Yes, the number of children identified with autism has risen dramatically over the past two decades. In 2000, about 1 in 150 eight-year-olds were identified with autism spectrum disorder. By 2022, that figure reached 1 in 31, according to the CDC’s most recent surveillance data. Whether this reflects a true biological increase, better detection, or both is one of the most debated questions in developmental medicine.
How the Numbers Have Changed
The CDC’s Autism and Developmental Disabilities Monitoring Network has tracked prevalence in eight-year-olds since the year 2000. The trajectory is striking:
- 2000: 1 in 150
- 2004: 1 in 125
- 2008: 1 in 88
- 2012: 1 in 69
- 2016: 1 in 54
- 2020: 1 in 36
- 2022: 1 in 31
That’s roughly a fivefold increase in identified prevalence over 22 years. The climb has been consistent, with no plateau in sight. An estimated 2.21% of U.S. adults also have autism, though adult surveillance is far less developed and that figure likely undercounts people who were never diagnosed as children.
Why the Numbers Keep Climbing
No single explanation accounts for the entire increase. Researchers point to several factors working simultaneously, and disentangling them is genuinely difficult.
Broader Diagnostic Criteria
Before 2013, conditions like Asperger’s syndrome and pervasive developmental disorder were classified separately. The shift to a single “autism spectrum disorder” umbrella in the DSM-5 folded these into one category. Critics had already argued that earlier editions of the diagnostic manual widened the criteria enough to inflate prevalence figures. When researchers applied the newer DSM-5 criteria to people who already had a clinical diagnosis under the old system, only about 61% met the revised threshold, suggesting the new criteria are actually somewhat narrower. But the cultural effect of a broader “spectrum” concept has likely encouraged more evaluations overall.
Better Screening and Earlier Detection
The American Academy of Pediatrics recommends screening all children at 18 and 24 months using a validated tool. Among pediatricians who screen for autism, about 90% use the Modified Checklist for Autism in Toddlers (M-CHAT). This kind of universal screening catches children who might have been overlooked a generation ago, particularly those without intellectual disability or obvious speech delays. The result is more diagnoses, not necessarily more autism.
Closing Gaps Across Race and Sex
For years, autism was diagnosed far more often in white boys than in any other group. That disparity has been narrowing. Black and Hispanic children are now identified at rates much closer to white children, which adds to the overall count without representing a new wave of the condition. It represents children who were always there but previously missed.
The gender gap is narrowing too, and the speed of that shift is remarkable. A 2025 study published in The BMJ found that by 2022, the cumulative male-to-female ratio for autism incidence had dropped to 1.2 by age 20. Among those diagnosed between 2020 and 2022, there was no male excess at all in individuals older than 15. The study’s projections suggested the ratio could reach full parity by 2024. Girls and women are more likely to mask social difficulties by mimicking peers’ speech and expressions, which can delay recognition until adolescence, when social demands intensify. Historically, diagnostic tools and criteria were also built around how autism presents in boys, creating a systematic blind spot.
Environmental and Biological Contributors
While much of the rise is attributable to detection, researchers have not ruled out a genuine increase in incidence. Several environmental and biological factors are associated with higher autism risk, and some of these factors are themselves becoming more common in the population.
Parental age is the clearest example. Each 10-year increase in a father’s age is associated with a 22% increase in the odds of autism in offspring, according to a large North American study. Average parental age at first birth has climbed steadily in recent decades, meaning more children are being born to older parents than in previous generations.
Prenatal exposures also matter. Air pollution, certain pesticides (including DDT), maternal obesity, diabetes, and immune system disorders during pregnancy are all linked to higher risk. Extreme prematurity, very low birth weight, and birth complications involving oxygen deprivation round out the list. Maternal fever during pregnancy and certain maternal antibodies that may interfere with fetal brain development have also been flagged. Prenatal and early childhood exposure to lead is another concern: studies have found higher lead levels in autistic children during specific developmental windows.
None of these factors alone would explain a fivefold increase. But as a group, shifting in the same direction over the same time period, they could contribute a real, if modest, increase layered on top of the much larger detection effect.
How Much Is “Real” Increase?
This is the question researchers are still working to answer, and honest estimates vary. Most experts agree that the majority of the rise reflects improved awareness, wider criteria, and better screening reaching previously underserved groups. The consistent pattern of racial and gender gaps closing with each new surveillance cycle supports this view. If autism were simply becoming more common biologically, you’d expect increases to be more uniform across demographics rather than concentrated in groups that were historically underdiagnosed.
At the same time, dismissing the entire increase as an artifact of better counting is hard to justify. The environmental exposures linked to autism are real, measurable, and in many cases increasing. Advanced parental age is a population-level trend that applies broadly. It is plausible, and many researchers believe likely, that some portion of the increase reflects a genuine rise in occurrence layered beneath the larger detection effect.
What’s clear is that 1 in 31 is the current reality for eight-year-olds in the United States. Whether the underlying rate is still climbing, stabilizing, or was always closer to this level than anyone realized, the practical implications are the same: more children and adults need support, services, and systems designed to accommodate a neurodevelopmental profile that is far more common than anyone imagined 25 years ago.