We tested the hypothesis that a high-fat diet increases the risk of breast cancer in a population-based study of 590 women aged 40-79 years who were without known breast cancer when they provided a quantitative 24-hour diet recall. Fifteen postmenopausal women were diagnosed with incident breast cancer during the next 15 years (approximately 7600 person-years of follow-up). These women had significantly higher age-adjusted intake of all fats (monounsaturated, polyunsaturated, and saturated), and oleic, linoleic, and linolenic acids, with a stepwise increase in risk across tertiles of intake. Fat intake was associated with total calories, protein, and carbohydrates, and women with incident breast cancer consumed more calories, protein, and carbohydrates than did other subjects. When each nutrient variable (calories, fats, protein, and carbohydrates) was adjusted for age, body mass index, age at menopause, parity, and alcohol consumption, the strongest risks for incident breast cancer were associated with total calories (relative risk per standard deviation = 2.72, 95% confidence interval = 1.51-4.89, p = 0.002) and total fats (relative risk per standard deviation = 2.01, 95% confidence interval = 1.19-3.41, p = 0.01). Fat composition of the diet, expressed either as percent of energy or as fat intake adjusted for calories by regression analysis, was not significantly associated with risk of breast cancer. These results support the hypothesis that total calorie consumption, as well as dietary fat consumption, is a risk factor for breast cancer in postmenopausal women, and parallel observations in animal models.