Low-density lipoprotein (LDL) cholesterol is the primary lipid parameter targeted to prevent myocardial infarction. Alternatively, non–high-density lipoprotein (HDL) cholesterol includes LDL cholesterol and other atherogenic particles but does not require a fasting sample. Non-HDL cholesterol and LDL cholesterol as predictors of first nonfatal myocardial infarction were compared in 303 patients and 297 controls matched for age, gender, and community within the Boston Area Health Study. Patients were white men or women aged <76 years living in the Boston area, without a history of myocardial infarction or angina pectoris, in whom symptoms of confirmed myocardial infarction began during the 24 hours before admission. After multivariate adjustment for coronary risk factors in unmatched analyses, the corresponding odds ratios (ORs) of a first nonfatal myocardial infarction for non-HDL cholesterol in the second, third, and fourth quartiles were 1.83 (95% confidence interval [CI] 1.07 to 3.14), 2.07 (95% CI 1.23 to 3.49), and 2.33 (95% CI 1.39 to 3.90) (p trend <0.01). For LDL cholesterol, the ORs were 1.10 (95% CI 0.67 to 1.81), 0.87 (95% CI 0.52 to 1.46), and 1.45 (95% CI 0.90 to 2.35) (p trend = 0.16). Including HDL cholesterol in the model increased the ORs and strengthened the test for a trend for LDL cholesterol, whereas the ORs were decreased and the test for a trend was weakened for non-HDL cholesterol. In conclusion, given that non-HDL cholesterol accounts for LDL cholesterol plus other atherogenic particles but does not require a fasting sample, this study suggests that non-HDL cholesterol may be at least as useful as LDL cholesterol to initially screen patients for risk of a first nonfatal myocardial infarction.