Portal de Periódicos da CAPES

The Moderating Role of Parents' Dysfunctional Sleep-Related Beliefs Among Associations Between Adolescents' Pre-Bedtime Conflict, Sleep Quality, and Their Mental Health

2019; American Academy of Sleep Medicine; Volume: 15; Issue: 02 Linguagem: Inglês

10.5664/jcsm.7630

1550-9397

Jack S. Peltz, Ronald D. Rogge,

Youth Substance Use and School Attendance

Free AccessSleep Quality - Sleep Latency - Sleep Duration - Infants - Adolescents - Anxiety - Insomnia - Light - Environment - Cognition - Bedtime - Behavior - Scientific InvestigationsThe Moderating Role of Parents' Dysfunctional Sleep-Related Beliefs Among Associations Between Adolescents' Pre-Bedtime Conflict, Sleep Quality, and Their Mental Health Jack S. Peltz, PhD, Ronald D. Rogge, PhD Jack S. Peltz, PhD Address correspondence to: Jack S. Peltz, Daemen College, 4380 Main Street, Amherst, NY 14226(716) 839-8228 E-mail Address: [email protected] Daemen College, Amherst, New York The University of Rochester Medical Center, Rochester, New York , Ronald D. Rogge, PhD University of Rochester, Rochester, New York Published Online:February 15, 2019https://doi.org/10.5664/jcsm.7630Cited by:4SectionsAbstractPDF ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTStudy Objectives:The current study's aim was to examine the indirect effect of parent-child pre-bedtime arguing about the bedtime process on adolescents' symptoms of anxiety and depression via the mediating role of adolescents' sleep quality. In addition, this study sought to test this mediation model across different levels of both parents' and children's dysfunctional sleep-related beliefs (ie, moderated mediation).Methods:A total of 193 adolescent (mean age = 15.7 years, standard deviation [SD] = .94; 54.4% female) and parent dyads completed both baseline, online surveys, and online 7-day, twice-daily sleep diaries. Parents (mean age = 47.6 years, SD = 5.4; 80% female) reported daily for 7 days on the intensity of any conflict regarding the adolescents' bedtime process, and adolescents completed daily reports of their sleep duration and quality (morning diary) and their anxiety and depressive symptoms (evening diary).Results:Results suggested that adolescent sleep quality mediated the indirect association between parent-child pre-bedtime arguing and adolescents' anxiety and depressive symptoms. Furthermore, this mediation model was moderated by parents' dysfunctional sleep-related beliefs. Only in families with parents reporting either average or above-average (+1 SD) levels of dysfunctional beliefs did this mediation model emerge as significant.Conclusions:Results provide further evidence for the essential role of the family environment in adolescent sleep and well-being, and they suggest that parents' dysfunctional sleep-related cognitions put adolescents at risk for a negative cascade stemming from arguing over bedtime to poor-quality sleep and its negative consequences on their mental health.Citation:Peltz JS, Rogge RD. The moderating role of parents' dysfunctional sleep-related beliefs among associations between adolescents' pre-bedtime conflict, sleep quality, and their mental health. J Clin Sleep Med. 2019;15(2):265–274.BRIEF SUMMARYCurrent Knowledge/Study Rationale: Family conflict has been shown to influence both adolescents' sleep quality and mental health. We sought to examine how both parents and adolescents' dysfunctional sleep-related cognitions moderated these associations.Study Impact: Given the epidemic of adolescent sleep problems and its consequences, parents' dysfunctional sleep-related beliefs must be considered because of the risks that they pose to promoting a negative cascade from family conflict to poor adolescent sleep and mental health problems. These findings suggest new avenues, such as parents' beliefs or attitudes about sleep, that must be addressed to support healthy sleep and mental well-being in adolescents.INTRODUCTIONThe Role of Dysfunctional Sleep-Related Beliefs in Adolescent Sleep and Mental HealthThere is extensive evidence documenting that adolescents' insufficient and poor-quality sleep plays a critical role in their academic, behavioral, and psychosocial functioning.1–5 Highlighting the proposals by Owens and Mindell,6,7 the evidence linking adolescent sleep problems and their negative sequelae has created a greater need for research on not only the predictors of adolescent sleep problems (ie, modeling sleep problems as a mediator linking those predictors to adverse adolescent functioning), but also the conditions under which such problems might emerge (ie, identifying broader family factors that might moderate when those associations emerge). Carskadon has referred to the intersection of the biological, social, cultural, and psychological changes associated with adolescence and their effect on adolescents' sleep as “the perfect storm.” Not only do the hormonal changes due to puberty affect the sleep/wake cycle (eg, delayed sleep phase) and chronotype, but behavioral (eg, greater use of pre-bedtime electronic media) and environmental (eg, early school start times) influences often result in adolescents' sleep deprivation.8 One key predictor to emerge in this “perfect storm” of adolescent sleep problems is conflict within the family.9–11 In fact, it has been shown that family environments marked by high levels of conflict can negatively influence adolescents' sleep.10–12 The current study sought to build on this previous research by examining a moderated mediation model in which family conflict (the predictor) promotes adolescent anxiety and depressive symptoms (the outcome) by reducing adolescent sleep quality (the mediator), particularly in families with dysfunctional sleep-related beliefs (the moderator).Family Conflict and SleepOne way of explaining the incompatibility between family conflict and sleep comes from evolutionary psychology.12 This perspective suggests that a stable and safe family environment, in which an individual no longer requires the vigilance typically maintained in daily life to protect oneself from external threats, is essential to healthy and restorative sleep.13–15 This association has been established in multiple studies involving adolescents, and the evidence suggests that family environments that support better sleep tend to involve order or routines,16,17 the establishment of parent-set bedtimes,18 positive parental involvement,19 and the absence of family conflict.11 For instance, Gregory and colleagues showed that higher levels of expressed anger and aggression between family members experienced by children (ages 9 to 15 years) predicted their symptoms of insomnia at age 18 years.11 In their study, the authors asserted that conflict within the family potentially resulted in the development of youth's poor sleep habits and ultimately their insomnia.11 It is also possible that the pubertal changes associated with adolescence and their subsequent impact on the adolescent's chronotype might play a role in increasing parent-child conflict around bedtime.8 Specifically, adolescents' delayed sleep phase might conflict with parents' bedtime expectations that their child should be getting ready for bed, especially if it is a school night. In the current study, we investigate such a model of indirect effects by examining the amount of parent-child conflict regarding adolescents' bedtimes or bedtime routines (reported by parents each evening) as influences of adolescents' sleep quality (reported by adolescents each morning) and, subsequently, their sleep quality's potential effect on their anxiety and depressive symptoms over the course of the day (reported by adolescents each evening).Dysfunctional Sleep-Related Beliefs as a Potential ModeratorIn addition to examining this process-focused mediation model (ie, the indirect effect of parent-child pre-bedtime arguing on adolescent mental health), the current study also sought to test for whom such processes would most likely occur (ie, what might moderate the indirect effects model). Specifically, for both adolescents and adults, dysfunctional beliefs about one's own sleep are associated with poorer quality sleep.20,21 In contrast to older individuals, however, adolescents appear to have more distorted sleep-related beliefs, such as the belief that one can recover lost sleep by sleeping for a long time.20 Other dysfunctional beliefs include remaining in bed and trying harder to fall asleep when one has trouble getting to sleep and the fear that one might lose control over one's ability to fall asleep.22 Furthermore, dysfunctional sleep-related beliefs not only have been shown to differentiate insomnia sufferers from good sleepers, but it has also been demonstrated that individuals with more dysfunctional sleep-related beliefs are more at risk of stress-related sleep disturbance.21,23 Given these associations, it is not surprising that the effects of cognitive behavioral therapy for insomnia have been shown to be mediated by changes in dysfunctional sleep-related beliefs.24Beyond its role as a predictor in models of adolescent sleep, we assert that dysfunctional beliefs and attitudes about sleep might exert their strongest effects at the contextual level as a moderator, as those dysfunctional beliefs could have the power to not only shape an adolescent's or parent's own behavior, but also adversely influence their reactions to the behaviors of others (eg, parenting efforts around bedtime). As a result, parents or adolescent-aged children with dysfunctional sleep-related beliefs might be more prone to react to issues around bedtime in a negative and dysregulated manner. From this conceptualization, dysfunctional sleep-related beliefs serve as a contextualizing factor that can predispose families to greater conflict and negative sequelae surrounding adolescent sleep. Building on this, recent research has shown that parents and adolescents' sleep habits evidence high levels of concordance,25 and could therefore potentially create family environments poised for chronic conflict surrounding adolescent sleep.To date, no reported studies have examined how a parent's sleep-related beliefs might be related to their adolescent-age children's sleep, nor have sleep-related beliefs of parents or adolescents been conceptualized as a contextual factor moderating models of sleep and adolescent functioning. Although research supports the notion that parenting beliefs influence parent-child interactions and children's behaviors (see Sigel et al.26), this area of research has not been extended specifically to models of adolescent sleep, although exceptions exist (for exceptions from the infant sleep literature, see refs.27–29). For example, Teti and Crosby found that higher levels of maternal worries about their infants' needs at night predicted mothers' increased presence in the infants' rooms, which, in turn, predicted higher rates of infants' night awakenings.28 Although adolescence presents a distinct set of parental concerns and behaviors regarding their children's sleep, parents' sleep-related beliefs remain a salient feature of the adolescent's sleep environment to be explored in the current study.The Current StudyDespite the demonstrated links between the sleep habits of parents and their adolescent-aged children,25 research has not previously provided evidence that parents' dysfunctional sleep-related beliefs might serve to differentiate how processes linking conflict and adolescent mental health problems emerge in families. Accordingly, the aim of the current study was to examine a moderated mediation model within an online sample of 193 parent-adolescent dyads who had completed 7-day morning and evening, sleep and mood diaries. In the model, parents' dysfunctional sleep-related beliefs were allowed to moderate the indirect effects of parent-child pre-bedtime arguing on adolescents' anxiety and depressive symptoms via their sleep quality. In our mediation model, we first hypothesize that (1) higher levels of pre-bedtime arguing between the parent and child will predict poorer adolescent sleep quality (Figure 1, path i); (2) poor adolescent sleep quality will predict higher levels of adolescents' anxiety and depressive symptoms (Figure 1, path ii); and (3) pre-bedtime arguing will be indirectly associated with adolescent mental health symptoms via adolescents' sleep quality (Figure 1, top panel). Furthermore, in our second hypothesis, we expect that both parents' and children's dysfunctional sleep-related beliefs will moderate these indirect associations such that this model will be more likely to emerge in families with parents and adolescents with more dysfunctional beliefs about sleep (Figure 1, bottom panel).Figure 1: Conceptual models of mediation (top) and moderated mediation (bottom).Download FigureMETHODSParticipants and RecruitmentParticipants were adolescent-parent dyads (n = 193), who were recruited through direct solicitation (eg, receiving a study brochure following a brief presentation at school), emails to distribution lists (eg, parenting groups), and through ResearchMatch, a national health volunteer registry that was created by several academic institutions and supported by the US National Institutes of Health as part of the Clinical Translational Science Award program. ResearchMatch has a large population of volunteers who have consented to be contacted by researchers about health studies for which they may be eligible. In order to participate, adolescents had to be in 9th through 11th grades in either a public or private day school within the United States, between the ages of 14 and 17 years, living 7 days per week in the participating family's household, and both parent and child had to agree to participate. Adolescents (and their families) were excluded from the study if their parents reported that the adolescent had a severe cognitive limitation (eg, intellectual disability) that would prevent him/her from completing the surveys and any major medical problems that might impair sleep, such as sleep apnea. Recruitment was targeted at parents, and any parent expressing interest was taken through an initial verbal informed consent process and screened for eligibility via a phone call. After parents gave verbal consent and confirmed their eligibility, we then spoke to the adolescents in those households to begin the verbal assent process as part of those same phone calls. Interested parent-adolescent dyads provided personal email addresses so that they could be sent invitations to complete both the baseline surveys and the daily diaries.A total of 193 adolescent (mean age = 15.7 years, standard deviation [SD] = .94; 54.4% female) and parent (mean age = 47.6 years, SD = 5.4; 80% female) dyads completed the baseline and 7-day sleep diary surveys, and parents provided data from the baseline survey on their sleep-related beliefs. The adolescents reported being in 9th (37%), 10th (32%), or 11th (31%) grade. Most of the adolescents and parents identified as Caucasian (71% and 79% respectively), with another 14% and 14% (respectively) identifying as African American, 8% and 2% (respectively) identifying as multiracial, 3% and 2% (respectively) identifying as Latino/a, 2% and 2% (respectively) identifying as Asian American, and 2% and 1% (respectively) identifying as “other.” Parents had relatively high levels of education, with approximately 42% reporting a graduate degree, 35% with a BA/BS, 19% with some college or an associate's degree, and 4% with a high school diploma or General Equivalency Diploma or less. Parents reported an average annual income of $81,600 (SD = $27,800), and 17.6% of families reported incomes equal to or less than $45,000.ProcedureThe study was approved by the local Institutional Review Board and informed consent from parents and assent from adolescents was obtained prior to participation. The baseline survey took approximately 20 to 25 minutes to complete; respondents were compensated $10 each as an incentive. During the baseline survey, parents set a start date for the dyad to concurrently complete the 7-day sleep diary (targeting what would be a typical school week for that family). The morning diary survey (ie, sleep-related assessments) was completed within an hour of waking up, and the evening diary survey (ie, adolescents' reports of mood-related assessments and parents' reports of pre-bedtime arguing) was completed within an hour of going to sleep. As incentives for completing daily diaries, respondents received $15 for completing a minimum of 4 morning and evening diary entries, an entry to win a lottery prize (an iPad mini) for every diary entry completed, and brief feedback on their sleep (eg, average bed/waketimes based on the diary data they provided) following the conclusion of the data collection.AttritionA total of 178 adolescents (92.2%) and 176 parents (91.2%) completed at least 4 days of the daily diaries, with participants (parents and children) completing on average approximately 11.7 diary entries out of a possible 14 (SD = 2.8). Analysis of variance and χ2 analyses suggested that the respondents participating in the daily diaries did not differ from participants who only completed the baseline survey across all primary variables and demographic covariates.MeasuresDysfunctional Beliefs About SleepTo assess both parents' and adolescents' dysfunctional beliefs about sleep, respondents completed the 10-item Dysfunctional Beliefs About Sleep scale (DBAS-10) during the initial assessment.22,30 The items of the DBAS-10 assess sleep-disruptive cognitions that include maladaptive beliefs, expectations, attitudes, and attributions about the respondent's sleep (eg, “When I have trouble getting to sleep, I should stay in bed and try harder,” “I am worried that I may lose control over my abilities to sleep,” “When I sleep poorly on one night, I know it will disturb my sleep schedule for the whole week.”). Originally, the questionnaire used a visual analog scale on a 100-mm line to indicate one's level of agreement with each.30 To capture the original intent while using an online format, the items were rated on 100-point sliding scales (strongly disagree – 0 to strongly agree – 100), with higher scores reflecting overly strong or rigid endorsements of sleep-related beliefs, which can impede the sleep process.31 Scores were averaged so that higher scores indicated more maladaptive sleep-related beliefs, and the scale demonstrated reasonable internal consistency (αparents = .75; αadolescents = .69).Pre-Bedtime ArguingThe amount of arguing about bedtime or having the child get ready for bed was reported by parents each evening across the 7-day sleep diary with a 1-item measure (ie, “How much did you and your child argue about bedtime or getting ready for bed?”). The item was rated on a 5-point response scale (not at all – 1 to a lot – 5), and scores were calculated by taking the average across the 7 days, with higher scores indicating higher levels of arguing before the child went to bed.Sleep QualitySleep quality was self-reported by the adolescents with a four-item measure in the morning for each of the seven consecutive days on the daily diary. Adapted from the recommendations of Ohayon and colleagues,31 the measure included overall sleep quality (“Indicate how you slept.”) that was rated on a five-point response scale (very restless – 1 to very sound – 5), sleep latency (approximate number of minutes before falling asleep – reverse-scored), the level of restedness (“Indicate how rested you felt upon awakening.”) that was rated on a five-point response scale (exhausted – 1 to very refreshed – 5), and the number of wake-after-sleep episodes (reverse-scored). Scores were standardized and summed with higher scores indicating better sleep quality (α = .63).Sleep DurationSleep duration was assessed in the morning diaries and was calculated by taking the average across the 7 days of the daily differences between the recorded time the child reported going to sleep and waking up the next morning, with both sleep latency (minutes) and wake after sleep onset episodes (minutes) having been subtracted from each night's sleep duration.Psychological FunctioningTo assess adolescents' psychological functioning, respondents self-reported adapted versions of the Patient Health Questionnaire-4 in the evening portion of their daily diary.32 This measure has demonstrated strong reliability and validity in adolescent samples.33,34 Respondents reported how much they had been bothered by the following symptoms since waking up that morning: “feeling nervous, anxious or on edge,” “not being able to stop or control worrying,” “little interest or pleasure in doing things,” and “feeling down, depressed, or hopeless.” The items were rated on four-point response scales (not at all to nearly all day), were summed so that higher scores indicated higher levels of anxiety and depressive symptoms, and were then averaged across the 7 days of diary reports (α = .83).Analytical PlanTo test both our mediation and our moderated mediation models (Figure 1), we used the PROCESS macro (models 4 and 59, respectively) for SPSS (v. 23).35 Our mediation model allowed us to test our first set of hypotheses regarding the indirect effect of parent-reported pre-bedtime arguing on adolescents' self-reported anxiety/depressive symptoms via adolescents' self-reported sleep quality. Following best practices, our moderated mediation model allowed us to evaluate if parents' dys-functional beliefs about sleep moderated any of the paths, both direct and indirect, within the mediational model simultaneously.36 Both models tested the significance of those indirect paths by using bootstrapping to accommodate the asymmetry in their confidence intervals.37 Bootstrapping assesses the effects of variables in a manner that maximizes power and is robust against non-normality; in the current study, we used a 95% bias-corrected bootstrap model with confidence intervals resampled 10,000 times for each analysis. The variables age, sex, diary-reported average daily sleep durations, and family-level socioeconomic status were included as covariates (ie, additional predictors) in the model to help control for their potentially confounding effects.RESULTSPreliminary AnalysesDescriptive statistics for the sample and intercorrelations among the key variables are presented in Table 1. Parents in the sample reported a range of parent-child arguing about bedtime or getting ready for bed, with approximately 31% of parents reporting at least some level (ie, greater than “none at all”) of these pre-bedtime arguments across the 7 days of the diary—most of them being at a fairly low intensity. Furthermore, across the 7 days of the daily diary, adolescents gave moderately high average ratings of sleep quality (mean = 3.99 on a 5-point scale; SD = 1.1), reported an average of 22.20 minutes (SD = 26.1) of sleep latency, an average of 0.62 (SD = 1.0) episodes of wake after sleep onset, and moderate average levels of restedness (mean = 3.13 out of 5; SD = 1.2). This suggests that the adolescents in the study got moderately satisfying and refreshing sleep across the daily diary assessments but also evidenced a reasonable range of sleep quality. Although we did not specifically include either bedtimes or waketimes in our analyses (apart from their inclusion as components of adolescent sleep duration), adolescents reported an average bedtime of 11:03 pm (SD = 1.1 hours) and an average waketime of 7:14 am (SD = 1.0 hours) across the daily diary.Table 1 Psychometrics and bivariate correlations between study variables.Table 1 Psychometrics and bivariate correlations between study variables.Regarding bivariate correlations between study variables across the sample (Table 1), adolescents' dysfunctional beliefs about sleep were positively associated with both their average levels of anxiety and depressive symptoms and female sex (male = 0; female = 1) and negatively associated with their average sleep quality during the week. Parents' reports of parent-child arguing about bedtime was positively associated with adolescents' reports of their anxiety and depressive symptoms and negatively associated with their sleep quality. In addition, adolescents' anxiety and depressive symptoms were negatively associated with both their sleep quality and their family-level socioeconomic status (SES) and positively associated with female sex. Finally, adolescents' average sleep duration was negatively correlated with both their age and their family-level SES and positively associated with their sleep quality.Mediation Model (Figure 2)Figure 2: Model of results from mediation analyses.*** = P < .001. CI = confidence interval.Download FigureAs shown in Table 2, the mediation model predicted approximately 11.3% of the variance in adolescents' self-reported average sleep quality over the course of the week, F5,1296 = 33.0, P < .001. Specifically, parent-reported parent-child arguing over bedtime (B = −1.0, P < .001), child's age (B = −.23, P < .01), and child's female sex (B = −.91, P < .001) significantly predicted lower adolescent sleep quality, whereas longer sleep duration (B = .58, P < .001) predicted higher sleep quality.Table 2 Results from mediation model of pre-bedtime arguing on adolescent mental health symptoms.Table 2 Results from mediation model of pre-bedtime arguing on adolescent mental health symptoms.The mediation model also predicted approximately 27.8% of the variance in adolescents' self-reported average anxiety and depressive symptoms over the course of the week, F6,1295 = 82.9, P < .001 (Table 2). Specifically, higher levels of parent-reported arguing (B = .58, P < .001) and child's female sex (B = .58, P < .001) predicted higher anxiety and depressive symptoms, whereas better sleep quality (B = −.29, P < .01), longer sleep duration (B = −.10, P < .05), and higher family-level SES (B = −.25, P < .001) predicted lower anxiety and depressive symptoms. In support of hypothesis 1c, adolescents' sleep quality significantly mediated the association between parent-reported pre-bedtime arguing and adolescents' anxiety and depressive symptoms as evidenced by a significant indirect effect (B = .30, standard error [SE] = .07; 95% confidence interval [CI] .16, .45). Taken together, these results suggest that higher levels of arguing between parents and their adolescents before bed were associated with lower levels of adolescent sleep quality, which, in turn, were associated with higher levels of their anxiety and depressive symptoms.Moderated Mediation Model: Parents' Dysfunctional Sleep Related Beliefs as ModeratorDirect EffectsAs shown in top half of Table 3, the results further suggested that parents' dysfunctional beliefs about sleep moderated the associations examined, F7,1273 = 24.4, P < .001. Specifically, for parents with either average (B = .48, SE = .16; P < .01) or above-average (+1 SD; B = .83, SE = .18; P = .001) levels of dys-functional sleep-related beliefs, higher levels of parent-child pre-bedtime arguing directly predicted significantly higher levels of adolescents' anxiety and depressive symptoms. However, this direct association was not significant for parents reporting below- average (−1 SD) levels of dysfunctional beliefs (B = .16, SE = .26; not significant). These results serve to highlight how parents' dysfunctional sleep-related beliefs can function as a key contextual factor informing models of adolescent functioning by helping to identify families at risk for bedtime arguing adversely affecting adolescent mental health.Table 3 Results of the direct and indirect paths in the moderated mediation models.Table 3 Results of the direct and indirect paths in the moderated mediation models.Indirect EffectsIn support of our second hypothesis, the results also suggested that parents' dysfunctional sleep-related beliefs moderated the mediational paths within the model (Table 3). Controlling for the direct effects previously described, the model suggested that significant mediation emerged only in those families with parents with average (B = .27, SE = .09; 95% CI .09, .45) or above-average levels (B = .31, SE = .10; 95% CI .13, .53) of dysfunctional sleep-related beliefs. Consistent with our second hypothesis, in families with parents holding average or above-average levels of dysfunctional beliefs about sleep, parent-child pre-bedtime arguing significantly predicted lower daily sleep quality, which, in turn, predicted higher levels of daily anxiety and depressive symptoms.Moderated Mediation Model: Children's Dysfunctional Sleep Related Beliefs as ModeratorWe further examined the associations between these direct and indirect associations between pre-bedtime arguing and adolescents' mental health using adolescents' dysfunctional sleep-related beliefs as a moderator in order to highlight the potential for children's maladaptive beliefs to influence these processes.Direct EffectsAs shown in the bottom half of Table 3, results suggested that adolescents' dysfunctional beliefs about sleep moderated the associations within this mediation model, F7,1280 = 35.1, P < .001. Specifically, for adolescents with either average (B = .47, SE = .14; P < .001) or above-average (+1 SD; B = .96, SE = .16; P < .001) levels of dysfunctional sleep-related beliefs, the direct effect of parent-child pre-bedtime arguing was significantly associated with higher levels of adolescents' anxiety and depressive symptoms. However, this direct association did not emerge as signifi-cant for adolescents' reporting below-average (−1 SD) levels of dysfunctional beliefs (B = −.14, SE = .20; not significant).Indirect EffectsThe results, however, failed to show that adolescents' dysfunctional sleep-related beliefs moderated the mediational paths within the model. Controlling for the direct effects of parent-child arguing before bedtime on adolescents' mental health symptoms, the model suggested that adolescents' sleep quality mediated the association between parent-child pre-bedtime arguing and adolescents' anxiety/depressive symptoms across all levels of adolescents'