| ObjectivesIn order to describe the prevalence and changing patterns of the number and pattern clustering of health risk behaviors(HRBs)among Chinese adolescents.To clarify the strength and direction of the association between circadian rhythm disturbances and the number and pattern clustering of HRBs.MethodsThe multi-stage cluster sampling method was adopted to conduct the survey from October to December 2021.A total of 22 868 questionnaires were collected from 8economically developed cities(Shenzhen City,Chongqing City,Zhengzhou City,Shenyang City)and economically underdeveloped cities(Kunming City,Xuzhou City,Taiyuan City,Nanchang City)with balanced consideration of geographical location distribution and economic development level.The survey included general information,chronotype,sleep duration,social jet lag(SJL),HRBs(Smoking,drinking,excessive screen time on weekdays,excessive screen time on weekends,having fast food consumption,skipping breakfast,insufficient fruit intake,insufficient vegetable intake,insufficient physical activity,excessive sugar-sweetened beverage intake,having suicidal ideation,having suicidal behavior,having suicide attempt,having non-suicidal self-injury),anxiety and depressive symptoms.The Morning and Evening Questionnaire5(MEQ-5),the Munich Chrono Type Questionnaire(MCTQ),the modified Questionnaire of Adolescent Health Risk Behaviors,the Patient Health Questionnaire(PHQ-9)and the Generalized Anxiety Disorder(GAD-7)were used to evaluate the chronotype,sleep duration,SJL,HRBs,depression and anxiety symptoms of adolescents at baseline and follow-up.From October to December 2022,based on the baseline survey,two schools were randomly selected from the two districts for the typical analysis.The students from the four schools who participated in the baseline survey were followed up.After matching the baseline sample with the follow-up sample,1 769 valid questionnaires were included.SPSS 23.0 software was used for data statistical analysis,the rate of categorical variables(%)was expressed,and the difference between groups was compared by chi-square test.then,latent class analysis(LCA)was used to analyze the clustering pattern of HRBs in adolescents.Multivariate logistic regression model was established to clarify the strength of the association between circadian rhythm disturbances and the number and pattern of HRBs clusters after controlling for confounding factors.For the follow-up data,latent transformation analysis(LTA)was used to explore the change pattern of HRBs clustered in adolescents.The autoregressive cross-lag model was used to analyze the direction and correlation coefficient of the association between circadian rhythm disturbances and the number of HRBs clustered in adolescents.ResultsAt baseline survey,the prevalence of morning types,neutral types,and evening types were 21.0%,62.8%,and 16.2%,respectively.The detection rates of sleep duration< 7 h,7-9 h and > 9 h were 16.0%,68.8% and 15.2%,respectively.The detection rates of social jet lag ≥1 hour and social jet lag < 1 hour were 46.0% and 54.0%,respectively.The detection rates of HRBs at baseline were between 6.5% and 63.8%.In this study,the number of clustered HRBs was divided according to the quartiles.After adjusting for confounding factors,students who had eveing types and neutral types were associated with having 2~3 kinds of HRBs,4~5 kinds of HRBs,and 6~14 kinds of HRBs.Students who had sleep duration > 9 hours were positively associated with 6~14 kinds of HRBs.students who had sleep duration < 7 hours were positively associated with having 2~3 kinds of HRBs,4~5 kinds of HRBs,6~14kinds of HRBs.Students who had SJL≥1 h were positively correlated with having 2~3 kinds of HRBs,4~5 kinds of HRBs and 6~14 kinds of HRBs.The LCA model divided 14 HRBs into low risk group,smoking-drinking and screen group,as well as self-injury risk group,the detection rates were 70.8%,9.9%,and 19.3%,respectively.Multiple logistic regression analysis with moring types,7-9hours sleep and social jet lag < 1 hour as reference groups showed that students who had morning types and neutral types were associated with smoking-drinking and screen group and self-injury risk group.Students who had sleep duration > 9 hours was positively associated with smoking-drinking and screen group and self-injury group,while students who had sleep < 7 hours were only associated with self-injury risk group.Students who had SJL > 1 h were positively associated with smoking-drinking and screen group and self-injury risk group.Stratified analysis of circadian rhythm disturbance indicators and the number of HRBs clusters showed that adolescents who had morning type and neutral types with sleep duration < 7 h were positively correlated with 2~3 kinds of HRBs,4~5 kinds of HRBs,and 6~14 kinds of HRBs.However,adolescents who had neutral types with sleep duration > 9 h were only positively correlated with 6~14 kinds of HRBs.There was no significant difference among different sleep duration in adolescents with eveing types.Stratified by chronotype,compared with adolescents with SJL < 1 h among chronotype,adolescents with SJL≥1 h were positively associated with 2 ~ 3kinds of HRBs,4~5 kinds of HRBs,and 6~14 kinds of HRBs.Stratified by sleep duration,compared with adolescents with SJL < 1 h among sleep duration group,SJL≥1h was positively associated with 2~3 kinds of HRBs,4~5 kinds of HRBs,and 6~14kinds of HRBs.Stratified analysis of circadian rhythm disturbance indicators and HRBs clustering patterns showed that the morning types with sleep duration < 7 hours was positively correlated with self-injury risk group,while the neutral types with sleep duration < 7hours was positively correlated with smoking-drinking and screen group.Adolescents who had neutral types and evening types with sleep duration > 9 h were positively correlated with self-injury risk group.When stratified by chronotype,compared with adolescents with SJL < 1 h among chronotype,those with SJL≥1 h were positively correlated with smoking-drinking and screen group and self-injury risk group.In adolescents stratified by sleep duration,compared with those with SJL < 1 h among sleep duration group,SJL≥1 h was positively correlated withmoking-drinking and screen group and self-injury risk group.In the follow-up survey,after comparing the number of clustered behaviors between baseline and follow-up,339 adolescents(19.2%)had no change in the number of clustered behaviors.701 adolescents(39.6%)showed increased risk behaviors;729adolescents(41.2%)reported a reduction in risk behaviors.In the follow-up survey,after comparing the pattern of clustered behaviors between baseline and follow-up,Adolescents in the low-risk group at baseline were more likely to maintain a relatively healthy lifestyle and have fewer HRBs.Adolescents in the self-injury risk group were more likely to transition to smoking-drinking and screen group(72.7%)than to low-risk group(18.2%).The probability of transition to the three behaviors of the smoking-drinking and screen group was relatively balanced.The results of the autoregressive cross-lag model of circadian rhythm disturbances and clustered HRBs showed that the score of chronotype at baseline(T1)had marginal significance in the negative prediction of the number of cluster behaviors at the follow-up(T2)(β=-0.15,P<0.05),and the number of cluster behaviors at T1 had a negative prediction effect on the score of chronotype at T2(β=-0.35,P<0.05).Social jet lag at T1 had a positive predictive effect on the number of clustered behaviors at T2(β=0.04,P<0.05),and the number of clustered behaviors at T1 had the strongest positive predictive effect on social jet lag at T2(β=0.11,P<0.05).Sleep duration at T1 had a negative predictive effect on the number of cluster behaviors at T2(β=-0.03,P<0.05),and the number of cluster behaviors at T1 had a marginal significance on the positive prediction of sleep duration at T2(β=0.01,P<0.05).ConclusionsThe circadian rhythm disturbances was significantly associated with higher number and worse patterns of clustered HRBs,and there was a bidirectional predictive effect between circadian rhythm disturbances and the number of clustered behaviors.Interventions targeting unhealthy chronotype,sleep duration and social jetlag in adolescents should be taken in the future to reduce the number and severity of multiple HRBs. |
PDF Full Text Request