Abstract
BACKGROUND: Physical Activity (PA) occurs in several behavioral domains (e.g., sports, active transport), and is affected by distinct environmental factors. By filtering objective PA using children's school schedules, daily PA can be separated into more conceptually meaningful domains. We used an ecological design to investigate associations between "playability" of 21 school-environments and children's objectively measured after-school PA. We also examined to what extent distinct time-periods after-school and the distance from children's residence to their school influenced this association. METHODS: PA was measured in 587 8-11 year-old children by accelerometers, and separated in four two-hour time-periods after-school. For each school-environment, standardized playability-scores were calculated based on standardized audits within 800 m network buffers around each school. Schools and children's residences were geocoded, and we classified each child to be residing in 400, 800, 1600, or >1600 m crow-fly buffers from their school. The influence of network-distance buffers was also examined using the same approach. RESULTS: Playability was associated with light PA and moderate-to-vigorous PA after-school, especially in the time-period directly after-school and among children who lived within 800 m from their school. Playability explained approximately 30% of the after-school PA variance between schools. Greater distance from children's residence to their school weakened the association between playability of the school-environments and after-school PA. CONCLUSIONS: This study demonstrated that relationships between the conceptually matched physical environment and PA can be revealed and made plausible with increasing specificity in time and distance.