BACKGROUND Few methods exist to identify physicians who might benefit from depression education. OBJECTIVE To develop a measure of physicians' confidence or self-efficacy in caring for depressed patients and assess it's reliability and validity. METHODS A national sample of primary care physicians were surveyed and exploratory factor analysis (EFA) was used to identify factors underlying physicians' responses to 26 items. We named the factors, selected items with factor loadings > or = 0.50 for final scales, and tested a priori hypotheses about self-efficacy. METHODS 1) Random cross-sectional sample of family physicians, internists, obstetrician-gynecologists, and pediatricians (n = 5,369) and 2) 49 general internists and family physicians participating in a prepost evaluation of a depression workshop. RESULTS In the national sample, 3,712 physicians were eligible and 2,104 responded. Forty-six percent were female, and 51% were family physicians and general internists. EFA identified 5 factors, the first of which was called Self-Efficacy (4 items, alpha = 0.86). More family physicians (64%) had confidence (self-efficacy) in caring for depressed patients compared with general internists (33%), obstetrician-gynecologists (16%), and pediatricians (6%) (P < 0.001). Few physicians intended to change their care of depressed patients (10%) or take CME on depression (24%). Of the 49 physicians attending a depression workshop, 76% reported high self-efficacy after the workshop versus 50% before it (P = 0.013). CONCLUSIONS This study supports the reliability and validity of the Self-Efficacy scale as one method to identify physicians who might benefit from interventions. New approaches are needed because physicians are unlikely to change.