Widome · JAMA pediatrics 2020 · prospective controlled cohort study · n=455

Association of Delaying School Start Time With Sleep Duration, Timing, and Quality Among Adolescents.

Cited 96 times in the scientific literature.

Level 3 - non-randomized controlled study

Non-randomized controlled cohort study analyzing a natural experiment

PubMed 32338727 · doi:10.1001/jamapediatrics.2020.0344 · record verified 2026-08-27

What was done

This prospective observational cohort study evaluated district-initiated school start time changes across 5 public high schools in the Minneapolis–St. Paul metropolitan area. A total of 455 students were followed from grade 9 (baseline in 2016, when all schools started at 7:30 am or 7:45 am) through grade 11 (2018). Two schools delayed their start times by 50 and 65 minutes, while three comparison schools maintained a 7:30 am start time. Wrist actigraphy was used to objectively assess sleep duration, timing, and quality. Changes were evaluated using linear mixed-effects difference-in-differences models.

What was found

Compared with students in comparison schools, students at delayed-start schools gained an additional mean 41 minutes (95% CI, 25 to 57) of objectively measured night sleep at follow-up 1 and 43 minutes (95% CI, 25 to 61) at follow-up 2. Delayed start times did not shift sleep onset later on school nights. Weekend catch-up sleep decreased in delayed-start schools relative to comparison schools (difference-in-differences: -24 minutes [95% CI, -51 to 2] at follow-up 1; -34 minutes [95% CI, -65 to -3] at follow-up 2). Differences were minimal for school-night sleep onset, weekend sleep onset latency, sleep midpoints, sleep efficiency, and sleep fragmentation index.

Why it matters

This study provides objective, multi-year evidence that delaying high school start times durably increases adolescent sleep duration on school nights and reduces weekend catch-up sleep without delaying bedtimes.

Limits

The study was non-randomized and observational, leaving potential for unmeasured school- or household-level confounding. Data were gathered from 5 high schools in a single metropolitan area, which may limit generalizability. The abstract reports no data on daytime sleepiness, academic performance, or mental health outcomes.

Cited by