Aim: To compare and analyze the effects of obesity and non-alcoholic fatty liver disease (NAFLD) on the incidence of type 2 diabetes mellitus (T2DM) in Chinese subjects., Methods: In 2008, a population of 4847 subjects was randomly sampled from 17 medical units for enrollment in this cohort study. Baseline information was obtained via a questionnaire on general information, physical examination (height, weight, and blood pressure), laboratory tests (triglycerides, total cholesterol, fasting blood glucose, alanine aminotransferase (ALT), uric acid, and creatinine), B-mode ultrasound, and ECG screening. The incidence of T2DM after four years of follow-up was calculated. Numeric variable data was tested for normality, with the data expressed as mean ± SD. Kaplan-Meier analysis was performed to calculate the cumulative incidence. The Cox proportional hazards model was used to analyze the relative risk (RR) of different body mass index (BMI) levels and NAFLD on T2DM, as well as analyzing the RR adjusted for age, sex, blood pressure, lipids, transaminases, uric acid, and creatinine., Results: A total of 4736 (97.71%) subjects completed 4-year follow-up, with a median follow-up time of 3.85 years, totaling 17223 person-years. 380 subjects were diagnosed with T2DM, with a cumulative incidence of 8.0%. The cumulative incidence of T2DM in the NAFLD and control groups was 17.4% vs 4.1% (P < 0.001), respectively, while the incidence in overweight and obese subjects was 11.0% vs 15.8% (P < 0.001), respectively. The incidence of T2DM increased with an increase in baseline BMI. Cox regression analysis showed that the risk of T2DM in the NAFLD group (RR = 4.492, 95%CI: 3.640-5.542) after adjustment for age, sex, blood pressure, lipids, ALT, uric acid, and creatinine was 3.367 (2.367-4.266), while the value (RR, 95%CI) in overweight and obese subjects after adjustment for age, sex, BMI, blood pressure, lipids and other factors was 1.274 (0.997-1.629) and 1.554 (1.140-2.091), respectively. Stratification of three BMI levels (BMI < 24 kg/m(2), 24 kg/m(2) ≤ BMI < 28 kg/m(2), BMI ≥ 28 kg/m(2)) showed that the risk of T2DM in the NAFLD group was significantly higher than that in the control group (RR = 3.860, 4.049 and 3.823, respectively)., Conclusion: Compared with BMI, NAFLD could be better at forecasting the risk of T2DM in Chinese subjects, and may be a high risk factor for T2DM, independent of overweight/obesity.