Feasibility of Consumer Wearable–Based Physical Activity Monitoring Across the School–Summer–School Transition in Middle School Students: A 208-Day Prospective Cohort Study (Preprint)

Le résumé fourni par la source

BACKGROUND Consumer wearables could enable continuous assessment of adolescents’ physical activity (PA) across school and summer periods, overcoming limitations of studies that rely on brief assessments at selected time points. However, little is known about whether sufficient data completeness can be maintained across school-calendar transitions or when personally owned or study-provided devices are used in the same study. OBJECTIVE This study evaluated the feasibility of collecting consumer wearable–derived PA data from middle school students across a 208-day school–summer–school transition (April 7 to October 31, 2025). METHODS A convenience sample of students in Grades 7 and 8 was recruited from Physical Education courses at one Midwestern US middle school. Participants who owned an Apple Watch used their personal device, whereas those without one were provided with a Fitbit Charge 6. The primary feasibility outcome was completeness of Fitbit moderate-to-vigorous PA (MVPA) or Apple Watch Exercise Minutes data. Completeness was summarized descriptively overall, by device group, and across school-calendar phases. Weekly PA trajectories were examined descriptively. Participants received weekly synchronization reminders via the study platform, follow-up outreach via email after prolonged interruptions, and cumulative incentives based on data completeness. RESULTS The feasibility sample included 59 participants (mean age 13.2, SD 0.6 years): 46 received a Fitbit and 13 used their own Apple Watch. Overall, 56 (94.9%) contributed at least one valid primary-outcome day. Median completeness was 42.3% (IQR 11.8%-75.0%), and 25 participants (42.4%) achieved at least 50% completeness. Median completeness was descriptively higher in the Apple Watch group than in the Fitbit group (68.3%, IQR 50.0%-86.1% vs 31.0%, IQR 8.2%-66.2%). Overall median completeness declined from 87.2% (IQR 44.7%-97.9%) at the end of the school year to 52.5% (IQR 0.0%-88.8%) during summer break and 0.0% (IQR 0.0%-61.1%) following the return to school. CONCLUSIONS Sustaining wearable-derived PA data collection among adolescents over 208 days was challenging, with fewer than half of participants achieving 50% completeness and roughly one in five achieving 90% completeness despite reminders and incentives. Participants using their own Apple Watches contributed more complete data than those provided with Fitbits, although this difference may reflect prior device familiarity, established use routines, or participation motivation. Extended adolescent PA monitoring may require simplified data transfer, rapid detection of interruptions, and adaptive support around school-calendar transitions.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.