0135 Association of Circadian Misalignment with Internalizing Symptoms and Externalizing Behaviors in Adolescents
Pura Ballester-Navarro,Natasha Morales-Ghinaglia,Susan Calhoun,Jason Liao,Alexandros Vgontzas,Duanping Liao,Edward Bixler,Julio Fernandez-Mendoza
DOI: https://doi.org/10.1093/sleep/zsae067.0135
IF: 6.313
2024-04-20
SLEEP
Abstract:Abstract Introduction Biological changes during adolescence naturally delay the circadian sleep-wake cycle. Adolescents struggle between the imposed timing of their sleep-wake cycle by social demands and their desired one, resulting in circadian misalignment. Circadian misalignment has been associated with a higher risk for mental health problems. We examined whether objective and subjective metrics of circadian misalignment are associated with internalizing symptoms (IS) and externalizing behaviors (EB) in adolescents. Methods We studied 376 adolescents from the Penn State Child Cohort (median 16 years; 47% female; 21% racial/ethnic minority) who had a minimum of 3-nights of at-home actigraphy (ACT), and 9-h in-lab polysomnography (PSG). ACT-measured sleep midpoint (SM) was calculated as the central point of the sleep period, while sleep regularity (SR) was calculated as the intra-individual standard deviation of the SM. Circadian preference (CP) was measured with the Morningness-Eveningness Questionnaire. IS and EB were measured by parent- or self-report on Achenbach System of Empirically Based Assessment using developmentally-appropriate forms and norms. Stepwise linear regression models adjusted for demographics (sex, race/ethnicity, age, BMI percentile), sleep disorders (insomnia, PSG-apnea/hypopnea index) and insufficient sleep (ACT-mean sleep duration, ACT-sleep duration variability). Results Both CP (r=-.118, p=.022) and SR (r=.116, p=.024) were associated with IS, however, they competed between each other in regression models resulting in marginally significant associations (β=-.094, p=.079 and β=.092, p=.084, respectively) that were further attenuated after adjusting individually for covariables (β=-.072, p=.190 and β=.076, p=.163, respectively). Both CP (r=-.230, p<.001) and SR (r=.140, p=.006) were associated with EB; however, only CP (β=-.153, p=.004) remained significantly associated with EB after adjusting for covariables (β=-.130, p=.016). Conclusion Assessing self-reported eveningness is a useful and inexpensive tool to identify adolescents in whom circadian misalignment contributes to mood and behavioral problems. However, it appears that its combination with objective assessments, particularly irregularity of the sleep-wake cycle, may yield the most predictive results. Support (if any) NIH R01HL136587, R01MH118308, UL1TR000127
neurosciences,clinical neurology