Associations of time in bed and sleep timing with dementia and cognitive decline among Chinese rural‐dwelling older adults: A cohort study

Background The longitudinal associations of time in bed (TIB) (a composite indicator sleep duration, latency, and fragmentation) timing marker circadian rhythms) with dementia cognitive decline older adults are unclear. Method This population-based cohort study used data from 1982 participants the Shandong Yanggu Study Aging Dementia, who were aged ≥60 years, free dementia, living rural communities western Province, China. At baseline (2014) follow-up (2018) examinations, parameters assessed using Pittsburgh Sleep Quality Index Epworth Sleepiness Scale. Global function was measured Mini-Mental State Examination (MMSE). Dementia diagnosed following DSM-IV criteria, NIA-AA criteria for Alzheimer disease (AD). Data analyzed restricted cubic splines, Cox proportional-hazards models, general linear models. Result During mean 3.7 (standard deviation = 0.3) 97 (68 AD). Restricted spline curves showed J-shaped TIB, rise risk, a reverse association mid-sleep time. When continuous categorized into tertiles, multivariable-adjusted hazard ratio (HR) incident 1.69 (95%CI 1.01-2.83) duration >8 hours (vs. 7-8 hours), 2.17 (1.22-3.87) bedtime before 9PM 10PM or later), 2.00 (1.23-3.24) 1AM 1-1.5AM). Early significantly associated AD (multivariable-adjusted HR range: 2.25-2.51; P<0.05). Among individuals at follow-up, long early time, late prolonged TIB advanced to greater MMSE score (P<0.05). These statistically evident mainly among men <75 years. Conclusion Long an increased risk AD, their only young-old people men.