About 1 in 31 children in the United States has autism spectrum disorder (ASD), based on the most recent data from the CDC’s monitoring network. That translates to a prevalence of 3.2% among 8-year-olds, the age group the CDC uses as its benchmark. The number has risen sharply over the past two decades, driven by a combination of broader diagnostic criteria, better screening, and increased awareness.
How the Numbers Have Changed Since 2000
The CDC’s Autism and Developmental Disabilities Monitoring (ADDM) Network has tracked autism prevalence every two years since 2000. In that first estimate, the rate was 1 in 150 children. By 2020, it had climbed to 1 in 36. The 2022 data, published in 2025, pushed the figure to 1 in 31, a 22% jump over just two years. At nine of the eleven tracking sites, prevalence was between 14% and 37% higher than the previous reporting cycle.
That trajectory looks dramatic, but a large share of the increase reflects changes in how autism is identified rather than a proportional rise in the number of autistic people being born. Diagnostic criteria have broadened, clinicians are better trained to recognize autism across a wider range of presentations, and public awareness has made parents more likely to seek evaluation. The shift from DSM-IV to DSM-5 diagnostic criteria in 2013 actually narrowed the definition slightly. A retrospective analysis found that about 81% of children diagnosed under the older criteria also met the newer ones, and estimated prevalence for a given year would have been roughly 10 per 1,000 under DSM-5 compared to 11.3 per 1,000 under DSM-IV. In practice, though, clinicians adapted their documentation and the overall trend continued upward.
Children Are Being Identified Earlier
One of the clearest patterns in the recent data is that younger children are being identified at much higher rates than previous generations were at the same age. Children born in 2018 had 1.7 times the rate of autism identification by age 4 compared to children born in 2014. That gap was consistent across most tracking sites, ranging from 1.4 times higher in Arizona and Georgia to 3.1 times higher in Puerto Rico.
This matters because earlier identification typically means earlier access to support services. Across 13 of the 16 monitoring sites, identification by age 4 was 40% to 300% higher in the younger group compared to the older one at the same age. That kind of improvement suggests screening practices are catching children who, a decade ago, might not have been diagnosed until school age or later.
Sex Differences in Diagnosis
Autism is diagnosed far more often in boys than in girls. The ratio has historically hovered around 3 to 4 boys for every girl. That gap has narrowed slightly over time as clinicians have become better at recognizing autism in girls, who often present differently. Girls are more likely to mask social difficulties or develop compensatory strategies that delay diagnosis. The DSM-5 criteria change did not appear to affect boys and girls differently; the percentage meeting the newer criteria was similar across sexes.
Racial and Ethnic Patterns Are Shifting
For years, white children were diagnosed with autism at higher rates than Black, Hispanic, and Asian children. That disparity was largely a product of unequal access to diagnostic services rather than a true difference in how common autism is across racial groups. Recent ADDM data show that gap has closed considerably, and in some communities it has reversed. Identification rates among Black and Hispanic children have risen faster than among white children as screening has become more consistent across healthcare settings. The remaining disparities likely reflect ongoing differences in access to developmental evaluations, insurance coverage, and culturally responsive care.
Global Prevalence Estimates
Outside the United States, the World Health Organization estimates that about 1 in 127 people worldwide had autism in 2021. That figure is substantially lower than the U.S. rate, but the gap reflects differences in surveillance infrastructure more than biology. Many low- and middle-income countries have little or no population-level data on autism. In countries with well-resourced tracking systems, reported prevalence tends to be much higher and closer to U.S. figures. The WHO notes that some well-controlled studies report rates substantially above the 1-in-127 average.
Why the Numbers Keep Rising
The steady climb in autism prevalence raises an obvious question: are more people actually autistic, or are we just getting better at counting? The honest answer is that both are likely true, but the measurement side accounts for most of the increase. Several forces push the numbers up with each reporting cycle.
Broader awareness means more parents request evaluations. School systems have expanded their autism eligibility criteria. Pediatricians now screen routinely at 18- and 24-month well-child visits using tools like the M-CHAT (Modified Checklist for Autism in Toddlers), which catches children who previously would have been missed or labeled with a different condition. The sensitivity of the M-CHAT varies across studies, from about 67% to 95%, meaning it flags most but not all children who will eventually be diagnosed.
At the same time, researchers cannot rule out a genuine increase in autism itself. Environmental factors, parental age at conception, and other biological variables are active areas of investigation. Disentangling a true rise from better detection is one of the hardest problems in autism epidemiology.
The Economic Scale of Autism
The rising prevalence carries significant economic implications. A widely cited projection estimated that the combined costs of direct medical care, non-medical services like behavioral therapy and special education, and lost productivity would reach $268 billion annually in the U.S. by 2015 and $461 billion by 2025. The upper-range estimate for 2025 was over $1 trillion, representing as much as 3.6% of GDP. These figures reflect lifetime support needs that vary enormously from person to person, with the highest costs concentrated among individuals who need full-time assistance.