This study compares indices of insulin sensitivity derived from fasting and oral glucose tolerance test (OGTT) glucose and insulin measurements, with respect to the reference measure (M/I), obtained from the euglycemic-hyperinsulinemic clamp, in postmenopausal women with varying glucose tolerance status. Fasting plasma insulin index, homeostasis model assessment index, and OGTT-derived indices (insulin 120-minute, Matsuda, metabolic clearance rate [MCR] of glucose, insulin sensitivity [ISI], and Cederholm indices) were calculated and compared with the M/I value in 112 postmenopausal women. All indices examined were significantly correlated with M/I (0.28 < or = r(2) < or = 0.56). Association studies revealed that on average, 48% of women were grouped in the same tertile of insulin sensitivity when using M/I and fasting plasma insulin index, and 54% when using M/I and insulin 120-minute index. However, concordance with M/I tertiles were 57%, 58%, 64%, 64%, and 68% for homeostasis model assessment, Matsuda, MCR, ISI, and Cederholm indices, respectively. Finally, correlation coefficients between M/I and insulin sensitivity indices were generally lower in women with normal glucose tolerance compared with women with impaired glucose tolerance or type 2 diabetes mellitus. These results suggest that in postmenopausal women, surrogate indices of insulin sensitivity obtained from OGTT data and incorporating a measurement of body weight or body mass index) (Cederholm, ISI, and MCR indices) appear to be superior to those without OGTT data or body weight-body mass index measurements and, therefore, could offer a better estimate of insulin sensitivity, allowing an improved clinical evaluation of this population at higher risk of cardiovascular disease and type 2 diabetes mellitus.