Background: Liraglutide can effectively reduce the weight of patients with type 2 diabetes. Nonetheless, its weight loss effect was highly heterogeneous in different patients in the clinical practice. Objective: To identify the factors most associated with the weight loss effect of liraglutide in obese or overweight patients with type 2 diabetes with poorly controlled oral medication in northeast China. Design: A prospective study. Methods: A prospective study was performed in subjects with type 2 diabetes who were taking oral medication and had a body mass index (BMI) of ⩾24 kg/m2. Liraglutide was administered for at least 12 weeks, while the original hypoglycemic regimen was kept unchanged (Phase I). Later, liraglutide treatment was continued or stopped as necessary or as subjects thought fit in the 13–52 weeks that followed (Phase II), and the potential factors affecting the effect of weight loss of liraglutide were analyzed. Results: Of the 127 recruited subjects, 90 had comprehensive follow-up data at week 12. In Phase I, the subjects’ blood sugar levels and weight decreased significantly( P < 0.001). Among all the significant factors, the gastrointestinal adverse reactions score (GARS) was more correlated with BMI change (ΔBMI; r = 0.43) and waist circumference change (ΔWC; r = 0.32) than the baseline BMI (BMI0) and WC (WC0). At week 12, linear regression showed that BMI0 independently affected ΔBMI and ΔWC, whereas WC0 only affected ΔWC. The GARS was significantly associated with ΔBMI and ΔWC, and this association continued until week 52, even after most subjects had discontinued liraglutide treatment. Conclusion: The degree of obesity and gastrointestinal adverse reactions were the most promising predictors of weight loss in liraglutide treatment.