Purpose: To identify surgeon characteristics associated with mortality or morbidity, following carotid endarterectomy (CEA). Methods: Data on all inpatient discharges from the 284 nonfederal Pennsylvania hospitals were obtained from the Pennsylvania Health Care Cost Containment Council for the period from 1994 to 1995. Physician data were obtained from the Physicians List of the American Medical Association, including name, gender, specialty, year of birth, board certified, and year of licensure. Cases were selected if any of six procedures codes were ICD-9-CM rubric 38.12, indicating CEA. Results: Among the 12,725 cases studied, in-hospital mortality was 0.7%, nonfatal morbidity was 3.0%, and the total bad outcome rate was 3.7%. Surgeons who performed I to 2 CEAs over 2 years had the highest mortality (2.0%) and total bad outcome (9.2%) rates. For surgeons performing three or more cases in 2 years, increased volume was not associated with better outcomes. A greater number of years since the surgeon was licensed was associated with greater mortality (p = 0.001), but not with morbidity or bad outcome rates. In regression analyses that adjusted for patient risk, both years since licensure and specialty predicted surgical mortality rate, but only volume predicted surgical bad outcome rate. Conclusions: More years since licensure and very low patient volume are associated with worse patient outcomes following CEA.