Observing children's playground activity levels at 13 Illawarra primary schools using CAST2

Declining levels of children’s physical activity may contribute to Australia’s increasing childhood obesity epidemic. School recess is an underutilized opportunity to increase children’s physical activity. Methods: Thirteen regional Australian public primary schools participated in the study (2946 children). The Children’s Activity Scanning Tool 2 (CAST2) collected observational playground physical activity data. The research also addressed: length of break, socioeconomic status (SES), gender, number of scanning days, and instrument calibration. Results: The proportions of Moderate or Vigorous Physically Activity (MVPA) children at the observed schools ranged from 0.4 to 0.7. The odds ratio of boys being MVPA relative to girls ranged from 0.8581 to 2.137. There were significant differences between the mean proportions of 3 days of activity (range P = .001 to P = .015) and no association between SES school groupings (deviance ratio: 0.48; P = .503). Interrater reliability for instrument calibration using Spearman correlations coefficients ranged from r = .71 to r = .99. Conclusions: There were significant differences between proportions of MVPA children at the 13 schools and between male and female populations. There was no association between playground physical activity and SES. The monitoring period for CAST2 should be at least 3 days. Interrater reliability indicates that correlations between observers were consistently high.