0906 Sleep Health and Depression in Teens: Which Sleep Variables Are the Most Relevant? – Data from the National Sleep Foundation Sleep in America® Poll

Abstract Introduction One in 5 adolescent Americans have experienced at least one major depressive episode. This study aimed to explore which aspects of sleep health and sleep satisfaction contribute most to depressive symptoms in teens. Methods Data were collected from the National Sleep Foundation Sleep in America® Poll. The random, population-based sample included N=1,097 teens (age 13-17) who provided complete data. Depression symptoms were assessed with the PHQ8 scale, evaluated as a continuous variable with the sleep item removed (sum of 7 items). Sleep variables included sleep quality (excellent, very good, good, fair, poor), weekday and weekend total sleep time, bedtime, and wake time, days/week feeling rested, with difficulty falling asleep, staying asleep, sleepiness, taking sleep-promoting medications, and sleep impacting daytime function (coded 0-7), diagnosed sleep disorder, talked to doctor about sleep, and perceived sleep need. Regression analyses adjusted for age, gender, race/ethnicity, and socioeconomics. Bonferonni correction (0.05/18) was used. Results The range of depressive symptoms scores was 0-21, reflecting the full possible range; however, scores were generally low, with mean 3.7 (SD=4.2) and IQR 0-6. Higher depression scores were associated with fair (+3.28 pts) and poor (+3.84 pts) sleep quality, weekday sleep duration (-0.45 pts/hr), feeling rested (-0.69 pts/day), difficulty falling asleep (+0.63 pts/day), difficulty staying asleep (+0.54 pts/day), sleep impacting daytime function (+0.71 pts/day), sleepiness (+0.37 pts/day), taking sleep-promoting medications (+0.38 pts/day), diagnosed sleep disorder (3.54 pts), and sleep need (+0.34 pts/hr). A forward stepwise model indicated that the variables that contributed the most unique variance were (in order): days difficulty falling asleep, days feeling rested, days sleep impacts functioning, diagnosed sleep disorder, sleep quality, and weekend waketime (all p< 0.05). These results suggest that adolescent depression risk is driven more strongly by subjective sleep experience and daytime impairment than by sleep duration alone. Conclusion Higher depression scores were associated with worse sleep quality. In particular, subjective sleep satisfaction and next-day experiences were especially important. Further studies should aim to assess the effectiveness of public health interventions targeting these specific outcomes. Support (if any) R01MD011600, R01MH135978