Summary: The perceived “autism epidemic” is largely driven by broadened diagnostic criteria and increased identification of mild, non-impairing cases rather than a true increase in severe autism. Cultural and institutional factors—like overdiagnosis and shifting norms in child psychiatry—play a significant role. Public health efforts are being misdirected by alarmist narratives and should instead focus on genuine health crises and more consistent, clinically meaningful diagnostic standards.

For years, public health debate has often fixated on a supposed rise in the prevalence of autism. Various culprits have been named, including the well-investigated but unsubstantiated claim that vaccines cause autism. More recently, additional risk factors have been proposed — many by Health Secretary Robert F. Kennedy Jr. — including maternal Tylenol use, food dyes and additives, chemical manufacturing agents and other possible stressors affecting perinatal development. Concerns about autism have been spotlighted within the larger Make America Healthy Again movement, motivated by a well-founded alarm over the nation’s devastatingly high burden of chronic disease and psychiatric illness. But there is a bigger problem with the autism epidemic: It doesn’t exist.

Autism diagnoses have indeed risen dramatically in recent decades. The most recently released Centers for Disease Control and Prevention report on autism, which was published in April last year, revealed a nearly five-fold increase in the prevalence of autism between 2000 and 2022, from 67 to 322 cases per 10,000 children. However, diagnostic criteria can change even when the underlying health phenomenon remains unchanged. A large-scale study published in December, drawing on CDC data from 24,669 8-year-olds across the country, suggests that the dramatic rise in autism diagnoses may be entirely driven by children with mild symptoms and no significant functional impairment. Between 2000 and 2016, there was a 464 percent increase in diagnoses among children with no significant functional impairment whatsoever. In fact, during the same time period, there was a 20 percent decrease in the prevalence of moderate or severe autism, from 15 to 12 cases per 10,000 children.

There is often a lag of several years before such epidemiological datasets are released, and years more for researchers to perform statistical analyses, publish the findings and enter public policy discussions. We do not yet have data more recent than 2016 breaking down symptoms by severity level while controlling for other psychological factors such as intellectual disability. However, given the trends observed between 2000 and 2016, it is highly unlikely that the additional 74 percent increase in autism diagnoses between 2016 and 2022 reflects a sudden surge in severe, functionally impairing autism. Rather, it is more likely a continuation of the same problem of overrepresentation among children with mild symptoms and no significant functional impairment.

Despite that, some advocates support the narrative that autism is on the rise, because an ever-expanding “spectrum” that produces more diagnoses draws more attention and research funding — even if children’s underlying psychology remains unchanged.

Some of the CDC’s data documenting the supposed rise in the characteristics of autism, meanwhile, comes not from gold-standard in-person psychiatric assessments but from parent-reported surveys such as the Social Responsiveness Scale. The SRS includes statements such as “Would rather be alone than with others,” “Has difficulty making friends,” and “Is regarded by other children as odd or weird,” which parents rate from “Not true” to “Almost always true.” In my own doctoral research on adolescent mental health, I included the SRS to account for the extent to which other psychological outcomes were explained by social difficulties. However, I was always careful to use hedging language — these are behavioral traits known to be associated with autism, not diagnostic markers. Unfortunately, many studies use high scores on the SRS as a substitute for clinical assessment of autism — accounting, for example, for at least 12 percent of “suspected cases” in the 2022 CDC data.

We should be concerned about the rising number of quirky children “on the spectrum,” but not because they are being exposed to neurotoxins that older generations were insulated from, nor because a growing number of children face clinically–significant social impairment. Rather, as Abigail Shrier argues in her 2024 book “Bad Therapy,” the more pressing concern may be a cultural and institutional drift toward overdiagnosis across child psychiatry. Like the rise in attention-deficit/hyperactivity disorder, anxiety and depression diagnoses among young people, the surge in autism labels may reflect shifting norms, looser diagnostic criteria and excess therapeutic attention directed toward ordinary struggles. If autism were truly increasing because of a new environmental insult, we would expect to see increases across all levels of severity. But that is not the case.

This reality should fundamentally reshape our national conversation. Policymakers and public health officials have rallied around dramatic claims fueled more by fear than by evidence. Yes, America faces a real crisis of chronic disease — including obesity, metabolic dysfunction and autoimmune disorders — which plausibly could be impacted by environmental toxins. Yes, many children face real mental health challenges that warrant increased attention and psychiatric support. But neither of these narratives survives scientific scrutiny when applied to the rise in autism diagnoses.

When public discourse starts from an alarming headline — “Autism rates have quadrupled” — even careful scientists can be pressured into chasing explanations for a biological phenomenon that doesn’t exist. The result is a misallocation of scientific effort and a blurring of the real signals of environmental harm. In many cases, the kid labeled “on the spectrum” is the same train‑obsessed third‑grader your grandfather knew, only now he’s been assigned a diagnosis. Let’s instead direct public health toward real, ongoing health crises and insist on psychiatric criteria that are consistent, unexaggerated and clinically meaningful.

A version of this article was published at the Washington Post on 2/10/2026.