MedPageToday: "Autism Rates Climbing"

In a large autism monitoring network, an estimated one out of every 88 8-year-olds had an autism spectrum disorder (ASD) in 2008, the most recent year with data available, CDC researchers reported.

That's a relative increase of 23% from a previous analysis of data from the same network for 2006, when the estimated prevalence was one out of 111 children, and a 73% relative increase from 2002, according to a surveillance summary in Morbidity and Mortality Weekly Report.

But getting a firm grasp on the prevalence of ASD is tricky because of a lack of objective diagnostic markers and changes in clinical definitions over time, so it's unknown how much of the increase is real and how much is related to changing diagnostic criteria and better identification of cases.

"Given substantial increases in ASD prevalence estimates over a relatively short period, overall and within various subgroups of the population, continued monitoring is needed to quantify and understand these patterns," Jon Baio, EdS, of the CDC's National Center on Birth Defects and Development Disabilities, wrote in the report.

The researchers collected data from the 14 sites of the Autism and Developmental Disabilities Monitoring (ADDM) Network, which conducts active surveillance system of children who are 8 years old, the age at which ASD prevalence peaks.

Prevalence is estimated not on the basis of professional or family reporting of a diagnosis but on analysis of the children's evaluation records from multiple sources, including general pediatric health clinics, specialized programs for children with developmental disabilities, and special education programs in public schools.

Children are considered to have an ASD if they display behaviors consistent with autistic disorder, pervasive developmental disorder-not otherwise specified (PDD-NOS), or Asperger disorder.

The methods within ADDM have remained consistent every year, allowing comparisons over time.

The 2008 prevalence varied widely among the ADDM sites, from 4.8 per 1,000 children in Alabama to 21.2 in Utah.

As seen in previous studies, boys had a substantially higher rate than girls (18.4 versus 4 per 1,000).

In addition, non-Hispanic white children had a higher rate than non-Hispanic black children or Hispanic children (12 per 1,000 versus 10.2 and 7.9, respectively).

The researchers noted that the ADDM sites are not nationally representative and that the results should not be generalized to children throughout the country.