Multilevel examination of diabetes in modernising China: what elements of urbanisation are most associated with diabetes?

Aims/hypothesis: The purpose of this study was to examine the association between urbanisation-related factors and diabetes prevalence in China. Methods: Anthropometry, fasting blood glucose (FBG) and community-level data were collected for 7,741 adults (18–90 years) across 217 communities and nine provinces in the 2009 China Health and Nutrition Survey to examine diabetes (FBG ≥7.0 mmol/l or doctor diagnosis). Sex-stratified multilevel models, clustered at the community and province levels and controlling for individual-level age and household income were used to examine the association between diabetes and: (1) a multicomponent urbanisation measure reflecting overall modernisation and (2) 12 separate components of urbanisation (e.g., population density, employment, markets, infrastructure and social factors). Results: Prevalent diabetes was higher in more-urbanised (men 12%; women 9%) vs less-urbanised (men 6%; women 5%) areas. In sex-stratified multilevel models adjusting for residential community and province, age and household income, there was a twofold higher diabetes prevalence in urban vs rural areas (men OR 2.02, 95% CI 1.47, 2.78; women, OR 1.94, 95% CI 1.35, 2.79). All urbanisation components were positively associated with diabetes, with variation across components (e.g. men, economic and income diversity, OR 1.42, 95% CI 1.20, 1.66; women, transportation infrastructure, OR 1.18, 95% CI 1.06, 1.32). Community-level variation in diabetes was comparatively greater for women (intraclass correlation [ICC] 0.03–0.05) vs men (ICC ≤0.01); province-level variation was greater for men (men 0.03–0.04; women 0.02). Conclusions/interpretation: Diabetes prevention and treatment efforts are needed particularly in urbanised areas of China. Community economic factors, modern markets, communications and transportation infrastructure might present opportunities for such efforts. Electronic supplementary material The online version of this article (doi:10.1007/s00125-012-2697-8) contains peer-reviewed but unedited supplementary material, which is available to authorised users