Objectives: Coronary artery bypass grafting performed off-pump has emerged in recent years as a less morbid alternative to on-pump bypass grafting. However, the impact of hospital volume on the outcomes of off-pump relative to on-pump bypass grafting has not been evaluated. Methods: We conducted a retrospective study of patients undergoing off-pump (n = 26,011) and on-pump (n = 99,344) coronary artery bypass grafting during 2000 through 2004 in 124 California hospitals, using the California Patient Discharge Database. Generalized linear mixed models were used to compare in-hospital mortality and postoperative complications in patients undergoing on-pump versus off-pump bypass grafting, accounting sequentially for differences in patient characteristics and hospital-level effects. The relative mortality and complication rates for patients undergoing on-pump versus off-pump coronary bypass were evaluated across hospital volume quartiles. Results: Mean length of stay was lower for patients who underwent off-pump compared with on-pump bypass grafting (8.7 vs 9.6 days; P < .001), as were unadjusted mortality and complication rates (2.2% vs 3.3%; 10.1% vs 11.6%, respectively; P < .001). For hospitals in the highest percent off-pump bypass quartile, adjusted mortality and complication rates for patients having off-pump bypass were significantly lower than for the on-pump group (odds ratio [OR] = 0.50; 95% confidence intervals [CI], 0.41-0.61; OR = 0.73; 95% CI, 0.66-0.81, respectively; P < .001); by contrast, for hospitals in the lowest percent off-pump bypass quartile, mortality and complications were similar in off-pump and on-pump groups (OR = 1.10; 95% CI, 0.75-1.63; OR = 0.92; 95% CI, 0.72-1.16, respectively; P > .05). Conclusions: Outcomes were significantly better for off-pump compared with on-pump coronary artery bypass grafting. Although the benefit of off-pump bypass grafting increased as the relative use of the procedure at a hospital increased, off-pump bypass grafting can be safely implemented across numerous hospitals.