OBJECTIVES: Recent literature suggests an increased risk of diabetes following smoking cessation. Our objective was to compare the risk of developing diabetes among obese smokers using bupropion vs. varenicline as well as other predictors during the first 3 years post-cessation. METHODS:  A population-based retrospective cohort study was conducted using the General Electric (GE) electronic medical record database (2006 – 2011). The cohort consisted of obese adult smokers newly initiating use of a smoking cessation medication:  bupropion vs. varenicline without a diabetes diagnosis at baseline. The outcome variable was time to developing diabetes following first prescription with a 3-year follow up. Chi-square tests were conducted to assess the frequency distribution of sample characteristics and association with diabetes development. Univariate survival analyses using Kaplan-Meier survival curve were conducted and log-rank test was used to assess significance. Cox Proportional Hazard (PH) regression model was carried out after evaluating PH assumption by Schoenfeld residual test. Interaction terms were included in the PH regression model if assumption was violated. RESULTS: The sample comprised of 91,899 individuals. A total of 3,668 (crude diabetes incidence rate: 13.3 per 1,000 person-years) obese smokers developed diabetes in 3 years. Abstinence at 12 months did not meet the PH assumption and interaction was created for this variable in the final Cox model. There was no statistically significant difference in diabetes risk using bupropion versus varenicline (Hazard Ratio: 1.408 95% Confidence Interval: 0.963 - 2.058). Non-primary care group (1.271 [1.069 – 1.511]), male (1.174 [1.004 – 1.373]), non-white (1.197 [1.015 – 1.410]), age group (40 – 64) (0.803 [0.656 – 0.984]), abstinence at 12 months (0.758 [0.604 – 0.952]), hypertension (1.239 [1.017 – 1.509]),and lung cancer (2.023 [1.024 – 3.999]) were significant predictors of developing diabetes. CONCLUSIONS: There is no significant difference in the diabetes risk three years post-cessation among obese adults using varenicline versus bupropion.