Background— Few studies have examined the association of race and outcomes after coronary artery bypass graft (CABG) surgery while controlling for both patient and hospital effects. Methods and Results— We retrospectively analyzed data on a cohort of 566 785 white and 24 354 black Medicare beneficiaries 65 years old and older undergoing CABG in 1091 US hospitals from 1997 to 2000. Mortality and repeat revascularization rates were examined after sequential adjustment for patient and hospital differences by use of generalized estimating equations. Unadjusted mortality was higher ( P <0.001) in black than in white patients at 30 (6.4% versus 5.2%), 90 (8.3% versus 6.6%), and 365 days (13.5% versus 9.8%) after surgery. Black patients were more likely ( P <0.001) to undergo CABG at hospitals with the highest mortality (56% versus 47%) and at hospitals in the lowest volume quintile (24% versus 20%). Adjusted only for patient characteristics, mortality was 8%, 11%, and 25% higher in black patients at 30, 90, and 365 days. After adjustment for hospital effects, 30 and 90 day mortality was similar but 17% higher in black patients at 365 days. Racial differences in mortality were greater in men than in women. On adjustment for patient and hospital effects, repeat revascularization rates were similar in black and white patients. Conclusions— Racial disparities in CABG outcomes are sensitive to the effects of sex and duration of postsurgical follow-up. The increasing disparity in outcomes as follow-up increased is consistent with the hypothesis that black patients have less access to secondary prevention and rehabilitation services after surgery.