BACKGROUND Cocaine use is associated with myocardial ischemia and infarction, but it is unclear whether this is only because of the acute effects of cocaine on heart rate, blood pressure, and vasomotor tone or whether accelerated atherosclerosis from long-term exposure to cocaine also contributes. METHODS We sought to measure the association between cocaine exposure and coronary calcification, a marker for atherosclerosis, among participants in the CARDIA Study who received computed tomography scanning and answered questions about illicit drug use at the year 15 examination in 2000-2001. RESULTS Among 3038 CARDIA participants (age 33-45 years, 55% women and 45% black), past cocaine exposure was reported by 35% and was more common among men, smokers, drinkers, and participants with less education. Powdered cocaine exposure was more common among whites, crack cocaine among blacks. Before adjustment, cocaine exposure was strongly associated with coronary calcification. After adjusting for age, sex, ethnicity, socioeconomic status, family history, and habits, however, these associations disappeared: adjusted odds ratios for coronary calcification were 0.9 (95% CI 0.6-1.3) for 1 to 10, 1.2 (95% CI 0.8-1.7) for 11 to 99, and 1.0 (95% CI 0.6-1.6) for > or =100 lifetime episodes of cocaine use, in comparison with none. Sex, tobacco, and alcohol use appeared to be primarily responsible for the confounding we observed in unadjusted models. CONCLUSIONS We found no evidence of a causal relationship between long-term exposure to cocaine and coronary calcification and conclude that acute nonatherogenic mechanisms probably explain most cocaine-associated myocardial infarction.