Autism is one of the most common developmental conditions in the world. Globally, about 1 in 127 people have autism, according to 2021 estimates from the World Health Organization. In the United States, the numbers are higher: the CDC’s most recent data, from surveillance year 2022, found that 1 in 31 children aged 8 has been identified with autism spectrum disorder.
U.S. Prevalence in Children
The CDC tracks autism prevalence through its Autism and Developmental Disabilities Monitoring (ADDM) Network, which surveys 8-year-olds across 16 sites around the country. The 2022 data put the rate at 32.2 per 1,000 children, or about 3.2%. That 1-in-31 figure is the highest the network has ever reported and reflects a steep climb from earlier estimates. Two decades ago, autism prevalence in the U.S. was closer to 1 in 150.
The upward trend has been consistent from one reporting cycle to the next, and several factors help explain it. Diagnostic criteria have broadened over time, folding in conditions like PDD-NOS (pervasive developmental disorder, not otherwise specified) that were previously tracked separately. Awareness among parents and pediatricians has increased dramatically, and many states now require developmental screening at well-child visits. A formal autism diagnosis also serves as a gateway to therapy, school accommodations, and insurance coverage, which gives families a practical reason to pursue evaluation.
How Many Adults Have Autism
Autism doesn’t end at childhood, but tracking it in adults is harder because many people were never diagnosed as children. A 2017 modeling study estimated that roughly 2.2% of U.S. adults, about 5.4 million people aged 18 to 84, were living with autism. That figure includes both people who received a childhood diagnosis and those who were never formally identified. The true number is likely higher today, as more adults seek evaluation after recognizing traits in themselves or their children.
Gender Differences in Diagnosis
Autism has long been considered far more common in boys, with older data suggesting a male-to-female ratio around 4 to 1. Newer research paints a more nuanced picture. A large population-based study published in The BMJ tracked birth cohorts over time and found that the ratio narrows dramatically with age. In childhood, the male-to-female ratio was roughly 3 to 1. But for individuals older than 15 in birth cohorts from 2000 onward, the ratio dropped to 1 to 1 or even slightly favored females. By age 20, the cumulative ratio was just 1.2 males for every female.
The reason is timing. Boys tend to be diagnosed earlier, with peak diagnosis rates occurring between ages 10 and 14. Girls are more often identified later, with their peak falling between ages 15 and 19. This delay likely reflects the way autism can present differently in girls and women, who may mask social difficulties more effectively during childhood. The gap isn’t a difference in how common autism actually is between sexes. It’s a difference in when and whether it gets recognized.
Why the Numbers Keep Rising
The steady increase in autism prevalence raises an obvious question: is autism actually becoming more common, or are we just getting better at finding it? The honest answer is that it’s probably both, though improved detection accounts for a large share of the increase.
Changes in how autism is defined have had a significant effect. When the diagnostic manual shifted from the DSM-IV to the DSM-5 in 2013, several previously separate diagnoses (Asperger’s syndrome, PDD-NOS) were consolidated under one umbrella: autism spectrum disorder. That single change brought more people into the count. At the same time, researchers have noted that the DSM-5 criteria are more specific but somewhat less sensitive, meaning they’re better at confirming autism when it’s present but may miss some people on the broader end of the spectrum.
Greater awareness plays a role too. Pediatricians screen more routinely now. Schools flag developmental differences earlier. Parents have access to checklists and information that didn’t exist a generation ago. Communities that were historically underdiagnosed, including Black and Hispanic families, are now being identified at rates closer to their white peers, which also pushes overall numbers up.
Whether there is also a genuine increase in the underlying rate of autism, potentially driven by environmental or biological factors, remains an open question. But the bulk of the trend can be traced to broader criteria, better screening, and wider access to evaluation.
Global Variation
The WHO’s 1-in-127 global estimate is an average, and it comes with a significant caveat: prevalence varies enormously from one study to the next. Some well-controlled studies report substantially higher figures. In many low- and middle-income countries, autism prevalence data simply doesn’t exist, which means the global average almost certainly underestimates the true number. Countries with robust screening programs and accessible diagnostic services consistently report higher rates, not because autism is more common there, but because the infrastructure to identify it is in place.
The U.S. figure of 1 in 31 is among the highest in the world, driven largely by its extensive surveillance network and relatively broad access to developmental evaluations. Countries with fewer resources for screening tend to report lower numbers, creating a misleading impression that autism is rare in those populations.