Workout videos matched teachers in class. Only about half got done.

Randomised controlled trial (RCT). 468 children. one academic year (planned 2 years, cut to 1 by COVID-19); accelerometer in fall (baseline) and spring; ~20 weeks of tracked delivery.

A trial. Chance decided who did what.

Who: 2nd and 3rd graders (mean age 7.6-7.8 y) in 11 elementary schools (6 districts, rural and urban) near Lawrence/Kansas City, Kansas, USA; 468 of 1,425 eligible students (33%) consented to assessments; accelerometer subsample of 376, valid data 321 in fall and 171 in spring

• Video-led breaks+9.1 minactive minutes/day
• Teacher-led breaks+9.8 minactive minutes/day
• Video vs teacher gapnoneequal within 5 min

More to know, from our read of the paper:

• No group without breaks: both arms got breaks, so the ~9-10 min/day rise in school-day activity could partly be season, age or other changes; authors say it should not be read as robust evidence of effectiveness
• The solid result is the comparison: video and teacher formats were statistically equivalent within +/-5 min/day (TOST). Say 'video did as well as a teacher', not 'breaks add 9 minutes'
• Implementation was low: no teacher delivered the requested two breaks a day; ~50% of the intended daily dose; only 7% of teachers reached five breaks a week
• Big loss of accelerometer data: 321 valid in fall, 171 in spring; analysed with FIML under missing-at-random; the tracked kids differed from the rest at baseline (more girls, fewer with learning disabilities or free lunch)
• Activity measured only during school hours (hip ActiGraph, 8 am-4 pm), not whole-day activity, fitness, grades or health

This cluster randomized trial put 11 Kansas elementary schools (468 second and third graders) on two 10-minute classroom activity breaks a day for one school year, led either by the teacher or by a video. Kids wore activity trackers in fall and spring: active minutes during the school day rose by about 9 to 10 a day in both groups, and the two formats were statistically equivalent. But teachers delivered only about half the planned activity, none ran two breaks a day, and there was no group without breaks, so the rise itself is not proven to come from the breaks. Only 171 kids had valid tracker data in spring.

We read the abstract and the full text of this paper and checked every number on this page against it (17 facts, checked on 2026-10-29). The study design is named above. Cohort and other observational studies are described as linked, not proven. Found an error? Email theis@onestudyaday.com.