For the Public Good: Highlights from the Institute of Medicine, 1970-1995 (1995)

by Avedis Donabedian, M.D., M.P.H.

Nathan Sinai Distinguished Professor Emeritus of Public Health, University of Michigan School of Public Health

While many things change, certain fundamentals endure. And it is a pleasure to see that during the quarter-century so ably reviewed by Kathleen Lohr, the Institute of Medicine (IOM) has remained abreast of the former without compromising the latter. Always, it has placed the health and welfare of people first, in all other things remaining flexible.

It is appropriate, therefore, that the review of IOM's contributions begins with a definition of quality itself, for it is this that embodies our societal values and purposes, while also shaping our efforts to assess and advance quality. Almost everything one would like to see is either manifest or latent in that definition.

The relation between science and technology, on the one hand, and the quality of care, on the other, is another of the fundamentals that the Institute has been attentive to. But perhaps the definition of technology could be expanded to include the technologies of management and behavioral change. Often, it is these, rather than medical science itself, that determine success or failure in the pursuit of quality in health care.

Besides the rapid pace of technological invention, perhaps the most important progressions during the quarter-century under review have been in the social commitment to paying for health care; in the more organized ways of delivering it; and as the power of payers and purveyors has grown, in the greater attention of consumers to their own interests. To all of these, the IOM has been responsive.

Government has been, of course, the largest single player in this arena, and as its obligations for ensuring access to care have grown, so also have its concerns for the cost of care, its quality, and for a balance of the two. Consequently, as government has had to assess the performance of its programs, to discover their weaknesses, and to find remedies for these, it has repeatedly called on the IOM for help. Because good policy requires good information, the Institute has perhaps made its greatest contribution as a source of impartial expertise whenever called upon for investigation, analysis, and advice.

Delivering care and paying for it under organized auspices, governmental and other, have improved access to care for many (though not all), while also offering prospects of significant improvements in quality as care is stripped of superfluities and becomes more subject to institutional supervision and public scrutiny. At the same time, payment per case and per capita, as well as, perhaps, the resurgence of for-profit enterprise, may have posed a risk to consumers. On the one hand, as efforts to control costs gain dominance, there is a temptation to skimp on some important aspects of care, of which attention to the niceties of care, so cherished by consumers, is perhaps the most vulnerable. On the other hand, as care is shifted to less equipped, less supervised sites, the threat to quality is bound to grow unless special efforts are made to anticipate the danger.

Consumers, individually and in association, have reacted rather incoherently to these developments. Some have become so disenchanted with the medical establishment, so suspicious of its motives, so skeptical about the efficacy of its methods that, alienated, they have sought alternatives apparently more responsive to their needs. Others, sometimes expecting more of the medical system than it can realistically accomplish, have challenged it by demanding more attention to their preferences, more participation in the system's operations, and more access to information necessary for informed choice.

Happily, the IOM, as Lohr's review of its activities amply demonstrates, has responded to these developments, cognizant of the currents and cross-currents they have set in motion. The Institute has explored the relation between quality and cost. It has defined quality to include patient preferences, while providing tools for measuring quality of life (the outcome of care) in ways meaningful to everyone, In assessing access to information, it has balanced nicely the need to know with the need to offer protection to both customer and provider. And it has advanced quality by holding it always as the most compelling goal, by setting standards that serve it, and by developing methods by which it may be assessed and, where needed, improved.

No doubt in obedience to her charge, much of Lohr's review deals with the strategies and methods of quality assessment and assurance. Here, perhaps the two most important developments the Institute has had to face have been the remarkable recent advances in informational technology and the infiltration

into the health care system of theories and methods of management relevant to industrial quality control.

There is ample documentation in Lohr's review of the attention the Institute has paid to the accuracy of information pertinent to quality assessment. As to the ostensibly new, as compared to the old, in the theories and methods of quality management, the Institute has adopted a judicious middle ground, welcoming what seems useful in the new without abandoning what appears useful in the old. The same judicious balancing of seeming opposites has also characterized the Institute's response to the process–outcome debate. It has helped to develop criteria and measures of both. By refusing to be duped by spurious dichotomies and by recognizing kinship in concepts and methods, the Institute has stood for continuity and sanity in a sometimes confused arena of contending interests.

As to the future, one could easily say, “Continue as before.” But each of us has views of what is more or less relevant, demanding greater or lesser attention, and these the commentator has been asked to reveal. In doing so, I do not mean to imply that the Institute is obligated to do all that I think needs to be done. It must, of course, pick and choose, governed by its own institutional purposes and obligations.

Concerning the more fundamental concepts of quality in health care, it seems that the issues that call for greater attention are the means for bringing about greater, more informed participation by patients in decisions concerning their own care; a better understanding of the implications of the cost–benefit balance to public policy and the behavior of individual practitioners; and the pursuit of equity, with its attendant implications.

As of now, we do not even have a near-complete understanding of how various cost control initiatives have influenced the quality of care in its subtler aspects. In particular, we need to know more about how shifts in the site of care have influenced quality, and what safeguards, through reorganization and supervision, need to be put in place.

As to the more specific methods of quality assessment and assurance, there is a need to validate clinical practice guidelines by empirical testing and to understand what factors contribute to their successful adoption. I also feel that the tracer method the Institute was so instrumental in seeing developed has not received the attention it deserves, when properly applied, as a means for methodically exploring the strengths and weaknesses of a health care system in specific areas of performance. It may be true, as some maintain, that these strengths and weaknesses are already known to system participants, if only they were to be asked; but that is something that deserves testing.

More generally speaking, if one were asked what methods for protecting and advancing quality are now in operation throughout this land, and how successful they have been, the answer might be difficult to find. A mere inventory, periodically updated, would be interesting and would generate a concep-

tual categorization and a corresponding vocabulary. But, more fundamentally, we do not know what such efforts accomplish in either altering behavior or improving health. We need to learn what factors influence the adoption of one method of quality control and improvement in preference to another and, subsequently, what factors contribute to success or failure. My own hunch is that success or failure is determined not so much by the “technical” features of a method as by the motivation of those who operate it and are subject to it. There is, of course, much speculation about what the determinants of motivation are, but the kind of empirical testing required to establish these seems to have lagged behind, at least with regard to quality assurance. The steps needed to resolve these issues are the usual ones: the establishment of a sound conceptual model, followed by observational studies and, eventually, by experimental evaluation.

The persuasive, but often unproved, claims made by what is variously called the “new philosophy,” the “industrial model,” “total quality management,” and the like, lend particular urgency to the task of rigorous, unbiased evaluation. Is it true that there is a single, optimal style of management? Are system failures rather than individual predispositions the dominant contributors to differences in quality? How efficient and successful are quality improvement teams in identifying the more important obstacles to better quality? To what extent, if any, has attention to the efficiency of supportive services detracted from attention to the clinical process itself? What is the desirable balance of centralized, managerially controlled activities as compared to those that are decentralized and delegated to the rank and file? And what is the appropriate balance of external and internal controls, of regulatory versus anticipatory and rehabilitative components? These and a myriad of other questions seem to require impartial exploration. What more suitable arbiter than the IOM to help elucidate such problems?

Greater involvement in management matters would seem to require greater attention to managerial philosophies, styles, and techniques, as well as the elaboration and testing of alternative system designs. The quality assurance enterprise should, itself, be viewed as a system of interrelated parts, comprising ethical, conceptual, motivational, regulatory, managerial, and technical components, Moreover, it is a system located in an environment (social, economic, political, regulatory, legal, etc.) that affects its institution and implementation at every turn. To view the quality assurance enterprise in this light is a challenge worthy of the IOM.

I have left to last the important subject of “consumerism.”

The IOM seems to be, by virtue of its role and membership, preeminently a source of medical expertise. As such, its undoubted sensitivity to consumer interests is fiduciary, in the best traditions of the health care professions. Would the next step be an actual partnership with consumers, as equals, in

serving the public interest? Whether that happens or not, the Institute will no doubt have much more to do in its advocacy on behalf of consumers.

As I have already said, much needs to be learned about how patients can participate in decisions about their own care; both patients and practitioners must desire this to an equal degree. We still do not know what to tell consumers, when, and how, so they can choose among alternative sources of care. More directly germane to quality assurance, we have, by and large, failed to involve consumers in the process of performance review itself. Moreover, consumers, as a whole, need authoritative, unbiased information so as to participate more intelligently in the public debate on health care policy.

Quite clearly, the IOM, during the last quarter-century, has had a remarkable record of service to quality in health care. It knows, of course, that much more remains to be done. As it faces the challenges ahead, I wish it well.

by Robert H. Brook, M.D., Sc.D.

Director, RAND Health Sciences Program, and Professor of Medicine and Health Services, UCLA Center for Health Sciences

If the 1994 U.S. elections are a marker of things to come, transformation of the health care system using the competitive marketplace as a model will likely be pushed forward. Under present and predictable circumstances, the Institute of Medicine (IOM) must play a central role, over the next 25 years, in keeping quality of care in that system from deteriorating. No other organization is situated as well as the IOM to keep quality on the policy agenda. Its membership of scientists, clinicians, and policy and administrative professionals makes it the preeminent organization to tackle this mission. (Given its membership, it can even provide this function on an international level.) The fact that it does not provide funding for medical care makes it a better organization to accomplish that mission than, for example, the federal government, which has an interest in seeing that medical costs are reduced or at least contained.

If the IOM is to fulfill this role, however, the future demands changes in how it conducts its work. Doing projects through its traditional committee structure, which results in peer-reviewed consensus documents on topics requested by external funders or, less often, that arise as self-initiated projects reflecting the Institute's own interests has served the IOM and the nation well in the past. This approach must now, however, be modified so that a strategic plan for addressing key questions systematically is used to decide what projects are taken on and, thus, what reports are written. In addition, the IOM must

establish a process for evaluating the efforts of its study committees. The key questions include the following: Do these committees produce specific, relevant recommendations that can be implemented? If so, is that advice implemented? Do positive results occur? These are the points I address in this commentary.

As Dr. Lohr has documented, over the past 25 years the IOM has played an impressive role in undertaking studies that affect how we think about and measure quality of care, as well as how we improve the quality of care.

As a member of some of the committees she describes, I can attest that the discussions about how to measure and how to improve quality of care were intense, provocative, and intellectually stimulating. Having spent 25 years of my life developing measures of quality of care and using those measures as a means to improve the average level of quality of care in the population, I can also affirm that the IOM's work has been useful to researchers.

For example, almost 15 years ago, the IOM agreed upon a definition of primary care (IOM, 1978, 1994) and identified the attributes that distinguish high-quality from lower-quality primary care. That framework provided the basis for developing a set of measures and tools by which we could evaluate access to care, processes of care, outcomes of care, the structural characteristics of the primary care site, and physician and patient satisfaction. I was part of a team of researchers and medical directors of 15 academically based, internal medicine group practices that used the IOM's definition of good primary care as the basis of an evaluation of the care that they provided in their settings (Kosecoff et al., 1985; Brook et al., 1987).

We determined that even academic primary care teaching sites failed to meet many components of the IOM's definition of good primary care. The medical directors' willingness to agree with this conclusion was facilitated, if not made possible, because a prestigious IOM committee, representing various points of view, had developed and agreed upon the way in which good primary care could be identified. The prestige attached to the IOM's committee process made the critical research results, which were based on recommendations in the report, acceptable to clinicians who wanted to improve quality of care in their own practices. Thus, in answer to my key questions in this commentary, this work demonstrated that an IOM committee could agree upon specific recommendations that could be used to design sound empirical research to identify problems in the quality of primary care and point the way to their resolution.

The health care system now faces a radically different environment, as does the IOM. Thus, after 25 years of committee-based work in the quality-of-care field that has been dictated largely by the interests of external funders, the IOM needs to ask itself, “How can the Institute ensure that the quality of health care improves rather than deteriorates under various health system reforms?” and “How can the IOM help reduce the variation in quality of care by gender, race, socioeconomic class, and hospital or provider?” To accomplish these goals, the IOM must rethink its structure.

First of all, the studies outlined by Dr. Lohr represent mostly windows of opportunity; they were not the result of a strategic planning process. Given its charter and funding structure, the IOM must respond in a cooperative manner to requests from the federal government and other organizations. Many of these requests represent important studies that should be done. In the future, however, the IOM needs to do more than respond: It must develop a strategic vision of where clear questions asked and specific recommendations produced by a committee representing multiple viewpoints about quality of care, such as the committee I cited earlier, will make a critical difference in terms of both improving the average level of quality and reducing its variation.

Second, once the IOM has identified the critical questions that should be addressed, it must create a process to evaluate whether it has accomplished its strategic objectives. This process of accountability would answer such questions as these: Did the IOM address the central issues of the field, or did the funds available to the Institute drive it to consider what are basically marginal issues? If the IOM committees addressed these central issues, did they also produce specific, usable recommendations? Was that advice heeded? If the above questions are answered affirmatively, then an analysis of the benefits produced by IOM recommendations should conclude that cost-effective solutions to maintaining or improving quality have been implemented in health systems throughout the world. The IOM should work with its funders to facilitate the completion of such analyses by independent research organizations. The results of those analyses should be used to further refine the IOM process.

Presently, the best that can be said of the past is that a wide array of interesting and productive studies were performed: Committees addressed important issues; action-oriented recommendations were produced; and anecdotal evidence, such as the evidence I provided earlier in this commentary, suggests that the recommendations influenced care. However, the world of health services research and policy analysis has come a long way in 25 years. Although the anecdotal paradigm may once have been appropriate, the next quarter-century requires that we apply both a more sophisticated strategic vision to the field and different standards of accountability.

The IOM has played a vital role in keeping quality of care on the agenda since its founding. If, as seems probable, cost pressures continue to dominate the U.S. health care system in the future, then the IOM will have to double or triple its activity in the quality-of-care area. If that is to be done, and if society is to believe that it gets value for the money invested in the IOM committee study process, then that process needs to address the questions listed above. Only by answering such questions can we demonstrate the value added by IOM's committee-based approach to its work in maintaining and improving the quality of health care in the United States.

As the IOM celebrates its 25 years of exceptional contributions to the field of quality of care, I hope its membership takes the time to ensure that, during the next 25 years, the IOM can issue even better, and more timely, products that will make health care better for all Americans. Its new quality initiative, as discussed by Dr. Lohr, is a step in this direction. If the IOM is successful, we will be able both to answer questions such as those below and to provide specific, usable recommendations regarding how to protect and improve quality.

• Is quality increasing, decreasing, or staying the same? Is the variation in quality by patient characteristics (age, gender, ethnic status) getting smaller or larger? Do we have an adequate data base to address these issues?

• Are differences in quality by state or city becoming larger or smaller?

• Are differences in quality by health care provider becoming larger or smaller?

• Which health systems provide better outcomes? For example, if I have heart attack in Washington, D.C., Paris, or London, where will I have the best chance of surviving?

• Are the financial incentives in the health system structured to ensure high-quality care or to undermine it?

• How does the governance structure or training of the leadership of managed care organizations affect health system quality?

• What information will help consumers choose the best hospital or physician or make a treatment choice that reflects their preferences and priorities?

• Do we know how to make trade-offs between improving quality of care for the population as a whole and improving quality for those who actually seek care?

• At the margin, is it better to invest in more medical care, greater quality of care, or more education, housing, or jobs?

In the next 25 years, the IOM must become the conscience of the nation in the quality arena. It must deal with the big issues and demand that the information needed to address them be produced. It must do work that will yield specific recommendations that are geared to maintaining quality while all the pressures on the health system are geared to reducing it. Most of all, the IOM must succeed—the consequences of failure will be too high.

Brook, R.H., Fink, A., and Kosecoff, J. 1987. Educating physicians and treating patients in the ambulatory setting: Where are we going and how will we know when we arrive?Annals of Internal Medicine 107:392–8.

Institute of Medicine. 1978. A Manpower Policy for Primary Health Care. Washington, D.C.: National Academy of Sciences.

Institute of Medicine. 1994. Defining Primary Care: An Interim Report,Molla S. Donaldson, Karl Yordy, and Neil A. Vanselow, eds. Washington, D.C.: National Academy Press.

Kosecoff, J., Fink, A., and Brook, R.H. 1985. General medical care and the education of internists in university hospitals: An evaluation of the teaching hospital general medicine group practice program. Annals of Internal Medicine 102:250–7.