ObjectivesTo explore associations between type and number of abnormal glucose values on antenatal oral glucose tolerance test (OGTT) with postpartum diabetes in South Asian women diagnosed with gestational diabetes (GDM) using International Association of the Diabetes and Pregnancy Study Groups criteria.MethodsThis post‐hoc evaluation of the Lifestyle Intervention IN Gestational Diabetes (LIVING) study, a randomized controlled trial, was conducted among women with GDM in the index pregnancy, across 19 centers in Bangladesh, India, and Sri Lanka. Postpartum diabetes (outcome) was defined on OGTT, using American Diabetes Association (ADA) criteria.ResultsWe report data on 1468 women with GDM, aged 30.9 (5.0) years, and with median (interquartile range) follow‐up period of 1.8 (1.4–2.4) years after childbirth following the index pregnancy. We found diabetes in 213 (14.5%) women with an incidence of 8.7 (7.6–10.0)/100 women‐years. The lowest incidence rate was 3.8/100 women years, in those with an isolated fasting plasma glucose (FPG) abnormality, and highest was 19.0/100 women years in participants with three abnormal values. The adjusted hazard ratios for two and three abnormal values compared to one abnormal value were 1.73 (95% confidence interval [CI], 1.18–2.54; p = .005) and 3.56 (95% CI, 2.46–5.16; p < .001) respectively. The adjusted hazard ratio for the combined (combination of fasting and postglucose load) abnormalities was 2.61 (95% CI, 1.70–4.00; p < .001), compared to isolated abnormal FPG.ConclusionsRisk of diabetes varied significantly depending upon the type and number of abnormal values on antenatal OGTT. These data may inform future precision medicine approaches such as risk prediction models in identifying women at higher risk and may guide future targeted interventions.image