|| Checking for direct PDF access through Ovid
The objective is to examine the association between physical activity level (PAL) and incident type 2 diabetes among middle-age and older Chinese men and women in urban China.This prospective study included 6348 participants (age 35 to 74 yr) who were free of diabetes and cardiovascular disease at baseline. PAL was estimated on the basis of self-reported overall physical activity on a typical day. According to PAL, participants were classified into four groups: sedentary (PAL, 1.00–1.39), low active (PAL, 1.40–1.59), active (PAL, 1.60–1.89), and very active (PAL, >1.89). The association of PAL with incident diabetes was examined by Cox proportional hazards model.During 7.9 yr of follow-up (50,293 person-years), 478 incident cases of type 2 diabetes were identified. After adjustment for age, sex, geographic region, educational level, smoking, alcohol use, and family history of diabetes, the HR (95% CI) values for type 2 diabetes across increasing categories of PAL were 1.00 (reference), 0.82 (0.62–1.09), 0.63 (0.47–0.83), and 0.47 (0.36–0.61), respectively (P for trend <0.0001). Additional adjustment for baseline body mass index or waist circumference attenuated the magnitude of risk reduction, but it remained significant. The inverse association between PAL and risk of incident diabetes was persistent in subgroup analyses according to age, sex, hypertension, smoking, body mass index, waist circumference, and fasting plasma glucose level.Higher PAL is associated with substantial reduction in risk of type 2 diabetes. Our findings suggest the importance of a physically active lifestyle in the prevention of diabetes.