Aims: To investigate the dose-response relationship of total sedentary time with incident diabetes in Chinese middle-aged and older adults. Methods: The present study followed 100,525 participants aged >= 40 years old from the China Cardiometabolic Disease and Cancer Cohort (4C) Study, which was a prospective study conducted in 25 communities across mainland China. Associations between sedentary time and incident diabetes were assessed with Cox regression and restricted cubic splines. Results: During a median follow-up of 3.8 years, 7,529 participants developed diabetes. After adjustment for multiple variables, high levels of sedentary time (>= 30 h/week) was associated with increased risk for developing diabetes (hazards ratio, 1.08; 95 % confidence intervals 1.02, 1.14) compared with low levels of sedentary time (< 20 h/week). Restricted cubic spline analyses revealed an inverted U-shaped relation between sedentary time with diabetes. Subgroup analyses found that the observed association remained significant in subgroup of individuals with body mass index (BMI) >= 25 kg/cm(2) or diastolic blood pressure (DBP) >= 90 mm Hg. However, the significant association was diminished in participants with sufficient physical activity (PA) (P = 0.22). Conclusions: The multicenter, population-based, prospective study suggested an inverted U-shaped relation be-tween sedentary time with diabetes. PA alleviated the deleterious effects associated with sedentary time.