Our aim was to evaluate and compare the diagnostic performance of three methods commonly used for GDM screening: fasting plasma glucose (FPG), two-step 50 g glucose challenge test (GCT), and 75 g glucose tolerance test (GTT) in a randomized study design to predict GDM in the first trimester and determine the best approach in predicting GDM. In a non-blind, parallel-group prospective randomized controlled study; 736 singleton pregnant women underwent FPG testing in the first trimester and randomly assigned to two groups; two-step 50 g GCT and 75 g GTT. GDM diagnosis was made according to Carpenter-Coustan or ADA (American Diabetes Association) criteria in two-step 50 g GCT and 75 g GTT groups, respectively. Subsequent testing was performed by two-step 50 g GCT at 24-28 weeks for screen negatives. After excluding the women who were lost to follow-up or withdrawn as a result of pregnancy loss, 486 pregnant women were recruited in the study. The FPG, two-step GCT, and one-step GTT methods identified GDM in 25/486 (5.1 %), 15/248 (6.0 %), and 27/238 (11.3 %) women, respectively. Area under ROC curves were 0.623, 0.708, and 0.792, respectively. Sensitivities were 47.17, 68.18, and 87.1 %, respectively. Specificities were 77.37, 100, and 100 %, respectively. Positive predictive values were 20.33, 100, and 100 %, respectively. Negative predictive values were 92.29, 97, and 98.1 %, respectively. Until superior screening alternatives become available, the 75 g GTT may be preferred for GDM screening in the first trimester.