Teun Remmers

Playability of school environments and afterschool PA | 115 procedure (computation of means from subscale items), and normalization procedures (37, 38). The concept of playability was introduced in the studies of Frank and Roberts (6, 41). As the study of Roberts solely reported on a protocol of developing a playability- index, to date no direct comparisons can be made with the current study’s SPACE observation instrument. The study of Frank et al. (6) derived the quantity and quality of public parks using the Environmental Assessment of Public Recreational Spaces (EAPRS). With the exception of the quality-concept ‘shade’ and trails, all concepts and methodology of EAPRS (i.e. independent audits of two trained observers) were also represented in the current study. One may argue that our results are not influenced by differences in playability, but by differences between schools in active transport. However, we found no indication for a ASPA-increase in children who lived more than 800 meters from their school at later time- periods (potentially to compensate for motorized transport). In addition, children reported the number of days per week they walked or cycled to school and the mean duration of those trips, and sensitivity analyses revealed that additional adjusting for active transport did not alter our results (data not shown). As our sample size did not allow further segregation for active transport use, we recommend future research to address the influence of active transport in this relationship. Our time-specific analyses showed that LPA performed after 20:00 hours was significantly related to playability. This was unexpected because in the Netherlands, during this time of the year it was dark. As we had no diaries, we were unable to confirm to what extent and where children were active at that time. Apart the possibility that relationships in the late evening may be influenced by the relatively small sample sizes because of non-weartime periods, we can only speculate that that this behavior may be merely related to LPA inside their houses, or that potential differences (by chance) between schools in bed-times may have influenced this association, rather than the actual influence of playability of the environment. The same explanation may suffice for unexpected statistically significant relationships between MVPA and playability in children living more than 1600 meters from their school. In addition, we observed that the MVPA percentages increased between 18:00–20:00, but strength of relationships between MVPA and playability did not increase accordingly. This potentially means that the observed increase in mean MVPA percentages was explained by other factors than playability of the environment, such as sports participation (organized forms of after-school activity often occur during evening-hours). Future studies are advised to include some diaries about sleep times and main activities after school (e.g. organized sports participation in the evenings), and are warranted to examine potential if attenuation of playability by organized sports participation would persist in spring, and whether breaks in organized sports (e.g. during summer recess) would relate to a stronger relationship between playability and PA after- school-time.