The project team conducted a multi-year, broad-based evaluation of how changing school start times impacts health and well-being for students in a diverse district in suburban Denver, Colorado. This study took advantage of a natural experiment where the district changed school start times, with high school students starting at 8:20 a.m. (70 minutes later), middle school students starting at 8:50 a.m. (40-60 minutes later), and elementary schools starting at 8:00 a.m. (60 minutes earlier).
- Do later school start times result in increased sleep duration and improved physical and mental health outcomes for middle and high school students?
- Do earlier school start times have a neutral impact on physical and mental health outcomes for elementary school students?
- Do later school start times result in improved community level outcomes for adolescents, including fewer vehicle crashes and fewer incidents of loitering, vandalism and shoplifting after school?
- Inform decision-making at the district, state, and national levels about the individual and community-wide impacts of changing school start times.
Shifting school start times to 8:30am or later for middle/high school students is a critical health policy that can quickly and effectively address significant adolescent sleep debt, with minimal impact on younger students, who often are required to start earlier in order to accommodate later secondary school start times.
Additionally, with later school start times, parents who had at least one student in middle school or high school reported later wake times and increased sleep duration, averaging an additional 20 minutes of sleep per night. More parents of middle and high school students reported good sleep quality and fewer parents reported feeling tired during the day with later start times. Parents who only had elementary school children reported earlier bedtimes and wake times, which aligned with the shift to earlier school start times. No changes were found for sleep duration, sleep quality, or daytime experience of feeling tired in this group.
Health: Sleep routine and schedule; diet; exercise; caffeine use; depression; anxiety; nurse-documented student health visits for sleepiness, headaches, nutrition, and stomachaches; mental health utilization of support services; motor vehicle crash rates; and pedestrian safety
Other: School attendance, tardiness, and grades
The research team took a longitudinal analytic approach to compare quantitative data from before and after the policy change occurred. Data were drawn from a number of different sources, including online surveys, district information, and community statistics. Linear mixed models were used to fit continuous outcomes, while generalized linear mixed models were used for non-normal (i.e. count or binary) outcomes. The research team also collected qualitative data through open-ended survey questions and focus groups to provide additional contextual data.