0624 Sleep Duration and Insomnia Predict Increases in Depression in Adolescents at Subsequent 3-Month Intervals: A Longitudinal Study

Abstract Introduction Adolescence is a critical developmental period for mental health, with the incidence of depression increasing significantly during this period. Both short sleep duration and insomnia are common and associated with depression risk. Longitudinal studies examining the relationship between sleep and depression in adolescents are scarce; therefore, the current study investigates the influence of sleep duration and insomnia on future increases in depressive symptoms. Methods A community sample of 151 never-depressed healthy adolescents (56.3% female; mean age = 14) were enrolled in 7th-9th grade (ages 11-15 years old) and followed for up to two years. On-line self-report questionnaires were collected every 3 months, including the Center of Epidemiology Studies Depression Scale (CES-D), the Insomnia Severity Index (ISI), and the Sleep Timing Questionnaire (STQ). The sleep item was excluded from the CES-D score used for analysis. Sleep duration was estimated through reported usual time in bed on the STQ (weighted average of weekdays and weekends). Separate mixed effects analyses examined whether time in bed or ISI predicted CES-D scores 3 months later, controlling for gender, age, and lag CES-D. Results Depression was associated with both time in bed (t(736)=-2.8, p< 0.005) and Insomnia Severity Index (t(736)=2.3, p=0.02). For every one standard deviation decrease in time in bed, depression was predicted to increase by 0.15 points 3 months later. For every one standard deviation increase in Insomnia Severity Index, depression was predicted to increase by 0.12 points 3 months later. Conclusion The findings suggest that lower time in bed and higher insomnia predict increases of depression in the following 3 months. Given that sleep is a modifiable risk factor for depression, improving sleep health may help ameliorate depression during this high-risk period. Support (if any) Source of support is the National Institutes of Health (R01 MH118312).