Nearly 17 percent of 8-year-old autistic children in the United States meet the criteria for profound autism, according to a new estimate by the U.S. Centers for Disease Control and Prevention (CDC).
The new figure is a decrease from the CDC’s previous estimate of 27 percent, published in 2023. The new estimate relies on an updated, more precise definition of profound autism that includes adaptive scores, which contributed to the drop, according to Matthew Siegel, chief of clinical enterprise in the psychiatry and behavioral sciences department at Boston Children’s Hospital, who was not involved in this study.
Another factor: The overall pool of autistic children has grown, so children with profound autism make up a smaller proportion. “We’ve got the definition or the group that meets profound autism getting a little smaller, more precise, and at the same time the denominator is getting bigger,” Siegel says.
But despite the decline, Siegel still finds the new estimate sobering. “That’s a big group of children and families who need intensive support and services,” he says.
The Lancet Commission on the Future of Care and Clinical Research in Autism coined “profound autism” in 2021 to describe autistic people who require 24-hour care as adults and, in most cases, have an IQ score below 50 or limited verbal abilities. The commission also said that the term isn’t appropriate for children under 8 years old. The CDC’s 2023 estimate relied on that definition in its analysis of data from 2000 to 2016.
But not everyone agreed with the Lancet measures, and the criteria vary across different studies. Earlier this year, a Delphi consensus group of more than 70 researchers, caregivers, autistic people and clinicians settled on a new definition: adaptive functioning far below age level and either an IQ score below 50 or minimal to no verbal abilities. The federal Interagency Autism Coordinating Committee, which helps the government decide how to spend its autism research budget, also proposed its own definition this year that does not mention IQ, contrary to both the Delphi consensus and the Lancet Commission.
“What is considered profound autism for one person might not be considered profound autism for another person,” says Joshua Anbar, an investigator on the new work and clinical assistant professor of healthcare administration and policy at Arizona State University. “Coming to an agreed-upon definition becomes really important so we can describe this population of people with significant needs and who require significant support.”
Study investigator Kelly Shaw, of the CDC’s National Center on Birth Defects and Developmental Disabilities, forwarded The Transmitter’s request for comment to a CDC spokesperson. Another CDC spokesperson responded to the request: “This work contributes to an important issue in the autism community by describing the current state of evaluation practices and the impact of using different profound autism definitions. Practices and definitions can evolve to better understand and meet needs of individuals with autism.”
I
n the new study, published earlier this month in the Journal of the American Academy of Child & Adolescent Psychiatry, Anbar and his colleagues applied the Delphi definition to 2022 surveillance data for 8-year-old children at 13 sites in the CDC’s Autism and Developmental Disabilities Monitoring Network (ADDMN). They also calculated profound autism prevalence using competing definitions to address the ongoing debate among scientists and advocates about how to best identify this population.Much of the data for IQ, verbal status and adaptive functioning were missing from the dataset, the team found. Among 7,084 children with autism, only 61 percent had recorded IQ information, 74 percent had data to determine their verbal status, and 56 percent had adaptive score data. In all, 54 percent had all the information the team needed. The team used a statistical method that predicts missing values to fill in the gaps.
“The missing data is a huge problem,” says Catherine Lord, distinguished professor of psychiatry and education at the University of California, Los Angeles, who participated in the Lancet and Delphi groups but was not involved in the new study. Those missing data likely aren’t random, she says, because IQ scores are harder to capture in children who are more likely to fall into the profound autism category.
“It’s hard, because there’s this trade-off of getting a diagnosis, which you can do from a psychiatrist or a pediatrician, and actually getting a full picture of what kids actually are capable of, which you’re less likely to do with a physician because they are under much more time pressure,” she adds.
Across sites, the prevalence of autistic children who met the criteria for profound autism ranged from 10 percent in Utah to 35 percent in Laredo, Texas. Evaluation and diagnostic practices varied across communities, potentially contributing to the wide-ranging results, Anbar says.
To test other definitions, the team analyzed the data using the 2021 Lancet criteria. They also tested the Delphi definition with a higher IQ cutoff of 70, as well as a definition using only verbal status and adaptive scores, without IQ.
The prevalence of profound autism was 34 percent higher when using the Lancet definition than with the updated Delphi definition. Raising the IQ score to 70 resulted in a 71 percent higher prevalence, and using only verbal status and adaptive scores without IQ lowered prevalence by 33 percent.
“It makes the point that depending on how you define it, depending on what you have access to, you’re going to come out with different rates,” Lord says.
Siegel, a member of the Delphi consensus, adds that none of the competing definitions include measures that get at the heart of the issue: whether the person with autism needs round-the-clock access to an adult caregiver.
“This study and all others are using proxy measures of that,” he says. “That’s a key gap in the field that’s begging to be filled.”
