A new study from the US Center for Disease Control (CDC) puts the prevalence of Autism Spectrum Disorders (ASD) at 0.9% (one in every 110) of 8-year-old children. Writing in Morbidity and Mortality Weekly Report, Catherine Rice of the CDC’s National Center on Birth Defects and Developmental Disabilities reported the results of her study 18 December.
In what is called a surveillance study, Ms. Rice examined educational and medical records of children age 8 in 2006, because that age appears to be a peak age for prevalence of ASD. Research teams from sites that participate in the Autism and Developmental Disabilities Monitoring (ADDM) Network monitor public health records. Ms. Rice’s report aggregated the data from 10 sites.
At each site, teams collected assessments conducted by other community professionals (i.e., not members of the research teams), including clincians and educators (pediatricians, psychiatrists, neurologists, psychologists, or speech-language pathologists). Team specialists reviewed those records using a coding guide based on diagnostic criteria and classified children as meeting or not meeting criteria for ASD.
Children… were classified as having an ASD if they displayed behaviors documented in evaluation records by a community professional that were consistent with the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision (DSM-IV-TR) criteria (15) for the subtypes of autistic disorder, PDD-NOS (including atypical autism), or Asperger disorder at any time through age 8 years.
The researchers analyzed the data by dividing the number of cases by the estimated number of children living in the area studied by the ADDM sites. They then conducted additional analyses by geographic area, ethnicity, gender, special education status, whether the child had received an earlier diagnoses from a qualified professional, etc.), but the main story that will be reported in the press is the prevalence rate of 1:110.
Here’s the first paragraph of the results section:
In 2006, the overall identified ASD prevalence per 1,000 children aged 8 years varied across ADDM sites (range: 4.2 [Florida]–12.1 [Arizona and Missouri]) (Table 2). The average across all 11 sites was 9.0 (CI = 8.6–9.3) per 1,000 children. Overall estimated ASD prevalence per 1,000 chil- dren aged 8 years was significantly lower (p<0.01) in Florida (4.2; CI = 3.5–5.0) than in all other sites, whereas overall estimated ASD prevalence in Alabama (6.0; CI = 5.3–6.9) was significantly lower (p<0.05) than identified ASD prevalence in each of the remaining sites except Colorado (7.5; CI = 5.7–9.8).
Ms. Rice’s report also compares the rates from the 2008 study with the rates from a comparable 2002 study. She found that that all 10 sites with comparable data reported increased numbers of children classified as ASD in the more recent study. This finding calls into question one of the explanations for reported increases in the prevalence of ASD. That explanation, the ascertainment bias (that the rates are increasing because of heightened public awareness of ASD), does not fit these data well, because the ADDM methods were consistent across time (i.e., relied on the researchers’ classification systems, which were essentialy the same in both studies).
Of particular interest to me is the variation in the classification according to geographic area. Why are the rates different in Florida and Alabama? (Note that variation by state in the identification rates for children with Learning Disabilities has long been used as an argument that LD is poorly defined.)
Although the results are surely going to be intrepreted in ways that are consistent with interpreters’ existing views about Autism, I think it’s important to keep a rational perspective on this study. It is a very good study, but it is not the most definitive examination of the prevalence of ASD.
It is employs existing records rather than individual interviews.
The results depend on the availability and quality of the records.
Although the the classification-by-records data and the identified-previously data are correlated, the ADDM-identification rates are conistently higher than the community-identification rates.
The denominator in the prevalence rates is an estimate, not a count (though it’s probably a very good estimate).
Rice, C. (2009). Prevalence of autism spectrum disorders—Autism and Developmental Disabilities Monitoring Network, United States, 2006. Morbidity and Mortality Weekly Report, 58(SS10), 1-20.