Predictive Modeling of Autism Spectrum Disorder: Socioeconomic, Prenatal, and Environmental Influences

Abstract

Autism spectrum disorder (ASD) arises from socioeconomic, prenatal, perinatal, and environmental influences. Using data from the 2022–2023 National Survey of Children’s Health (N=82,068, ages 2 to 17), we built predictive models with logistic regression, random forest, support vector machines, gradient boosting, and neural networks. Gradient boosting and neural networks achieved the best performance across accuracy, sensitivity, specificity, precision, F1-score, and AUC-ROC. Key predictors spanned socioeconomic, prenatal, and environmental domains, highlighting the value of multifactorial modeling for ASD risk prediction.