Skip navigation

Calls to Action

Shift in diagnosis only partly explains rise in autism prevalence

New report shows increase in special education services for autism partly offset by decrease in services for intellectual disability
July 23, 2015


A new report shows that a three-fold increase in autism among special education students between 2000 and 2010 is partly offset by a decrease in diagnoses for intellectual disability.

The report, by researchers at Penn State University, appears in the American Journal of Medical Genetics. The researchers analyzed eleven years of U.S. special-education enrollment information on around 6 million children per year.

Intellectual disability is generally defined as involving an IQ below 70. It often co-occurs with autism and other developmental disabilities and can “mask” them in the absence of a skilled evaluation.

Increased awareness of autism has helped increase recognition and diagnosis of the disorder – with or without intellectual disability, experts agree.

“For years, we’ve known that a proportion of the increase in autism prevalence is attributable to increased awareness of autism and changes in diagnostic criteria,” comments epidemiologist Michael Rosanoff, Autism Speaks director of public health research.

“However, these factors alone don’t account for the entire increase in autism prevalence reported over the past two decades,” Rosanoff says. “It’s vitally important that we conduct research that teases out the effects of diagnostic changes on changes in autism prevalence over time to understand the factors that are truly causing a rise in autism.”

The Penn State researchers used data from the United States Individuals with Disabilities Education Act (IDEA) for students enrolled in special-education programs. Under IDEA, individuals are classified into one of thirteen disability categories including autism spectrum disorders, intellectual disability and other learning disabilities.

Although many of these categories can co-occur, children can be classified under only one category.

“Because this sample included only children enrolled in special education, it represents only a portion of the autism population,” Rosanoff adds. “Many less severely affected students wouldn’t be enrolled in these programs and thus not included in this analysis.”