Abstract
BACKGROUND AND OBJECTIVES: There is growing evidence that sleep duration is associated with future risk of new-onset diabetes mellitus. However, discussions of gender differences have yielded inconsistent results. The aim of this longitudinal study was to explore this issue in a large group of Chinese adult population. METHODS: 13,142 participants (6,366 men and 6,776 women) without diabetes at baseline from China Health and Nutrition Survey (CHNS) were included. They participated in at least two rounds of the CHNS during 2004-2015. Multivariate Cox proportional hazards regression models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for new-onset diabetes. In addition, non-linearity relation of sleep duration and new-onset diabetes was analyzed by restricted cubic splines (RCS). RESULTS: During the follow-up period from 2004 to 2015, there were 222 new cases of diabetes in men (5.34 per 1000 person-years) and 234 cases in women (5.26 per 1000 person-years) participants had newly developed diabetes. In women, there was a U-shaped association between sleep duration and new-onset diabetes with the lowest risk for diabetes in individuals sleeping 8-9h per day after adjusting for covariates. Compared with the reference (8-9h/day), the HRs for participants who slept <6h/day, >10h/day were 2.47 (1.22-4.99), 2.65(1.14-6.16) after adjustment for covariates. Among men <60 years old, compared with subjects who slept <7 hours per day, those slept 7-<8h/day (HR = 0.73, 95%CI 0.42-1.24), 8-<9h/day (HR = 0.57, 95%CI 0.33-0.99), and ≥9 h/day (HR = 0.35, 95% CI 0.14-0.90) had lower risk of diabetes after adjusting for all potential confounders (p-trend < 0.001). No significant relationship between sleep duration and diabetes was observed in men over 60 years of age. CONCLUSIONS: There was a U-shaped association between sleep duration and the risk of diabetes in women, with the lowest risk for diabetes at approximately 8-9h/day. For men, risk for new-onset diabetes decreased significantly with increasing sleep duration only among participants < 60 years. The effect of sleep on older men was not statistically significant.