Cut-off points of homeostasis model assessment of insulin resistance, beta-cell function, and fasting serum insulin to identify future type 2 diabetes: Tehran Lipid and Glucose Study

Aims To determine cut-off points of homeostasis model assessment of insulin resistance (HOMA-IR), beta-cell function (HOMA-B), insulin sensitivity (HOMA-S), and fasting insulin for identifying the subjects with type 2 diabetes mellitus (T2DM) in Iranian adults using data from a prospective population-based study. Methods From participants of Tehran Lipid and Glucose Study, 4942 Iranian subjects, aged 20-86 years, were followed for incident T2DM. Fasting serum insulin was determined by the electrochemiluminescence immunoasaay. The associations between HOMA-IR, HOMA-B, HOMA-S, and fasting insulin and incident T2DM were evaluated using Cox proportional hazards models. The receiver operator characteristic curve analysis was used to determine the cut-off points of HOMA-IR, HOMA-B, HOMA-S, and fasting insulin. Results After 9.2 year follow-up, 346 (7.0 %) incident cases of T2DM were identified; the risk-factor-adjusted hazard ratios for HOMA1-IR, HOMA2-IR, HOMA1-B, HOMA2-B, HOMA1-S, HOMA2-S, and insulin were 1.15, 1.70, 0.732, 0.997, 0.974, 0.986, and 1.01 in women and 1.37, 1.67, 0.588, 0.993, 0.986, 0.991, and 1.06 in men, respectively (all p < 0.05 except for HOMA2-B in women). Optimal cut-off points for HOMA1-IR, HOMA2-IR, HOMA1-B, HOMA2-B, HOMA1-S, HOMA2-S, and insulin were 1.85, 1.41, 86.2, 72.5, 54.1, 63.7, and 11.13 A mu U/ml in women and 2.17, 1.18, 67.1, 74.6, 46.1, 74.1, and 9.16 A mu U/ml in men, respectively. Conclusions HOMA-IR, HOMA-B (except for HOMA2-B in women), HOMA-S, and fasting insulin were independent predictors of T2DM. Optimal cut-off points of HOMA-IR, HOMA-B, HOMA-S, and fasting serum insulin were determined from a population-based study for identifying incident T2DM.