Adolescent Sleep as a Transdiagnostic Factor: Associations Between Actigraphy-Derived Night-to-Night Sleep Metrics and Adolescent Psychopathology

Eric M. Phillips, Emily L. Goldberg, Rebecca L. Brock, Emily Hamburger, Jennifer Mize Nelson, W. Alex Mason, Kimberly Andrews Espy, Timothy D. NelsonJAACAP 2024 · DOI

TL;DR

This study reveals that inconsistent sleep duration (not just average duration) is a significant transdiagnostic factor for adolescent psychopathology, particularly for the general factor of dysregulation. This suggests that targeting sleep consistency could be a more effective approach for interventions across a broad spectrum of mental health issues in adolescents.

What Problem Does This Paper Address?

Adolescence is a critical period for the onset of psychiatric disorders, and these disorders often co-occur, suggesting shared underlying mechanisms. Researchers have proposed a "p-factor" (general psychopathology factor) to explain this comorbidity, often linked to emotion dysregulation. While sleep problems are known to be prevalent in psychiatric disorders, previous research has largely focused on average sleep duration and specific disorders in isolation. This leaves a gap in understanding if sleep, particularly its variability, acts as a transdiagnostic factor – a common underlying mechanism that cuts across various psychiatric conditions. This paper aims to address this by examining how night-to-night sleep metrics, including variability, relate to the general factor of psychopathology and specific internalizing/externalizing problems in adolescents.

Methods

The researchers employed a sophisticated statistical approach: Multilevel Structural Equation Modeling (MSEM) using a Markov Chain Monte Carlo (MCMC) procedure. This method is crucial because it allows for the disaggregation of variance into within-person (daily fluctuations) and between-person (individual differences) components. This is particularly well-suited for analyzing repeated measures data like nightly sleep.

Key methodological aspects include:

  • Bifactor s-1 Model: This psychopathology model was used to simultaneously estimate a general factor of dysregulation and psychopathology (GF-DP) and specific internalizing and externalizing factors. Emotion dysregulation was designated as the reference indicator for the GF-DP, grounding it in a strong theoretical foundation.
  • Actigraphy: Nightly sleep was objectively measured using actigraphy watches over a 14-night period, providing detailed data on sleep duration, bedtime, and waketime.
  • Daily Sleep Surveys: Participants completed daily surveys to supplement actigraphy data and resolve discrepancies.
  • Bayesian Full-Information Likelihood: This approach was used to handle missing data, minimizing bias and preserving statistical power by utilizing all available data.
  • Focus on Variability: The study specifically modeled intraindividual variability (night-to-night fluctuations) in sleep duration, bedtime, and waketime, moving beyond simple averages.

Datasets

The study utilized data from a longitudinal investigation of cognitive development.

  • Participants: A sample of 238 adolescents (aged 14-18 years) was included. This sample was oversampled for socioeconomic risk.
  • Data Collection: Actigraphy data was collected over a 14-night period, supplemented by daily sleep surveys. Psychopathology was measured using caregiver and adolescent reports on various standardized scales (e.g., CBCL, YSR, Conners 3, ERI).

Key Insights and Ideas

  • Sleep Inconsistency as a Transdiagnostic Marker: The most striking finding is that variability in sleep duration (how much sleep fluctuates from night to night) is significantly associated with the general factor of psychopathology. This suggests that it's not just about getting enough sleep, but about getting a consistent amount of sleep.
  • Beyond Average Sleep: The study highlights that average sleep duration, bedtime, and waketime were not significantly associated with general or specific psychopathology. This challenges the common focus on these average metrics in sleep research and clinical practice.
  • Robustness Across Demographics: The association between sleep duration variability and psychopathology was consistent across different ages, sexes, maternal education levels, and income-to-needs ratios, underscoring its broad relevance.
  • Transdiagnostic Nature Confirmed: Supplemental analyses using a correlated factor model confirmed that sleep duration variability was associated with internalizing problems, externalizing problems, and emotion dysregulation independently, reinforcing its transdiagnostic role.

Key Results

  • A significant positive association was found between the general factor of dysregulation and psychopathology (GF-DP) and intraindividual sleep duration variability (β = 0.18, 95% CI = 0.013, 0.335). This means adolescents with higher general psychopathology tended to have less consistent sleep duration.
  • This association was not significant for specific internalizing or externalizing problems when controlling for the general factor.
  • No significant associations were found between psychopathology and intraindividual variability in bedtime or waketime, nor with average sleep duration, bedtime, or waketime.
  • Exploratory analyses found no significant moderation effects by sex, age, maternal education, or income-to-needs ratio.

Limitations and Future Work

  • Sample Composition: The sample was predominantly non-Hispanic White, limiting the generalizability of findings to more diverse populations.
  • Cross-Sectional Design: The study's cross-sectional nature prevents inferring the directionality of the association between sleep variability and psychopathology. The authors acknowledge the bidirectional relationship between sleep and mental health.
  • Potential for Subtle Moderation: While no significant demographic moderators were found, the authors suggest that future research with larger sample sizes might uncover subtle moderating effects.
  • Future Directions:
    • Longitudinal Studies: Future research should employ longitudinal designs and daily assessments to better understand the dynamic interplay between sleep and psychopathology and to infer directionality.
  • Methodological Refinement: Further exploration of the minimum number of nights required for reliable sleep variability estimates using MSEM is encouraged.
  • Intervention Development: The findings strongly support the development and testing of interventions that specifically target sleep consistency, such as the Transdiagnostic Intervention for Sleep and Circadian Dysfunction (TranS-C), for a broad range of adolescent mental health issues.

Summary updated

Abstract

• Clinicians should assess for consistency of adolescent sleep in addition to the amount of time an adolescent is sleeping. Sleep inconsistency may be more relevant to adolescents' mental health problems than total sleep time.• Interventions aimed at improving the consistency of young people's sleep, such as the Transdiagnostic Intervention for Sleep and Circadian Dysfunction (TranS-C), may be beneficial in reducing diverse forms of mental health problems (eg, depression, anxiety, attention problems, oppositionality, and conduct problems).• Improving consistency of sleep may be particularly beneficial for adolescents with comorbid mental health problems.
BibTeX
@inproceedings{phillips2024adolescent,
  author = {Eric M. Phillips and Emily L. Goldberg and Rebecca L. Brock and Emily Hamburger and Jennifer Mize Nelson and W. Alex Mason and Kimberly Andrews Espy and Timothy D. Nelson},
  title = {{Adolescent Sleep as a Transdiagnostic Factor: Associations Between Actigraphy-Derived Night-to-Night Sleep Metrics and Adolescent Psychopathology}},
  booktitle = {JAACAP},
  year = {2024},
  doi = {10.1016/j.jaacop.2024.06.001},
}
Figure