CDC: Autism Up 23% From 2006 to 2008 as Rates Continue to Rise
By Daniel J. DeNoon
WebMD Health News
Reviewed by Laura J. Martin, MD
March 29, 2012 -- One in every 88 U.S. children -- and one in 54 boys -- has autism, the CDC now estimates.
The latest analysis, from a 2008 survey, shows autism is up 23% since 2006 and 78% since 2002.
"This is a large number of children and families affected by autism," study leader Marshalyn Yeargin-Allsopp, MD, chief of the CDC's developmental disabilities branch, tells WebMD.
Why the increase? That isn't clear. A number of factors likely contribute to the increase, says Coleen A. Boyle, PhD, director of the CDC National Center on Birth Defects and Developmental Disabilities.
"We know that some of it is due to how children with autism are identified and served in their local communities," Boyle tells WebMD. "We do feel doctors are getting better at diagnosing autism. ... But we don't know how much is due to better identification and diagnosis, how much is due to availability of services, and how much is a true rise in prevalence."
One hint comes from data showing that autism prevalence is higher in areas where doctors are better at diagnosing autism in children with relatively high intellectual ability.
The CDC study -- the Autism and Developmental Disabilities Monitoring (ADDM) study -- is based on data from more than 337,000 8-year-olds in 14 states. That's 8.4% of all U.S. 8-year-olds. The study first used health and education records to identify children who may have autism. Then all of the records were analyzed by autism professionals to identify kids who fit the current autism diagnosis.
Autism rates varied widely across states. Autism prevalence was one in 47 kids in Utah, but only one in 210 children in Alabama. Study sites that relied only on health records to identify children with autism had significantly lower autism rates than sites that had both health and education records.
In Colorado, for example, there was a single county with access to both education and health records. The autism rate there was twice as high as the rate in six Colorado counties with health records only.
Despite the different autism rates across sites, the overall autism prevalence detected is similar to that estimated by other national health surveys.
"This method is really the gold standard for tracking autism," Boyle says. "One thing we do know is we don't overestimate autism prevalence."
Another study, the CDC's huge multi-year Study to Explore Early Development (SEED), exploring various autism risk factors, began in 2008. The very first results should start coming out later this year. But since SEED follows kids from the time of their mother's pregnancy, it will take time for the study to mature.
It's known that autism results from a complex interaction between genetic and environmental influences. But it's not known which types of autism are most closely linked to which factors.