BACKGROUND: The contribution of apolipoprotein A-I (apoA-I) to coronary heart disease (CHD) risk stratification over and above high-density lipoprotein cholesterol (HDL-C) is unclear. We studied the associations between plasma levels of HDL-C and apoA-I, either alone or combined, with risk of CHD events and cardiovascular risk factors among apparently healthy men and women. METHODS AND RESULTS: HDL-C and apoA-I levels were measured among 17 661 participants of the EPIC (European Prospective Investigation into Cancer)-Norfolk prospective population study. Hazard ratios for CHD events and distributions of risk factors were calculated by quartiles of HDL-C and apoA-I. Results were validated using data from the ARIC (Atherosclerosis Risk in Communities) and WHS (Women's Health Study) cohorts, comprising 15 494 and 27 552 individuals, respectively. In EPIC-Norfolk, both HDL-C and apoA-I quartiles were strongly and inversely associated with CHD risk. Within HDL-C quartiles, higher apoA-I levels were not associated with lower CHD risk; in fact, CHD risk was higher within some HDL-C quartiles. ApoA-I levels were associated with higher levels of CHD risk factors: higher body mass index, HbA1c, non-HDL-C, triglycerides, apolipoprotein B, systolic blood pressure, and C-reactive protein, within fixed HDL-C quartiles. In contrast, HDL-C levels were consistently inversely associated with overall CHD risk and CHD risk factors within apoA-I quartiles (P<0.001). These findings were validated in the ARIC and WHS cohorts. CONCLUSIONS: Our findings demonstrate that apoA-I levels do not offer predictive information over and above HDL-C. In fact, within some HDL-C quartiles, higher apoA-I levels were associated with higher risk of CHD events, possibly because of the unexpected higher prevalence of cardiovascular risk factors in association with higher apoA-I levels. CLINICAL TRIAL REGISTRATION: URL: https://www.clinicaltrials.gov. Unique identifier: NCT00000479.