Why estimates of physician supply and requirements disagree

To review the major forecasts of physician supply and physician requirements.

English-language medical literature on physician manpower published between 1980 and 1990, identified from journal articles selected through searches of the MEDLINE and Health Planning and Administration databases using the search formulations physician supply, health manpower--physicians, and physician manpower.

Sources were selected if they were repeatedly cited and provided quantitative projections.

Each study was reviewed for its quantitative projections and to identify its methodological assumptions.

All forecasts point to a steadily increasing physician supply and, with one exception, suggest that supply will exceed requirements in the year 2000. The estimates of physician supply across studies varied primarily because of differing assumptions about entrance rates into the profession and the size and clinical work load of specific categories of physicians: researchers, teachers, administrators, residents, and women. The estimates of physician requirements were more volatile because of differences in the basic approach and assumptions about future growth and market structure.

Even though the measurement of entrance and exit rates from the profession is a generally accepted approach to forecasting supply, apparently minor disagreements on assumptions create large discrepancies between forecasts over time. There is no accepted approach to forecasting physician requirements.