Trends in Birth Prevalence of Autism Spectrum Disorders (ASD) in California from 1990 to 2010, By Race-Ethnicity and Income
Objectives: To characterize trends in ASD birth prevalence in California over two decades by race-ethnicity and income.
Methods: Data on ASD diagnoses were obtained using statewide Department of Developmental Services (DDS) data of clients receiving services for autism as of February 2018. DDS data were linked to birth certificate data for live births to California resident mothers from 1990 to 2010, excluding infant deaths. The linked data therefore represent children who were at least 7 years old at the time of entry into the DDS system. Annual ASD birth prevalence (cases per 1,000 births) was calculated overall and for non-Hispanic white, non-Hispanic black, and Hispanic mothers, and further stratified by insurance type during pregnancy or delivery, with public insurance used as a marker of low income. Average annual percent change (AAPC) in birth prevalence and change in AAPC were assessed using Poisson regression.
Results: A total of 89,107 cases were identified from 13,014,703 births. ASD birth prevalence increased from 1.9 to 11.7 between 1990 and 2010, or 7.5% annually. The AAPC was highest for children of Hispanic mothers (9.2%, vs. 8.0% for blacks and 6.3% for whites; p<.001) and for low-income (publicly insured) families regardless of race-ethnicity (10%, vs. 6.1% for privately insured; p<.001). Trends in ASD birth prevalence differed by time period. For children born from 1990 to 2000, ASD prevalence was similar for children of white and black mothers, and lower for Hispanics, with annual increases of 11.2%, 9.5% and 14.2%, respectively. Over the next decade, the AAPC in ASD prevalence decreased to 3.5%, 7.1% and 7.0% among children of white, black and Hispanic mothers, respectively. The largest decrease in the AAPC between decades was for white mothers who were privately insured (AAPC from 10.1% to 1.8% after birth year 2000; P<.001). By 2010, ASD prevalence among children of black mothers was significantly higher than among children of white mothers (14.9 vs. 10.7), and Hispanic prevalence (11.4) was similar to white prevalence.
Conclusions: In this large, population-based analysis over 21 years, ASD prevalence increased for white, black and Hispanic children in California, but at different rates so that Hispanic prevalence matched that of whites, and black prevalence exceeded that of whites. Increases were particularly noteable among low-income families. These increases may reflect improved access to screening, diagnosis and/or services.