ABSTRACT
Objective To investigate the characteristics of sleep patterns and their effects on the risk of diabetes mellitus (DM) among adults in Guangzhou. Methods A retrospective study was conducted using data collected by the lifestyle survey and health outcomes of 5 666 employees from Guangzhou who underwent physical examination between November 2012 and October 2013 at Guangdong Provincial People's Hospital. Sleep patterns and their distribution profiles were analysed using latent class analysis (LCA). Multiple logistic regression models were performed to investigate the association between sleep patterns and DM. Results LCA identified five sleep patterns: “very short sleep duration with insomnia” (class 1, 5.6%), “sleep insufficiency with mild daytime dysfunction” (class 2, 20.4%), “normal sleep” (class 3, 47.7%), “sleep insufficiency with daytime functioning complaints” (class 4, 4.7%) and “sleep insufficiency with poor nocturnal sleep” (class 5, 21.6%). After adjustment for the confounding factors, subjects of class 1 (OR=2.28, 95% CI: 1.51-3.43, <0.001), class 4 (OR=2.48, 95% CI: 1.54-4.00, P<0.001) and class 5 (OR=1.31, 95% CI: 1.01-1.71, P=0.045) had a higher risk of DM when compared with those of class 3. Conclusions There were significant associations between various sleep-related factors, leading to distinct sleep behavioral patterns among adults. Poor sleep patterns could increase the risk of DM.