We Know What’s Causing a Rise in Autism Diagnoses — and it’s not Vaccines.
3 well-established trends likely account for it
At the risk of giving more attention to a claim that should really go away, I thought it might be helpful to offer an explanation that’s based on three pretty well-established trends. Though I don’t believe there’s anything to the claims of a link between vaccines and autism, the belief in that link seems to be impacting public health for the worse, so it’s worth addressing.
Perhaps just as pernicious as its impact on public health is its impact on critical thinking and support for sound, science-based medicine. That impact is harder to measure, but for anyone who spends time on social media, impossible to miss.
Let’s be clear: People claiming a link between autism and vaccines are not engaged in science. Though their language may mimic that of empirical research, they do not do real statistical analysis, ignoring confounding variables, confusing correlation with causation, repeating rejected claims, and dismissing data that don’t fit their narrative. This presents a misleading picture.
Stephanie Seneff is a good example. She travels the world calling herself an “MIT researcher” and pointing at glyphosate as a major cause of autism. But Seneff’s post at MIT is in computer science, not biology or epidemiology. She wants people to believe she has some kind of authority or expertise to speak about autism, but in fact when it comes to autism she is a layperson.
Yes, glyphosate rates and vaccination rates have increased over the past several decades, and so has autism. But then again, so have the number of homeruns per baseball game, the median household income, the sales of organic food, and the number of miles Americans commute by bicycles.
If pesticides were having this kind of impact, why just autism? Wouldn’t we see this trend show up with other conditions? But we don’t.
So if it’s not vaccines and it’s not pesticides, what is causing the rise in autism diagnoses?
First of all, much of the increase in diagnosis is not actually an increase in incidence. That is, there aren’t really that many more people who have autism, but we have gotten much better at identifying them.
Public education and more funding for mental health mean that kids (and, less frequently, adults) with autism are actually being diagnosed with autism. In the bad old days — right up to the turn of the century — many kids were incorrectly labeled, with anything from attention deficit or mental retardation to schizophrenia, oppositional defiant disorder, or just being a “bad kid.”
Before the 1990s, rates of autism were in the area of 1 in 1,000 — extremely rare. Compare that to well-established rates of schizophrenia, bipolar disorder, and ADHD, for example, and the current estimated rate of 1 in 68 starts looking pretty predictable. Current rates are similar to what we see with many mental illnesses.
CHANGE IN DEFINITION
Even more important than better identification, though, is the change in diagnostic criteria. Until the last decades of the 20th century, an autism diagnosis almost always meant a near total lack of communication, severe social deficits, and repetitive movements that would be obvious to any observer. Kids with autism were impossible to mistake for a “neurotypical” person, and many children with milder symptoms were simply not identified.
Though not made official until 2013, when the Fifth Edition of the Diagnostic and Statistical Manual of Mental Disorders was published by the American Psychiatric Association, a much broader view of autism began to take hold in the 1980s and is now fully integrated into the mental health and education systems. The autism spectrum takes in many times more people than the old autism diagnosis did.
Of course any time you expand the criteria for a condition, you are going to increase the number of people who have that condition. With autism, this effect was dramatic.
Meanwhile, there is one established cause for any actual increase in autism incidence, as opposed to just an increase in diagnosis. Several studies have shown a clear link between the age of a father and the odds of his children developing autism, as well as other conditions.
With one in six first time fathers over the age of 40, and a general increase in the age of parents, this is a very strong explanation for a significant portion of the rise in autism diagnosis.
Perhaps the best evidence against vaccination or pesticides as a major cause of autism is the fact that twin studies show a strong genetic component. Although genes don’t seem to hold promise as an explanation for any increase in diagnoses, the established genetic influence should give us pause as we rush to blame environmental factors.
But why, then, some will ask, do so many kids develop autism at the time of vaccination? And why do some studies show vaccinated kids more likely to be diagnosed with autism?
Autism shows up during early childhood. Part of the course of autism is often that children begin speaking and playing normally, and then — sometimes quite suddenly — seem to regress, losing the ability to speak and communicate. This tends to happen during the same stage of development that vaccines are administered. A scary causal connection?
I don’t think so. Every day, thousands of kids get vaccines. And every day, thousands of kids start to show signs of autism. So just by sheer coincidence, many children must show the first signs of autism on the same day, or on the day after, they are vaccinated. There is no way around that eventuality; it has to occur.
Many kids also fall off their bikes, hit their little sister with a stick, refuse to eat their vegetables, and have an allergic reaction to strawberries within a day or two of vaccination. By chance, this has to occur.
But the vast majority of kids with autism do not show the first signs immediately after vaccination — and the vast majority of vaccinated kids never develop autism.
Some activists have also claimed that children who are not vaccinated are less likely to have autism than children who are. But the studies they cite — like this one — are virtually useless, for two reasons:
First, they rely on self-report, a grossly unreliable method for measuring the prevalence of almost anything. But second, even if we assume parents are reporting both honestly and accurately about their kids’ health, we would expect vaccinated kids to have more diagnosed conditions, not because they are truly sicker, but because they are more likely than unvaccinated kids to see a doctor.
Families who do not vaccinate often mistrust the medical establishment in general and avoid visits to medical professionals unless absolutely necessary. Kids who aren’t vaccinated may go many years without appearing in front of a physician. Vaccinated kids are seeing a doctor regularly, usually at least annually. That makes any medical or psychiatric condition they have much more likely to be identified.
Epidemiology is complicated, and so is mental health. A perfect understanding of autism and its causes is not easy to establish; disagreement is normal and healthy; our knowledge evolves; dissent and debate is a valuable part of that process.
But there’s a limit to this. Solid empirical research, controlling for variables, and listening to feedback from peers in the field are crucial, and the anti-vax community not only fails to engage in these activities, it seems to openly scoff at them, demonstrating the very closed-mindedness of which it accuses the medical establishment.
We need debate and dissent, but many of the claims of anti-vax advocates are so unfounded and so easily dismissed, we simply get noise.
As always, the answer is to dig deeper.