Three-year behavioural, health-related quality of life, and body mass index outcomes from the RESPOND randomized trial

Objectives: Evaluate RESPOND, a community-based systems intervention to prevent childhood obesity and non- communicable diseases (NCD). Study design: Cluster randomized trial of building community capacity to use systems science for child obesity and NCD prevention in 10 local government areas in northeast Victoria, Australia. Four-year stepped wedge trial, adapted due to COVID-19 restrictions Methods: Cluster randomized trial of building community capacity to use systems science for child obesity and NCD prevention in 10 local government areas in northeast Victoria, Australia. Four-year stepped wedge trial, adapted due to COVID-19 restrictions. Data from 31 primary schools participating at both March to June 2019 (60 % school participation rate), and March to August 2022 (30 %) analysed using linear mixed models. Primary outcome was age-sex-adjusted body mass index z-scores (BMIz) calculated from measured height and weight (children in grades 2, 4 and 6 [aged 7-12 years]). Secondary outcomes were health-related quality of life (HRQoL) and self-reported health behaviours (grades 4 and 6). Results: Non-significant intervention effects were observed for BMIz (-0.10; 95 % CI: 0.30, 0.11), and percentage with overweight or obesity (-5.4 %; 95 % CI: 13.6 %, 2.7 %). HRQoL deteriorated between 2019 and 2022, except where intervention effects for HRQoL were observed in boys' psychosocial health summary score (7.4; 95% CI: 3.5, 11.2) and total scale score (9.9; 95% CI: 5.5, 14.2). There was an intervention effect for the proportion of boys consuming >= 5 glasses of water per day (15.2 %; 95% CI: 0.9-29.6). Conclusion: RESPOND protected overall and psychosocial health and had positive effects on BMIz.