This study examines black/white differences in the utilization of three cardiac procedures (coronary angiography, coronary artery bypass graft, and coronary angioplasty) for patients hospitalized with coronary artery disease in New York State in the first 6 months of 1987. In contrast with previous studies, disease stages are used to control for severity of illness in addition to various severity proxies. Another methodological difference is that patient episodes (a fixed period of time after an initial hospital admission) are used as the unit of analysis rather than discharges to accurately account for patients whose initial visit is to a hospital not certified to perform the procedure. After controlling for severity using logistic regression analysis, whites were found to undergo significantly more of each of the procedures than blacks (odds ratios of 1.25, 2.06, and 1.69 for angiography, bypass graft, and angioplasty, respectively). These significant differences existed for most levels of the various control variables.