Background: Physicians are important in the early detection of melanoma. We investigated whether primary care physician supply and the supply of dermatologists were related to stage at diagnosis for malignant melanoma. Methods: From the state tumor registry in Florida in 1994, we identified incident cases of malignant melanoma for which stage at diagnosis was available (N = 1884). Data on physician supply was obtained from the 1994 American Medical Association Physician Masterfile. Logistic regression determined the effects of physician supply (at the ZIP code level) on the odds of early-stage diagnosis controlling for patients’ age, gender, race/ethnicity, marital status, education level, income level, comorbidity, and type of health insurance. Results: Each additional dermatologist per 10,000 population was associated with a 39% increased odds of early diagnosis (odds ratio = 1.39, 95% confidence interval [CI] 1.09-1.70, P = .010). For each additional family physician per 10,000 population, the odds of early diagnosis increased 21% (odds ratio = 1.21, 95% CI 1.09-1.33, P < .001). Each additional general internist per 10,000 population was associated with a 10% decrease in the odds of early-stage diagnosis (odds ratio = 0.90, 95% CI 0.83-0.98, P = .009). The supplies of general practitioners, obstetrician/gynecologists, and other nonprimary care specialists were not associated with stage at diagnosis. Conclusions: Increasing supplies of dermatologists and family physicians were associated with earlier detection of melanoma. In contrast, increasing supplies of general internists were associated with reduced odds of early detection. Our findings suggest that the composition of the physician work force may affect important health outcomes and needs further study. (J Am Acad Dermatol 2000;43:211-8.)