In 1995,784 of this country's 3,082 counties were wholly designated as Primary Care Health Personnel Shortage Areas (PCHPSAs). This analysis, however, focused on the other 2,298 counties that had only a partial PCHPSA designation or none at all (i.e., those considered to have a more adequate number of primary care physicians). The 1999 Area Resource File (ARF) was the data source used. It contains information for each county in the United States pertaining to PCHPSA designation, its general population and the number of non-federal patient care physicians practicing there, by specialty. A lack of required detail in the ARF characterizing osteopathic physicians prevents a comprehensive analysis for any year since 1995. Thus, 1995 is the most recent year for which all required data were available.

Family practice (FP/GP), general pediatrics, general internal medicine and obstetrics-gynecology currently constitute primary care specialties for the purpose of federal designation of PCHPSAs. The principal criterion for awarding such designations to counties or smaller geopolitical units is that there are more than 3,500 people per primary care doctor. Unusually high need for medical services and other variables (e.g., proportion medically indigent, ethnic minority) may also be considered to justify designating a special population or part of a county.

The geographic distribution of physicians is known to vary substantially by medical specialty. We assessed every county's reliance on each of the specialties on which PCHPSA designation depends. A measure of reliance was constructed by withdrawing, in turn, the physicians of each of the respective primary care specialties, then determining if the county met the population-to-physician criterion.

In 1995, 2,298 U.S. counties were not fully designated PCHPSAs and had population-to-primary care physician ratios less than 3,500 to 1. For 1,332 (58.0 percent) of these, this ratio would deteriorate to more than 3,500 people per primary care physician if FP/GPs were withheld from the calculation. Few counties were similarly dependent on the services of other subspecialties, even in aggregate. For the majority of all U.S. counties that would warrant whole county PCPHSA designation with the loss of their general internists (40 of 45; 88.9 percent), pediatricians (nine of 11; 81.8 percent), or obstetricians (eight of nine; 88.9 percent), withdrawal of just their FP/GPs would also justify that designation (see accompanying table).

This table portrays a national picture, including urban and rural areas. Restricting the analyses to rural counties shows an even greater reliance on family physicians. Rural regions (Non-Metropolitan Statistical Areas) accounted for 1,548 of the 2,298 counties not wholly designated as PCHPSAs (67.4 percent). Among these rural counties, the ratio of population to primary care physicians without FP/GPs exceeded 3,500 to 1 for 1,050 (67.8 percent). When general internal medicine, pediatrics and obstetrics-gynecology were not included in computing the ratio for these rural counties, it was greater than the 3,500 to 1 threshold for 33 (2.1 percent), eight (0.5 percent), and eight (0.5 percent) of counties, respectively.