10
Issues and Insights Regarding Research, Education, and Training



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 189
--> 10 Issues and Insights Regarding Research, Education, and Training

OCR for page 189
This page in the original is blank.

OCR for page 189
--> Introduction Deborah A. Freund Medical schools and the hospitals in which predoctoral and postdoctoral physicians train are important institutions today, and they will continue to be in the future. However the financial support that they have to train physicians and to undertake research, both basic and clinical, will continue to erode. Traditional sources of support for internally funded training and research include the teaching pass-through of diagnosis-related group (DRGs) and Medicaid payments, fellowship and research grants from the public and private sectors, and reimbursements from private insurers that are higher than marginal costs. Together, these streams of funding have allowed training programs to offer treatments that ''do not pay for themselves'' to all in need. Such monies historically have provided plentiful opportunities for physicians at various stages to observe the most cutting-edge therapies, do research on them, or refine them. Meenan discusses how the new managed care environment may affect the training of physicians, opportunities for research, and the potential impact of these effects on outcomes. He contends that when all of these sources are cut, it is questionable if the number of physicians who can train and the opportunities to practice with expensive therapies can be maintained at current levels. It also is unlikely that managed care and integrated care entities, whether they eventually affiliate with teaching hospitals or not, will continue to provide the revenue now given in other forms to make cross-subsidies for care provision and research possible. Meenan contends that dollars made available by managed care are likely to come with the contingency that the focus be on cost-efficient practices rather than those simply improve or refine current techniques but do not improve patient outcomes.

OCR for page 189
--> Meenan foresees a world in which medical education and the research performed along with it occur in very different configurations than they do today. He contends that it is unclear whether our teaching hospitals as configured today can respond quickly enough to supply managed care entities with the appropriate mix of physicians (e.g., more generalists and fewer specialists) having the necessary clinical competencies in cost-effective medical practices. If not, managed care organizations may start their own training programs alone or in concert with affiliated hospitals. For example, Kaiser has had its own residencies for a long time. Teaching hospitals that are unable to respond may shrink in size or close. Medical schools may decide to affiliate with managed care to form vertically integrated health systems and use managed care practices as training sites for students and residents, or as pointed out by Meenan, they may elect to be free-standing educational enterprises without their complicated missions of today. In his mind the ultimate questions then boil down to the following: (1) What is the evidence that the questions tackled by researchers and the medicine practiced in the future under managed care will actually result in less favorable outcomes for the population? Why should we presume that the current fee-for-service system really is better than or preferable to the alternative?, (2) How will medical schools and other training institutions actually be configured to do research and take care of patients when the "dust settles"?

OCR for page 189
--> Invited Address Robert F. Meenan Uwe Reinhardt has pointed out that what is generally referred to as "managed care" actually consists of three relatively distinct elements of health care system change. The first, managed care, involves the oversight and proctoring of medical decisions. Such oversight is not new: it has been prevalent for decades in the form of utilization review and related approaches. Practice guidelines and practice profiles are simply new approaches to the management of care. Reinhardt's second element of health care system change is financing, which increasingly emphasizes capitation and risk bearing. These approaches essentially reverse the financial incentives in health care from over utilization to underutilization. The third element of health care system change is market competition based on price and coverage. This competition is increasingly characterized by classic market behaviors. It is not in any real sense "managed competition" because the competition is not seriously constrained by government regulation. Most of the impacts of health care system change on medical care process and outcome for people with chronic rheumatic diseases will result from two of Reinhardt's factors: management of care and risk-based financing. The major impacts of change on medical research, education, and training, however, will derive primarily from market competition and its direct and indirect effects on academic medical organizations. Three key factors will be especially critical in determining the future status of medical research, education, and training: (1) the new form of the academic medical organization; (2) the underlying philosophy driving government policies; and (3) the market power of managed care organizations. Competition in health care will play the primary role in shaping each of these factors.

OCR for page 189
--> The prevailing form of today's academic medical organization is the academic medical center, essentially a medical school linked to a teaching hospital. This form will undoubtedly change in the face of the dual pressures of declining clinical incomes and stagnant research funding. In the future, some medical schools will break their close links with clinical service organizations and will come to look more like business schools and law schools. Other will become components of large, integrated health care delivery systems. Although the second model is appealing to many in academic medicine, it is not clear that such an organization can be successful in the increasingly competitive environment of health care. It is also unclear to what extent such an entity would actually pursue a classical academic mission of research, education and teaching. On the other hand, if the "medical school as medical school" model becomes the prevailing form of academic medical organization, it is not clear how medical research, education, and training will be supported. Government policies of various types have played a major role in the growth of medical education, research, and training over the past five decades. These policies include Hill-Burton funds for postwar hospital construction in the 1950s, Medicare support for graduate medical education beginning in the 1960s, capitation support for increased medical school class sizes in the 1970s, and consistent support for medical research funded by the National Institutes of Health (NIH) over four decades. In the 1990s and beyond, government policy will continue to play a critical role in determining the nature and magnitude of the impacts of health care system change on medical education, research, and training. If the traditional view of these activities as public goods prevails, government policies may substantially blunt the impact of health system change. If the newer view prevails—that these activities, particularly education and training, should and can be determined by the market—then government policies (or the lack of them) may exacerbate the detrimental effects of health system change. Finally, medical research, education, and training will be affected in major ways by the growing market power of managed care organizations in an increasingly competitive health care system. As these entities become major customers for academic medical organizations and as they become progressively larger corporations with growing influence on a range of public and private policies, they will increasingly define their own agendas for research, education, and training. As these large corporations utilize capital markets to finance their growth, they will have little or no interest in subsidizing the traditional activities of academic medical organizations, because they will derive no clear stockholder benefit from doing so.

OCR for page 189
--> Research Different types of medical research will be affected differently by health system change, but there is no reason to believe that research into rheumatic diseases will be affected differently from medical research in general. Basic biomedical research in rheumatology and in all areas of medicine will shrink. Medical schools currently underwrite basic research with cross-subsidies from clinical income. As the clinical incomes of academic medical organizations continue to decrease, this cross-subsidization will be markedly attenuated, if not eliminated. Stagnating federal support for basic research, decreases in grant overheads to academic organizations, and limits on the percentage of a faculty member's salary that can be charged to research grants will all reinforce the shrinking of biomedical research. There is little likelihood that managed care organizations will fund basic research because it produces no short-or intermediate-term payoffs for them. Although a convincing argument can be made that basic biomedical research is a public good, it is unlikely that government support for such research can be substantially expanded in the face of increasing pressures to cut government spending. Academic medical organizations will increasingly turn to private industry, particularly the pharmaceutical and biotechnology industries, as sources of support for biomedical research. Such support will focus more on applied than on basic research, and it will increasingly raise issues about the appropriate boundaries and relationships between academic and business organizations. Regardless of the new models that might develop, it seems inevitable that total support for basic biomedical research will decline and the number of faculty involved in this activity will decrease. This will necessarily delay the discovery of markedly better treatments for rheumatoid arthritis and systemic lupus erythematosus, but it is impossible to predict if this postponement will be a particularly lengthy one. The effects of health system change on clinical research are difficult to predict. On the negative side, there are a number of reasons why support for clinical research is apt to decrease in an increasingly competitive health care system. Clinical incomes that previously supported such research will continue to fall, and this will be accelerated by the ongoing shift of public health care plans (Medicare and Medicaid) to a managed care approach. The pressures to maintain clinical incomes will also decrease the amount of time that clinical investigators can devote to research and to the training of future clinical investigators. On the positive side, managed care organizations have the potential to be excellent sites for clinical research because of their large size and their increasingly sophisticated data systems. Certain forms of clinical research may also produce short-term benefits for managed care organizations in terms of better patient care. Academic medical organizations will also turn to business corporations for

OCR for page 189
--> the support of clinical research. The obvious targets for such efforts will be drug and device manufacturers. Once again, these companies will emphasize very applied, product-oriented clinical research. To the extent that they do support clinical research, they will do so only in academic organizations that possess certain key attributes. These will include large patient populations that support both the conduct of clinical research and the eventual marketing of approved products; efficient research processes that facilitate the rapid approval and conduct of clinical research; and sophisticated data systems that identify potential subjects and permit the efficient collection of high-quality data. Since academic medical organizations possess these attributes in varying degrees, it is likely that clinical research will grow substantially in some academic medical organizations while declining dramatically in others. Support for health care research may actually increase under managed care. Managed care organizations have fairly direct incentives to carry out health care research that examines the cost and effectiveness of alternative approaches to medical care. Furthermore, managed care organizations can increase their competitiveness by supporting research into cost-effectiveness, clinical pathways, outcomes measurement, and other aspects of health care. The results of such studies may well provide proprietary advantages to the sponsoring company. Unfortunately, this point has been used as a primary argument for decreasing government support for health care research through agencies such as the Agency for Health Care Policy and Research (AHCPR). In fact, the health care research needs of managed care organizations are so substantial that alternative arrangements are developing to support such studies. These include major in-house units in larger managed care organizations and numerous private sector companies that are building their health care research capacities. Both of these developments represent direct competition with academic medical organizations in the search for health care research dollars. Once again, patient populations, efficient research operations and information technology capacity will determine which academic medical organizations are able to compete for health care research dollars. Education Medical education, as distinct from medical training, refers to the preparation of medical students that culminates in the awarding of an M.D. degree. Medical school education is generally a four-year program, with the first two years focused on classroom-based, didactic education and the second two years emphasizing clinic-based practical training. This general approach to medical education dates from the Flexner report, issued early in this century, which transformed medical education from an apprenticeship model to an academic model. Since the 1960s, the didactic components of medical

OCR for page 189
--> education have increasingly been taught by basic science faculty, while the clinical elements have increasingly been taught by specialist physicians in hospital-based settings. Health system change has profound implications for medical education. Once again, the most immediate and obvious impact of such change will be a progressive, competition-driven decrease in clinical income to medical schools and their clinical faculties. A combination of competition from nonacademic medical care providers and decreases in reimbursement levels from payers will cost academic medical organizations billions of dollars per year in clinical income. These losses will soon produce substantial decreases in the number of clinical faculty as well as reductions in the time that remaining faculty can devote to the teaching of medical students. There is no reason to expect that either managed care organizations or government will take steps to mitigate these financial strains on medical education. In fact, the basic premise in government policy toward medicine and other professions is that the choice to pursue professional education should be made by each individual based on classic market considerations of cost and eventual income. Government, accrediting bodies, and managed care organizations all have an interest in the content of medical education. Each is interested in making medical education more relevant to the practice of medicine in the evolving health care system, making the content of medical education more relevant to the evolving demographic and disease burdens of society, making medical education more oriented toward preventive care, and making medical education more balanced in its attentions to population health and individual patient health. Changes in these directions would all represent positive steps for medical education in general and for rheumatology education in particular. In fact, it is arguably more important to change the content of medical education than it is to change the number of medical students. However, it remains to be seen whether medical schools can respond effectively to the pressures of the new medical care market by making medical education more outpatient oriented, more chronic disease oriented, more cost control oriented, and more prevention oriented. Training Training refers to the post graduate education of physicians. In the current system, the vast majority of medical school graduates pursue clinical training for three or more years in order to become medical specialists or subspecialists. In rheumatology, the training program is five years long, requiring three years of training in internal medicine followed by two years of training in rheumatology. For most trainees, the two years of rheumatology training are a combination of clinical and research experiences. The modest

OCR for page 189
--> salaries of rheumatology trainees have traditionally been supported by a mix of training grants, research grants, and clinical income. Health system change has already had substantial impact on rheumatology training. Once again, the impact has been driven primarily by financial factors. The clinical incomes of rheumatology divisions have decreased in line with the overall decrease in academic clinical incomes. Tighter research budgets do not allow for support of trainees. Funds for training grants have been cut. Medical care competition also has decreased the incomes of practicing rheumatologists, which in turn has decreased the economic returns to rheumatology training. As a result, the number of applicants for rheumatology training has dropped steadily and markedly over the past decade. This decrease in the number of rheumatology trainees will not necessarily have adverse effects on the health care outcomes of people with rheumatic diseases. Managed care organizations increasingly strive to have specialists serve only as specialists. In this model, rheumatologists are utilized primarily for the care of complex rheumatic disorders such as rheumatoid arthritis and systemic lupus erythematosus. Substantially fewer rheumatologists are needed in this model than in the old fee-for-service model. Thus, a reduction in the number of rheumatology trainees and practitioners need not produce access or outcome problems for the most pertinent patients. Like medical education, the content of medical training should change to reflect the changing needs of the competitive health care system. Rheumatology training should increasingly emphasize clinical skills and the care of complex rheumatic disorders. Given the realities of a competitive health care system, the number of training slots should be decreased to alleviate an impending oversupply of rheumatologists. It remains to be seen, however, whether rheumatology training programs will have the foresight and the will to make such changes. Managed care organizations have an interest in well-trained physicians. To the extent that academic medical organizations do not change their approaches to training, it is possible that managed care organizations will set up their own training programs, particularly for generalist physicians. It is unlikely, however, that managed care organizations will ever directly support rheumatology training. Rheumatologists will continue to be trained in the academic medical organization of the future, but the organization and the content of training will both be substantially different. Conclusion The changes occurring in the health care system are profound, and they will be permanent. Health care competition represents a new era in American medicine, not just a minor adjustment in the old system. As the new,

OCR for page 189
--> competition-driven system takes form, approaches to research, education, and training will arise that are most appropriate for the new system. The question is not how current approaches to research, education, and training can be supported under the new system. They cannot and should not be. The onus is on the old system to change and adapt, not on the new system to preserve.

OCR for page 189
This page in the original is blank.

OCR for page 189
--> Invited Reaction Robert Mechanic We all agree that markets are changing. The question many are focusing on is whether academic medical centers will prosper or even survive in this new environment and, if so, what that means for research and education. I will give you a few numbers with which to gauge the magnitude of what is going on right now and help separate some of the rhetoric from the current reality. Then I will talk a bit about how managed care plans and academic medical centers might come together in productive ways. Federal grants for research have been growing at about 9 percent a year over the past decade. The NIH received an increase of about 6 percent for FY 1996. Although the rate of growth in federal grants may slow down, clinical cross-subsidy of research is the funding source that is most threatened by the changing market environment. If you look at medical schools, about half the funding today is clinical revenue. Government policy is important, perhaps more important than what is going on in the private market. Here are some of the things that we need to think about. First is the rate of growth in federal research funding. Second is consideration of an all-payer indirect medical education (IME) and direct medical education (DME) funding pool for teaching hospitals. Even though you hear less about this concept in Washington, many states, driven by either their medical centers, their universities, or their state legislators, are looking into dedicated medical education funding pools, although to my knowledge no legislation has actually been passed. Medicare payment is still very important. For the 225 major teaching hospitals, about one-third of their DRG payment is based on indirect medical education and disproportionate share hospital (DSH) payments. In the most

OCR for page 189
--> recent year for which data are available, 1993, major teaching hospitals did well under Medicare, with an average margin of 11.7 percent compared with the average for all hospitals of 0.3 percent. The congressional budget resolution that was passed but vetoed early in 1996 proposed a 35 percent cut in IME and DSH payments. A cut of that size would have reduced Medicare payments to major teaching hospitals by about $1.3 billion and would have lowered the average Medicare margin from 11.7 percent to less than 2 percent. In terms of market impacts, two things are happening. Over the last 10 years, there has been a reduction in inpatient utilization of about 30 percent. A number of scholars, including The Lewin Group, have done projections for the year 2000, and the number that seems to fall out—a midrange estimate—is a further decline in inpatient utilization of about 35 percent. That could have a profound impact on academic health centers (AHCs). The Association of American Medical Colleges (AAMC) shows AHC total margins in the 4 percent range for 1991. This is higher than the historical average, so it is difficult to say that AHCs are in dire financial straits right now. On the other hand, I think there will be a great deal of variation in AHC financial status on a market-by-market basis, and I agree with Meenan's suggestion that it will be the ''marketers'' that do well in this new market. Being a "seller" will be a losing proposition. Now I want to very quickly talk about a relevant study we just completed. We did seven site visits to assess the question of how managed care is affecting clinical research. Two hypotheses were investigated. One was that managed care plans are denying more research-related care. The second was that market change is affecting the overall ability of institutions to attract patients and fund research. We found, in general, that managed care was not denying more research-related care than fee-for-service systems, but there was a lot of pressure due to aggressive price negotiations in competitive markets. Many plans also try to direct people into low-cost provider networks. These general financial pressures affect the level of support available for faculty. What managed care plans want is efficient, patient-friendly care; a work force that can practice in a managed care setting; and research results that help them produce better outcomes at lower cost for their enrollee populations. If academic health centers are going to stay in the forefront of research as power shifts to managed care organizations, they will have to devote greater attention to those goals.

OCR for page 189
--> Discussion WILLIAM HAZZARD: What about international medical graduates and how they perhaps confound or complicate the issue of a rational plan for graduate medical education? ROBERT MEENAN: They do complicate things. This is one of those balloon-like problems where you press on one end to reduce it and it simply pops out someplace else. International medical graduates not only raise issues about rational planning for graduate medical education, but also raise central questions about the purpose of medical schools in the evolving health care system. If you utilize the market model for the evolving health care system, you might ask who the customers of medical schools are. One of the products they produce is medical students. What attributes should those medical students have? Obviously, it depends on the customer. If the customer is a hospital-oriented residency training program, medical schools should produce students with certain skills. If the customer is a managed care organization, the content of medical school education ought to be different. If graduates actually went from medical schools to their ultimate places of employment, the content of medical school education probably would be very different. Jerry Barondess has talked about the difficulty of teaching clinical skills when students are overwhelmed with the mass of biomedical knowledge they are being fed. I think everyone recognizes that there must be changes in terms of where medical students are taught, who teaches them, and what they are taught. We currently have a specialist-driven clinical education system and a basic scientist-driven first two years. Neither one is an appropriate way to

OCR for page 189
--> teach medical students what they need to know to be effective care providers in the evolving health care system. Rheumatology training is not unique in its dependence on less-than-appropriate teachers, settings, and content. One issue that must be addressed in any restructuring of medical education and training is the consistent instillation of a mind-set that physicians should maximize total benefit without regard for marginal benefit and the cost of the next increment of benefit. A key point I make when I talk to medical students about clinical decision making is that one can look at medical practice as a giant two-by-two table, in which "positive/positive" and "negative/negative" are the preferred boxes to be in. When it comes to choosing between the other two boxes, however, every medical student and trainee is taught to avoid the "disease present/no action" box, which represents an error of omission. Students and trainees are taught that it is alright to do many, many tests in which nothing is found, but it is a terrible thing to miss a diagnosis when every possible test was not done. Teaching students and trainees that it is acceptable to look at the issue of marginal benefits and the cost of marginal benefits is relevant because, as health care continues to evolve, we are likely to lose some ground in terms of the intensity of evaluation and therapy we can provide in our medical care system. We will undoubtedly move back on Dr. Tarlov's S-shaped curve. It drives most physicians insane to contemplate that: to think that they might be giving up some benefit at the margin by not using every test or therapy. As we train physicians for the future, we must make them understand and become more comfortable with the issues of marginal benefit and marginal cost. This will help them function in an environment in which we may not be able to give a hip replacement to that next person in line or train that next rheumatology specialist. MATTHEW LIANG: Let me introduce the notion of a type III error, an extension of types I and II, which involves giving the right answer to the wrong question. As one who has tried to do economic analyses of health practices, I realize that assigning a dollar value sometimes is a excuse for not assigning a moral value, which is much more difficult to do and may be more contentious. I also realize that in many cases we are trying to sweep out the sea in attempting to resist change, but some of the changes that we are trying to resist concern things of real value, public health concerns. How do we introduce these issues into the public debate and decision making, when the entire debate is centered around dollars and assignment of dollars? ROBERT MEENAN: I think we must accept that this is really a paradigm shift that will be fundamental and long-lasting. We must recognize that the dynamics and values of the market will determine much of what happens in health care, so we should try to focus public policy on aspects of health care

OCR for page 189
--> that we know will not be handled well in a system dominated by market characteristics. The market has some positive aspects, including driving excess costs and capacity out of the system and forcing us to ask relevant questions about costs and benefits. The challenge will be to build on some of the positive aspects of a market-oriented system—to craft a synthesis between the thesis of traditional care and the antithesis of managed care. If we as clinicians had a drug that we thought had modest positive effects but also had substantial toxicity, would we simply throw it out? Of course we wouldn't. In rheumatology, we faced this issue with the early evidence about methotrexate for rheumatoid arthritis. We handled it the way we handle many new drugs for rheumatic diseases, where treatment options are limited and not very effective. We studied it, tried various dosing regimens, and eventually developed a major new treatment for rheumatoid arthritis. We should do the same thing with managed care: look at it as a treatment for the ills of our health care system that is effective but has toxicity. We need to study it, refine it, and use it to build an effective new approach to a complex problem. Let us never forget that the good old days were not all that good. The approaches that prevailed have led us to the point at which we are paying 14 percent of our gross national product for health care. As economists point out, it is not so much that we are paying 14 percent, but that we are not getting acceptable value for those dollars. If one looks at the results of managed care, it is very clear that there is substantial excess capacity in the clinical care system. If we translate this insight into research, education, and training, we can easily conclude that the good old days have led to excess capacity in those systems as well. The problem is to decide on a new balance point between quantity and quality in our medical education and research systems. One of the key insights in adapting to a market-oriented system is recognizing that we must stop being believers in medical care or sellers of medical care and that we must become marketers of medical care. The believer is one who feels that the work that he or she does should be funded because it is valuable. It is work with intrinsic value (in some sense, God's work) that should be supported. Many, if not most, medical researchers are believers. A seller is someone who decides what people ought to buy. Usually what people ought to by is what the seller is willing to offer, and the whole idea of selling is to convince people that they want what you make. Most professionals and organizations in health care have been sellers. They decide what tests and treatments the customer should buy, and they determine the time, place, and price at which these services are available. The health care marketer, on the other hand, asks what people out there are looking for. The health care marketer asks what service, approach, or product can I produce that will have enough value for the market to buy it? One positive result of the evolving health care system would be to produce more marketing and less selling and believing.

OCR for page 189
--> I also want to point out that public policy is likely to be far more important in the areas of research, education, and training than it has been or will be in the clinical delivery area. We need to help policy makers recognize those aspects of research, education, and training that will not survive well in a market-oriented system. They are the aspects that have elements of a public good, like a national park, in which group benefits outweigh the benefits to any individual. Public policy must be developed to support these aspects. JEREMIAH BARONDESS: Let me take a minute to tell you an illuminating story about paradigm shift. When I was a "puppy" in medicine, the state-of-the-art treatment for acute myocardial infarction was strict bed rest. People were fed. They were not even permitted to turn over in bed by themselves; two people came and turned them. They lay in bed until their sedimentation rate was normal, this being the index that the inflammatory process had subsided. The treatment took six weeks on average. I can't tell you with what awe we approached those extraordinarily fragile people. They were literally not allowed to cut their own meat. When the insurers said they would no longer pay for care of that duration, the system responded, and hospital stays for acute myocardial infarctions were cut in half fairly rapidly. Like Kuhn said, progress is not made in a logical, sequential order in which each step is based on the revelations of the prior step. It comes in jumps, and sometimes we jump for surprising reasons. BEVRA HAHN: Do you think we should be teaching two separate groups in medical school: one group for efficient outpatient care in a managed care setting and the other for intensive inpatient care, so that the physician groups are different, as they are in Britain? ROBERT MEENAN: I don't know, Bevra. That is one step further than I would be prepared to go at this point. I think there are some basic steps that we need to take before we consider a two-track system. If we look at the current system with its basic science-driven first two years and its clinical specialist-controlled second two years, there are many aspects of it that we could change in terms of content, teachers, and learning sites. Such changes would go a long way toward producing medical school graduates who are prepared more appropriately for work in the new health care system. I would be concerned that the approach you suggest would spark debate and resistance that would actually make it more difficult to implement these other, more general changes in medical education. JEREMIAH BARONDESS: I would add one codicil to that, in relation to medical education and especially residency training. I think that complex challenges in differential diagnosis and triage arise with every complaint

OCR for page 189
--> presented by patients. We should be very careful, as we train the next generations of generalists, to be sure of two things. The first is that they should be trained rigorously for the complicated real-world challenges presented by ambulatory patients. The second is that generalists should expect of themselves a capacity to handle clinical problems of more than minimal levels of complexity. Whether we can effectively train people to do this by concentrating heavily on ambulatory care training environments is a serious question. Implicit in the idea of training people oriented around lesser clinical challenges, if you will, is the pressure to continue to train large numbers of subspecialists to whom generalists can refer people with more complex disorders. The cost implications of excessive numbers of subspecialists have been commented on very widely. The impact on the quality of care—especially that deriving from fragmentation and the resulting lack of coherence—has not achieved as high a profile.

OCR for page 189
This page in the original is blank.