Why Is Autism More Common Now? What's Actually Driving the Rise

Introduction

If you've noticed that autism seems more common today than it used to be, you're noticing something real, but probably not the thing you might think you're noticing. Current data from the CDC's Autism and Developmental Disabilities Monitoring Network (April 2025) puts autism prevalence in the United States at 1 in 31 children, up from roughly 1 in 150 two decades ago.

That's a substantial shift. But the honest answer about what's driving it isn't "something in the environment causing more autism." Most of the rise reflects better identification of autism that was always there, through expanded diagnostic criteria, increased awareness, improved screening, and recognition of populations that were previously missed. Some real increase may also be occurring, but it's a smaller portion of the picture than the headlines often suggest.

And critically, since this question gets repeatedly tangled with misinformation: vaccines do not cause autism. We'll address this directly because the topic deserves a clear answer.

This piece focuses on what's actually driving the rise right now. For a broader look at how autism understanding has evolved over the decades, see our companion piece on autism being more common than 20 years ago.

The Current Picture

A few specific data points worth grounding the conversation:

US prevalence (CDC ADDM 2025): 1 in 31 children (3.2%)
Boys diagnosed at higher rates than girls: roughly 3-4:1 in current data, though research (Loomes 2017 meta-analysis) suggests the true ratio is closer to 3:1, meaning many autistic girls are still being missed
Global prevalence (GBD 2021): roughly 1 in 127 people, or about 61.8 million autistic individuals worldwide
Intellectual disability co-occurrence (CDC ADDM 2025): approximately 38% of autistic children

These are the current, reliable figures. Some commentary cites older numbers like "1 in 36" — that was the 2020 surveillance year data, since updated.

What's Actually Driving the Rise

The increase in prevalence reflects several specific, established factors. Each of these accounts for a meaningful portion of what looks like rising autism rates.

Diagnostic Criteria Have Expanded Substantially

This is the single largest factor in the apparent rise. The diagnostic categories that define autism have changed substantially over the decades:

Pre-1980: Autism was rarely diagnosed and required severe presentations matching narrow criteria
DSM-III (1980): First formal autism diagnostic criteria, still narrow
DSM-III-R (1987) and DSM-IV (1994): Broadened criteria; DSM-IV added Asperger's syndrome and PDD-NOS, capturing presentations that didn't fit the narrow autism criteria
DSM-5 (2013): Consolidated previously separate diagnoses (autism, Asperger's, PDD-NOS, others) into a single Autism Spectrum Disorder diagnosis with specified support levels

Each expansion meant more people qualified for autism diagnosis under the new criteria who wouldn't have under the old ones. This isn't disease prevalence changing, it's how we define what we're counting.

For more on the terminology evolution and what diagnostic categories are used now, see our piece on why "high-functioning autism" is no longer used.

Awareness and Screening Have Improved Substantially

Pediatricians now routinely screen children for autism at well-child visits. The American Academy of Pediatrics recommends autism-specific screening with tools like the M-CHAT-R at both 18 and 24 months. This wasn't standard practice 20 years ago.

Beyond clinical screening:

Parents have more access to information about autism signs
Teachers and early childhood educators are increasingly trained to recognize autism characteristics
Pediatric subspecialty access has expanded
Public awareness campaigns and autism representation in media have made autism more visible

When you screen more, you find more. This isn't a problem; it's how detection of any condition works.

Populations Previously Missed Are Increasingly Identified

Several populations were historically underdiagnosed and are now being identified more accurately:

Girls and women — Autistic girls present differently on average than autistic boys (more internalizing patterns, more effective masking, special interests that don't match the stereotypical autism picture), and diagnostic criteria built primarily on observations of boys missed many of them. As recognition of female autism presentations has grown, more girls and women receive diagnoses. (For more, see our piece on autism in girls.)

Black and Hispanic children — Historically underdiagnosed compared to white children. Recent data shows narrowing of this gap, though disparities remain. Better identification in these populations contributes to overall rising rates.

Children with average or above-average IQ — In earlier eras, autism was strongly associated with intellectual disability. Children with autism but no intellectual disability — what was previously called "high-functioning autism" — were often missed.

Adults discovering their own autism — Many adults, particularly women, are now being diagnosed as adults after their child's diagnosis prompts them to recognize their own traits, or after they encounter information that helps them recognize themselves.

Diagnostic Substitution

In earlier decades, children who would be diagnosed as autistic today were often diagnosed with other conditions instead, intellectual disability, learning disabilities, language disorders, or general "behavioral problems." As autism diagnostic criteria have expanded and clinical understanding has improved, many of these children are now correctly identified as autistic.

This isn't autism replacing other conditions; it's autism being recognized where it was previously labeled something else.

Research has documented this substitution effect specifically.

Service Availability Creates Diagnostic Incentives

Practical reality: when autism diagnosis opens access to services (special education accommodations, insurance-covered therapies, ABA, speech-language therapy, occupational therapy), families and clinicians have more reason to pursue diagnosis when appropriate. The expansion of autism insurance mandates in most US states over the past two decades has made diagnosis substantially more meaningful for access to support, which has increased diagnosis rates among children who genuinely meet criteria.

Some Real Increase May Also Be Occurring

Researchers don't entirely attribute the rise to identification changes. Some specific factors may contribute to modest actual increases in autism prevalence:

Advanced parental age: Both maternal and paternal advanced age are associated with slightly increased autism risk, partly through de novo (new) mutations. As parents have children at older ages on average, this contributes modestly.
Premature birth survival: Children born very prematurely have higher autism risk, and improved neonatal care has increased survival rates for very premature infants.
Possible environmental factors under research: Research continues into whether specific environmental factors (certain prenatal medications, prenatal infections) may contribute, but the established environmental contributors are limited.

These factors likely account for a small portion of the rise. The much larger portion is identification rather than incidence.

What Doesn't Cause Autism

Because this topic frequently gets tangled with misinformation, several things deserve direct answers.

Vaccines Do Not Cause Autism

This is the most important misconception to address directly.

Vaccines do not cause autism. The scientific consensus on this is robust, supported by multiple large-scale studies involving millions of children, and endorsed by the CDC, the American Academy of Pediatrics, the World Health Organization, and essentially every major medical and scientific body in the world.

The misinformation traces to a single 1998 paper by Andrew Wakefield published in The Lancet, which proposed a link between the MMR vaccine and autism. That paper was found to be fraudulent, was retracted by The Lancet, and Wakefield lost his medical license.

Multiple large studies since have found no association:

Madsen et al. (2002), New England Journal of Medicine, 537,000+ Danish children, no association
Jain et al. (2015), JAMA, 95,000+ children, no association
Hviid et al. (2019), Annals of Internal Medicine, 657,000+ Danish children, no association

Collectively, these studies have examined over a million children with no evidence of a vaccine-autism link. The rise in autism prevalence is not driven by vaccines. For more on the specific way "regressive autism" became tangled with vaccine misinformation, see our piece on autism regression.

Vague "Environmental Toxins" Are Not the Established Cause

You may have encountered claims that the rise in autism is caused by various toxins, pollutants, foods, or environmental exposures. The honest framing: research continues into whether specific environmental factors contribute modestly to autism risk, but no specific environmental toxin or pollutant has been established as a cause of autism. The strong heritability of autism (64-91% per twin studies, Tick 2016 meta-analysis) means most autism risk is genetic, not environmental.

Parenting Style Doesn't Cause Autism

The "refrigerator mother" hypothesis, that emotionally cold parenting caused autism, was thoroughly debunked decades ago. Parenting doesn't cause autism. Both parents can be wonderful, attentive, loving caregivers and have an autistic child. The cause is overwhelmingly genetic and developmental.

Specific Diets and Supplements Don't Cause or Prevent Autism

Various claims circulate about diet (sugar, gluten, dairy) causing or preventing autism. No specific dietary intervention has been established to cause autism, prevent autism, or "cure" autism. For autistic children with specific food sensitivities or co-occurring conditions, individualized nutritional support may help with those specific issues, but not as autism causation or treatment.

What This Means for Families

If you're a parent reading this, particularly if you've recently received an autism diagnosis for your child, or are pursuing evaluation, a few things worth knowing:

Your child being diagnosed isn't part of an epidemic. It's part of better identification of autism that's likely been at similar prevalence in earlier generations, just not recognized as autism.

Better identification is genuinely good news. Children identified early have access to support, speech-language pathology, occupational therapy, early intervention, special education accommodations, that improves outcomes substantially. For Texas families specifically, Texas Early Childhood Intervention (ECI) provides services for children under 3 without requiring formal diagnosis first.

Your child's autism isn't caused by something you did. Not by vaccines, not by parenting, not by environmental exposures. Autism is overwhelmingly genetic and developmental. Both parents typically contribute genetic factors, though no single "autism gene" exists, autism is polygenic, with hundreds of genes contributing.

Future siblings have modestly elevated autism risk. About 20% sibling recurrence per current AAP guidance (Hyman et al. 2024), compared to about 3.2% in the general population. This is information for monitoring development, not for reproductive decisions.

Some adults discover their own autism through their child's diagnosis. This is common and meaningful, particularly for women whose autism wasn't recognized in childhood due to era-specific diagnostic limitations.

For more on the genetics specifically, see our pieces on which parent carries the autism gene and genetic testing for autism.

Conclusion

Autism is more visible today than it used to be, mostly because we've gotten better at seeing it. Expanded diagnostic criteria, improved screening, recognition of populations previously missed, and reduced stigma have all contributed to more accurate identification of autism that was likely present at similar rates in earlier generations. The "rising rates" largely reflect this improvement, not an environmental crisis.

Vaccines do not cause autism. Parenting style does not cause autism. The honest causes are genetic and developmental, with most of the apparent increase in prevalence reflecting better identification rather than actual rising incidence.

At Steady Strides ABA, we work with autistic children across Texas, and approach autism as a way of being to be supported, not a condition caused by something to be feared.

If you've recently received a diagnosis for your child and want to talk through what kind of support might fit, contact us for a conversation with a BCBA.

Frequently Asked Questions

What is the current autism prevalence rate?

In the United States, current data from the CDC's Autism and Developmental Disabilities Monitoring Network (April 2025 release covering the 2022 surveillance year) puts autism prevalence at 1 in 31 children, or about 3.2%. This is up from roughly 1 in 150 in 2000. Global prevalence per the Global Burden of Disease 2021 study (Lancet Psychiatry) is approximately 1 in 127 people, or about 61.8 million autistic individuals worldwide. The current US figure of 1 in 31 has replaced older figures like 1 in 36 that appeared in 2023 reporting.

Do vaccines cause autism?

No. The scientific consensus on this is clear and supported by multiple large-scale studies. The original 1998 paper proposing this link was fraudulent and was retracted; its author lost his medical license. Major studies since, Madsen 2002, Jain 2015, Hviid 2019, have collectively examined over a million children and found no association between vaccines and autism. Continuing childhood vaccinations on schedule protects children from genuine serious illnesses (measles, mumps, rubella, pertussis, others) without any associated autism risk. The rise in autism prevalence isn't being driven by vaccines.

Is the rise in autism diagnoses an "autism epidemic"?

The framing of "epidemic" is generally misleading. Most of the rise reflects better identification of autism that was always there, through expanded diagnostic criteria, improved screening, recognition of populations previously missed (particularly girls and women), and diagnostic substitution (children now correctly identified as autistic who would have been labeled differently in earlier eras). Some modest real increase may also be occurring related to advanced parental age and premature birth survival, but it's a much smaller portion than identification changes. "Epidemic" framing tends to drive misinformation about supposed environmental causes; the more accurate framing is "improved identification."

Why are more boys diagnosed than girls?

Current data shows roughly 3-4 boys diagnosed for every 1 girl, but a 2017 meta-analysis (Loomes et al., JAACAP) estimated the true ratio is closer to 3:1, meaning many autistic girls are being missed. Reasons for this gap include: diagnostic criteria originally built primarily on observations of boys; many autistic girls presenting with internalizing patterns (anxiety, withdrawal) rather than externalizing patterns more associated with the typical autism picture; autistic girls often masking more effectively than boys; and "special interests" in girls often looking more "typical" (specific people, fictional characters, social topics) than the stereotypical autism interests. Better recognition of female presentations is contributing to rising overall prevalence as more autistic girls receive diagnoses.

What role do genetics play?

A significant role. Twin studies consistently show autism heritability at 64-91% (Tick 2016 meta-analysis), meaning most autism risk is genetic. But autism isn't caused by a single "autism gene", it's polygenic, with hundreds of genes each contributing small amounts of risk in complex combinations. De novo (new) mutations in egg or sperm cells contribute to some cases. Specific genetic syndromes with stronger inheritance patterns (Fragile X, Rett syndrome, tuberous sclerosis, others) account for a portion of autism cases. For more, see our piece on which parent carries the autism gene.

Should families be worried about the rising rates?

Generally no. Higher prevalence numbers mean better identification, which means more children getting access to support, more accommodations in schools and workplaces, more research, and more recognition of autistic people as a substantial part of human variation. Autistic people have always existed; we're just identifying them more accurately now. The exception is concern for specific situations: families with one autistic child considering future children may want to know about the ~20% sibling recurrence risk (per AAP Pediatrics 2024); families noticing developmental signs in young children should pursue evaluation early because early intervention improves outcomes. But the general rise isn't cause for alarm, it's evidence of an autism-aware society correctly identifying neurological variation that's been present all along.