To evaluate the therapeutic effects of exercise interventions in children and adolescents with autism spectrum disorder (ASD) using an umbrella meta-analysis. We systematically searched Embase, CINAHL, the Cochrane Library, Web of Science, PubMed, SinoMed, CNKI, Wanfang, and VIP for systematic reviews and meta-analyses of exercise-based interventions in ASD. Two reviewers independently assessed methodological quality and extracted data. We included 20 systematic reviews and meta-analyses comprising 14,245 participants. Exercise interventions were associated with enhanced social skills (SMD = 0.52; 95% CI 0.28-0.76; P < 0.001), improvements in repetitive stereotyped behaviors (SMD = 0.49; 95% CI 0.29-0.69; P < 0.001), and improved motor skills (SMD = 1.16; 95% CI 0.64-1.68; P < 0.001) in individuals with ASD. Exercise interventions are associated with improvements in social skills, repetitive stereotyped behaviors, and motor skills in children and adolescents with ASD. However, the overall certainty of evidence is low, and substantial heterogeneity and moderate-to-high overlap across meta-analyses warrant cautious interpretation. Exercise may represent a scalable, low-cost neuromodulatory adjunct within neurodevelopmental rehabilitation; future high-quality trials with long-term follow-up and clearly defined intensity and dose–response relationships are needed. • This umbrella meta-analysis synthesizes evidence from 20 systematic reviews and meta-analyses involving 14,245 participants. • Exercise interventions are associated with improvements in social skills, repetitive stereotyped behaviors, and motor skills in ASD. • Overlap across included meta-analyses is slight to high, but sensitivity analyses support the robustness of the overall pattern. • Exercise may serve as a scalable, low-cost adjunct—particularly relevant for resource-limited settings—pending higher-quality, long-term trials.
Wang et al. (Sat,) studied this question.