The Association Between Restless Legs Syndrome And Suicidal Behaviors Among Rural Adolescents In Shandong Province

1.BackgroundHundreds of thousands of adolescents around the world die by suicide every year,imposing heavy burdens and great loss on families and societies.Adolescent suicide is a major public health concerns across the world.Suicidal behaviors including suicidal ideation(SI),suicide plan(SP),and suicide attempt(SA)are the significant predictors of future suicide attempts and suicide death.It is of great significance to carry out research on suicidal behaviors among adolescents to prevent suicide and improve the strategies of adolescent suicide prevention.Restless legs syndrome(RLS)is characterized by uncomfortable feeling or pain in limbs,such as ant walking,tingling,numbness,burning,etc.,which are more common in legs.The feelings urge sufferers to move their limbs,especially at night,when rest and sleep are most desired,which affects sleep or life quality.RLS can also cause anxiety,depression,cognitive decline and other problems.The influence of RLS may aggravate among adolescents due to their physiological and psychological characteristics,leading to risk behaviors.The association between RLS and suicide or self-harm may be relative to the mediating effects of depression and sleep problems.To our knowledge,there are no studies on associations between RLS and suicidal behaviors among adolescents.Using baseline data from Shandong Adolescent Behavior and Health Cohort(SABHC),the current study aimed to preliminarily explore the associations between RLS and suicidal behaviors among adolescents.In addition,we assumed that insomnia and depression played roles in the path of the associations and analyzed the mediating effects of depression and insomnia separately on the relationship between RLS and suicidal behaviors among adolescents.2.ObjectivesIn the study,we aimed:(1)To explore the association between RLS and suicidal behaviors among adolescents.(2)To analyze the mediating effects of depression and insomnia separately on the relationship between RLS and suicidal behaviors among adolescents.3.Method3.1 SubjectsThe Shandong Adolescent Behavior and Health Cohort(SABHC)baseline survey was conducted in three counties(Lijin Yanggu,and Zoucheng)of Shandong Province,China in 2015.Eight high schools(five junior schools and three senior schools)were selected in the 3 counties.All 7th graders and 10th graders were selected to participate in the study,and half of students from the 8th,9th,and 11th grades were randomly selected using class as sampling units.12th graders were not selected because they were preparing for college entrance exams.3.2 MeasurementThe survey was conducted during normal school hours.The Adolescent Health Questionnaire(AHQ),a self-administered structured questionnaire,was used to collect information including history of RLS,suicidal behaviors,depression,sleep problems,and adolescent and family demographic information.Before the survey carried out,we got the permission of the principals of target schools,the head teachers of target classes and participants.Our trained interviewers administered the AHQ to the participants in the target classes during their school time.Before the students filled out the survey,they were informed the purpose of the study and the procedure.They were also told that the survey was anonymous,and their participation is voluntary and independent of school evaluations.The interviewers instructed students to read the instructions carefully and answered their questions about the AHQ during the survey.The questionnaires were completed in about 45-50 minutes.3.3 Statistical analysesAll statistical analyses were performed using SPSS version 24.0 and Stata version 15.0.For quantitative variables,if it was normally distributed,(means±standard deviation)was used to describe data and t-test was used to make statically comparison.If it was not normally distributed,the median(P25,P75)was used to describe data and non-parametric Mann-Whitney U test was used to make statically comparison.For the categorical variables,frequency and proportion were used to describe data and ?2 test or Fisher's exact probability method was used to make statically comparison.Logistic regression analyses were performed to explore the association between RLS and suicidal behaviors.Odd ratio(OR)and 95%confidence interval(95%CI)were calculated.We established 2 models with RLS as the independent variable and suicidal behaviors as the dependent variables.In Model 1,we examined the associations between RLS and suicidal behaviors after adjusting for adolescent and family demographic covariates.In Model 2,we added insomnia,daytime sleepiness and depression to Model 1.The mediating effects of depression and insomnia on the relationships between RLS and suicidal behaviors were tested by the KHD method based on Logistic regression model.Differences were tested using two-tailed tests and P values less than 0.05 were considered statistically significant.4.Results4.1 Sample characteristicsOf the 12301 participants in target classes,465 students were not in school or did not submit questionnaires and 11836 students participate in the study.The response rate was 96.2%.Five questionnaires with blank information were excluded,so 11831 questionnaires were included for statistical analysis.Of 11831 students,mean age was 14.97(SD=1.46)and 50.9%were boys.4.2 Restless Legs SyndromeAmong the 11831 participants,9.5%(1121)met the diagnostic criteria for RLS,7.3%(863)reported RLS symptoms<3 times/week and 2.2%(258)reported RLS symptoms?3 times/week.Adolescents with RLS are more likely to be older,girls,ever smoking,ever drinking,insomnia,daytime sleep,depression,the parents'divorce/one partner dies,poor physical health and poor family economic status(all P<0.05).There was no significant difference in parental education level(all P>0.05).4.3 Suicidal behaviors among adolescentsAmong the 11831 adolescents,the prevalence of past-year SI,SP and SA was 12.5%,3.3%and 1.5%,of lifetime SI,SP,and SA was 20.5%,9.6%and 3.9%,respectively.Univariate Logistic regression analyses of suicide behaviors show that female,older age,ever smoking,ever drinking,insomnia,depression,daytime sleep,poor family economic condition and poor physical health are associated with past-year SI,SP,and SA(all P<0.05)and female,older age,ever smoking,ever drinking,insomnia,depression,daytime sleep,poor family economic condition,high parental education level and poor physical health are associated with lifetime SI,SP,and SA(all P<0.05).4.4 Association between RLS and suicidal behaviorTaking the RLS(yes or no)as the independent variable,Model 1 showed that RLS was significantly associated with past-year and lifetime SI,SP,and SA(all P<0.05).Model 2 showed that RLS was significantly associated with past-year SI(OR=1.23,95%CI=1.03-1.46),SP(OR=1.3 8,95%CI=1.03-1.83)and lifetime SI(OR=1.18,95%CI=1.01-1.37).After stratification by gender,Model 2 showed that RLS was significantly associated with past-year SI(OR=1.44,95%CI=1.11-1.87),and lifetime SI(OR=1.33,95%CI=1.06-1.66),SP(OR=1.37,95%CI=1.04-1.80),and SA(OR=1.54,95%CI=1.03-2.29)among male adolescents.There are no associations among female adolescents(all P>0.05).Taking the frequency of RLS(no,<3 times/week or?3 times/week)as the independent variable,Model 2 showed that the frequency of RLS<3 times/week was not associated with suicidal behaviors(all P>0.05).and the frequency of RLS?3 times/week was significantly associated with past-year SI(OR=1.61,95%CI=1.18-2.21),SP(OR=2.03,95%CI=1.30-3.14),and lifetime SI(OR=1.64,95%CI=1.23-2.18),SP(OR=1.50,95%CI=1.08-2.09),SA(OR=1.59,95%CI=1.02-2.49).After stratification by gender,Model 2 showed that the frequency of RLS?3 times/week was significantly associated with past-year SI(OR=2.01,95%CI=1.33-3.02),SP(OR=2.31,95%CI=1.29-4.13)and lifetime SI(OR=1.96,95%CI=1.36-2.85),SP(OR=1.86,95%CI=1.22-2.86)and SA(OR=1.92,95%CI=1.07-3.43)among male adolescents.There are no associations among female adolescents(all P>0.05).4.5 Mediating effectAfter controlling for other covariates,the results show that depression and insomnia had significant mediating effects on the relationship between RLS and past-year SI,which accounted for 20.62%and 6.15%of the total effects,respectively.Depression and insomnia had significant mediating effects on the relationship between RLS and lifetime SI,which accounted for 21.41%and 9.15%of the total effects,respectively.Depression and insomnia had significant mediating effects on the relationship between RLS and past-year SP,which accounted for 12.54%and 7.87%of the total effects,respectively.Depression and insomnia had significant mediating effects on the relationship between RLS and lifetime SP,which accounted for 19.26%and 13.13%of the total effects,respectively.Depression and insomnia had significant mediating effects on the relationship between RLS and lifetime SA,which accounted for13.05%and 11.21%of the total effects,respectively.5.Conclusion(1)RLS is related to suicidal behaviors among adolescents.(2)Depression and insomnia can play mediating effects in the association between RLS and suicidal behaviors.6.Suggestion(1)Schools and families should actively guide the adolescents with RLS to prevent them from sleep and emotional problems and suicidal behaviors.(2)Schools should strengthen the monitoring of the symptoms of restless leg syndrome among students.Adolescents with frequent RLS need to be paid more attention.