for Health As the Internet becomes a more important tool for health improvement, providing access to the Internet, especially in people's homes, will be critical to the achievement of the Healthy People goals and objectives. Objective 11-1 in Healthy People 2010 explicitly calls for an increase in the proportion of households with access to the Internet at home because "access to the Internet and subsequent technologies is likely to become essential to gain access to health information, contact health care organizations and health professionals, receive services at a distance, and participate in efforts to improve local and national health" (U.S. Department of Health and Human Services, 2000, pp. 11-13). The issue of equal access typically refers to the need to ensure that individuals have information technologies at hand so that they can be used when needed. The federal government has numerous initiatives, described below, that are designed to extend technology access to those who have not been able to afford to purchase access in the marketplace. Equal access has other facets as well, such as equal access to the technical training and support that allow consumers and health professionals to use technologies effectively, and equal access to health information materials of appropriate literacy levels. Health literacy is the "capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions" (U.S. Department of Health and Human Services, 2000). Persons who find it difficult to read and understand complex health information materials are often doubly constrained when these materials are put online and require technical skills to gain access. Both the Medical Products Safety and the Health Communication focus areas of Healthy People 2010 highlight health literacy as a key barrier to improving the health of many individuals and population groups (U.S. Department of Health and Human Services, 2000), and the Health Communication chapter includes a specific objective to improve health literacy (U.S. Department of Health and Human Services, 2000). The gap between those with easy access to the electronic information infrastructure and those without is commonly referred to as the Digital Divide (U.S. Department of Commerce, 1999). The Clinton administration

376

PUBLIC POLICY

has placed great emphasis on closing the divide. In December 1999, the Department of Commerce hosted a national Digital Divide Summit that focused on expanding access for underserved populations to information technologies (U.S. Department of Commerce, 1999). Summit participants from the federal government, technology industries, civil rights and nonprofit communities, grassroots community organizations, and the general public examined public and private initiatives to close the technology gap and discussed how to expand and coordinate these efforts. In February 2000, the Secretary of Commerce began a multi-city Digital Divide tour to draw attention to the problem and to new solutions to closing the technology gap. The Department of Commerce's National Telecommunications and Information Administration (NTIA) issues an annual report on the Digital Divide that highlights disparities in the penetration of telephone, computer, and Internet technologies (U.S. Department of Commerce, National Telecommunications and Information Administration, 1999). The analysis is based on Census Bureau data, and the latest report indicates that the Digital Divide is widening. The "have" groups—whites, Asian and Pacific Islanders, those with higher income and higher education—are far more connected than the "have nots"—younger people, people with lower income or education, certain racial and ethnic groups, and people living in rural areas and central cities. The gap between white and black households grew 39% between 1994 and 1998 for ownership of personal computers (PCs) and 38% between 1997 and 1998 for Internet access. The gap between white and Hispanic households grew 43% for ownership of PCs and 38% for Internet access. As of November 1999, African Americans and Hispanics use the Internet less than whites and Asian and Pacific Islanders; only 19% of African Americans and 16% of Hispanics use the Internet compared with 38% of whites and 36% of Asians and Pacific Islanders (U.S. Department of Commerce, National Telecommunications and Information Administration, 1999). Numerous federal programs exist to increase access to the information infrastructure. The Telecommunications Act of 1996 directed the Federal Communications Commission (FCC) to establish a universal service fund for rural health care providers (http://www.fcc.gov/ccb/apd/ruralhealth/). The fund provides discounts on telecommunications services for rural, not-for-profit health care facilities using telemedicine. The program is in its third year of operation. Numerous administrative problems have hampered the program, however, which has yet to provide much rate relief. The

ehealth

377

Department of Commerce operates the Technology Opportunities Program (TOP), formerly known as the Telecommunications and Information Infrastructure Assistance Program (TIIAP) (http://www.ntia.doc.gov/ otiahome/tiiap/index.html). TOP gives grants for model projects demonstrating innovative uses of network technology. Since 1994, TIIAP has made matching grants to state, local, and tribal governments; health care providers; schools; libraries; police departments; and community-based nonprofit organizations, that total $135.8 million. The Department of Housing and Urban Development (HUD) addresses the issue of the Digital Divide with a program called Neighborhood Networks (http:// www.hud.gov:80/nnw/nnwindex.html). Launched by HUD in September 1995, Neighborhood Networks is a HUD community-based initiative that encourages the development of resource and computer learning centers in privately owned HUD-assisted and/or -insured housing. These centers serve lower-income families and seniors. As of August 2000, more than 600 Neighborhood Networks centers were open in cities and rural towns throughout the United States. The Department of Education funds Community Technology Centers (http://www.ed.gov/offices/OVAE/CTC/). The purpose of the Community Technology Centers program is to promote the development of model programs that demonstrate the educational effectiveness of technology in urban and rural areas and economically distressed communities. These Community Technology Centers provide access to information technology and related learning services to children and adults. The program began in 1999 with $10 million in grant money for eligible agencies and organizations.

TRENDS AND IMPLICATIONS FOR THE NEAR FUTURE Many of the policy issues described in the previous sections will continue to be debated in the near term. The proliferation of commercial online health information Web sites will motivate continued attention on issues of information quality and the protection of privacy. Current public policy in the area of privacy protections is a mix of legislative proposals, promotion of technology tools, private-sector-generated ethical codes of conduct, and, in the health sector, impending HIPAA privacy regulations. As noted elsewhere in the chapter, the FTC and the Secretary of HHS are on record recommending national privacy legislation. Without such legisla-

378

PUBLIC POLICY

tion, HIPAA will provide the clearest guidance on protections for electronically exchanged health information, although HIPAA's applicability in many Internet-based uses of personal health information has not be clarified and tested. Also, in the absence of comprehensive national privacy legislation, many policymakers appear willing to give the private sector the opportunity to demonstrate the utility of disclosure of Web site practices, seal-ofapproval programs, and technology tools that users select for themselves based on their personal privacy requirements (U.S. Senate Judiciary Committee, 2000). Policymakers are likely to require concrete evidence of the ability of the Web sites and seal programs to follow their own guidelines and to enforce commitments to consumers. Two recent two initiatives— the e-Health Ethics Initiative and Hi-Ethics—have developed ethical guidelines for operators of health Web sites (Health Ethics Initiative, 2000; Hi-Ethics, 2000). These are in addition to the HON Code of Conduct (http://www.hon.ch/HONcode/), which already operates on the Web, and the American Medical Association's (ΑΜΑ) ethical guidelines for its own sites (http://www.ama-assn.org/about/guidelines.htm). They represent the most current and widely accepted guidance on quality standards or developers of health Web sites. It is likely that these codes and corresponding seals of approval will be implemented in either late 2000 or early 2001. Multiple seals of approval raise questions about redundancy, gaps, and possible competition among them to become recognized as "the standard" for health Web sites (Baur & Deering, 2000). The utility of disclosure and seals will be tested when more sites post clear policies and adhere to seal programs. Consumers will need to decide if Web sites can be trusted to do what they say, including provide quality information and safeguard personal information, and seal programs can be trusted to verify Web sites' claims. Shedding the light of day on the practices of Web sites may indeed raise the quality of those practices, but disclosure may also serve to remind online users that commercial sites still set their own rules about such matters as the corporate use of information. There are important limits to disclosure and seals of approval that should be acknowledged (Baur & Deering, 2000). The disclosure approach carries a risk of information overload and consumer confusion. The codes of ethics for health Web sites contain dozens of items that an informed user might need to know to evaluate a site's operations. Seals of approval may provide consumers a false sense of reassurance if the seals are not sup-

ehealth

379

ported by rigorous verification of a site's adherence to a code. Even though verification of some Web site practices maybe difficult, it is critical that the private sector develops clear and verifiable criteria, realistic mechanisms to enforce adherence to standards, and specific sanctions for violators. Noncommercial health information providers, such as government agencies, nonprofit organizations, and professional associations, are likely to position themselves as credible alternatives to commercial sites and to emphasize their roles as providers of unique, reliable information. The partnering of for- and nonprofits, such as Aetna-US Healthcare and Johns Hopkins University in Intelihealth, may be an attractive proposition for those nonprofits with name recognition and authority and who want to leverage that onto the Web. Questions about the nature of these partnerships and the influence of commercial interests that have been raised, for example, in the New York Times's coverage of the drkoop.com site, may percolate in policy debates (Noble, 1999), but will likely be overshadowed in the short term by concerns about abuses of personal information online. Ecommerce has become one of the defining features of the Internet. Entrepreneurs will continue to seek out new categories of online sales and services, and the areas of online fee-for-service consults with physicians, confidential management of personal health information, personalized health risk assessments, and chronic disease management may be among the growth sectors. Concerns about the ability of states or federal entities to monitor the online practice of medicine are likely to increase when more physicians go online, not only to connect with their existing patients but also to recruit new patients. The potential risks from the practice of online medicine may become more real if off-shore sites set up shop and attract customers with promises of low prices, convenience, and "giving the customer what she wants," regardless of the efficacy of the treatment or medication. Currently, evaluation of Internet health applications is sparse. What little evaluation there is occurs in the research environments of universities and academic medical centers. Healthy People 2010 contains an objective to increase research and evaluation of health communication activities that would include Internet health activities (U.S. Department of Health and Human Services, 2000), as well as an objective to establish Centers for Excellence that will include technology training and evaluation of technology use (U.S. Department of Health and Human Services, 2000). The National Cancer Institute has budgeted for these centers in 2001 (National

380

PUBLIC POLICY

Cancer Institute, 1999). Both the Science Panel on Interactive Communication and Health and the Institute of Medicine have issued substantive reports with comprehensive frameworks to guide the evaluation of interactive health communication applications and telemedicine systems, respectively (Eng and Gustafson, 1999; Field, 1996). The challenge will be to find ways to get the private sector online health industry to adopt these recommendations. Last, as more and more activities migrate to the Internet, the need to evaluate the impact of these activities on individuals' health behaviors and outcomes will receive a lot of attention. Policymakers, consumers, health care purchasers, and providers will all want to know more about how online information and services change individuals' sense of risks and rewards and expectations for health and well-being. The current trend in online health is for Web sites to emphasize that individuals should use the Web at their own risk and consult with their own physicians about any health information from the Internet before they make major health decisions. This implies that, as a result of information they find on the Internet, individuals may actually end up visiting their physicians more often, which is likely to increase, not decrease, utilization and costs. In addition, if clinical services become more freely available over the Internet, then overall utilization may also increase if individuals who are dissatisfied with the response from their own physician use the Internet to seek out clinicians anywhere in the world who may be more willing to comply with requests. If so, then we may see a very interesting set of benefits evolve. Individuals will certainly be better informed but perhaps not better off if online health tools fail to assist them in applying information to maintain health, improve decision making, and seek care when it can do the most good for their health. As a society, we may benefit from having consumers know more about their health; at the same time, we may be worse off if individuals use the information primarily to demand a greater amount of increasingly expensive services that do not provide any better health outcomes than less intrusive and costly alternatives. If preventive and self-care activities and evidence-based practice increase as a result of the Internet, then the net gain will be positive for individuals and society.

REFERENCES Baur, C , & Deering, M. J. (2000, Sept. 26). Proposed frameworks to improve the quality of health Web sites: Review. Retrieved October 4, 2000 from the World

ehealth

381

Wide Web: http://vv7vw.medscope.com/Medscope/GeneralMedicine/JouiTial/ public/mgm.journal.html California HealthCare Foundation & Internet Healthcare Coalition. (2000). Ethics survey of consumer attitudes about health Web Sites. Oakland: California HealthCare Foundation. Clinton, W. J. (1999, December 17). Presidential memorandum for the heads of Executive Departments regarding the use of information technology to improve our society. Washington, DC: Office of the White House Press Secretary. drkoop.com. (1999, November 4). Leading e-healthcare companies form alliance to benefit Internet consumers [Press release]. Retrieved November 30,1999 from the World Wide Web: http://www.drkoop.com/aboutus/media/press_releases/ 1999/110499_we gotethics.html Eng, Τ. R., & Gustafson, D. (Eds.), for the Science Panel on Interactive Communication and Health. (1999). Wired for health and well-being: The emergence of interactive health communication. Washington, DC: Government Printing Office. European Parliament. (1995). Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data. Retrieved February 24, 2000 from the World Wide Web: http://www2.echo.lu/legal/en/dataprot/ directiv/directiv.html Federal Trade Commission. (1997). North Amencan Health Claim Surf Day targets Internet ads, hundreds of e-mail messages sent. Retrieved February 18, 2000 from the Worldwide Web: http://www.ftc.gov/opa/1997/9711/hlthsurf.htm Federal Trade Commission. (1999a). Five years: Protecting consumers online. Washington, DC. [Online]. Available: http://www.ftc.gov/opa/1999/9912/ fiveyearreport.htm Federal Trade Commission. (1999b). "Operation Cure.all" targets Internet health fraud. Retrieved February 18, 2000 from the World Wide Web: http:// www.ftc.gov/opa/1999/9906/opcureall.htm Federal Trade Commission. (1999c). Self-regulation and privacy online: A report to Congress. Washington, DC. [Online]. Available: http://www.ftc.gov/os/1999/ 9907/index.htm#13 Federal Trade Commission. (2000a). Advisory Committee on Online Access and Security [Online]. Available: http://www.ftc.gov/acoas/index.htm Federal Trade Commission. (2000b). Privacy online: Fair information practices in the electronic marketplace, a report to Congress. Washington, DC: Author. Federal Trade Commission Advisory Committee on Online Access and Security. (2000, May). Final report. Retrieved May 17, 2000 from the World Wide Web: http://www.ftc.gov/acoas.finalreport.htm Ferguson T. (1996). Health online. Reading, MA: Addison-Wesley. Field, M. J. (Ed.), for the Committee on Evaluating Clinical Applications of Telemedicine. (1996). Telemedicine: A guide to assessing telecommunications in health care. Washington, DC: National Academy Press. Goldman, J., Hudson, Z., & Smith, R. M. (2000). Privacy: Report on the privacy policies and practices of health Web sites. Oakland, CA: California HealthCare Foun-

382

PUBLIC

POLICY

dation. [Online]. Available: http://ehealth.chcf.org/priv_pol3/index_show. cfm? doc_id=33 Harris, L. (Ed.). (1995). Health and the new media. Mahwah, NJ: Lawrence Erlbaum. Health Ethics Initiative. (2000). eHealth code of ethics. Retrieved May 24,2000 from the World Wide Web: http://www.ihealthcoalition.org/ethics/ethics.html Health Privacy Working Group. (1999). Best principles for health privacy. Washington, DC: Health Privacy Project, Institute for Health Care Research and Policy, Georgetown University. [Online]. Available: http://www.healthprivacy.org/ resources/index, shtml Hi-Ethics. (2000). Ethical principles for offering Internet health services to consumers. Retrieved May 9, 2000 from the World Wide Web: http://www.hiethics.org/ Principles/index, asp Hodge, J. G., Jr., Gostin, L. O., & Jacobson, P. D. (1999). Legal issues concerning electronic health information. Journal of the American Medical Association, 282, 1466-1471. Internet Healthcare Coalition. (2000). e-Health ethics initiative draft code. Washington, DC: ehealthethics summit. Retrieved February 22, 2000 from the World Wide Web: http://www.ihealthcoalition.org/community/draft.html Mabley, K. (1999). Privacy vs. personalization. Cyber Dialogue [Online]. Available: http://www.cyberdialogue.com MedicineOnline.com. (2000). Bid for surgery—Coming end of February 2000. Retrieved February 18, 2000 from the World Wide Web: http://www. medicineonline.com/bidforsurgery/ National Committee on Vital and Health Statistics. (1998). Assuring a health dimension for the National Information Infrastructure. Retrieved February 22,2000 from the World Wide Web: http://www.ncvhs.hhs.gov/hii-nii.htm National Cancer Institute. (1999). The nation's investment in cancer research. Washington, DC: National Institutes of Health. [Online]. Available: http:// 2001.cancer.gov/ National Health Information Infrastructure Work Group. (2000, June). Toward a National Health Information Infrastructure. Interim report. Washington, DC. Retrieved Sept. 21, 2000, from the World Wide Web: http://www.ncvns.lins.gov/ NHII2KReport.htm Noble, Η. B. (1999, September 5). Hailed as a surgeon general, Koop is faulted on Web ethics. New York Times, pp. AI, A20. Office of the White House Press Secretary. (1999, December 28). The Clinton administration unveils new initiative to protect consumers buying prescription drugproducts over the Internet. Retrieved December 29,1999 from the World Wide Web: http:// www.fda.gov/oc/buyonline/onlinesalespr.html President's Working Group on Unlawful Conduct on the Internet. (2000). The electronicfrontier: The challenge of unlawful conduct involving the use of the Internet. Retrieved June 28,2000 from the World Wide Web: http.7/www.usdoj .gov/criminal/ cybercrime/unlawful, htm Pritts, J., Goldman, J., Hudson, Z., Berenson, Α., & Hadley, E. (1999). The state of health privacy: An uneven terrain. Washington, DC: Health Privacy Project, Institute for Health Care Research and Policy, Georgetown University. Retrieved

ehealth

383

August 24, 1999 from the World Wide Web: http://www.healthprivacy.org/ resources/Statereports/contents.html Rustad, M. (1999). Bid-for-surgery Weh site seeks to he "Medical Ε-Bay." Retrieved November 30,1999 from the World Wide Web: http://www.drkoop.com/news/ stories/November/surgical_auctions.html Shortliffe, Ε. H., & Perreault, L. E. (Eds.). (1990). Medical informatics. Reading, MA: Addison-Wesley. Strecher, V. J., Kreuter, Μ., Den Boer, D. ]., Kobrin, S., Hospers, H. J., & Skinner, C. S. (1994). The effects of computer-tailored smoking cessation sensations messages in family practice settings. Journal of Family Practice, 39, 262-270. U.S. Department of Commerce. (1999). Digital Divide summit [Online]. Available: http: //www.ntia. doc. gov/ntiahome/digitaldi vide/summit/ U. S. Department of Commerce. (2000). International Safe Harbor Privacy Agreement. Retrieved September 27, 2000 from the World Wide Web: http://www.ita.doc. gov/td/ecom/shprin.html U.S. Department of Commerce, National Telecommunications and Information Administration. (1999). Falling through the Net: Defining the Digital Divide. (Nov. ed.). Washington, DC. Retrieved February 24, 2000 from the World Wide Web: http: / /www. ntia. doc. gov/ntiahome/fttn99/ c on tents. html U.S. Department of Health and Human Services. (1997, September 11). Confidentiality of individually-identifiable health information. Recommendations of the Secretary of Health and Human Services, pursuant to section 264 of the Health Insurance Portability and Accountability Act of 1996. Washington, DC. Retrieved February 25, 2000 from the World Wide Web: http://aspe.hhs.gov/admnsimp/ pvcrecO. htm U.S. Department of Health and Human Services. (1999, October 29). HHS fact sheet: Protecting the privacy of patients' medical records. Retrieved May 11, 2000 from the Worldwide Web: http://www.hhs.gov/news/press/1999pres/991029a. html U.S. Department of Health and Human Services. (2000, November). Healthy people 2010 (2nd ed., in two volumes). Washington, DC: Government Printing Office. www.health.gov/healthypeople U.S. General Accounting Office. (1996). Consumer health informatics: Emerging issues (GAO Publication No. GAO/AIMD-96-86). Washington, DC: Author. U.S. Government Working Group on Electronic Commerce. (1999). Towards digital eQuality. Second annual report. Retrieved December 27, 1999 from the World Wide Web: http://www.ecommerce.gov/annrpt.htm U.S. House of Representatives. Committee on Commerce. Subcommittee on Oversight and Investigations. (1999, July 30). Drugstores on the Net: The benefits and risks of on-line pharmacies. Retrieved August 24,1999 from the World Wide Web: http://com-notes.house.gov/cchear/hearingsl06.nsf/HearingExpand? OpenView&StartKey=E19D9E0A8CCDE825852567AF005DF9DE U.S. Senate Judiciary Committee (2000, Sept. 20) Pnvacy in the Digital Age: A resource for Internet users. Retrieved October 4, 2000 from the World Wide Web: http: //judiciary, senate. gov/privacy. htm

17 Old Wine in Silicon Prescription Bottles Some Legal Issues, Benefits, and Disadvantages Associated With Internet Pharmacies BARRY D. BAYER

N

ew technologies always seem to present new problems, but often the new problems are really older problems recast into the language and social context of the new technology. The practice of the professions of medicine and pharmacy in the United States has traditionally been matters of state licensing and control. A doctor licensed in New York cannot, without following the licensing rules of California, open an office to practice medicine in that state. A pharmacist licensed in Illinois cannot, without further registration, open a store to sell prescription drugs in Florida. The development of the Internet and its current user-friendly overlay, the World Wide Web, as well as the almost universal availability of electronic mail, have made it easy for physicians and pharmacists to contact, communicate with, and service patients and customers throughout the nation, indeed throughout the world. In the past year alone, there were at 385

386

PUBLIC POLICY

least 47 Web sites that would arrange for online consultation by physicians who write prescriptions to be filled by online pharmacies on the same Web site, selling prescription drugs to customers in any part of the country, with seemingly little control by state licensing agencies (Bloom & Iannocone, 1999). Sometimes it is not even clear where the online pharmacy is located and which state has jurisdiction over the company. Is it the state where the Internet server is located? The state from which the drugs are shipped? The state from which the pharmacist accesses the server, obtains the prescription, and fills the order? Or the state to which the drugs are ultimately shipped? While each of these questions is interesting, they are not particularly new. Although the Internet has facilitated and speeded the easy contact between pharmacist and customer, there is little difference between an out-of-state pharmacist filling prescriptions ordered by a customer by mail or over a toll-free telephone line. Over the years, state licensing agencies have cooperated in developing national standards for pharmacists and pharmacy licensing; the new freedom of communications possible over the Internet merely makes such cooperation more important.

PRE-INTERNET LICENSING AND PROCEDURES In the United States, the state, and not federal, authorities permit physicians to practice medicine. While most states practice reciprocity and do permit physicians licensed in other states to practice their profession in an emergency or for their own patients, when licensing a physician currently licensed in another state without requirements of prior residency or additional training or license examinations, registration with the additional state is usually required for any permanent practice. Before the Internet, a traditional doctor-patient relationship was usually established by taking a medical history and usually some personal contact. Although "ask a doctor" features are common on many of the health Web sites (see Rice, Peterson, & Christine, Chapter 10, this volume), it is a rare physician who will attempt to treat a patient without the taking of a medical history, current vital signs such as blood pressure and weight, and a personal interview.

Issues Associated With Internet Pharmacies

387

Before the Internet, prescriptions were filled by pharmacists licensed by the state in which they practiced in local facilities that, in most states, are also required to be licensed. While prescriptions could not be filled by email, prescriptions by mail order or, of late, through toll-free long distance telephone numbers, were not uncommon. Most states require a pharmacy that ships prescription drugs to its residents from outside of its borders to have a nonresident pharmacy license; the pharmacist filling the prescription has to have a license issued by the sending state, but not a license for the receiving state. The typical pharmacy that was set up to fill prescriptions for out-of-state customers had to obtain a nonresident license for each state requiring one, as well as a resident license for its home state. Pharmacists are required, by state law, to fill only valid prescriptions issued by licensed physicians. While a typical mode for a local transaction is for the patient to bring the original written prescription form to the pharmacist, prescriptions received by fax or over the telephone are not uncommon. (If the prescription is for a controlled substance, an original must be mailed to the pharmacist, although this usually happens after the prescription is filled.) If the pharmacist is familiar with the prescription issuer and the prescription seems in order, the pharmacist fills the requested drug and hands it to the customer. If the pharmacist is unfamiliar with the issuing doctor, detects something he or she deems incorrect or unusual in the order, or becomes aware of some other reason to question the prescription, he or she can telephone the doctor or check the doctor's credentials in a local directory or a national directory of holders of federal narcotics licenses issued by the Drug Enforcement Administration.

PROCEDURES OVER THE INTERNET The Internet makes it easier to establish immediate long distance relationships between customer and pharmacist, but the rules seem not to have changed. A pharmacist and pharmacy must still be licensed in the home state. To send drugs into most states, the pharmacy must still have a nonresident license. A pharmacist must still do what is necessary to verify the physician's order although, quite obviously, it is unlikely that the pharmacist will be familiar with the physician and that the prescription itself is probably presented by telephone or fax. Although the rules haven't changed, consumers considering purchases of prescription drugs over the Internet may well have more information

388

PUBLIC POLICY

about the Internet pharmacy than they do about their local corner drugstore. In 1904, an organization consisting of the pharmacy licensing authorities of the various states, the National Association of Boards of Pharmacies (NABP), came together to work toward standardization and uniformity of licensing standards for pharmacists in the various states. Today, the NABP sponsors standardized national examinations, maintains a national database of complaints and actions taken against licensed pharmacists, and continues with other programs in accordance with its mission. Increasing drug sales over the Internet have spurred the NABP to adopt a voluntary certification plan known as Verified Internet Pharmacy Practice Sites or VIPPS. As of February 2000,4 Internet-based pharmacies had qualified under the VIPPS program, and 19 were working through the certification process. According to the NABP, To be VIPPS certified, a pharmacy must comply with the licensing and inspection requirements of their state and each state to which they dispense pharmaceuticals. In addition, pharmacies displaying the VIPPS seal have demonstrated to NABP compliance with VIPPS criteria including patient rights to privacy, authentication and security of prescription orders, adherence to a recognized quality assurance policy, and provision of meaningful consultation between patients and pharmacists. Once certified, an Internet-based pharmacy is entitled to post on its Web site a VIPPS banner that includes a "verify" hyperlink back to the NABP Web site. The linked page includes the name, address, and phone number of the pharmacy's CEO; toll-free telephone numbers for reporting business problems and drug- or device-related problems; a listing of the license, if any, maintained by the pharmacy in each state; and for most states, a hyperlink to the state's pharmacy licensing authority. This is, in most instances, far more information than can be easily obtained about the local "brick-and-mortar" pharmacy.

A CASE STUDY Drugstore.com is one of the largest of the Internet-based pharmacies and one of the first to be certified under the NABP's VIPPS program. The site sells typical drugstore nonprescription items including beauty aids and

Issues Associated With Internet Pharmacies

389

Band-Aids, but also handles prescription drugs. (The site makes it clear that it does not sell Class II controlled substances, but does sell Class III and IV items as well as noncontrolled substances.) The company is affiliated with the RiteAid chain; RiteAid customers can order prescriptions through drugstore.com and pick them up at RiteAid, and can purchase RiteAid prescription refills through drugstore.com. E-mail and telephone customer assistance are available 24 hours a day, 7 days a week. A typical user points a Web Browser to http://www.drugstore.com and clicks on the "Pharmacy" button. The user can then search for a particular drug or select a drug from a list, read descriptions and side effects, as well as determine the price to be charged. If the user wishes to purchase the drug, the user is led through a series of screens asking about existing medical conditions, including allergies and drugs currently being taken. If all is well, the customer is invited to order the drugs in question by mailing the original of the physician's prescription, required for Class III to Vitems, or having the doctor call the pharmacy or the pharmacy call the doctor if the items are not controlled substances. The Web site requires a prescription from the customer's physician and does not make referrals or have some resident physician who writes a prescription. Payment is made by charging a credit card, although the company will deal with health insurance companies that will deal with them. Shipping by U.S. Postal Service Priority Mail is done at no additional cost; overnight delivery was quoted at $11.95 additional. Altogether, we found the drug purchasing experience at drugstore.com to take less ordering time, and be more convenient, than using a traditional local drugstore. We were able to get more information about the drugs in question and ask more questions than we would have at our local drugstore, without standing in line to speak with the pharmacist. The local pharmacy may be open 12 to 18 hours each day; drugstore.com is open all of the time. Finally, drugstore.com, unlike most local pharmacies, delivers. The major advantage that a local pharmacy has is that it is local and can get a drug needed immediately to the customer immediately. Drugstore.com cannot do that, except to RiteAid customers. Next-day or 3-day delivery is likely adequate for most maintenance drugs with refills that can be ordered in advance. The drugstore.com customer is likely not to have the same personal relationship that local customers used to have with the friendly neighborhood pharmacist, but as chains have driven out pharmacist-owned stores and as pharmacists are given less time to do their jobs, that personal contact may be absent even from the local store. In fact,

PUBLIC POLICY

390

drugstore.com greeted us by name the second time we accessed the site; the typical local pharmacy probably won't match that level of personal contact even after several visits.

COMPLIANCE EFFORTS Web-based pharmacies are sufficiently new that there appear to have been few compliance problems thus far. One large area of exception surrounds several unlicensed companies that are alleged to have been selling the phosphodiesterase inhibitor Viagra (Sildenafil), referring a customer's medical history, as entered by the customer, to an affiliated physician who, practicing "telemedicine," issues a prescription to the pharmacy, which proceeds to fill it and ship it to the customer (Illinois Attorney General, 1999). A reading of the complaints indicates that these practices would be actionable even if business had not been transacted over the Internet, as the pharmacies were not licensed and the prescriptions filled do not, with the pharmacist's knowledge, result from a bona fide doctor-patient relationship, and therefore would have been illegal even if the pharmacy had been licensed.

NEAR-TERM TRENDS AND IMPLICATIONS As in many areas of consumer commerce, the World Wide Web model presents a pharmacy that can be more efficient and less expensive for the consumer. Overnight or 2-day shipping is remarkably inexpensive and may be more convenient than actually venturing out to a local brick-andmortar pharmacy. Further, technology exists today that would enable physicians to sign their prescription orders with a digital signature that confirms both the content of the prescription and the identity of the physician. A state licensing authority or the DE A could issue a time-limited digital certificate that not only identifies the physician but guarantees the pharmacist that the physician is duly licensed. The certificate could even be made to refer back to the issuing authority, ensuring that the license in question is still in effect. This would provide a level of protection not available today from pen or telephone, would cost little, and in substituting a computer-based text message for a difficult-to-decipher handwritten scrawl, would reduce communication errors.

Issues Associated With Internet Pharmacies

391

The prescription itself need not require that a physician learn to type, but could be generated from an on-screen menu from which the physician could select the drug, the medication dosage, the frequency and times at which it should be taken, and even the purpose for which the drug was prescribed. All of these can be validated for internal consistency of dosage and purpose and checked against the patient's database file for suitability for the specific patient, including possible allergies and interactions with other medications previously prescribed. After the validation process is complete, the digitally signed prescription is e-mailed to the pharmacy. Because the digitally signed e-mail contains its own validation as to the physician's credentials and the message contents, there can be no question that it was ordered by a properly licensed physician and that the order was exactly what the physician intended. Properly structured, the order could even be checked for suitability before it arrives on the pharmacist's screen.

REFERENCES Attorney General of the State of Illinois. (1999). Ryan sues to protect consumers against illegal sales of prescription drugs over the Internet [Press Release by Attorney General Jim Ryan, October 21] [Online]. Available: http://cait.wiu.edu/ press/pressfrm.html Bloom, B. S., & Iannocone, R. C. (1999). Internet availability of prescription pharmaceuticals to the public. Annals of Internet Medicine, 7(December 13), 830833.

18 Networked Communication Practices and the Security and Privacy of Electronic Health Care Records JAMES E. KATZ PHILIP ASPDEN With research assistance from Eleanor Wynn

First and foremost, we have to safeguard our citizens' privacy. Last year, we proposed to protect every citizen's medical record. This year, we will finalize those rules. President Bill Clinton in his State of the Union Address, January 21, 2000

I

n this chapter, we highlight how the Internet and other computer technologies have introduced new complexities into the management of the privacy and security aspects of personal health information. Our data are drawn largely from interviews with health care professionals whose expertise is security or privacy. We began investigating this topic in 1995 in response to a request from a large telecommunications company that 393

394

PUBLIC POLICY

wanted to understand the unmet security and privacy needs of this sector. Once under way, we continued our inquiry on a less formal basis. We have conducted more than a score of structured interviews with members of the health care community (as Table 18.1 summarizes), as well as many less formal ones. We take as a point of departure that, when examining user needs and commercial opportunities in the health care sector, it is insufficient to identify only the demand characteristics for system performance. It is also of central importance (but also easily overlooked) to include an analysis of the social and cultural contexts within which such communication takes place. Hence, this is the focus of our analysis. One reason we have adopted this perspective is that we have been disappointed by "social engineering fixes" to deal adequately with complex communication environments. These complex environments include not only health care, but education, religion, and scientific research as well (Katz & Graveman, 1991).

UNIQUE ASPECTS OF HEALTH CARE INFORMATION Medical information is distinct in character from the information on which contemporary security models are based. The primary function of medical information may be thought of as routine, which is to say that even though it is important for treatment and saving lives in real-time treatment situations, often on an emergency basis, there is seldom great value in the possession of much of the data themselves. Yet there are also times when the data are not only extremely important but also extremely valuable (U.S. Congress, 1993). This, of course, is in stark contrast to military or financial records where the value is intrinsic. Medical records are characterized by contradictory demands on the design of medical records security: On the one hand, to "make available the right information at the right time to the right people for the right purpose," and on the other to protect information from increasing numbers and kinds of intrusions that will be occasioned by both the proprietary value and importance in legal proceedings, as well as the sheer mass, and hence potential "value," of records to the "wrong people for the wrong purpose."

Security and Privacy of Electronic Records

395

Examples of Professions and Role of Interviewees Information Security Officer, Kaiser Permanente, Augusta, G A Chief Information Officer, Oregon Health Sciences University, Portland, O R Information Systems Staff, Brigham and Women's Hospital, Boston, MA Quality & Medical Staff Services, Providence Medical Center, Portland, O R Information Architect, Sunnybrook Health Science Center, Toronto, Canada Physician, Providence Medical Center and BICC/Oregon Health Sciences University, Portland, O R Internet medical chat and support group participant Oncology Records Technician, Portland Adventist Medical Center, Portland, O R IS Applications Developer, Blue Cross-Blue Shield of Oregon, Portland, O R Researcher, Royal Cancer Hospital, Sutton, Surrey, UK Telemedicine Practitioner and Researcher, Biomedical Information Community Center, Oregon Health Sciences University, Portland, O R Researcher, Ontario Northland Telecom, Ontario, Canada Information Scientist, Oklahoma Medical Research Foundation's Telemedicine Research Program, Oklahoma City, O K Information Scientist, Center for Clinical Computing, Beth Israel Hospital, Boston, M A Computer Staff, W i l l i a m Beaumont Hospital, Troy, Ml Radiation Oncologist (and chair of electronic medical records committee), Salem, O R Medical privacy legislation writer, A C L U Massachusetts, Boston, M A

The improper use of patient information creates an open-ended liability with severe consequences for every stakeholder involved: the patient, the practitioner, the institution, the insurer, and the payer. Local, regional, and national committees have been formed to sort out the requirements for appropriate use of data, which must reflect laws, standards, and best practices established from multiple perspectives. The complexity of the medical records situation in a networked environment requires a different approach to security than the relatively straightforward applications in military and financial domains.

Practitioner Concern Uneven Over Evolving Practices Until quite recently, the bulk of medical records has been maintained on paper. Each record had one copy. Information was selected or abstracted

396

PUBLIC POLICY

from extensive paper records and transferred manually to electronic administrative systems for billing and other basic administrative functions. Generally, a patient would have records distributed over many practitioners and institutions, with no means of linking the information in them other than paper or fax transfer and the manual copying of data from selected form fields. Perhaps the most comprehensive records were those maintained on claims to insurance companies and information communicated to the Medical Information Bureau on applications for insurance. But even insurance records lose track of individuals when they change providers. Moreover, the sheer bulk of the physical paper militated against the records being readily accessed. Due to the dispersed nature of paper records, the culture of security surrounding them, while based in a strong professional ethic of patient privacy, was also controlled by informal social means based in the community of practitioners and the norms reinforced among them. Only in the past few years has the term unified patient record become common. The unified patient record carries many implications beyond the simple transfer of information from paper to electronic media. First, as a business process, it eliminates redundant copying of records and the concurrent loss of information, especially billable information, which occurs with every transfer. Second, in terms of physician practice, it means that a patient's history can be linked from beginning to end so that far more context for treatment is available. A third consequence is that the records now become material for a wide range of statistical data mining, which promises valuable information both in medical outcomes research and in public health efforts in prevention (Dayhoff & Maloney, 1992). Hence the unified medical record, through its accumulation into local, regional, and national databases, has the ability to transform medical business processes, practices, and knowledge. However, by the very fact of being an accumulation of both longitudinal life histories and of contemporary masses of individual information, these records can readily become a security vulnerability of major proportions because of the many forms of value they will represent to parties not directly involved in medicine nor concerned about patients' rights. The community of practitioners stands to lose control over the information (Department of Commerce, 1997). Most MDs interviewed were still unaware of the implications for invasion of privacy that medical records represent. They tended to be more concerned with how security measures would affect their practices. To the extent that they were concerned, they worried about protecting their prac-

Secunty and Pnvacy of Electronic Records

397

tices from unwanted legal intrusion or efforts to recruit patients away from them. A common statement from MDs was that they could not understand why anyone would bother to browse patient records without a good reason. They tend to think of the records in terms of the medically useful information in them, ignoring the collateral information that is also contained in them that, when properly agglomerated, becomes valuable in its own right. Thus one physician who worked in a hospital setting told us, "We never indicate on a patient's chart whether he is HIV positive. We just mention it to one another. If we wrote it down, then it could become seen by too many people and the quality of care would suffer."

Significance of Health Information Security and Privacy Despite the grand potential and the broad interest, none of the benefits cited above will be possible if health databases are easily open to abuse or if they violate the interests of any of the participants through damaging use of information (Dalander, Wilner, & Brasch, 1997). In part this will be due to the fact that many who feel that their privacy might be compromised will avoid the medical establishment. This obviously would make it more difficult to wage public health campaigns against diseases like HIV. At the most elementary level, then, security and privacy are at the foundation of any benefit that could arise from large medical databases. Below, we attempt to set forth some major issues concerning data security and privacy, why these problems are particularly nettlesome for the health care community, and to suggest possible quadrants from which solutions might be drawn.

VARYING PERSPECTIVES ON THE NEED FOR HEALTH CARE INFORMATION CONFIDENTIALITY Abuses of patient privacy can severely impact three main groups: patients themselves, health care providers, and administrators. Patients generally believe they have a right to confidential treatment of their medical records. The first and most visible victim is the patient whose privacy has been invaded. However, surveys show that few think there are laws ensuring such treatment. Indeed, there are no laws at the federal level

398

PUBLIC POLICY

saying that physicians must do so. While state laws protect patients in some jurisdictions, these often tend to be a patchwork with many exceptions. For instance, even though a state law may require a hospital to adhere to privacy regulations, it may exempt dentists. Clearly, there are important reasons why patients might be concerned about their medical records becoming public or semi-public information. In particular, sensitive information (e.g., past alcohol problems, results of psychiatric tests, abortions) in the hands of insurance companies and employers (current or potential) could lead to restrictions in health insurance, problems in obtaining both health and life insurance, and difficulties in retaining or getting jobs. Moreover, the implications for self-esteem and social relationships could be considerable. Health care providers are concerned that unrestricted access to information on their patients will aid or stimulate malpractice suits, will lead to unfavorable comparisons with the performance of other physicians, and, in extreme cases, will lead to poaching of their patients. Administrators have major concerns about being sued for violations of patient data confidentiality as well as about having their institutions analyzed or compared unfavorably with other institutions. In addition, some practices are considered proprietary information that gives the practitioner or institution a competitive advantage, and this competitive edge must be preserved.

Communication Concerns Regarding Patient Privacy In terms of public policy, the first level of concern to address is that of patient privacy and confidentiality of records. While hospital administrators and practitioners are able to manage or lobby for their own needs, patients remain an unorganized and predominantly unaware "interest" group. However, emerging events will draw the public's attention to this area, as records become more available and abuses of access become more frequent, without adequate protection. This may be highlighted in the occasional revelation about misdeeds that find their way into newspapers. For example, the Congressional campaign of Nydia M. Velazquez was shaken when it was revealed that Velazquez had sought emergency treatment at a hospital after a suicide attempt. Velazquez went on to win the election and then sued a Manhattan hospital for $10 million for failing to protect her privacy ("Who's Reading Your Medical Records?" 1994). A

Secunty and Privacy of Electronic Records

399

convicted child rapist working at a hospital near Boston rifled through nearly 1,000 files on young female patients for telephone numbers. The violator used the numbers to place obscene calls to girls. He had obtained a password from a former employee at the hospital (Privacy Journal, 1995). Perhaps one of the most famous cases was the one in which a banker who served on the board of a hospital obtained records on his clients, and then pulled the loans of those who were being treated for cancer (Crowley, 2000). The public mood concerning health care privacy on the internet was reflected in a national poll released in January 2000 by the California Health Care Foundation. The survey revealed that one in five U.S. adults believes that a health care provider, insurance plan, government agency, or employer has improperly disclosed personal medical information. Half of these people say the disclosure resulted in personal embarrassment or harm. The survey also found that 75 % of those seeking health information on the Internet are concerned or very concerned about the sites' sharing their personal health information with a third party without their permission (http://www.healthprivacy.org/resources/exposed.pdf). Clearly the problem spills over from questions of privacy to questions of probity. Hence, following revelations that he had been quietly accepting revenue from products sold on his branded Web site (http:// www.drkoop.com), former surgeon General C. Everett Koop suggested that a national summit be conducted to investigate ethical practices on health-oriented Web sites.

Communication Confidentiality Concerns Regarding Patient Records There are numerous views and contingencies about who owns and who has access to patients' records (Institute of Medicine, 1991), yet it is clear that many have access to various levels of them, and that patients have only a dim understanding of the extent to which access is available ("Why Your Privacy Is Threatened," 1997). The patient's actual health record or health history information carries implications for a range of opportunities for discriminatory actions, regardless of how trivial the incident is. Thus any past health problem can affect a candidate for political office, regardless of the applicability of the condition to the patient's present health or job requirements. Prior treatment for alcohol abuse, any mental health counseling, abortion, and HIV are only the most obvious of a host of facts

400

PUBLIC POLICY

about patients that carry implications about an individual patient's character or reputation. Patients disclose large amounts of personal information to practitioners. It is not easy to predict what will be relevant to a diagnosis, and often a search process is involved that uncovers many aspects of the patient's personal life. This remains in the record forever. Abuses of patient privacy can thus carry major consequences, such as the following: 1. Sensitive information (e.g., past alcohol problems, results of psychiatric tests, abortions) can impact careers, insurance eligibility, employability, marriageability, and suitability for child custody. 2. There have already been cases of attorneys browsing records for malpractice claims, practitioners soliciting another physician's patients, and predatory use of patient telephone numbers and addresses by criminals. 3. Invasion of patient privacy can lead to legal actions against institutions for violations of fiduciary confidentiality.

Growing Threats to Confidentiality of Health Care Information In the context of health-related information systems, the increase in anxiety about threats to personal privacy can be explained by the two major factors. First, electronically stored information is easy to collect, analyze, and transmit in bulk. As a result, the accountability for protection of privacy is more dispersed, and it has increasing impact on people's careers, ability to get insurance, or retain jobs. Second, the tradition of confidentiality of patient information, while still of great concern, is more difficult to control by traditional means in the large health care communities that are emerging. Figures 18.1 and 18.2 illustrate this situation. Figure 18.1 depicts the typical perception of the consumer of health care when meeting with a doctor about, say, a sore throat. The doctor orders some tests and writes a prescription. The patient has the tests conducted and has the prescription filled. The requisite forms are filed with the insurance agency. A more realistic, but still incomplete, picture of data flows, including personal information about the patient, is shown in Figure 18.2. Given this much more complex flow of information, it would be useful to see what technologies might be available to enhance security and privacy of records in this environment, especially across boundaries (see Katz & Rice, Chapter 19, this

Secunty and Pnvacy of Electronic Records

401

Lab Claim

•

Explanation of

Benefit /payment Imaging Center

Order • Results

Doctor Pharmacy Pick-up scrip Patient

i

Figure 18.1. Typical Patient's Perception of Personal Information Flow Caused by Minor Office Visit to a Physician During the Course of Which a Routine Test Is Performed S O U R C E : JanLori Goldman. N O T E : Aggregate information flows indicated by thick dotted lines.

volume). Lest the reader think we are putting too fine a point on this matter, a Congressional Research Service study in 1996 estimated that during a hospital stay, perhaps as many as 400 people could view at least part of a patient's medical record (Blevins, 1999). The problem of confidentiality, rather than getting better as time passes and policy matures, seems to be getting worse. The vaunted possibility that the Internet could provide safe, secure, and anonymous surfing is being turned on its head. In fact, ever more detailed data can be collected and profiles constructed; friendship and communication patterns can be captured in superlative detail. Table 18.2 shows ways this can and does happen with health care-related activities of individuals; it also includes a hypothetical illustration showing precisely how the Internet could intrude on someone's health issues. The Clinton-Gore administration has not been insensitive to the problem. Vice President Gore himself has been a consistent advocate for consumer privacy protections, especially on the Internet. When a story broke about a national pharmacy chain selling lists of clients by type of drug purchased to a pharmaceutical marketing firm, a storm of protest was unleashed (though it did not result in any new laws being passed). When asked during an interview about this incident, the Vice President said:

402

Re-insurer, 1 risk arbitrager |

Employer

f

-^Enrollment

ρ

~

E

Paymen

-

d

*

e

t

|

^

—

Γ Public

J

\

*

insurance & Health

j

Programs

|

f

Ϊ

agengies

|

incldg,

g

I J

ι ι |

b

\\

0nus

|

\1

Encounter

S

/

I /

*y\ Report

Q

C

Center

m0

|

Public Comm. on I Health Quality § Service, 1 Assurance | Soc. Svcs

B

Ύ

4

ffr—

Patient

, B

·«»

•

Pharmacy

Pharm.

1

/

I

r—

j

1

1

1

[

L___ 1

Co-payment

tx

\ * j

t

at practice 1I

, \

v

* ' Pick-up

|

Nationat

f

/ γ L

Results / 4

1 1

i

a

• ./ / / / P ' \ ^ // / |> payment \

Doctor

rder

™

uti

/ ' '%

/

ι

L

Λ

1 /

V

Association

Board

Pharmaceut

L^f™!^ J

Lab / 4

' ;

|

——— —i 3rd Party

/

"

1

Bureau

Admin J

' '

| Local records

5

^

I

j payment \ \

ϊ

Indep. Practit.

1

™_ϊ_"

/

*

L^In 4

/

/

a

§ EOB / \\

Imaging

£

\

/

Ί>