Background. Strengthening the role of primary care is considered a key element in improving the outcomes and efficiency of health care delivery. Previous studies of the benefits of primary care have primarily examined the association of primary care physician supply and mortality, generally in cross-sectional analyses using counties or states with limited patient-level data. The results of these studies have been mixed. More importantly, adult primary care physician supply is usually measured by combining primary care physicians across specialties, despite known differences in training and care delivery patterns of family physicians and general internists.

Objectives. To measure Medicare beneficiary health outcomes associated with differing primary care physician supply, measured as combined and by specialty.

Methods. Three studies were designed to test our overall hypothesis that greater primary care physician supply is associated with improved patient-level outcomes, and that this effect is greatest with family practice. Study 1 compared different metrics of primary care physician supply. Study 2 (cross-sectional) and Study 3 (longitudinal Medicare cohorts) examined the association of supply and Medicare health outcomes including mortality, ambulatory care sensitive condition admissions, preventive care services, and practice patterns using multi-level Poisson models with patients as the units of analysis.

Results. Area supply of general internists was inversely correlated with that of family physicians. Higher area adult primary care physician supply was weakly and inconsistently associated with mortality. Moreover, association of supply with mortality was not observed at the one-year follow-up. Lower levels of utilization were observed at higher supply levels, particularly in family practice but not for general internal medicine, which had opposite effects.

Conclusions. Our findings did not demonstrate that primary care physician supply was associated with better quality of care, lower admission rates and improved survival, although small specialty-specific effects were observed. In general, the practice patterns of family physicians appear to be more efficient than general internists, i.e., they used fewer resources with similar outcomes. Results of these studies provide evidence that simply adding more doctors to regions may be less important to improve patient care than investments that directly influence that quality and efficiency of medical care.