The Sleep Fitness Study: A Universal School-Based Intervention to Promote Sleep and Prevent Substance Use

Introduction: Sleep is essential for positive youth development, yet most adolescents get insufficient sleep. Mounting evidence for the role of sleep in biological and emotional regulation suggests that sleep serves as a contributing factor in diverse outcomes including depression, psychiatric disorders, obesity, diabetes, and all-cause mortality. Notably, the transition to adolescence is a critical time for the incidence of internalizing and externalizing problems. We examined the impact of a novel school-based intervention, Sleep Fitness (SF), on several health outcomes in a cohort of California public high school students.

Methods: Using an experimental design, 691 high school students enrolled in elective health and career classes at three urban, public schools in one ethnically diverse California school district were randomized at the classroom level to the SF intervention (5-session interactive curriculum based in sleep science and cognitive-behavioral and motivational interviewing principles) or control (1 sleep science informational session). Surveys were administered at baseline with follow-up occurring at 30 days, 3 months, and 6 months post-intervention. We assessed the intervention’s impact on substance use, sleep quantity and quality, and mental health outcomes. We used linear regression for the continuous sleep quantity, substance use, and mental health measures and ordered logistic regression for the ordered categorical sleep quality measure.

Results: The impact of intervention was assessed using (1) unadjusted models, (2) baseline health outcome-adjusted models, and (3) models adjusted for teacher, age, grade, gender, school, and ethnicity. After adjustment for relevant covariates, caffeine use in the past 30 days was lower for the experimental group compared to the control group at 6 months post-intervention (beta: -0.16; 95%CI: -0.31, -0.01), as was weekend total sleep time at 30 days post-intervention (beta: -0.23; 95%CI: -0.40, -0.06). Similarly, social anxiety was lower for the experimental group at 30 days post-intervention after adjustment for relevant covariates (beta: -0.13; 95%CI: -0.24, -0.03). Subgroup and interaction analyses to identify variable patterns in effects will be presented.

Conclusions: In early follow-up, the intervention was associated with several targeted outcomes, but these did not persist. Limitations of the intervention design and methods for addressing long-term stressors and environmental barriers to promoting healthier sleep patterns will be considered, as will implications for future research and interventions focused on sleep as an outcome and as a potential mediator of substance use and mental health in diverse adolescent populations.