Abstract
OBJECTIVES: We investigated the association between menopausal hormone therapy (MHT) and incident type 2 diabetes in postmenopausal women, and explored the potential modifying role of body fat distribution on this association. METHODS: We included 2210 postmenopausal women without prevalent diabetes at recruitment (2000-2002) from the Multiethnic Study of Atherosclerosis. Cox proportional hazards models were used to examine associations of MHT and MHT types with incident diabetes, testing for variation according to body fat distribution. RESULTS: Over a median follow-up of 11.1 years, there were 226 incident cases of diabetes. There were no significant interactions with central or generalized body fatness. In fully adjusted models, current and past MHT use was associated with a greater risk of incident diabetes [HR: 1.66 (1.18-2.35) and 1.60 (1.11-2.30) respectively]. Estrogen only (ET) and combined progestin and estrogen (PET) formulations were similarly associated with a greater risk of incident diabetes [HR: 1.52 (1.03-2.24) and 1.77 (1.15-2.72) respectively]. CONCLUSIONS: In our observational study of middle-aged and older, non-diabetic postmenopausal women, a current or past use of MHT was independently associated with a greater risk of incident diabetes. ET and PET are associated with similar risks of incident diabetes in postmenopausal women. The association of MHT use with incident diabetes is the same irrespective of body mass index (BMI) or waist circumference.