2 years change of waist circumference and body mass index and associations with type 2 diabetes mellitus in cohort populations.

OBJECTIVE: Compare the association between 2 years change of waist circumference (WC) or body mass index (BMI) and incident type 2 diabetes mellitus (T2DM) in cohort populations.

METHODS: Prospective cohort study, based on a baseline investigation from 1999, we conducted the first follow-up for subjects whose follow time met 2 years, and the second follow-up for subjects whose follow time met 5 years. Two years change of WC or BMI was measured by the D-value between baseline and the first follow up values. The association between 2 years change of WC or BMI and incident T2DM were analyzed by using Poisson regression model.

RESULTS: Among 3461 participants without T2DM at baseline, 160 subjects developed T2DM during follow up period. Across quartiles of WC D-value, hazards ratio (HR) of incident T2DM increased, but there was no significant dose response relationship was found between BMI D-value and incident T2DM. Compared with subjects whose both WC and BMI were modified, there was no significant increase for T2DM risk in subjects whose WC was modified but BMI was not modified [HR(95%CI) = 1.66(0.70-3.97)], but the T2DM risk was significantly higher in subjects whose WC was not modified but BMI was modified [HR(95%CI) = 1.73(1.17-2.54)].

CONCLUSIONS: WC change was a better predictor for incident T2DM than BMI change. Intervention programs designed to reduce WC through lifestyle modification, including physical activity and diet, may have significant public health significance in preventing incident T2DM.