Accelerating autism prevalence in California with changepoint in the mid-2010s and divergence by county and socioeconomic indicators

Abstract

Background/Objectives The Autism and Developmental Disabilities Monitoring (ADDM) network has observed an ongoing increase in autism spectrum disorder (ASD) among U.S. children, with highest prevalence in California. Recent county-level trends in California were analyzed to provide insight into this increase. Methods ASD prevalence was estimated from the California Department of Developmental Services (DDS) ASD caseload in 2025, stratified by county, race/ethnicity, and age across birth years 1993-2020. Denominators were estimated from the mean of public school and U.S. census data. Results Diagnosed ASD prevalence increased among all race/ethnicity groups across this period, with a strong uptick in the mid-2010s, most commonly in birth year 2016, in many low and middle income counties. The estimated values varied widely by county and by race/ethnicity, ranging from 1.2% ± 0.2% in San Mateo to 9.5 ± 1.3% in Los Angeles among white and black children, respectively, born in 2019. The rate of change in diagnosed prevalence from birth year 2015-2019 was significantly correlated to county Medicaid (Medi-Cal) fraction for Hispanics and all races, although this ecological association could not distinguish individual-level risk. Comparison of diagnosed ASD prevalence in San Diego County between DDS and ADDM (for which California is represented by San Diego) suggested that DDS tends to underestimate total ASD, likely because it represents the more severely affected subset. Within DDS, diagnosed prevalence trends in San Diego captured statewide California trends relatively well but prevalence was about 20% higher in San Diego by birth year 2019. Conclusions Diagnosed ASD prevalence has risen unequally across California counties and race/ethnicity groups since the birth years of the mid-2010s, with the steepest increases in low and middle income counties. Understanding these patterns and disentangling true increases in ASD from diagnostic changes are urgent priorities for future research.