Prevalence of Autism Spectrum Disorders (ASD) Among Young Children, by Demographic Characteristics

Background: Standardized collection of data for monitoring trends has documented continuing increases in identified ASD prevalence in the U.S. among school-aged children (8-year-olds). Despite concerns about development before the ages of 2- and 3-years, most children are not diagnosed until after age 4 years.  Delays in early evaluation and diagnosis create challenges for monitoring ASD prevalence among young children.

Objectives: Establish a multi-source, active surveillance system to determine prevalence and demographic characteristics of children < 48 months identified with ASD in one-county in California for two birth years, 2005-2006.

Methods: Methods are based on those of the Autism and Developmental Disabilities Monitoring (ADDM) network established by the U.S. CDC. Records of children with a diagnosis or signs of ASD at health-related sources (Early Start Programs, CA Department of Developmental Services, Kaiser Permanente Medical Care Program, and clinics) were reviewed and abstracted. An expert review process was followed to determine surveillance case classification consistent with DSM-IV-TR criteria for autistic disorder, PDD-NOS, or Asperger’s disorder. Both birth and current prevalence were determined. Records were linked to birth certificates to identify resident births and obtain demographic information. Census data for the population of 2 and 3-year olds in the appropriate years were used for current prevalence. Prevalence ratios (PR) and 95% CI were calculated to compare sub-groups.

Results: For 2005 births, 200 children were ascertained as definite cases for a prevalence of 7.6/1000 births (95% CI 6.5-8.6). Prevalence was slightly higher in 2006 at 8.4/1000 births (CI =7.3-9.4). Combining the two years, prevalence was significantly higher among males than females with a PR of 3.7 (CI 2.9-4.7). Prevalence varied by race/ethnicity as well, with a significantly elevated PR among Asians compared to white, non-Hispanics (PR=1.3, CI 1.0-1.7) and slightly elevated among Blacks, although based on small numbers. Prevalence was significantly lower among Hispanics, particularly among children whose mothers were foreign-born (PR= 0.49, CI 0.36-0.69). Multi-variate modeling and additional sub-group analyses will be conducted.

Conclusions:  The identified ASD birth prevalence among young children (<4) is about 75% higher than prevalence in this area among 8-year old children a decade earlier, although case-finding methods differed slightly. Furthermore, although on the low side, rates are within the range of US ADDM Network sites reporting 8-year-old prevalence. These results indicate ASD prevalence can be estimated at younger ages on a population basis and provide evaluation of methods for monitoring at ages less than four.