Dec. 3, 2021 -- Childhood autism rates are at the highest level since the CDC began tracking the disorder in 2000, new data released today show.
The increase likely reflects improvements in diagnosis and identification of autism spectrum disorder (ASD), not an increase in the number of children actually having autism, study authors with the CDC’s Autism and Developmental Disabilities Monitoring (ADDM) Network say.
Using a new surveillance method, researchers found that 2.3% of 8 year olds in communities in 11 states were diagnosed with autism in 2018, up from 1.9% in 2016.
A separate report on early identification in 4-year-olds shows that children born in 2014 were 50% more likely to be diagnosed with autism or be given an ASD special education classification by 4 years of age than those born in 2010, which shows improvement in early diagnosis.
Taken together, the data suggests efforts to raise awareness about autism are working, though researchers were quick to say much work remains.
"It means we’re doing a better job of identifying children, which helps to get them into services earlier so they can achieve their best developmental outcome,” says Stuart Shapira, MD, PhD, associate director for science in CDC’s National Center on Birth Defects and Developmental Disability.
The studies, published online Thursday in Morbidity and Mortality Weekly Report, are the first to use a new ASD surveillance system that relies on ASD diagnosis or special education classification and medical billing codes, and eliminates comprehensive records searches.
The updated methodology was less labor-intensive and took less time, but it is not without its critics, some of whom claim it will undercount the number of children with ASD.
Created in 2000 and funded by the CDC, the ADDM Network is the only surveillance program in the U.S. that tracks the number and characteristics of children with autism spectrum disorder in multiple communities.
When ADDM released its first report in 2007 from six states and based on data from 2000, ASD prevalence was 6.7 per 1,000 children or 1 in 150 children.
In the latest report, which includes data from 2018, the autism prevalence rate across 11 states was 23.0 per 1,000 children, or 1 in 44 children.
That rate is closer to reported autism prevalence from the National Survey of Children’s Health and the National Health Interview Survey, both of which rely on parent-reported ASD diagnoses.
For the report, researchers analyzed medical and special education records of 220,281 children born in 2010 in Arizona, Arkansas, California, Georgia, Maryland, Minnesota, Missouri, New Jersey, Tennessee, Utah, and Wisconsin.
Children were counted as having autism if their records included an ASD diagnosis, a special education classification of ASD, or an ASD International Classification of Diseases (ICD) code. A total of 5,058 children met that criteria.
Rates of ASD ranged from a low of 1.7% in Missouri to 3.9% in California and were more than 4 times higher in boys than in girls. Just under half of the children with ASD were evaluated by the age of 3.
Although the overall ASD prevalence was similar among white, Black, Hispanic, and Asian/Pacific Islander children, the report highlighted a number of other racial disparities overall and in individual states.
For example, among those with ASD and data on cognitive ability, 35.2% had an intelligence quotient (IQ) score of less than or equal to 70. Black children with ASD were far more likely to have an IQ in that range than Hispanic or white children.
"The persistent disparities in co-occurring intellectual disabilities in children with autism is something that we continue to see and suggests that we need to better understand exactly what’s happening," Matthew Maenner, PhD, an epidemiologist and autism surveillance team lead with the CDC’s National Center on Birth Defects and Developmental Disabilities, says.
Another long-standing trend observed again low ASD prevalence among 8-yea-rold Hispanic children. While the overall estimate showed similar autism rates, a closer review of state-level data reveals a different picture.
"In almost half of the sites, Hispanic children were less likely to be identified as having ASD," he said. "This gets lost if you look only at the overall estimate."