Soy food intake has been shown to have beneficial effects on cardiovascular disease risk factors. Data directly linking soy food intake to clinical outcomes of cardiovascular disease, however, are sparse. We examined the relationship between soy food intake and incidence of coronary heart disease (CHD) among participants in the Shanghai Women's Health Study, a population-based prospective cohort study of approximately 75000 Chinese women aged 40-70 y at the baseline survey that was conducted from 1997 to 2000. Included in this study were 64915 women without previously diagnosed CHD, stroke, cancer and diabetes at baseline. Information on usual intake of soy foods was obtained at baseline through an in-person interview using a validated food-frequency questionnaire. Cohort members were followed biennially through in-person interviews. After a mean of 2.5 y (162277 person-years) of follow-up, 62 incident cases of CHD (43 nonfatal myocardial infarctions and 19 CHD deaths) were documented. There was a clear monotonic dose-response relationship between soy food intake and risk of total CHD (P for trend = 0.003) with an adjusted relative risk (RR) of 0.25 (95% CI, 0.10-0.63) observed for women in the highest vs. the lowest quartile of total soy protein intake. The inverse association was more pronounced for nonfatal myocardial infarction (RR = 0.14; 95% CI, 0.04-0.48 for the highest vs. the lowest quartile of intake; P for trend = 0.001). This study provides, for the first time, direct evidence that soy food consumption may reduce the risk of CHD in women.