| Background:Diabetes is the third largest non-communicable disease after cardiovascular diseases and tumors.Its incidence is rising worldwide,and the burden of diabetes-related diseases is also getting heavier.It is an important cause of death and health and economic losses.Among them,type 2 diabetes(T2D)accounts for 85-95%of adult diabetes cases.The risk of T2D is closely related to the environment and lifestyle.Early lifestyle changes to control known risk factors are effective measures to reduce disease morbidity and mortality.Therefore,it is very important to identify controllable factors that can reduce the risk of T2D.In addition to traditional risk factors such as high-energy diet,obesity,and sedentary lifestyle,sleep behaviors such as sleep duration,chronotype,insomnia,snoring and excessive daytime sleepiness are also closely related to the onset of T2D.However,most studies evaluate the relationship between a certain sleep behavior and T2D risk separately,and do not consider the relationship between integrated sleep behavior and the risk of T2D.In addition to behavior,genetic factors also have a great influence on the incident of T2D.Polygenic risk scores(PRS)is a tool to quantify individuals’ susceptibility to diseases based on individual genetic variation.Both genetic factors and sleep behaviors are associated to the incident of T2D,but there have not been studies analyzing the joint effect and interaction of sleep and genetic risk on the T2D.Aims:Considering the both sleep behavior and genetic risk may jointly cause the incidence of T2D,the purpose of this study is to analyze the relationship between sleep state combined with sleep duration,chronotype,insomnia,snoring and daytime sleepiness and the risk of T2D;and to construct a PRS that best predicts the onset of T2D;and to finally study the joint effect and interaction of sleep behavior and genetic risk on the risk of T2D,and explore whether the relationship between sleep and T2D risk was altered by genetic susceptibility.Methods:This study used the UK Biobank(UKB)and selected British white people as the research object.We combined sleep duration,chronotype,insomnia,snoring and daytime sleepiness to construct a sleep score,and divided participants into healthy,intermediate and poor sleep state,and used Cox regression model to analyze the relationship between sleep and the risk of T2D.Then we used the traditional method with odds ratio(OR)as the weight,the LDpred and the DBSLMM method to construct polygenic risk scores(PRS)of T2D,compared the predictive ability of these three methods for the onset of T2D,selected the optimal PRS,and estimated the relationship between PRS and T2D risk.Finally,we analyzed the combined effect and interaction of sleep state and PRS on the risk of T2D.Results:1.A total of 271,282 subjects were included in the T2D incidence cohort.From 2007 to 2017,a total of 6,047 people developed T2D,and the cumulative incidence of T2D was 2.23%.The incidence of T2D tends to increase with age,and the incidence of men is higher than that of women.2.Cox regression obtained the relationships between five sleep behaviors and the risk of T2D are as follows:(1)There is a U-shaped relationship between sleep duration and T2D risk,those who sleep 7-8 hours a day have the lowest risk of T2D.After adjusting for multiple factors,taking 7-8 hours of sleep a day as reference,the risk of T2D for those who sleep less than 7 hours or more than 8 hours a day is HR(95%CI)=1.27(1.21,1.34).(2)Early-rising chronotype is a protective factor for T2D.After adjusting for multiple factors,taking the early-rising chronotype as a reference,the risk of T2D in late-sleeping chronotype is HR(95%CI)=1.10(1.04,1.16).(3)Insomnia is a risk factor for T2D.After adjusting for multiple factors,taking never or rarely insomnia as reference,the risk of T2D for people with sometimes or frequent insomnia is HR(95%CI)=1.28(1.20,1.37).(4)Snoring is a risk factor for T2D.After adjusting for age and gender,taking non-snoring as reference,the risk of T2D in snorers is HR(95%CI)=1.49(1.42,1.57);however,after adjusting for multiple factors,the correlation between the two is weak,HR(95%CI)=1.03(0.98,1.08),the difference is not statistically significant.(5)Excessive daytime sleepiness is also a risk factor for T2D.After adjusting for multiple factors,taking the absence of excessive daytime sleepiness as reference,the risk of T2D for people with excessive sleepiness is HR(95%CI)=1.16(1.03,1.30).3.Cox regression shows that the risk of T2D gradually decreased with the sleep score increased.After adjusting for multiple factors,for every point increase in sleep score,the risk of T2D incidence was reduced by 13%.Taking the sleep score of 5 as a reference,the HR(95%CI)for sleep scores of 0 to 1,2,3,and 4 are 1.84(1.57,2.15),1.57(1.36,1.80),1.32(1.15,1.52),and 1.22(1.06,1.41),respectively.The sleep score is divided into three sleep states.After adjusting for multiple factors,the risk of T2D in mediate sleep state is HR(95%CI)=1.19(1.12,1.26),and the risk of T2D in poor sleep state is HR(95%CI)=1.54(1.40,1.70).4.The PRS constructed by the DBSLMM method has higher predictive ability of T2D incidence and lower time cost,AUC(95%CI)=0.686(0.679,0.693).Cox regression shows that for every one standard deviation increase in the PRS,the risk of T2D increases by 1.99(1.94,2.04).The PRS is divided into three categories according to the number of people.Taking the low genetic risk as a reference,the risk of T2D for people with high-risk PRS is HR(95%CI)=4.23(3.92,4.56).5.Cox regression shows that taking "low-risk PRS and healthy sleep state" as a reference,the T2D risk in those with low genetic risk and poor sleeper is HR(95%CI)=2.07(1.61,2.67),and the T2D risk in those with healthy sleep and high genetic risk is HR(95%CI)=2.66(2.29,3.08);the joint effect of poor sleep state and high PRS on T2D risk is HR(95%CI)=3.77(3.16,4.50).6.Stratified analysis shows that poor sleep state is associated with an increased risk of T2D among participants with low,intermediate or high genetic risk.Under the low genetic risk,with healthy sleep state as reference,the T2D risk in poor sleepers is HR(95%CI)=1.92(1.49,2.49);under the high genetic risk,with healthy sleep state as reference,the T2D risk in poor sleepers is HR(95%CI)=1.44(1.26,1.65).Compared with the low genetic risk,the risk effect of poor sleep on T2D in the high genetic risk is weakened(P=0.026),and genetic factors can modify the relationship between sleep and T2D.7.Cox regression shows that poor sleep state and high-risk PRS have a negative interaction with T2D,HR(95%CI)=0.68(0.51,0.91).Conclusions:1.Poor sleep(including long or short daily sleep,late chronotype,insomnia,snoring,and daytime sleepiness)and high genetic risk are risk factors for the incidence of T2D,and the two have a joint effect.Compared with low genetic risk and healthy sleep,people with high genetic risk and poor sleep have the highest risk of T2D.2.Under different genetic risks,poor sleep is a risk factor for T2D,but for people with low genetic risk,poor sleep has a stronger risk effect on T2D,and genetic factors can modify the relationship between sleep and T2D.3.Sleep state and genetic risk have a negative interaction with T2D,and the joint effect of the two is smaller than the product of their individual effects. |
PDF Full Text Request