Integrating executive function training with health behavior intervention for preschool obesity prevention in China: a cluster-randomized controlled trial

Background: Preschool is a critical period for obesity prevention. However, few real-world interventions have integrated executive function (EF) enhancing activities with health behavior promotion. Objectives: To evaluated the effectiveness of a multilevel kindergarten–family intervention among preschoolers in China. Methods: In this cluster-randomized controlled trial, 26 kindergartens in Wuhan were stratified by kindergarten educational tier and randomly assigned (1:1) to either the intervention or control group. The intervention comprised health education, EF-enhancing activities, environmental modifications, and family engagement strategies over 3 semesters. The primary outcome was change in body mass index z-score (BMIz); secondary outcomes included EF performance and health behaviors. Linear mixed models accounted for clustering and repeated measures; a bias-reduced linearization (CR2) cluster-robust small-sample-corrected sensitivity analysis was conducted. Results: Of 1239 children included, 638 were assigned to the intervention group and 601 to the control group; 21.2% were classified as having overweight or obesity at baseline. In the primary analysis, BMIz gain was smaller in the interventions than in controls [mean difference (MD), −0.08; P = 0.043]. EF performance improved at midline in Digit Span Test 1 (MD: 0.28; P = 0.002), Digit Span Test 2 (MD: 0.23; P = 0.15), and Corsi Block Tapping Test 2 (MD: 0.41; P < 0.001). At the endpoint, the simplified Stroop day–night task showed a nominal improvement (MD: 0.48; P = 0.009). The healthy dietary consumption scores changed favorably (MD: −0.73; P = 0.017). Exploratory subgroup analyses suggested possible heterogeneity by kindergarten tier at midline (P for interaction = 0.086) and by sex at endpoint (P for interaction = 0.052), with a larger estimated endline effect among boys. Conclusions: This kindergarten–family intervention showed directionally favorable but statistically uncertain effects on BMIz gain after small-sample correction, alongside modest improvements in selected EF outcomes. These findings support the feasibility and potential promise of this approach for early obesity prevention, whereas larger trials are needed to confirm effectiveness, durability, and equity across settings.The trial was registered at the Chinese Clinical Trial Registry as ChiCTR2200058452 (9 April, 2022; https://www.chictr.org.cn/hvshowproject.html?id=205233&v=1.1).