Longitudinal study of 1988 Chinese adolescents finds physical activity predicts fewer insomnia symptoms six months later

Researchers at five Hubei middle schools followed students averaging 13.79 years old across three assessments spaced six months apart. They measured self-reported physical activity, maladaptive cognitive emotion regulation including rumination and catastrophizing, and insomnia symptoms using cross-lagged panel models with full information maximum likelihood for missing data. The design allowed separation of within-person change from between-person differences.

Initial bidirectional associations emerged, but only the activity-to-insomnia direction persisted after adjusting for stable traits. Activity also reduced negative overthinking, which in turn linked to later insomnia; this mediation appeared stronger in girls. The effect size remained modest and no reverse effect from sleep to activity survived the stricter model.

Prior Western adult and college samples produced mixed longitudinal results, while this study adds data from a high-academic-pressure Chinese context where sleep deprivation prevalence reaches 26 percent. It highlights potential cognitive pathways but leaves open whether objective activity measures or interventions would replicate the pattern.

Next steps require randomized trials testing activity increases against active controls in diverse adolescent populations, with objective sleep tracking and longer follow-up to assess clinical insomnia incidence.