Although many case-control studies have suggested positive associations between carbohydrate intake and breast cancer incidence rates in both pre- and postmenopausal women, there is limited information available from cohort studies. We examined the effect of the intake of different carbohydrates, the glycemic index, and the glycemic load on breast cancer incidence in postmenopausal women taking into consideration tumor estrogen receptor status. Postmenopausal women (n = 23,870; aged 50-65 y) participated in the "Diet, Cancer, and Health" study, and provided information about diet and established risk factors for breast cancer. During follow-up, we identified 634 incidents of breast cancer. Associations between carbohydrate intake and breast cancer incidence were analyzed using Cox's regression models. There was no association for intake of glucose, fructose, sucrose, maltose, lactose, or starch and breast cancer incidence rate, and no association for glycemic index or glycemic load after adjusting for confounding factors. Intake of different carbohydrates was not associated with breast cancer incidence rates for either estrogen receptor positive (ER+) or (ER-) breast cancer. Similarly, glycemic index and glycemic load were not associated with ER+ breast cancer after adjusting for confounding factors. A borderline significant positive association between glycemic index and (ER-) breast cancer was observed (P = 0.05). In conclusion, we found no clear associations between intake of different carbohydrates, total carbohydrate intake, glycemic index, or glycemic load and breast cancer incidence in postmenopausal women. Furthermore, when ER+ and ER- breast cancer cases were analyzed separately, no clear associations were observed.