Advertisement

Fact check: Autism diagnosis criteria changes have led to increased rates

Bayliss Wagner, USA TODAY
·9 min read

The claim: Autism has increased by 30,000% in 50 years, is linked to vaccines

As states advance in their distribution of COVID-19 vaccines, Americans have continued to discuss what risks could be associated with gaining protection against the virus. One common belief espoused by parents and public figures who are against vaccination is that vaccines cause autism spectrum disorder, despite several studies showing no correlation between childhood vaccination and autism diagnoses.

A post popular on Facebook may represent a similar view. The post claims that autism "increased 30,000% in 50 years" and suggests there is a link between injections and this statistic. Originally posted on April 3, 2020, the post has accrued roughly 1,500 shares.

In a comment on the post, the author linked vaccines to the "injections" he mentions, suggesting he made the post, "Because they are trying to make us all take a shot of their NEW 'Vaccine'!" The post's comment functionality has since been turned off.

ADVERTISEMENT

Fact check: COVID-19 vaccines won't cause a zombie apocalypse

USA TODAY reached out to the poster for comment.

How autism spectrum disorder prevalence is estimated

As understanding of ASD has changed, so have the methods for diagnosing the disorder and estimating its prevalence in the general population.

It is critical to understand that any numbers measuring autism prevalence in the U.S. are estimates; scientists use a statistical method to extrapolate a subset of data to the country as a whole. Research methods for estimating autism prevalence across the country are much more accurate than they were in the '60s, as well as more broad, so they gloss over fewer cases in the general population.

In 2000, the CDC established the Autism and Developmental Disabilities Monitoring Network to estimate prevalence of autism. It detailed its method in a 2016 study on autism prevalence. Every two years, these researchers scan a representative sample of the school records of 8-year-old children for signs of autism, then extrapolate that number to estimate the number in the general population.

Drastic variations in the numbers of autistic children among different states — the same CDC study estimated a prevalence of 13.1 per 1,000 children in Colorado, and 31.4 per 1,000 in New Jersey — signal that even this method is affected by different levels of autism awareness and resources in different states. A study from March 2005, conducted by David S. Mandell at the Leonard Davis Institute of Health Economics, established a positive correlation between state educational spending and diagnoses of ASD. Additionally, the study found that diagnoses of ASD increase with the availability of health-related resources, such as pediatricians and school health clinics.

The influence of awareness and resources for ASD on prevalence estimates is clear: without schools screening children and keeping detailed records, and without parents inquiring about their children’s ASD status to doctors, less children are diagnosed. This applies to distinct periods of time as well as to distinct geographical areas.

How autism rates have increased

If the Facebook post meant "increase in autism rates," a "30,000% increase in autism" would require the rate of prevalence to be 300 times what it was 50 years ago. Autism rates have increased immensely in the last half-century, though not at the rate the post states.

In the 1960s, researchers estimated that 0.04% of European children, or roughly 1 in 2,500 children, had autism, according to HuffPost. Currently, 1.85% of all American children, or 1 in 54, are estimated to have autism, according to the CDC's Autism and Developmental Disabilities Monitoring Network. A comparison of the 1966 estimate and the 2020 estimate shows the prevalence estimate increased 45-fold (or 4,525%) in that time period.

The widely publicized increase in diagnoses has become a cause for concern among parents and public health experts. However, it does not mean that a child today is 45 times more likely to truly have ASD and its symptoms than a child 50 years ago was. Many studies confirm the increase is due not just to true biology, but rather broader diagnostic criteria and ASD awareness and resources.

Broadening of diagnostic criteria

There is no objective measure for diagnosing ASD, unlike body weight and blood pressure. When applied in the 1966 study, Scientific American explains, the definition of autism had not yet made it into the widely used DSM. The definition relied on Leo Kanner's 1943 observations of "infantile autism," a term he coined to identify children who had previously been described as "as feeble-minded, retarded, moronic, idiotic or schizoid," Spectrum News writes. Therefore, most diagnosed cases were severe.

Today's diagnostic criteria comes from the the DSM-5, which clinicians have continued to expand since 1980, and defines ASD as a developmental disorder that causes "Persistent deficits in social communication and social interaction across multiple contexts." People with ASD also exhibit "restricted, repetitive patterns of behavior or interests," which can include reactions of extreme sensitivity or indifference to sensory inputs such as pain and loud noises, intense interests, and inflexibility in routine.

This criteria is much broader than Kanner's definition because it includes milder forms of autism. As detailed in a 2012 study led by Judith S. Miller, a senior scientist and training director at the Children's Hospital of Philadelphia Autism Research Center, the definition broadened even further after the 1980s. Diagnoses of ASD now include disorders like Asperger's Syndrome and pervasive developmental disorder that are distinct from autistic disorder, hence the change in terms from "autistic disorder" to "autism spectrum disorders."

This change is significant enough that a literature review in the journal Research in Autism Spectrum Disorders in 2011 concluded, "Unless researchers are able to control for previously used diagnostic criteria, comparisons to previous prevalence estimates hold little information.” Miller's study did just that and concluded that 59% of children diagnosed "not autistic" in a landmark study on autism from the 1980s would meet today's criteria for ASD.

Other factors in the increase in autism prevalence

There are other reasons why changes in practices around autism make it hard to compare autism diagnosis rates over time.

Writing for Scientific American, Jessica Wright, PhD, a genetics specialist at the research partnership Spark for Autism, stated clinicians tend to diagnose children on the borderline of the clinical criteria with autism because it now qualifies those children for special resources in school, where it did not in the '60s.

In addition, a CDC study found that many African American and Hispanic children have historically had less early access to diagnostic services that would recognize and record the signs of autism, but more screening in schools has increased prevalence rates among those populations.

In the past two decades, the proportion of the child population the CDC has identified with ASD has increased 176%, from 6.7 cases per 1,000 in 2000 to 18.5 per 1,000 in 2016.

Finally, certain biological factors increase the chance that a child will develop autism. These include a child having older parents, in particular an older father, and a child being born prematurely. As modern health and medicine have enabled more premature babies to survive and parents of a greater age to have children, Scientific American estimates that these factors account for part of the rise in autism rates over time.

Vaccines and autism

The persistent fear of a link between autism and vaccination began with a 1998 study claiming the mumps and rubella vaccines caused autism. The article was later retracted and author Dr. Andrew Wakefield was stripped of his license by the UK's medical licensing board.

Fact check: Fauci says it's too early to know results of eased COVID-19 restrictions in Texas

According to the CDC, studies by the CDC and the Institute of Medicine have found that children who receive vaccines do not have a higher chance of developing ASD. The CDC also states that studies of particular ingredients in vaccines have shown there is no correlation between ASD and vaccine ingredients.

Our rating: False

We rate this claim FALSE, based on our research. Even the most generous comparison of autism prevalence rates between 1960 and now shows that rates have increased far less than the author of the post states. The claim misses important context about such comparisons, which show that significant overhauls of data collection methodology and definitions of autism make such comparisons inaccurate. In addition, the post's suggestion of a connection between vaccines and autism has been repeatedly disproved.

Our fact-check sources :

Thank you for supporting our journalism. You can subscribe to our print edition, ad-free app or electronic newspaper replica here.

Our fact check work is supported in part by a grant from Facebook.

This article originally appeared on USA TODAY: Fact check: Autism diagnosis changes over years account for high rate