Despite extensive research on obesity and breast cancer in recent decades, inconsistencies in the literature exist. The authors examined prospectively the relation between adult relative weight (weight (kg)/height (m)1.5) and breast cancer risk in a cohort of 54,896 women aged 31-89 years who had previously participated in the Breast Cancer Detection Demonstration Project. During a mean follow-up period of 7 years, 226 of the premenopausal women and 1,198 of the postmenopausal women developed breast cancer. Analysis was performed using Cox proportional hazards regression methods with age as the underlying time variable and adjusted for the effects of potential confounders. Among postmenopausal women, the risk of breast cancer increased with increasing relative weight (p < 0.05 for trend); relative risk for the highest compared with the lowest quintile for relative weight was 1.3 (95% confidence interval (CI) 1.1-1.6). This association was modified by age at diagnosis, with relative risks of 1.1 (95% CI 0.8-1.4), 1.2 (95% CI 0.8-1.7), and 1.8 (95% CI 1.3-2.5), respectively, for women aged < 60, 60-64, and > or = 65 years. The higher risk of breast cancer among the older and overweight women was largely confined to women whose weights were measured during the postmenopausal but not the premenopausal period. This risk pattern was observed among the naturally menopausal women, but was also apparent in the smaller group of women with bilateral oophorectomy or hysterectomy with one ovary retained. Among premenopausal women, adult relative weight was not associated with breast cancer risk. These findings suggest that the inconsistencies in the literature on obesity and breast cancer may be due in part to the differing age distributions of the populations studied. The authors conclude that prevention of obesity throughout adulthood, particularly after menopause, may help reduce breast cancer among older women.