Distinct Aspects of Sleep Predict Increases in Depression and Behavior Problems from Early to Mid-Adolescence

Background: Early adolescence is a time of significant increase in mental health problems (Steinberg et al., 2006).  Although national recommendations underscore the importance of more sleep for adolescents (National Sleep Foundation, 2017), too much sleep is associated with worse mental health (Fuligni et al., 2017).  Yet, many studies have focused exclusively on sleep duration or other sleep parameters in isolation.  Consequently, less is known about distinct aspects of sleep that may differentially predict increases in mental health problems such as depression, and internalizing/externalizing behavior from early-to-middle adolescence. 

Differential aspects of sleep timing are particularly important to examine.  Indeed, chronotype/circadian-based morning preference (i.e., predisposition to be a “morning lark” or “night owl”), timing (i.e., time period in which one sleeps), and social jet leg (i.e., discrepancy between one’s circadian preference and social demands such as school start times) have been found to be associated with mental health problems, but few studies have examined differential contributions and potential differences between boys and girls.  Yet, identifying aspects of sleep that produce greatest risk and determining whether they vary for boys and girls may have important implications for mental health intervention. 

To address this gap, we examine whether distinct aspects of sleep differentially predict increases in depression, and sleep, internalizing, and externalizing problems from age 12 to 15.

Method:  We used data from the NICHD Study of Early Child Care and Youth Development (n=929).  Sleep parameters and child depression were measured via child-report using adaptations of the Children’s Sleep Habits Scale (Owens et al., 2000), the Morning/Eveningness Questionnaire (Carkasdon et al., 1993), and the Children’s Depression Inventory (Kovacs, 1992), respectively. Internalizing and externalizing behaviors were assessed via maternal-report using the Child Behavior Checklist (Achenbach, 1991).

Results:  Results from multiple regression analyses demonstrated that morning preference/chronotype predicted reduced externalizing behavior and sleep problems from ages 12 to 15 (beta = -.05, p < .05 and beta = -25, p < .001, respectively), whereas later timing of sleep on school nights predicted increased sleep problems (beta= .06, p< .05). 

For girls, later timing of sleep on school nights predicted increased depression from age 12 to 15 (beta = .12, p< .01), whereas social jet lag predicted increased internalizing and externalizing behavior (beta = .09, p < .05, beta = .11, p < .01, respectively).  For boys, however, later timing of sleep on school nights predicted reduced internalizing behavior (beta=-.09, p < .05).  Importantly, when sleep timing parameters were included, weekend sleep duration did not predict depression or problems with sleep, internalizing, or externalizing behaviors from age 12 to 15.

Conclusions/Implications:  Our findings underscore the importance of attending to distinct aspects of sleep timing in the prediction mental health problems across adolescence. That varied aspects of sleep timing, but not weekend duration, forecasted increases in depression, sleep problems, and internalizing/externalizing problems, and varied by gender, suggests that research is needed to examine whether social work policies and interventions designed to target aspects of sleep timing may be particularly important for preventing adolescent mental health problems.