| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1995;60:1530-1533
© 1995 The Society of Thoracic Surgeons
Ohio State University College of Medicine, Columbus, Ohio
Abstract
United States physicians are grappling with a fundamental reorganization of the healthcare system. Although many remain skeptical of governmental efforts at reform, they seem to take as given an industrial efficiency model of change, including large and integrated managed care arrangements and performance measurement based on outcomes such as quality. This uncharacteristic acquiescence seems to derive in part from a confounding of concepts. This article makes three distinctions: among knowledge about practice, knowledge about quality, and outcomes research; between outcomes research and the outcomes movement; and between the outcomes movement and other options for healthcare reform. The suggestion that statistical measurement of specific variables ought not to have a priori precedence over other ways of thinking—and doing something—about healthcare is made.
Physicians are grappling with a fundamental reorganization of American healthcare. You have heard about managed care arrangements of various kinds—how to join them, how to own them, how to humor them. You have heard managed care linked to industrial efficiency and industrial efficiency linked to information about outcomes. You have been exhorted to become businessmen and statisticians, to circle the data points in self-defense or mobilize them to assault competitors and payers. And you have enjoyed veiled and not-so-veiled attacks on government, on healthcare reform, and on Hillary Clinton.
However, you have not heard a few simple distinctions that I believe generate some not-so-simple questions about the future of thoracic surgery and all of medicine. I make these distinctions to clear away some of the hype and to urge you to become critical thinkers, if not experts, about the current mania for measurement-based managed care. I will distinguish knowledge about practice from knowledge about quality from outcomes research, outcomes research as an information-gathering technique from the outcomes movement that privileges this information, and the outcomes movement from other options for healthcare reform.
The first distinction is among three kinds of knowledge: knowledge about practice, knowledge about quality, and outcomes research. Public discussion often treats these matters indiscriminately. It is assumed that quality is all there is to practice and measurement is all there is to quality. Sometimes speakers accept this equivalence, sometimes it is advantageous to act as if they did, and often it is simply easier not to bother teasing it apart. I believe the distinction is critical—that ultimately the confounding of practice with quality with outcomes will disempower clinical medicine.
Knowledge about practice is what you know, your colleagues know, scientists and social scientists and managers and patients know about what happens in the course of medical care. What physicians know has until recently enjoyed a special status in United States healthcare, and physicians' knowledge is itself woven of several epistemological threads [1]. Medical knowledge derives, for example, from both a realist and empiricist view of the world. That is, physicians posit actual entities related through cause and effect, as laboratory scientists do, and rely, as in the case of randomized, controlled trials, on probabilistic findings from statistical analysis of aggregate data. Physicians partake in some measure of both perspectives. When they think probabilistically, furthermore, they tend to read research findings through the lenses of their clinical experience [2].
Some of what physicians know is scientific theory; some comprises an accumulation of facts. Physicians also make clinical judgments and bear and remember the consequences (professional and personal) of judgments made. ``Tacit knowledge'' is a name that has been given to the highly personal but sorely reality-tested knowledge that physicians acquire, store, and rely on without acknowledging it as such [3]. The correspondingly unspecified capacity to use this knowledge has been compared to getting a joke or grasping a gestalt [4]. That physicians stray from formal decision models is evinced by the prominent role of storytelling in medical communication [5].
No matter how much physicians know about practice, they cannot be certain of what to do in an individual case. Neither realism nor probabilism predicts specific outcomes. Rather the practicing physician must ``reparticularize'' [6] what is prototypical or average to reflect the circumstances of the patient at hand. The enduring divide between knowledge of the aggregate and care of the individual is largely overlooked in the enthusiasm for practice guidelines based on outcomes research. In fact, the impetus for today's outcomes movement was an indictment of uncertainty by small-area variations researchers who found groups of physicians treating the same conditions differently. Physician uncertainty came to define the problem, and the hard data of statistical studies—the search for evidence [7]—became the solution. Yet statistical analysis can only yield probabilities, and certainty of what is probable may or may not be preferable to uncertainty of what is true. Your technical prowess is legendary, but as far as I know, you still operate on one patient at a time. In that case, you will always be playing the odds; the bridge from medical knowledge to your next patient will still be yours to build.
Accepting for the moment my sketch of a variegated medical knowledge, how shall we know about high-quality practice, that is, about quality? Most of today's speakers have referred to quality as if it were a given, as if we all know what it is or perhaps as if it is what one wants it to be, and in a sense, they would be right either way. At some level, we all know that quality is a matter of doing what is best, of knowing what you are doing, of devoting necessary resources, and of getting good results. At the same time, one means one of these more than the others, and defines them more or less broadly, or emphasizes technical sophistication, or compassionate care-giving, or increased life span, or enhanced functionality, or patient self-determination, and so on. In other words, none of you want to leave the sponge in, but one treated patient's invaluable reprieve is another patient's unwanted prolongation of illness.
If we know about medical practice in many different ways, surely we know about medical quality in more than just one. I would guess you have ways of judging quality in your own work. Some of you participate in peer review or the training of more junior surgeons; surely you discriminate higher quality and lower quality performance there. Patients make similar, but not identical, distinctions from their point of view, and even managers—at least before the outcomes movement—considered the quality of inputs and processes as well as outcomes.
I do not mean to belabor the point, but quality in medical practice is much larger than any one operationalization for regulatory or research purposes. This is not to say that quality need not be operationalized; it is only to put quality as a buzzword, as a research variable, as the legitimizer of organizational change in perspective. Quality as it has been used here today, for example, is surely important to the future of thoracic surgery. However, it is not synonymous with clinical quality, and to the extent that it claims to be, it tips its political hand.
In the context of current changes in the healthcare system, quality, like efficiency, is a term borrowed from industrial engineering. It refers to some subset of outcomes that is generally considered to be desirable and, perhaps more importantly, standardizable, measurable, and statistically analyzable. The ``quality'' of quality measurement and performance contracting, of outcomes research and practice guidelines is defined to be useful to those defining it. The nuances of this quality matter less than its impact, that is, its power to move medical practice in specific directions.
This first distinction, then, among knowledge about practice, knowledge about quality, and outcomes research is meant to help you think critically about the new standards to which you are being held. Although outcomes analysis can be useful and can tell you things you do not know but would like to know, it is not all there is to know, nor is it superior to other ways of knowing. Do not be cowed by the apparent sophistication of its practitioners.
The second distinction is between outcomes research and the outcomes movement in healthcare policy and management. If outcomes research is a category of statistical analysis, analysis of large databases to discover the probability of certain results under certain circumstances, the ``outcomes movement'' [8] is the public assertion of the pressing need for and basic superiority of this kind of analysis in healthcare. The well-funded Agency for Health Care Policy and Research is the federal agency charged with funding and disseminating outcomes research and practice guidelines. The Agency for Health Care Policy and Research is indeed an important locus of outcomes activity, but it symbolizes a more generalized ascendancy of medical probabilism as a policy and management tool. A recent review of the major, and quite divergent, proposals for healthcare reform last year found only one point of unanimous agreement: an expanded role for outcomes research [9]. In the meantime, managed care organizations, insurance companies, and all manner of utilization reviewers are making extensive use of outcomes analysis to judge performance and to shape it.
The popularity of outcomes research is representative of a recurrent irony in American healthcare. As other political scientists have pointed out [10, 11], United States health policymaking consistently rejects openly political solutions to perceived healthcare problems, choosing instead ostensibly technical responses, such as diagnosis-related groups or resource-based relative value scales. Furthermore, policymakers favor behavioral regulation, for example, utilization review, over overtly economic regulation, especially when the former can be delegated to professional peers or private-sector payers rather than a governmental entity. However, the American aversion to political struggles about economic desserts has empowered bureaucratic entities within and outside government. Utilization review, for example, is based on case-by-case assessment of medical activity; it requires levels of reviewers working from highly specialized criteria according to organizationally sanctioned procedures. In short, it requires a bureaucracy.
The outcomes movement—the coalescence around outcomes research as the gold standard—resembles earlier efforts to resolve the healthcare crisis through technically sophisticated behavioral regulation. Advocates reason that if outcomes research can determine what works and what does not, then enforcement of research-based practice protocols can only mean better medical care. Characteristically, behavioral regulation follows directly from technical expertise and requires new structures for producing knowledge and for enforcing its use. The outcomes movement is a bureaucratizing one, even when government is entirely absent. As I expect you know, private payers also practice utilization review, and whether micromanagement serves public policy or investor income, the impact on the physician is the same. As thoracic surgeons, you may be accustomed to expertise as the precondition for power, but in large organizations expertise resides in bureaucracy and not in professionalism or in politics.
This is not a public-sector–private-sector issue. This is not a Democratic-Republican issue. This is not about Bill Clinton, except perhaps in that he systematized his intended use of outcomes research. This is about our willingness to resolve the healthcare crisis and, beyond that, whether we can own up to the political, that is, the distributive nature of healthcare in advanced industrial society. It may surprise you to know that in Western Europe and Canada, where there are national healthcare systems with global budgets and negotiated fee schedules, practicing physicians face far fewer behavioral regulations and enjoy much greater clinical autonomy [12]. They may have fewer surgical suites, and they may not earn as much, but they do not clear their operations with insurance company nurses either.
So far, I have tried to distinguish medical knowledge from outcomes research and outcomes research from the outcomes movement. Finally, I want to distinguish outcomes-based reform from other reform possibilities and to urge you to reconsider your long-standing allies and allegiances.
In political terms, the outcomes movement as a reformist impulse disempowers some kinds of physicians, practicing physicians generally, and some kinds of patients, lots of patients, especially high-risk and socially marginal patients. You have been exhorted today to stay on top of your data, to gather and analyze and present your data, and in general, this is not a bad idea. But I do not believe you will ever beat the data analysts—private sector or public sector—at their own game. You will never have more nor better data, you will never have more market clout, especially than the large investor-owned companies that are busy buying up practices and hospitals, thereby getting bigger databases with more statistical power and more market clout. You should pay attention to the data game you are playing, but you should not imagine you can win it in the long run—not all of you anyway, and not those of you who are physicians first. Instead, I urge you to consider other ways to reform the healthcare system, or if you are not a reformer, other ways to participate in this undeniably tumultuous era.
As a start, I would have you rechart your atlases of the current healthcare system so the public and private sectors overlap in the bureaucratic sphere. I would have you recognize that nowhere are the forces of free enterprise simply arrayed against the forces of government regulation. The regulation of medical care is everywhere—even in the licenses you treasure—and ironically only regulation can keep enterprise free. The important distinctions are not these. The important distinctions are among kinds of regulation: of what, by whom, to protect what, at what cost; and kinds of freedom: to earn, to decide, to change. The practicing physician is no safer in the hands of industry than in the hands of government. I would even argue that he or she is less so because industry works for its shareholders, whereas ideally government works for everyone and even in practice it attends to organized professional and patient constituencies.
Only government, as one powerful buyer, can resolve the healthcare crisis and preserve your clinical autonomy. You may not get much richer much faster, but frankly, economic autonomy is fast becoming the least of your worries. Allow me to leave you with a few recommendations. First, think carefully about the extent to which what you do is captured in the outcome measures you agree to. Be sure you know what they mean and how they relate to other things you are certain you know. Then, make clear distinctions between outcomes information that is helpful, that enlightens you, patients, and policymakers, and information for information's sake—that overwhelming and unthinking enthusiasm for more information and regulation in information's clothing. Finally, look again at healthcare reform. Decide what is more or less intrusive in more or less important ways. If what you know is less important than what they know about you, you have suffered an intrusion at the heart of clinical medicine, and in spite of much of what I heard today, I believe that this is what matters to you most.
Footnotes
Presented at Preparing Your Practice for Change: Thoracic Surgery Into the Next Decade, Atlanta, GA, Sep 24–25, 1994.
Address reprint requests to Dr Tanenbaum, 246 SAMP, 1583 Perry St, Columbus, OH 43210–1234.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
