Global Prevalence and Covariates of Autism Spectrum Disorder: A Meta Analysis of the Past Two Decades (2004 to 2025)

BackgroundAutism spectrum disorder (ASD) is a neurodevelopmental condition characterized by difficulties in social communication and restricted, repetitive behaviors emerging in early childhood. Given rising diagnostic rates and substantial lifelong impacts, accurate global prevalence estimates are essential for health policy, resource planning, and early intervention. This systematic review and meta-analysis synthesized global observational evidence to estimate ASD prevalence across world regions and examine factors influencing heterogeneity. MethodsFollowing PRISMA guidelines, searches were conducted in PubMed, Scopus, Web of Science, Embase, and Google Scholar for observational and registry-based studies published from 2004-2025. Twenty-two studies met inclusion criteria. Random-effects models (REML) estimated pooled global and regional prevalence. Heterogeneity was assessed using Q, I{superscript 2}, {tau}{superscript 2}, Baujat plots, and influence diagnostics. Subgroup analyses and meta-regressions evaluated moderators including geographic region, mean age, and prevalence per 1,000. Sensitivity analyses assessed model robustness. ResultsThe pooled global prevalence of ASD was 1.8% (95% CI: 0.8-3.7%). Substantial heterogeneity existed across studies (Q p <.001; {tau} = 1.64), with prediction intervals ranging from 0.05% to 37.8%. Regional estimates ranged from 0.65% in the Middle East to 3.4% in North America, though subgroup differences were not statistically significant. Meta-regression suggested significant moderation by mean age and prevalence per 1,000 (R{superscript 2} = 59.35%), although sensitivity analyses revealed that these effects were largely driven by one influential study. Publication bias was minimal. ConclusionGlobal ASD prevalence remains around 1%-2%, consistent with contemporary epidemiological data; however, considerable heterogeneity indicates that methodological and contextual factors likely drive observed regional differences. Standardized diagnostic criteria, enhanced surveillance, and early detection systems are needed to improve global prevalence accuracy and inform public health planning.