Objective. —To assess prospectively the risk of coronary heart disease associated with elevated plasma levels of homocyst(e)ine. Design. —Nested case-control study using prospectively collected blood samples. Setting. —Participants in the Physicians' Health Study. Participants. —A total of 14916 male physicians, aged 40 to 84 years, with no prior myocardial infarction (Ml) or stroke provided plasma samples at baseline and were followed up for 5 years. Samples from 271 men who subsequently developed Ml were analyzed for homocyst(e)ine levels together with paired controls, matched by age and smoking. Main Outcome Measure. —Acute Ml or death due to coronary disease. Results. —Levels of homocyst(e)ine were higher in cases than in controls (11.1±4.0 [SD] vs 10.5±2.8 nmol/mL;P=.03). The difference was attributable to an excess of high values among men who later had MIs. The relative risk for the highest 5% vs the bottom 90% of homocyst(e)ine levels was 3.1 (95% confidence interval, 1.4 to 6.9;P=.005). After additional adjustment for diabetes, hypertension, aspirin assignment, Quetelet's Index, and total/high-density lipoprotein cholesterol, this relative risk was 3.4 (95% confidence interval, 1.3 to 8.8) (P=.01). Thirteen controls and 31 cases (11%) had values above the 95th percentile of the controls. Conclusions. —Moderately high levels of plasma homocyst(e)ine are associated with subsequent risk of Ml independent of other coronary risk factors. Because high levels can often be easily treated with vitamin supplements, homocyst(e)ine may be an independent, modifiable risk factor. (JAMA. 1992;268:877-881)