Prevalence of autism spectrum disorders--autism and developmental disabilities monitoring network, six sites, United States, 2000.

PROBLEM/CONDITION: Data from a population-based, multisite surveillance network were used to determine the prevalence of children aged 8 years with autism spectrum disorder (ASD) in six areas of the United States and to describe the characteristics of these children.

REPORTING PERIOD

METHODS

Children aged 8 years were identified as having an ASD through screening and abstraction of evaluation records at multiple sources, with clinician review of abstracted records to determine case status. Children whose parent(s) or legal guardian(s) resided in one of the six surveillance areas during 2000 and whose records documented behaviors consistent with the American Psychiatric Association's criteria for diagnosing 1) autistic disorder, 2) pervasive developmental disorder-not otherwise specified, or 3) Asperger disorder were classified as having an ASD.

RESULTS

For 2000, across six sites, a total of 1,252 children aged 8 years were identified as having an ASD. The overall prevalence of ASDs per 1,000 children aged 8 years ranged from 4.5 in West Virginia to 9.9 in New Jersey. With the exception of one surveillance site (Georgia), no statistically significant (p<0.05) differences were identified in the rate of ASDs between non-Hispanic black and non-Hispanic white children. The ratio of male-to-female prevalence varied (range: 2.8:1.0-5.5:1.0). The majority of children with ASDs received special education services and had a documented history of concerns regarding their development before age 3 years. The prevalence of children with a previously documented ASD classification varied across sites, but the median age of earliest documented ASD diagnosis was similar across sites (age 52-56 months). For three sites with sufficient data on intelligence quotient (IQ), cognitive impairment (i.e., IQ of </=70) was reported for 40%-62% of children whose conditions were consistent with the case definition for ASD.

INTERPRETATION

Findings from this first U.S. multisite collaborative study to monitor ASD prevalence demonstrated consistency across the majority of sites, with prevalence statistically significantly (p<0.001) higher in New Jersey. Average ASD prevalence across all six sites was 6.7 per 1,000 children aged 8 years. These results indicate that ASDs are more common than was believed previously.

PUBLIC HEALTH ACTIONS

Collecting data regarding prevalence of ASDs by associated characteristics (e.g., cognitive impairment, age of first documented concerns, and history of ASD diagnosis), race/ethnicity, and sex will provide important baseline standards that can be compared with follow-up surveillance data to track changes in ASD prevalence. Knowledge of these characteristics has implications for identification and intervention strategies and for medical and educational service planning for children with ASDs.