The recent call by a leader of a major medical professional organization for a "comprehensive study of national health professions workforce needs" harkens to similar forecasts over the past three decades.1 Attempts to predict national medical workforce requirements have been consistently marred by their inability to accurately forecast national requirements for health care providers, which in turn leads many to believe that such activities, however well intentioned, are inevitably doomed to failure.

A Surplus of Physicians?

Most PAs are barely aware of the infamous health workforce study put forth in 1980 by the Graduate Medical Education National Advisory Committee (GMENAC). This attempt at medical workforce prognostication sent shock waves through the medical workforce policy community and the medical professions when it confidently predicted that the United States would experience a surplus of physicians by 1990 and an even greater surplus by 2000.

The impact of the GMENAC report was profound on the PA profession. Many of those who readily accepted the prediction of a physician surplus asked: If we will have too many physicians, why do we need PAs? This was a body blow to the young PA profession, resulting in defections from the profession, a decline in applications and the closure of a number of PA programs.

But the most notable aspect of this federal-level, multimillion dollar effort to predict future health workforce needs was that it was dead wrong. The predicted surplus of physicians never occurred. GMENAC marked the first of several subsequent efforts to predict medical workforce needs that also became notable for their inaccuracy. GMENAC morphed in to the Council on Graduate Medical Education in the 1980s and it, too, became famous for its inability to accurately predict the need for physicians in the health system.

Workforce Needs: Predictably Unpredictable?

Why is this so difficult? For one, it is not easy to predict, in an industry as vast and complex as health care, what the need will be for medical providers, particularly physicians, in the future. Second, the variables are multiple, tough to measure and interacting. At present, any effort to estimate the future need for physicians and other health care professionals would need to take into consideration clinician output (productivity), lifestyle and gender variables, technology utilization, disease epidemiology, and case mix and intensity among many other variables, and to perform such an analysis for each practice specialty and setting. Needless to say, conducting such a complicated analysis in a rapidly changing system and expecting to emerge with a reasonably accurate prediction of the medical workforce we'll need in the next five years is, at a minimum, a daunting challenge. Some would say that it is nearly impossible to forecast with any reasonable degree of accuracy exactly how many medical providers we would need in the future.

The skeptic also asks: Why is this such a big deal? Who needs to know physician demand requirements? Why do we place so much importance on getting the numbers right in terms of the physician supply? It is ostensibly because, in a system that is so expensive and cost-conscious, having the "right" number of physicians in the system (i.e., one where supply and demand are roughly in balance), serves to conserve precious resources. Since physicians determine a great deal of medical costs though ordering tests, controlling use of technology, prescribing medications and so forth, having too many physicians would drive soaring costs even higher.

A Practical Proposal

In this most recent call, after reviewing the current state of the medical workforce in terms of supply and demand, Darrell G. Kirch, MD, who heads the Association of American Medical Colleges, states that "determining the ?correct' size of the future physician workforce involves deciphering an extremely complicated, multivariate equation that defies simple solutions." So true, and the reason why past attempts have been so miserably wrong. He notes that most experts now operate under the assumption that there will be a strong and increasing demand for physicians, as well as PAs and other "nonphysician professionals," in the future. This is progress, since in the past a number of these forecasting efforts have omitted the impact of PA and NP utilization, contributing even further to the perceptions of inaccuracy.

Thus it is interesting that physicians are once again calling for an effort "to attain a much better understanding of future national health care workforce needs." Throughout history, such efforts, particularly those conducted by the federal government, have not been known for their accuracy. Rather than waste money performing arcane studies very likely to produce erroneous results, perhaps a better workforce policy strategy would be to consider more practical approaches. To this observer, these would include:

• Enacting legislation to institute and expand programs for medical students and PA students to reduce their debt level in exchange for periods of service in medically needy settings;

• Creating incentive programs designed to promote the increased utilization of PAs and other clinicians in primary care, including increasing Medicare payments to primary care practices employing these providers; and

• Since we know that it is not just a matter of increasing numbers,2 developing more efficacious systems of delivery of medical services that maximize the utilization of PAs and other nonphysician clinicians.

References

1. Kirch DG, Vernon DJ. Confronting the complexity of the physician workforce equation. JAMA. 2008;299(22):2680-2682.

2. Goodman DC, Grumbach K. Does having more physicians lead to better health system performance? JAMA. 2008;299(3):335-337.

James F. Cawley is professor and director of the PA/MPH program at the School of Medicine and Health Sciences at The George Washington University in Washington.