Global Autism Spectrum Disorder Prevalence Estimates and Associated Covariates: A Systematic Review and...

Area of Science:

• Epidemiology
• Neurodevelopmental Disorders

Background:

• Global prevalence estimates for autism spectrum disorder (ASD) show significant variability.
• Factors contributing to this heterogeneity, including methodological and contextual differences, require quantification.

Purpose of the Study:

• To estimate the global prevalence of autism spectrum disorder (ASD).
• To identify methodological and contextual factors influencing ASD prevalence estimates.

Main Methods:

• Systematic literature search of major electronic databases (PubMed, Scopus, Web of Science, Embase, Google Scholar) from 2004-2025.
• Inclusion of 19 population-based studies with ~20.6 million participants.
• Application of random-effects modeling, Bayesian hierarchical modeling, meta-regression, and sensitivity analyses to assess prevalence and covariates.

Main Results:

• Frequentist pooled ASD prevalence was 0.8% (95% CI: 0.4%-1.7%) with high heterogeneity.
• Bayesian modeling yielded a posterior mean prevalence of 1.55% (95% CrI: 0.75%-4.1%), or 0.9% (95% CrI: 0.6%-1.8%) after sensitivity analysis.
• Regional variations observed: North America (1.9%), Africa (6.3%), Europe (0.4%), Latin America (0.2%), Middle East (0.6%).
• Meta-regression indicated diagnostic framework, study setting, and region significantly influenced prevalence estimates (R² ≈ 0.78).

Conclusions:

• Global ASD prevalence is highly variable, primarily due to methodological and contextual factors, not solely demographics.
• Standardized diagnostic practices and enhanced surveillance are essential for accurate and comparable global ASD prevalence data.
• Targeted investment in underrepresented regions is needed to improve data collection and understanding of the global ASD burden.