Latent Typologies of Sleep Patterns: Associations with Pubertal Development, Internalizing and Externalizing Problems

• In Event: Pubertal Change, Deficient Sleep, and Mental Health Problems: Understanding This Slow Moving Trainwreck

Abstract

Introduction: Sleep disparities are commonplace during adolescence and are often influenced by biological changes such as pubertal development. The detrimental effects of poor sleep on youth are evident in the development of internalizing and externalizing problems (Zimmerman, 2008). Yet most studies have examined the associations between pubertal development, sleep, and youth psychopathology via variable-center approaches (e.g., linear modeling), which may overlook the coexistence of sleep problems and fail to holistically capture individuals who exhibit multivariate sleep patterns (Yue et al., 2022). The present study used a latent profile analysis to identify subpopulations based on sleep duration, latency, efficiency, wake count, wake minutes and midpoint. The analytic sample includes a sample of 3326 youth (49.3% female; Mage = 9.5) with at least 7 recorded nighttime sleep events and who wore Fitbit on both weekdays and weekends. The racial-ethnic composition was 61.8% European American, 7.0% African American, 18.7% Latino(a), 2.7% Asian/Pacific Islander, and 9.7% Other. We hypothesize that 1) three to four classes ranging from no/mild sleep problems to high sleep problems groups would be identified. 2) Youth with higher puberty status would be more likely to be identified in high sleep problems groups. 3) High sleep problems group is at more risk for internalizing and externalizing problems.

Method: All hypotheses were tested using data from the Adolescent Brain Cognitive Development (ABCD) study. Sleep was measured using Fitbit Charge HR 2 devices at T5. Internalizing (e.g., somatic, anxiety, depression, and withdrawal) and externalizing problems (e.g., rule-breaking and aggression) were measured using CBCL at T7. Puberty status was measured using Peterson’s Pubertal Development scale (Petersen et al., 1988) using a 4-point Likert scale (0 = “not yet started,” to 4 = “seems completed”). Once the optimal latent profile solution was obtained, the R3STEP method was used to examine how puberty development correlates with sleep profile membership. Next, the manual BCH approach was used to examine prospective links between profile membership (T5) and each outcome at T7, controlling for relevant covariates (e.g., race, income, BMI, age, gender) and the T5 outcome variable (McLarnon & O’Neill, 2018).
Results: Four sleep classes were identified: average sleep group (40.39%), high sleep duration and low sleep efficiency group (28.58%), low sleep duration group (16.89%), and high sleep problems group (14.17%). Participants with higher puberty status are more likely to be classified in the “high sleep problems” group and the “low sleep duration” group (OR =1.36, p <.01; OR = 1.18, p <.01, respectively). The “High sleep problems” group had a higher level of externalizing problems and rule-breaking behaviors and anxious/depressed symptoms than the “average sleep” class and the “low efficiency, high sleep duration group.” The “High sleep problems” group had a higher level of social problems than the “average sleep” class. We did not see any group differences in internalizing problems, aggressive behaviors, withdrawn-depressed symptoms, attention problems, thought problems, and somatic complaints.

Conclusion: Our results highlight how puberty development is related to different sleep patterns and consequent implications for youth psychosocial outcomes.

Authors

• Linhao Zhang, University of Georgia
  Presenting Author

• Jeri Sasser, Arizona State University

• Jack S Peltz, SUNY Brockport

• Leah D Doane, Arizona State University

• Michelle Barth, University of Georgia

• Assaf Oshri, University of Georgia