Prospective associations of sleep duration and screen time with transition from overweight/obesity to normal BMI in U.S. adolescents.

Abstract

Shorter sleep duration and longer screen time are established risk factors for adolescent obesity. However, the extent to which these behaviors are prospectively associated with the transition back from overweight/obesity to a healthy status remains unclear. We examined whether sleep duration and screen time among adolescents with overweight/obesity are associated with the likelihood of transitioning to a normal body mass index (BMI). We used data from 3498 U.S. adolescents aged 9-11 years with overweight/obesity (45.1% female), from the Adolescent Brain Cognitive Development (ABCD) Study. Cox proportional hazards models examined the prospective associations of parent-reported sleep duration (9-11, 8-9, 7-8, and <7 h/day) and screen time (hours/day) with a shift from overweight/obesity (BMI percentile ≥85) to a normal (BMI percentile <85) accounting for key covariates including pubertal status. Over a median 657 days of follow-up, 643 (18.4%) adolescents transitioned from overweight/obesity to a normal BMI percentile. Compared with those sleeping 9-11 h, adolescents sleeping 7-8 h were less likely to transition to a normal BMI percentile (hazard ratio [HR]: 0.60, 95% CI 0.44, 0.82), with significant dose-response trend (p for trend = 0.003). The association remained significant in sex-stratified analyses for both females (HR: 0.55, 95% CI 0.30, 0.98) and males (HR: 0.59, 95% CI 0.41, 0.86), with similar significant trend in both groups (p for trend <0.05). Higher screen time was not associated with transitioning to a normal BMI overall (HR: 0.99, 95% CI 0.96, 1.02) or by sex (females, HR: 1.00, 95% CI 0.95, 1.05; males, HR: 0.99, 95% CI 0.95, 1.02). Short sleep duration was prospectively associated with a lower likelihood of transitioning to a normal BMI among adolescents with overweight/obesity. This association warrants further investigation as a potential intervention target.