β-Cell dedifferentiation, characterized by loss of glucose sensitivity (β-cell glucose sensitivity [βCGS]), has been reported to play an important role in the development of type 2 diabetes (T2D). Traditionally, βCGS was derived from C-peptide-based method. However, C-peptide was not routinely examined in normal subjects and diabetes never treated with insulin. Thus, the aim of the study was to evaluate the use of insulin in oral glucose tolerance test (OGTT) in estimation of β-cell glucose response ability. A total of 1,599 subjects including normal glucose tolerance (NGT), impaired glucose tolerance (IGT) and T2D were included in the study. A subgroup of NGT subjects (n = 591) were followed up for an average duration of 56.88 ± 20.76 months. Insulin release rate (IRR ) in the function of glucose (IRR response to glucose [IRRG]) during OGTT was compared with βCGS. Both βCGS derived from C-peptide by deconvolution approach and IRRG by insulin release progressively declined from NGT to IGT and T2D. Both βCGS and IRRG were associated with deposit of first-phase insulin secretion (DI ). After 56.88 ± 20.76 months, 32 (5.41%) NGT subjects had developed T2D. NGT subjects who progressed to diabetes after follow-up had lower IRRG and DI levels than those who did not (P < 0.01). Furthermore, multiple logistic regression analyses showed that decreased IRRG was a significant independent risk predictor for future diabetes after adjustment of age, body mass index (BMI), homeostasis model assessment (HOMA)-insulin resistance, DI and family history. NGT subjects with decreased IRRG during OGTT had defective early insulin secretion and were at higher risk of developing diabetes. IRRG could be a useful T2D predictor in NGT subjects. © 2017 IUBMB Life, 69(9):756-766, 2017.