The Relationship Between Sleep Duration And All-cause Mortality And Its Influencing Factors And Pathway

Objectives: Many studies suggest an association of both short and long sleep duration with all-cause mortality,but the effect of co-occurrence of sleep duration and other lifestyle risk factors or health status remains unclear.Thus,this study is aimed to explore the association between sleep duration and all-cause mortality,as well as the effect of the interaction between sleep duration and other risk factors on all-cause mortality,and to detect the mediating effect of health status in the association between sleep duration and all-cause mortality.Methods: All participants were stable residents(over 5 years)with age ?18 years in Tiemen town and Cijan town of Xin'an Country,Henan province.Basic unit of sampling was natural villages and a total of 20194 participants were selected by a cluster randomization method from July to August of 2007 and July to August of 2008.Questionnaires,physical examinations,fasting plasma glucose(FPG)testing,and lipid profile testing were conducted at baseline.Follow-up survey were conducted from July to August 2013 and July to October 2014.The mean follow-up duration was 6.0 years and the response rate was 85.5%.We excluded 3 participants who had no data on sleep duration and 77 participants with outlier of sleep duration(?3 or ?16 hr/d),finally 17,185 participants were eligible for the present analysis.Participants were divided into 5 groups according to sleep duration.Kruskal-Wallis test was used for analyzing both categorical and continuous variables.Cox regression models were used to estimate hazard ratios(HRs)and 95% confidence intervals(95% CIs)for short and long sleep duration and explore the effect of multiplicative interaction between sleep duration and lifestyle risk factors or health status on all-cause mortality.The additive interaction was calculated by Anderson Method.Vander Weele method was used to analyze whether blood pressure,FPG,or lipid index played a mediating role in the association between sleep duration and all-cause mortality.Generalized additive model was used to fit the dose-response relation between sleep duration and all-cause mortality.In addition,sensitivity analyses were conducted to evaluate the robustness of the main results by excluding participants died at the first two years of the follow-up duration and those who were suffering cancer,heart failure,stroke,and myocardial infarction at baseline.Results: 1.We identified 1101 deaths in 17185 participants.6.0-years cumulative mortality rate was 6.41% and the death density(/1000 person-years)was 10.89.As compared with participants in 6.5 to 7.5 hr/d of sleep,those with shorter duration(<6.5 hr)and longer duration(?9.5 hr/d)were less educated and more frequently widowed/divorced/separated,had a low level of work-related physical activity,and had a higher level of biochemical indicators such as systolic blood pressure(SBP),diastolic blood pressure(DBP),FPG,total cholesterol(TC),triglyceride(TG),and low density lipoprotein cholesterol(LDL-C)(all P<0.001).2.Risk of unadjusted mortality was significantly increased with <6.5 hr/d(HR=2.08,95% CI: 1.53-2.82),and ?9.5 hr/d(HR=2.37,95% CI: 1.85-3.03)as compared with 6.5-7.5 hr/d by the Cox regression models.After adjusting for potential confounding factors,the risk of all-cause mortality increased 35%(1.05-1.73)with ?9.5 hr/d and the mortality risk was nonsignificant with <6.5 hr/d.In the sensitivity analysis,multivariate Cox regression showed that both <6.5 hr/d(HR=1.45,95% CI: 1.01-2.10)and ?9.5 hr/d(HR=1.38,95% CI: 1.02-1.87)were associated with all-cause mortality.The multivariate Kaplan-Meier survival curve showed that the cumulative survival rate was the lower in the population with <6.5 and <9.5 hr/d than other sleep duration categories during the same observation period(P<0.001).3.sleep duration of <6.5 and <9.5 hr/d multiplicatively interacted with age(P<0.001),physical activity(P<0.001),SBP(P<0.001),DBP(P<0.001),hypertension(P<0.001),type 2 diabetes mellitus(T2DM)(P<0.001),TC(P=0.015),HDL-C(P=0.010),LDL-C(P=0.008),and dyslipidemia(P=0.006)in men.For women,sleep duration of <6.5 and <9.5 hr/d multiplicatively interacted with age(P<0.001),physical activity(P<0.001),SBP(P=0.010),DBP(P=0.09),hypertension(P<0.001),T2DM(P<0.001),TC(P=0.037),and LDL-C(P=0.031).The additive interaction was no statistically significant in both men and women.4.The mediation analysis showed that FPG played a mediating role in the association between sleep duration and all-cause mortality.The estimated parameter and 95% CI of total effect was 0.067(0.029-0.105),the direct effect was 0.062(0.024-0.100)and the indirect effect was 0.004(0.027-0.054),suggesting FPG partly mediated the sleep-mortality association.The mediating effect of FPG accounted for 5.97% of the total effect.Other indicators,such as SBP,TC,TG,HDL-C and LDL-C,were not mediators of sleep duration and all-cause mortality.In sensitivity analysis,the conclusion was still valid.5.After adjusting for age,gender,education,marital status,smoking,drinking,physical activity,body mass index(BMI),waist circumference(WC),FPG,SBP,DBP,TC,TG,LDL-C and the interaction between sleep duration with other factors(gender,age,physical activity,hypertension,T2 DM and dyslipidemia),the generalized additive model showed the relation between sleep duration and all-cause mortality was J-shaped.In sensitivity analysis,the conclusion was still valid.Conclusions: 1.In this study,the relation between sleep duration and all-cause mortality was J-shaped,and as compared with 6.5-7.5 hr/d,both short and long sleep duration could increase the risk of all-cause mortality.2.In this study,sleep duration multiplicatively interacted with age,physical activity,hypertension,and T2 DM in both men and women,and for men sleep duration also multiplicatively interacted with dyslipidemia,increasing the risk of all-cause mortality.3.In this study,FPG partly mediated the association between sleep duration and all-cause mortality.