Age at Menarche and Incidence of Diabetes: A Prospective Study of 300,000 Women in China

Previous studies of predominantly Western populations have reported inconsistent associations between age at menarche and risk of diabetes. We examined this relationship among Chinese women, who generally experience menarche at a later age than Western women. In 2004-2008, China Kadoorie Biobank recruited 302,632 women aged 30-79 years from 10 areas across China, and recorded 5,391 incident cases of diabetes during 7 years of follow-up among 270,345 women without baseline diabetes, cardiovascular disease or cancer. Cox regression models yielded adjusted hazard ratios for incident diabetes associated with age at menarche. Overall, the mean age at menarche was 15.4 years, and decreased across successive generations. Age at menarche was linearly and inversely associated with incident diabetes, with adjusted hazard ratio of 0.96 (95% confidence interval (CI): 0.94, 0.97) per year delay. Hazard ratios were greater in younger generations (for women born in the 1960s-1970s, hazard ratio (HR) = 0.93, 95% CI: 0.90, 0.97; for women born in the 1950s, HR = 0.95, 95% CI: 0.93, 0.98; and for women born in the 1920s-1940s, HR = 0.97, 95% CI: 0.95, 0.99). Further adjustment for adulthood body mass index significantly attenuated the association (HR = 0.99, 95% CI: 0.97, 1.00), especially among those born before 1950 (HR = 1.00, 95% CI: 0.97, 1.02). Much of the inverse association between age atmenarche and incident diabetes was mediated through increased adiposity associated with early menarche, especially in older generations.