Comparative effectiveness of non-pharmacological interventions for anxiety, depression, and quality of life in individuals with autism spectrum disorder: A systematic review and network meta-analysis.

Background: Individuals with autism spectrum disorder (ASD) commonly experience comorbid depression, anxiety, and impaired quality of life (QoL), significantly affecting daily functioning and social adaptation. Non-pharmacological interventions (NPIs), offering an alternative without drug-related side effects, have gained increasing attention for emotional improvement and health promotion. However, the comparative effectiveness of different NPIs remains unclear, and clinical decisions lack robust evidence. Methods: This study adhered to the PRISMA-NMA guidelines. Five databases (PubMed, Embase, Cochrane Library, Web of Science, and EBSCOhost) were systematically searched to identify randomized controlled trials (RCTs) published before March 2025. A total of 67 RCTs involving 3,604 ASD participants were included. A frequentist network meta-analysis using a multivariate random-effects model was conducted in Stata, along with pairwise meta-analyses, to compare the relative effects of mindfulness-based interventions (MBI), cognitive behavioral therapy (CBT), behavioral and functional training (BEHAVE), physical activity (PHYS), sensory therapies (SENS), technology- and family-based interventions (TAFI), and other interventions (OTH) on anxiety, depression, and QoL. Standardized mean differences (SMDs) with 95% credible intervals (CIs) were used to estimate effects, and SUCRA rankings were calculated to assess comparative efficacy. Results: MBI showed the greatest improvement in anxiety symptoms (SMD = -0.84, 95% CI: -1.32 to -0.36; SUCRA = 91.4%), CBT ranked highest for depression reduction (SMD = -0.77, 95% CI: -1.25 to -0.28; SUCRA = 90.1%), and PHYS performed best for enhancing QoL (SMD = 0.59, 95% CI: 0.20 to 0.98; SUCRA = 87.5%). The analyzed population primarily consisted of high-functioning male individuals. Subgroup analyses showed stronger effects in adults and with moderate-duration interventions (9-16weeks). No significant inconsistency or publication bias was detected. Limitations: Findings mainly apply to high-functioning ASD populations without intellectual disability. Heterogeneity in interventions and assessments should be considered. Conclusions: Different NPIs exhibit distinct advantages in improving emotional symptoms and QoL among individuals with ASD. MBI, CBT, and PHYS demonstrate relative superiority for anxiety, depression, and QoL respectively, supporting their targeted application in clinical and rehabilitative settings. Future studies should prioritize long-term follow-up, refined intervention designs, and personalized strategies tailored to ASD subgroups to enhance clinical utility and scalability. Systematic review registration: https://www.crd.york.ac.uk/prospero/, identifier CRD420251021423.