DUAL EFFECTS OF WEIGHT AND WEIGHT-GAIN ON BREAST-CANCER RISK

Context.-Breast cancer is a major cause of mortality among women. It i s important to identify modifiable risk factors for this disease. Obje ctive.-To examine body mass index (BMI) at the age of 18 years and at midlife and adult weight change in relation to breast cancer incidence and mortality. Design.-Cohort study. Setting.-A cohort of 95 256 US f emale nurses aged 30 to 55 years who were followed up for 16 years. Ma in Outcome Measure.-Incident and fatal breast cancer. Results.-During 1 203 498 person-years, 2517 incident breast cancers (60% postmenopaus al) were documented. Higher current BMI was associated with lower brea st cancer incidence before menopause and was minimally associated with incidence after menopause. However, a stronger positive relationship was seen among postmenopausal women who never used hormone replacement (relative risk=1.59 for BMI >31 kg/m(2) vs less than or equal to 20 k g/m(2); 95% confidence interval, 1.09-2.32; P for trend <.001). Higher BMI at the age of 18 years was associated with lower breast cancer in cidence both before and after menopause. Weight gain after the age of 18 years was unrelated to breast cancer incidence before menopause, bu t was positively associated with incidence after menopause. This incre ased risk with weight gain was limited to women who never used postmen opausal hormones; among these women, the relative risk was 1.99 (95% c onfidence interval, 1.43-2.76) for weight gain of more than 20 kg vs u nchanged weight (P for trend <.001). Current BMI and weight gain were even more strongly associated with fatal postmenopausal breast cancer. In this population, the percentage of postmenopausal breast cancer ac counted for by weight gain alone was approximately 16% and by hormone replacement therapy alone was 5%, but when the interaction between the se variables was considered, together they accounted for about one thi rd of postmenopausal breast cancers. Conclusions.-Avoiding adult weigh t gain may contribute importantly to the prevention of breast cancer a fter menopause, particularly among women who do not use postmenopausal hormones.