PURPOSE: Approximately half of previous studies on serum uric acid have reported it to be an independent risk factor for coronary heart disease (CHD). We tested this hypothesis in the Atherosclerosis Risk in Communities (ARIC) Study. METHODS: A total of 13,504 healthy middle-aged men and women were followed prospectively for up to eight years. We identified 128 fatal and nonfatal CHD events in women and 264 in men. RESULTS: The age-, race-, and ARIC field center-adjusted relative risk of CHD for sex-specific quartiles of serum uric acid were 1.0, 1.39, 1.08, and 2.35 in women (p for trend = 0.009) and 1.0, 1.03, 0.89, and 1.21 in men (p for trend = 0.44), respectively. However, serum uric acid was correlated positively with many risk factors, and after multivariable adjustment, there was little evidence of an association of uric acid with CHD in either sex. CONCLUSIONS: Our results are not consistent with serum uric acid being an independent risk factor for CHD.