The paper’s adult autism numbers are wildly overblown, and worse than useless.
By Jill Escher
Autism research has long been something of a spectacle. For every study that actually moves the field forward, it seems a dozen are fairly worthless, same-old same-old, or totally bizarre.
But I never expected that work from our own U.S. Centers for Disease Control (CDC) would fall into that last category.
This week the Journal of Autism and Developmental Disorders published a paper, “National and State Estimates of Adults with Autism Spectrum Disorder,” by the CDC’s Patricia Dietz, Charles Rose, Dedria McArthur, and Matthew Maenner. The news sites blared across the internet: “First United States Study Of Autism In Adults Estimates 2.2% Have Autism Spectrum Disorder,” or “CDC Researchers: Over 5 Million US Adults Have Autism.”
But the study estimating a supposed 2.2% prevalence in adults, was, how can I put it mildly, utterly preposterous.
It was, in essence, nothing more than a glorified back-of-the-envelope type of exercise that did not actually look at any data pertaining to autistic adults. Rather, it used data for children with ASD and, adjusting for mortality rates, projected what adult rates might be based on current childhood rates.
The obvious flaw? The method was based on the fundamental assumption that the birth year autism prevalence is constant over all ages considered, going back to 84 years before 2017. Yep, they assumed autism rates were the same in 1933 as 2017, and all years in between.
Never mind that there’s not a shred of evidence for this — not in medical, academic, institutional, military, court, social security or any other records, whether pegged under autism or any other label. And never mind that the CDC itself has published numerous studies indicating increasing autism prevalence over successive birth years.
But apparently at the CDC it’s cool to ignore actual facts. So, simply assuming current rates apply to all ages, even 87 year-olds, the paper estimates that in 2017 the U.S. adult prevalence of autism was 2.21%, or 5,437,988 U.S. adults (!) aged 18–84 years. Five and a half million autistic adults? What? How? Where? Anyone with eyes can see that’s just not true.
It also estimates the states with the greatest number of adults with ASD included California, at 701,669 cases, Texas with 449,631, New York, with 342,280, and Florida, with 329,131.
Validity check, anyone? These numbers are absurd on their face. For example, in California we obviously have nowhere near 701,669 cases of adult ASD.
California is known for maintaining the country’s best records on autism in the population. In our Department of Developmental Services (DDS) there are currently about 39,000 cases of autism for ages 18 and up. Outside of DDS, which would include milder cases ineligible for developmental services, the population has been estimated to add about 30% more than that number, so perhaps we have about 50,000-60,000 total adult ASD cases. It boggles the mind that any researcher who knew anything about autism in the real world would estimate that California has 701,669 diagnosable adult ASD cases.
Moreover, California has kept close tabs on autism for many decades. And what does that data say about autism by birth year? That the rates have skyrocketed beginning with births in the early 1980s, something even the California Department of Public Health has verified in prevalence studies. The graph to the left shows the DDS autism cases by birth year: about 200 per year through the early 1980s to more than 7,200 by 2014. In other words, real-world data show unequivocally a massive upsurge in cases over time.
In fact, DDS data show there are only 4,000 developmental disability-qualifying autistic adults born before 1980 in the entire state. If rates remained constant over time, as the CDC suggests is the case, the state would have about 250,000 such adults instead of the actual 4,000. But there’s not a shred of evidence our robust system has missed even a tiny fraction of the supposed 246,000 missed cases.
So, how did this new CDC paper get away with such absurdities?
The authors say they assume ASD prevalence among children and adults (of all ages) is similar because there is no evidence that environmental factors associated with ASD have changed over time. They say “few risk factors have consistently been associated with ASD and those that have been identified have accounted for a very small percent of increases in diagnosed ASD.”
Actually, they are correct, to a point. With rare exceptions (like fetal exposure to anti-seizure medication and prematurity), exposures investigated to date (and this includes vaccines, which obviously do not cause the dysregulated brain development of autism) cannot explain the colossal increases in autism witnessed across our country.
But — and I can’t believe this needs stating — it makes no sense to dismiss an increase by birth year simply because you don’t know what caused it. That’s like saying that people were not getting sick and dying of Covid-19 if we didn’t know what caused it. Or children were always born with high rates of microcephaly because we didn’t know about Zika.
Epidemiology works the other way around, of course. First you identify changes in disease incidence, which for disorders like autism means birth year prevalence, and you use that information as a clue as to causes. The CDC’s role is to identify risk factors driving pathologies afflicting Americans. But in a move that should outrage anyone who cares about autism, it is now waving the white flag of scientific surrender, saying basically, “We don’t know what is causing this autism increase and we assume there’s nothing more to discover.” An abdication of its core responsibility to the U.S. people.
In the good news department, it seems this paper had some noble intentions. “National and state-based estimates of adults living with ASD could inform planning for programs,” say the authors. And yes, we desperately need policymakers to understand the prevalence and number of cases so we can support the legions of disabled adults. But this study offers nothing of value, and, worse, signals an alarming level of apathy about causation of very serious neurodevelopmental pathology that has flooded our communities.
Jill Escher is founder of the Escher Fund for Autism, which promotes research in the genetic toxicology of autism, GermlineExposures.org. She is also President of the National Council on Severe Autism and Immediate Past President of Autism Society San Francisco Bay Area.
Disclaimer: All blogposts on NCSA represent the views of the authors and are not necessarily the views of NCSA.