LAUSR

Objective We aimed to examine the association between food and physical activity environments in primary schools and child anthropometric, healthy eating and physical activity measures. Design Observational longitudinal study using data from a childhood obesity prevention trial. Setting State primary schools in the West Midlands region, UK. Participants 1392 pupils who participated in the WAVES (West Midlands ActiVe lifestyle and healthy Eating in School children) childhood obesity prevention trial (2011–2015). Primary and secondary outcome measures School environment (exposure) was categorised according to questionnaire responses indicating their support for healthy eating and/or physical activity. Child outcome measures, undertaken at three time points (ages 5–6, 7–8 and 8–9 years), included body mass index z-scores, dietary intake (using a 24-hour food ticklist) and physical activity (using an Actiheart monitor over 5 days). Associations between school food and physical activity environment categories and outcomes were explored through multilevel models. Results Data were available for 1304 children (94% of the study sample). At age 8–9 years, children in 10 schools with healthy eating and physical activity-supportive environments had a higher physical activity energy expenditure than those in 22 schools with less supportive healthy eating/physical activity environments (mean difference=5.3 kJ/kg body weight/24 hours; p=0.05). Children in schools with supportive physical activity environments (n=8) had a lower body mass index z-score than those in schools with less supportive healthy eating/physical activity environments (n=22; mean difference=−0.17, p=0.02). School food and physical activity promoting environments were not significantly associated with dietary outcomes. Conclusions School environments that support healthy food and physical activity behaviours may positively influence physical activity and childhood obesity. Trial registration number ISRCTN97000586.