AbstractAim of the study: To determine the need to change the present glucose challenge threshold value in screening for gestational diabetes mellitus (GDM) among women in Dakshina Kannada and neighboring areas. Study Design: Retrospective and Cross Sectional study. Materials & Methods: 997 patients were analysed in the retrospective study having a GCT done between 24 – 28 weeks from October 2009 to October 2010 and complications of GDM observed in them. A mean GCT was identified above which complications (maternal and fetal/neonatal) were noted. Following this, in the cross sectional study done between November 2010 to September 2011, 209 patients were identified as having a GCT above that determined by the retrospective study. They were subjected to a 100g glucose threshold test to diagnose GDM (using Carpenter Couston values). Sensitivity, specificity, positive and negative predictive value was calculated for each GCT threshold value. Results: The mean GCT at which complications of GDM were observed in the retrospective study was 123.56 (maternal complications) and 124.15 (fetal/ neonatal complications). In the Cross Sectional Study, therefore, 124 mg/dl was taken as the cut off above which women were subjected to a 100g oral GTT. The ROC curve identified a value of 143mg/dl showing a sensitivity of 83.3% and specificity of 72.3%. A value of 135mg/dl, however showed a high sensitivity (91%) which is required in a high prevalence area. This value had a moderate specificity (52%). Conclusion: A high prevalence area like ours requires a GCT threshold with a high sensitivity which in our study was observed at a threshold value of 135mg/dl.