Read our COVID-19 research and news.

Monkey business. Young rhesus monkeys that receive human antibrain antibodies in utero act oddly in social situations.

Nancy Collins/Creative Commons

Autism rates are up, but is it really on the rise?

The number of U.S. school children placed in special education programs due to autism more than tripled from 2000 to 2010, to nearly 420,000. But a new study argues much of that increase likely came as educators swapped one diagnosis for another. The overall percentage of kids diagnosed with a collection of brain development problems that includes autism remained unchanged, suggesting that children who used to be labeled with conditions such as “intellectual disability” were in fact autistic.

“If you asked me, ‘Is there a real increase in the prevalence of autism?’ maybe there is, but probably much lower than the reported magnitude,” says Santhosh Girirajan, a geneticist at Pennsylvania State University (Penn State), University Park.

In the new study, Girirajan and colleagues combed through data collected in each state for approximately 6.2 million U.S. school children with disabilities who are enrolled in special education programs. The information is collected each year under the federal Individuals with Disabilities Education Act. Based on his or her diagnosis, each child was assigned to one of 13 broader categories, ranging from autism to physical challenges such as blindness.

Between 2000 and 2010, the number of children in the autism category more than tripled from 93,624 in 2000 to 419,647 a decade later. Yet nearly two-thirds of that increase was matched by a decline in the rate at which children were labeled as having an “intellectual disability.” The number of kids in that category fell from 637,270 to 457,478.

The data indicate that the autism rise is partly the result of students being moved from one category to another, Girirajan says.

For him, one lesson is that autism encompasses a potpourri of symptoms. It can also occur hand-in-hand with other conditions including intellectual disabilities, epilepsy, and attention-deficit hyperactivity disorder. That mixture makes it hard to pin down how common autism really is, and lends itself to shifting diagnoses, he says.

  	As autism rates have soared in U.S. special education programs, the number of children labeled with "intellectual disabilities" has fallen, suggesting that much of the increased autism comes from shifting diagnoses.

As autism rates have soared in U.S. special education programs, the number of children labeled with "intellectual disabilities" has fallen, suggesting that much of the increased autism comes from shifting diagnoses.

Penn State University

The findings, reported today in the American Journal of Medical Genetics Part B: Neuropsychiatric Genetics, underscore the growing acceptance within institutions and families of a condition once either ignored or avoided as a mark of shame, says Roy Richard Grinker, an anthropologist at George Washington University in Washington, D.C., who has studied autism rates and wrote a book about autism, Unstrange Minds. He points to a recent Washington Post article about the emergence of groups of autistic adults advocating for acceptance through events like “Autistic Pride Day.”

“What this paper is saying is that autism is increasingly being embraced as a useful, acceptable, less stigmatizing framework,” says Grinker, who has an autistic daughter.

There are some risk factors that might account for a small amount of the increase in autism cases, says Jon Baio, a Centers for Disease Control and Prevention (CDC) epidemiologist and principal investigator for an autism monitoring program. That includes children born premature or underweight. A recent study in the journal Molecular Psychiatry found that older fathers, and older or teenage mothers had a higher risk of having children with autism.

But Baio, believes much of the increase they have seen since 2000 comes from growing awareness of autism and more sensitive screening tools. For example, he says, there are now more autism cases with milder symptoms, such as normal or above-normal intellectual ability. At the same time, the number of children identified with autism by experts in the community—such as the school special education programs examined by the Penn State scientists—has come closer to matching the CDC’s more comprehensive screening methods.

The possibility that autism numbers have jumped as children’s' diagnoses are changed has been discussed, says Annette Estes, director of the University of Washington's Autism Center in Seattle. The Penn State study's use of the vast special education database makes a convincing case that is happening, she says. But it doesn't explain all of the increase. "People who are in the field are generally in consensus that the majority of the increase is due to progress in our ability to diagnose and identify people with autism in a broader spectrum than used to be possible," Estes says. "But then there is this portion of an increase that is not accounted for in a lot of statistical studies that are done."

*Correction, 23 July. 12:09 p.m.: The headline of this article has been changed. The original headline referred to autism as a disease.