Aim: This study aims to evaluate the trends and geographic variations of incident diabetes as well as the corresponding sex differences in China.
Methods: The open cohort study derived data of 16,610 individuals from the China Health and Nutrition Survey 1997-2015. Direct standardisation was employed to calculate the age-standardised diabetes incidence. Mixed effects logistic regression models with interaction terms were performed to examine variations in incident diabetes. Socio-demographic (age, sex, marital status, racial compositions and educational attainment) and lifestyle attributes (smoking history, BMI and waist circumference) were sequentially included as covariates.
Results: Overall age-standardised diabetes incidence increased from 2.94 per 1000 person-years (95% CI, 2.44-3.44) in 1997-2004 to 5.54 (95% CI, 4.94-6.14) in 2009-2015. Models with interaction terms suggest that the increase among men was higher than that among women (wave 2006-2009 × Female: OR = 0.45, 95% CI 0.28-0.72). Age-standardised incidence of diabetes varied across regions, ranging from 5.67 (95% CI, 4.95-6.40) in Eastern China to 2.69 (95% CI 2.19-3.19) in Western China. Subsequent modelling analyses suggest that the geographic variations could be mostly explained by the variations in the BMI and waist circumference across regions.
Conclusions: Results suggest that the incidence of self-reported diabetes approximately doubled during the study period. The increase among men was steeper than that among women. Public interventions reducing the population's obesity level hold promise to alleviate geographic variations and flatten the growth curve.
Keywords: diabetes mellitus; geography; incidence; medical; sex disparities; temporal trends.
© 2020 Diabetes UK.