The Nation's Physician Workforce: Options for Balancing Supply and Requirements (1996)

The size and composition of the physician workforce has been an intermittent policy issue in the United States. A century ago, the policy debate focused on the proliferating graduates of proprietary medical schools with dubious staff, facilities, and curricula. Experts of the time concluded that the nation had too many poorly trained practitioners. Following the publication of the Flexner report (1910), many schools closed and the quality of the remaining institutions improved. Thirty years ago, resources for health care were expanding as private health insurance continued to grow and the government extended access to the elderly and the poor through Medicare and Medicaid. The nation faced a potential physician shortage, and policymakers took steps to increase the physician supply.

As the twenty-first century draws near, physician supply once again has begun to trouble policymakers. The context, however, is quite different. The health care sector of this nation is undergoing radical, rapid, and unpredictable transformation. With the collapse of efforts between 1992 and 1994 to enact comprehensive health care reform, much of this restructuring is taking place through changes in the private sector, with as-yet-unforeseen consequences. What can be anticipated is a continued debate over how many and what mix of health care personnel the nation needs and judges affordable. More generally, the goals and means of expanding (or protecting) access to health services and improving (or maintaining) health and well-being will, as never before, be scrutinized through the lens of cost control.

In the drive to control health care costs, the role of physicians is a central focus because they have traditionally been viewed as the decisionmakers who

directly or indirectly account for most health care spending. Thus, a central strategy of the managed care plans and integrated health care systems that increasingly dominate health services in the United States is to limit the number of physicians available to health plan members and to manage patient access to physicians, especially specialists. As government officials and employers intensify their efforts to direct members of their respective health insurance programs into so-called managed care plans, questions about the appropriate supply and mix of physicians and other professionals have become increasingly acute for government policymakers, educators, professional organizations, and the public.

Clearly, these questions are complex and deserve careful, rigorous study. Notwithstanding the desirability of a more detailed and prolonged examination, the Institute of Medicine (IOM) concluded that a brief review of data on physician supply and requirements and an examination of options for dealing with the physician workforce would be a valuable—and timely—contribution in the current policy environment. (In this report, the term physician refers to both allopathic and osteopathic physicians.) The intention is to spark informed debate among all interested parties, with a view to encouraging appropriate regulatory or market-oriented steps (or both) to bring about a closer match between physician supply and the requirement for physician services. In so doing, the committee was mindful of the outlook stated by Pritchett (1910, p. xv) in the introduction to the Flexner report:

That is, this committee, like that of nearly a century ago, sought to be sensitive to the best interests of the nation's young people and cognizant of the public need for expert health care.

The active physician workforce, which numbered 226,000 after World War II (Rivo and Satcher, 1993) and 628,000 in 1992 (Kindig, 1994), has been the subject of intensive, systematic study and debate for nearly 50 years, principally in terms of the match between the expected supply of physicians and the estimated need or demand for physicians (or physician services). Various commissions and other groups have issued a number of reports, often conflicting: some forecast shortfalls between supply and demand; others predict excess

supply. During the past two decades, however, attention to questions of the match or mismatch between the supply of and demand for physician services has escalated dramatically. Some of this notice was prompted by the appreciable growth in the numbers of allopathic and osteopathic schools and their graduating classes beginning in the 1960s—phenomena themselves motivated by concerns that the nation would have a deficit in physician supply overall or in certain geographic areas (outside metropolitan areas or in isolated, rural and inner-city locales) if production were not increased. More recently, the focus on physician supply was intensified by attempts (now ended) to enact health care reform proposals that were intended, among other things, to extend access to health care.

Physician supply is a complex mix of the existing pool of physicians, current and projected enrollments of medical students, current and projected numbers of residents in hospitals, the flow of international medical graduates (IMGs), losses from the profession owing to deaths and retirements, and other factors. Residency training figures prominently in this equation. For instance, although the number of U.S. medical school graduates has remained constant in recent years, the number of physicians in residency training continues to increase. This results largely from the growth in the numbers of IMGs, most of whom are not U.S. born and most of whom remain in the United States to practice after completing their residency training. In the context of more general debate over U.S. immigration policy, the growth of IMGs is particularly controversial.

Other determinants of the supply-demand equation are also important. As a case in point: new schools of osteopathic medicine are in the planning stage, and one has just opened. Moreover, the potential surplus caused by an increased supply could well be magnified by a decreased demand for physician services as a result of the startling growth of managed care plans of various sorts. Ample evidence demonstrates that managed care plans use fewer physicians per unit of population served than the traditional fee-for-service system, although physician-population ratios vary considerably by type of managed care system.

Many other questions have been raised in recent years about the U.S. physician workforce. As a case in point: the geographic location and dispersion patterns of both primary care practitioners and specialists demand considerable attention. Similarly, a perceived deficit of primary care physicians (e.g., those in family practice, general internal medicine, and general pediatrics) has been roundly decried.

Especially in the past few years, the belief has grown that the United States has now, or shortly will have, a surplus of physicians. For example, for the past two decades the supply of ''active" physicians (those in patient care, teaching,

research, and administration) in the United States has grown one and one-half times as fast as the population at large. In 1970, active physicians numbered 151 per 100,000 persons; by 1992 the figure was 245. Of those engaged in direct patient care, the rates per 100,000 population were 109 in 1970 and 180 in 1992.

Several recent government and private sector studies have marshaled evidence to suggest that nothing but physician surpluses lie in the future. Increasing marketplace evidence also supports the conclusion of a physician surplus. Anecdotal information and informal surveys, for instance, report empty office calendars, early physician retirements, decreasing physician incomes, and even physician bankruptcies.

Numbers alone do not make the case that a surplus of physicians is undesirable. The question of whether a significant excess of physicians would have a positive or negative impact has been controversial.

Proponents of measures to restrict the U.S. physician supply have characterized a surplus as being wasteful of both human and financial resources. They stress the problems that young physicians who have devoted many years of their lives to medical education and training will face if they are unable to practice their profession—particularly if they have incurred significant debts in the process (and most do). Such advocates also believe that the billions of dollars the federal government now spends to support physician residency training—through payments from Medicare, Medicaid, and Title VII of the Health Professions Act—are ill-directed if that funding induces an even greater physician surplus; this is especially so at a time when the Medicare and Medicaid budgets for providing health care to vulnerable populations are under severe pressure. They also are convinced that a surplus could lead to a physician population that is less experienced and less competent because its members have less work to do in the areas in which they were trained. Others argue that large and growing numbers of physicians will increase health care expenditures overall and will not solve some of the nation's most pressing problems of access to care and geographic maldistribution of health care resources and services.

On the other side are those who argue that the physician supply should be controlled by the marketplace alone. These experts contend that a so-called physician surplus could lower health care costs via competition and that specialists might diffuse into areas in which they are now in short supply as a function of decreased demand in areas where they abound. They also see potential for improved quality of care, because managed care plans are able to take on only the best practitioners. Yet others take a longer-range view that the marketplace generally will tend toward equilibrium as regards the supply of newly trained physicians.

Much of this information converged in the mid-1990s to a view that some steps were needed to constrain overall supply, to redress the balance between primary and subspecialty care, and to consider the intended and unintended consequences of existing financial support for residency training. Thus, some

proposals for health care reform took these issues into consideration. Indeed, many believed that comprehensive health care reform would bring forth mechanisms by which the nation as a whole could address these questions in a systematic way.

In view of the failure of comprehensive federal health care reform, however, it appears likely that no coordinated nationwide mechanism is available with which to even examine these issues thoroughly, let alone take steps to remedy problems that are identified. Moreover, the number of physicians in residency training is increasing each year, and little evidence is at hand to suggest that the domestic production of physicians will drop any time soon. Many experts take the position that the physician supply will continue to increase at a rate faster than growth of the general population—with unknown but potentially negative repercussions for individual health professionals, the health care system, and the nation as a whole—unless new steps are taken by the public sector, the private sector, or both.

Concerned about the lack of a clear locus for informed debate ALDJL and decisionmaking, the IOM, through its Board on Health Care Services, decided to initiate a short but substantive review of existing data about the U.S. physician supply, to identify positive and negative implications of the possible mismatch between supply and requirements in coming years, and to lay out possible options for addressing any perceived problems. In May and June 1995, an expert committee co-chaired by Don E. Detmer, M.D., of the University of Virginia and Neal A. Vanselow, M.D., of Tulane University reviewed key publications, information, and analyses. It met in July 1995 to discuss and formulate its report, which was then submitted for review in accordance with IOM and National Research Council report review policies. The document was revised again based on that external review, and this monograph constitutes the committee's final report.

The primary audiences for this work are public and private sector policymakers. They include leaders of societies and associations representing both physicians and other health care professionals; medical educators at the undergraduate, graduate, and postgraduate levels; administrators and directors of all types of health care organizations; officials in the federal and state governments; and the public at large.

Cutting across the demand and supply questions are issues relating to the distribution of physicians, both by geography or practice location (e.g., rural or urban areas) and by specialty (e.g., primary or generalist care or subspecialty care). Although questions of distribution by specialty and locale are very significant, the committee had neither the time nor the resources to explore those matters adequately; thus, this report makes no attempt to do so. In addition, the committee recognizes that significant issues remain about the representation of minorities in the U.S. physician community, but it calls attention to another recent IOM report on this matter (Balancing the Scales of Opportunity: Ensuring Racial and Ethnic Diversity in the Health Professions, 1994) in preference to addressing the matter itself. The IOM is at present conducting a large-scale study on the future of primary care; it has begun a project on the applications of telemedicine; and a study on managed care and rural communities is planned. All of these projects can be expected to shed light on some of the topics not addressed in this report.

A fundamental tenet of this committee was that the nation can never achieve a perfect, steady-state match of physician supply and requirements, because different levels of supply will have different implications depending on the underlying structure of the U.S. health care system. For that reason, the committee accepted the proposition that a modest "oversupply" of physicians is not a bad thing per se; it concluded that if the nation has inevitably to err in matching supply and requirements, it would do better to err on the side of a surplus. In addition, the committee concluded that the characteristics of the overall system of health care organization, financing, and delivery are very important in determining both supply and requirements; thus, it judged that considerable attention must be directed at developing data and policymaking tools that will facilitate good analysis, flexibility, and timely actions in the future.

The committee ended its deliberations in agreement with three important principles. First, the nation should not tie national workforce policy or graduate medical education to the service delivery needs of selected parts of the health care system. Second, long-term physician workforce policy should be driven by aggregate requirements nationally, and meeting those requirements should be cued more to the output of U.S. allopathic and osteopathic schools than it is today. Third, opportunities in the United States for careers in the healing arts, such as medicine, should be reserved first for graduates of U.S. medical schools.

Chapter 2 of this report provides a brief overview of data about the supply of and requirements for physicians in the United States today. Although oriented chiefly to very recent publications (whose data are current up to about 1992), some information about trends since the 1950s (and especially since the 1970s) is presented. Chapter 2 also notes the complexities of analyzing the need or demand for physician services, as measured against the supply of those services, when intervening variables such as physician productivity are taken into account. The chapter generally concludes, in keeping with much of the contemporary literature, that the nation does have a substantial abundance of physicians now and will for the foreseeable future.

Chapter 3 asks and attempts to answer two questions: (1) What is the likely impact of the current or projected levels of physician supply? (2) Is a possible physician oversupply in the public interest? It explores these topics in terms of the costs of health care, access to that care, and the quality of that care. It also considers questions relating to the efficient use of human resources, trade-offs between opportunities in the medical profession for graduates of U.S. schools as set against opportunities for those coming from foreign schools, and the pressures on the nation's academic health centers.

Chapter 4 lays out the committee's judgments about the strategic options that might be considered for exploiting the possible benefits of very large numbers of physicians in this country and for overcoming the possible harms that may ensure from such an oversupply. It outlines practical steps that might be taken by the private or the public sector, or both, to align better the future physician supply with anticipated requirements, and it presents the committee's five major recommendations.