Abstract
This meta-analysis aims to systematically evaluate if different physical activities (PA) improve the social communication impairments (SCI) seen in children with autism spectrum disorder (ASD). For this meta-analysis, five databases (Web of Science, PubMed and Cochrane Library database in English, and CNKI and Wanfang Data Knowledge Service Platform in Chinese) were searched from database inception to September 11, 2024. The eligibility criteria included (1) study population comprised children with ASD, with no gender restriction; (2) experimental group consisted of a PA intervention; (3) control group consisted of nonPA interventions or routine activities; (4) outcomes were assessed using validated SCI scales (e.g., social cognition, social motivation, and/or social awareness); and (5) study design involved a randomized parallel group design. The quality of the evidence for each study was assessed using the Cochrane Risk of Bias tool. From a pool of 2714 potentially relevant articles, 17 were eligible for inclusion in this review. The results of overall response of PA intervention to SCI illustrated a Medium degree of statistical heterogeneity between studies (I(2) = 53.3 %, p < 0.01); the effect size of PA intervention on SCI (expressed as standardized mean difference) was -0.34 (95 % CI: -0.57, -0.11), which was statistically significant. SCI of autistic children could be potentially improved by PA interventions, though further research is needed to clarify if benefits achieve clinical significance in addition to statistical significance. Proper design of PA interventions (45-90 min per session, more than 3 times per week and for 8-12 weeks) may enhance their effectiveness in treating SCI in children with ASD.