A quasi-experimental study of the impact of school start time changes on adolescent sleep

Objective To determine whether simultaneous school start time changes (delay for some schools; advance for others) impact adolescents' sleep. Design Quasi-experimental study using cross-sectional surveys before and after changes to school start times in September 2015. Setting Eight middle (grades 7-8), 3 secondary (grades 7-12), and 8 high (grades 9-12) schools in Fairfax County (Virginia) public schools. Participants A total of 2017 (6% of ~34,900) students were surveyed before start time changes, and 1180 (3% of ~35,300) were surveyed after. Intervention A 50-minute delay (7:20 to 8:10 AM) in start time for high schools and secondary schools and a 30-minute advance (8:00 to 7:30 AM) for middle schools. Measurements Differences before and after start time changes in self-reported sleep duration and daytime sleepiness. Results Among respondents, 57.5% were non-Hispanic white, and 10.3% received free or reduced-priced school meals. Before start time changes, high/secondary and middle school students slept a mean (SD) of 7.4 (1.2) and 8.4 (1.0) hours on school nights, respectively, and had a prevalence of daytime sleepiness of 78.4% and 57.2%, respectively. Adjusted for potential confounders, students with a 50-minute delay slept 30.1 minutes longer (95% confidence interval [CI], 24.3-36.0) on school nights and had less daytime sleepiness (−4.8%; 95% CI, −8.5% to −1.1%), whereas students with a 30-minute advance slept 14.8 minutes less (95% CI, −21.6 to −8.0) and had more daytime sleepiness (8.0%; 95% CI, 2.5%-13.5%). Conclusions Both advances and delays in school start times are associated with changes in adolescents' school-night sleep duration and daytime sleepiness. Larger changes might occur with later start times.