RESUMO
Some studies have found associations between dietary quality and obesity and their concurrent changes were observed in a few interventions. The present study aimed to assess the effect of a multifaceted intervention for childhood obesity on dietary quality and examine the mediating effect of dietary quality between the intervention and changes in adiposity indicators. Based on the social ecological model, the cluster randomized controlled trial included five components (three targeted children and two targeted their environment). In total, 1176 children from three cities in China participated in a baseline (2018) and end-of-trial (2019) examination, including 605 children in the intervention group and 571 in the control group. Self-reported behavior and anthropometric measures were collected at both time points. The Diet Balance Index Revision (DBI-07) was calculated to assess dietary quality. Generalized linear mixed models were used to estimate the intervention effect on dietary quality and its mediating effects were examined. Compared to the controls, the proportion of sugar-sweetened beverage (SSB) intake (OR = 0.27, p < 0.001, corrected p < 0.001) decreased in the intervention group. Higher bound scores (HBS) of the DBI-07 indicating over-intake decreased in the intervention group compared to the controls (mean difference = −1.52, p = 0.005, corrected p = 0.015). Changes in the HBS partially mediated the associations between the intervention and changes in body mass index, waist circumference, and body fat percentage. Future intervention should promote knowledge, attitudes, and behaviors related to dietary quality.