Interventions for Preventing Obesity in Children and Adolescents Aged 5-18 Years: An Overview of Nonrandomized Study...

Community- and population-level and policy interventions are commonly evaluated using nonrandomized studies (NRS), rather than randomized trials (RCTs). Recent Cochrane reviews of interventions for preventing childhood obesity have been restricted to RCTs, so less is known about the effectiveness of these more upstream interventions. To address this gap, we conducted an overview of reviews of NRS interventions (NRSI), which assessed change in BMI in children and adolescents aged 5-18 years and compared NRSI findings with those from RCTs.

We searched five databases up to November 2024. Screening, data extraction, and quality assessment were performed using standardized tools.

We included 28 systematic reviews and identified 136 NRSI either based in school (n = 118), community (n = 4) or combined settings (n = 14) and evaluating policy (n = 48), education (n = 11), or a combined intervention (n = 77). Twenty-six reviews included both NRSIs and RCTs; of these, 12 reported meta-analyses. Findings were largely unchanged when we excluded the RCTs and re-ran analyses. Overall, study-level results from the NRSI favored the intervention group; a quarter favored the comparison group. The meta-analysis summary effects from NRSIs were consistent with two recently published Cochrane meta-analyses of RCTs of obesity prevention interventions.

The results from this overview of reviews suggest researchers and policy makers can be confident in considering the results of robust nonrandomized study designs (evaluating their impact on BMI) alongside RCTs in their decision making. Although we identified a significant number of NRSIs for review, very few evaluations of upstream interventions were eligible for inclusion.