Despite the fact that there are now more physicians per capita in the United States than there have been for at least 50 years, the Council on Graduate Medical Education (COGME) recently predicted a 10% shortfall of physicians by 2020. Public concern about access to care, reports of difficulties in recruiting physicians in many specialties, and discussion of the looming collapse of primary care all contribute to the sense of crisis. The Association of American Medical Colleges has responded with calls for a 30% expansion of U.S. medical schools and a lifting of the current cap on Medicare funding for graduate medical education so that federal dollars can support the expansion of the workforce.
Physician supply varies dramatically by region of the country; COGME is concerned about a 10% shortfall at a time when the regional supply of physicians varies by more than 200%. And the presence of more physicians doesn't translate into better care. Medicare beneficiaries' satisfaction with their care and perceptions of access are no better in high-supply regions than in low-supply regions. Nor does more physicians generally mean better care for hospitalized patients. Physicians in high-supply regions are more likely to report concerns about inadequate continuity of care, inadequate communication among physicians, and greater difficulty providing high quality care. And certainly most important, patient outcomes are not better in regions with a very large supply of physicians. Having more physicians does, however, mean more spending on health care -- a strong correlation that should not be surprising. Physicians' incomes are an important component of medical spending, and physicians order most clinical services. Taken together, these analyses contradict the notion that health care systems have inflexible physician requirements and call into question the significance of a 10% national "shortfall." They should also lead us to question the diagnosis of a crisis in the physician workforce.
Increasing the number of physicians will make our health care system worse, not better. First, unfettered growth is likely to exacerbate regional inequities in supply and spending; our research has shown that physicians generally do not choose to practice where the need is greatest. Second, expansion of graduate medical education would most likely further undermine primary care and reinforce trends toward a fragmented, specialist-oriented health care system. Current reimbursement systems strongly favor procedure-oriented specialties, and training programs would almost certainly respond to these incentives. Third, workforce expansion will be expensive. If outcomes and patients' perception of access improved as supply increased, then we could debate whether an expansion of training offers better value than investments in preventive care, disease management, or broader insurance coverage, which have known benefits. Instead, the costs of expansion will limit the resources available for necessary reform efforts without any evidence-based promise of a benefit.