Despite emerging evidence of the role of flavonoids in cardiovascular disease (CVD) prevention, the association remains unclear. We examined whether flavonoids and selected flavonols and flavones or their food sources are associated with CVD risk. Women (n = 38 445) free of CVD and cancer participated in a prospective study with a mean follow-up of 6.9 y. On the basis of a food-frequency questionnaire, total flavonoids and selected flavonols and flavones were categorized into quintiles, and food sources were categorized into 4 groups. Relative risks were computed for important vascular events (519 events; excluding revascularizations) and CVD (729 events), including myocardial infarction, stroke, revascularization, and CVD death. The mean flavonoid intake was 24.6 +/- 18.5 mg/d, primarily as quercetin (70.2%). For both CVD and important vascular events, no significant linear trend was observed across quintiles of flavonoid intake (P = 0.63 and 0.80, respectively). No individual flavonol or flavone was associated with CVD. Broccoli and apple consumption were associated with nonsignificant reductions in CVD risk: 25-30% and 13-22%, respectively. A small proportion of women (n = 1185) consuming > or =4 cups (946 mL) tea/d had a reduction in the risk of important vascular events but with a nonsignificant linear trend (P = 0.07). Flavonoid intake was not strongly associated with a reduced risk of CVD. The nonsignificant inverse associations for broccoli, apples, and tea with CVD were not mediated by flavonoids and warrant further study.