Factors associated with effectiveness of interventions to prevent obesity in children: a synthesis of evidence from 204 randomized trials

Abstract Objective To identify effective characteristics of behaviour change (physical activity and diet) interventions that prevent obesity in children aged 5 to 18 years. Design A Bayesian multi-level meta-regression analysis of randomized trial results, with intervention and trial characteristics coded according to an analytic framework co-developed with stakeholders. Data source Two Cochrane systematic reviews of the effects of interventions to prevent obesity in children, 5 to 11 years and 12 to 18 years, both updated in 2024. Main outcome measures Mean difference (MD) in change from baseline in age- and sex-standardized BMI measured as a Z-score (zBMI). Results that had been reported as (unstandardized) BMI or BMI percentile were converted to zBMI using bespoke mapping techniques. Results We included 204 trials (255 intervention arms) reporting data on at least one of the main outcome scales. Interventions were effective on average (MD in zBMI −0.037, 95% credible interval −0.053 to −0.022). The greatest effects were associated with medium term follow-up (9 to <15 months) and older children (12 to 18 years). We found evidence of small but beneficial effects for interventions targeting physical activity alone compared with diet alone (difference in MDs −0.227, −0.362 to −0.090) and small unfavorable effects for interventions that involved a change to the structural environment (the majority of changes were in the school food environment) (difference in MDs 0.05, 0.017 to 0.085). Accounting for interactions between covariates, we found that the most effective combination of intervention characteristics was to intervene in the school setting, with an individualized element to delivery, targeting physical activity, using multiple strategies of short duration and high intensity, and involving modification of behaviour through participation in activities. Conclusions The most effective characteristic to include in a behaviour change intervention to prevent obesity in children aged 5-18 years was targeting of physical activity. This should not be interpreted as evidence that attempts to modify diet are not beneficial. Being physically active and consuming a healthy diet during childhood offer many important benefits beyond contributing to healthy weight and growth. Our findings suggest that interventions to prevent obesity in children should consider focusing primarily on the promotion of physical activity and consider other effective characteristics we identify here. Key messages What is already known on this topic Rising population levels of childhood overweight and obesity present a global challenge. Many interventions have been developed and evaluated to try and prevent obesity in children and young people. The most effective characteristics of these interventions are not well understood. What this study adds This re-analysis of the results of 204 randomized trials of diverse interventions seeks to identify effective characteristics of behaviour change (physical activity and diet) interventions. The most effective characteristic to include in a behaviour change intervention may be targeting physical activity. Other useful features of interventions appear to be individualized delivery, using multiple strategies, being intense and of short duration, and involving participation in activities.