Abstract
OBJECTIVES: The study reviewed methods for measuring the specialty distribution of the US physician workforce. It was hypothesized that current databases and measurement conventions overestimate the number of generalist physicians. METHODS: A descriptive analysis of the American Medical Association (AMA) Physician Masterfile for California was done with different assumptions about the definition of generalists based on primary and secondary specialty information. RESULTS: A rigorous definition of generalist physician that excludes physicians with secondary practices in specialist fields resulted in an estimate of generalist physicians 25% lower than the number estimated by conventional workforce evaluation methods. Physicians who reported practicing in both generalist and specialist fields were more likely to be older, to be international medical school graduates, and to be in solo or duo practice compared with physicians who listed only generalist or specialist fields. CONCLUSIONS: The actual number of generalist physicians in the United States may be less than previously believed. Although the exact magnitude of the "hidden system" of specialists providing primary care is difficult to measure, at least a portion appear to already be counted as generalist physicians by current conventions.