Revisiting the Autism “Epidemic”

Anyone who cares about autism, and particularly the supposed spike in autism in recent years, would do well to read this very informative, cogent, and non-hysterical OpEd by Paul T. Shattuck and Maureen Durkin. It is written on the occasion of a case before the U.S. Court of Federal Claims that’s investigating whether autism is linked to childhood vaccinations, as many parents of autistic children believe.

Shattuck and Durkin’s argument, in a nutshell, is that:

1. Vaccines do not cause autism, according to the scientific consensus;

2. There surely has been an eruption of documented cases of “autism spectrum disorder,” but this does not mean that autism itself is actually on the rise, as many people believe.

3. There are a variety of reasons for the increase in documented cases of autism and related disorders, including:

a. better reporting/diagnosing;

b. more funds available for treatment, which incentivizes parents to have their children diagnosed;

c. a much broader definition in recent years of what constitutes “autism,” including the reclassification of other disorders into the autism umbrella label.

If you have an autistic child or know someone who does, you may not be satisfied with Shattuck and Durkin’s article, for it may be more comforting to have a tangible villain, like vaccines, to blame for this disorder. And a tangible villain would certainly make it easier to prevent autism in the future. But you should still read the article with an open mind.

There is one point that Shattuck and Durkin didn’t raise, which I sometimes wonder about when people discuss a link between childhood vaccines and autism. Autism usually begins to present itself at about three years of age, by which time kids have started to get a lot of vaccines. This proximity may naturally cause many parents to link the two events in their mind. But just because one event happens shortly before another does not mean that the first even caused the second — as comforting, in a warped way, as that may seem.

Here’s what we’ve written in the past about the subject.


The heritability of autism is >.8, which says it all, really. Source, e.g.:

Bailey, A., Le Couteur, A., Gottesman, I., Bolton, P., Simonoff, E., Yuzda, E., and Rutter, M. (1995). Autism as a strongly genetic disorder: Evidence from a British twin study. Psychological Medicine, 25, 63-77.

The vaccine scaremongers will argue that since heritability is expressed in terms of the proportion of the phenotypic (ie behavioural/expressed) variance that is attributable to genetic variance, things that don't vary won't be taken into account, such as vaccines that are given to every child.

This reasoning is specious. By this logic, any childhood universal that has emerged post-autism is equally likely to cause the increase in incidence - plastic diapers, baby gyms, TVs, you name it. And you would still have to explain why the condition runs in families. What is the vulnerability that causes the condition to be expressed so selectively?

Given what is known at present, it is more parsimonious to argue that autism is largely caused by inherited deleterious gene mutations, rather than it being caused by vaccine X that produces affects only certain inter-related people because of vulnerability Y.

Both accounts can explain the increase in incidence, so that's neither here nor there.



The heritability of autism is >.8, which says it all, really.

Well there you go. God works in mysterious ways. Only 2 hours and 31 minutes after I spoke, we have our answer.

God is banished from this subject. Science moves on to the next mystery. Life gets better.


For me, investigating the autism link was the tip of the iceberg in questioning the safety of the childhood vaccination program. I'd love to see the Freakonomics authors address other concerns, such as whether it makes sense to vaccinate five times by age five for diptheria when there are barely a handful of cases each year--but the reporting of "adverse events" from the vaccine number in the thousands. Or, what is the wisdom of vaccinating day-old infants for Hepatitis B, which is spread through intravenous drug use or sexual contact--when, again, their likelihood of contracting the disease (unless the mother was a carrier) is miniscule compared to the rate of "adverse events".

mandala oblongata

As a parent of a child with an autism spectrum disorder, I was annoyed by both the NYT OpEd and Dubner's blog post about it.

First, much time has been spent discussing whether there is an "epidemic" of autism spectrum disorders, with the focus being whether there is an increase in the number of kids getting ASDs or whether there is an increase in statistical reporting. This is often the lead point of stories. Who cares? Isn't the main issue that we now realize that there are many more kids on the autism spectrum than we did 15 years ago?

Second, neither the article nor Dubner try to tackle the issue of what the prevalence of kids with ASDs really is. Prevalence rates vary from state to state but are much higher in states such as New Jersey that have more sophisticated methods of identifying and tracking kids, more educators and therapists who are adept at identifying kids, and more awareness among parents (and other family and friends) who can point out the red flags. However, even in states like NJ where reporting rates are approaching 1 in 100, not all kids are diagnosed. Therefore, my guess is that the CDC estimate of 1 in 150 is significantly below the actual prevalence.

Third, the OpEd and Dubner spend nearly no time analyzing what the right amount of funding for treatment and research for kids on the spectrum would be. Since this is an area of diagnosis, treatment and study that is relatively nascent, my guess is that funding lags in all important areas. For example, even a minimal amount of research would show that the number of schools and placements for kids on the spectrum lags prevalence even in the more "innovative" geographies.

Fourth, people in the media seem to write a lot of articles about speculation about what causes autism spectrum disorders. Since the issue has only really been studied for a decade or so, there hasn't been enough research over a significant enough period of time to determine correlation let alone causation definitively. The early evidence suggests a strong family link that is hereditary, and genetics may end up being proven to be the only cause. But the short answer right now seems to be that there isn't enough evidence about cause and we need much more research.

Fifth, Dubner's wrote, "it may be more comforting to have a tangible villain, like vaccines, to blame for this disorder." Perhaps there are people out there who feel this way. But I've never met a parent of a child on the spectrum who spends much time on finding someone or something to blame. The vast majority of us have little time or energy for this, as it won't help our kids. Instead, to help our children as best we can, many of us completely upend our lives: some quit or change jobs to be more involved, some move closer to decent services, and most invest countless hours educating ourselves (and our families, friends, teachers and therapists) on how best interact with our kids to speed their development whenever possible.

Sixth, Dubner wrote, "you should still read the article with an open mind." Yes, and I should use my fingers rather than my elbows to type on a keyboard. Let me counter some pedantic commentary with some of my own. If you are a journalist and are going to write multiple posts on the topic on an internationally-read blog, you should interview some parents, psychologists or therapists to become better educated on autism spectrum disorders.



The issue with the general public hearing a sound bite of an entire body of scientific research is that it creates baseless fear. This fear leads to parents choosing to not vaccinate their children. Whether you want to believe it or not, vaccines have improved quality of life for the general public. Was it as recently as the 1940s and 1950s that children were crippled by polio? My own aunt was born deaf because while her mother was pregnant, she developed German measles. So, clearly the benefits of vaccinating your children far outweighs the risk of autism.

Additionally, thimerasol hasn't been in vaccines since the early 1990s and autism prevalence has continued to increase. Therefore, there is clearly not a link between the two.


browndog, you are so wrong. Thimerosal was not removed from vaccines in the early 90's - that's when it was at its hight.
Thimerosal content was added up in 1999, realized to be way over EPA limits, and instead of doing a recall, the manufacturers were recommended to phase it out. It's still in flu shots, which are recommended annually, and trace amounts are in other childhood vaccines.


Talk of the Nation on NPR did a whole segment today and one interesting point that came out was that its possible that autistic children have some kind of immune abnormality that causes them to be highly susceptible to environmental toxins. In other words, while in some cases a vaccination may be a direct trigger for autistic behavior to start occurring, its not exactly because of the vaccine. Its because the genetic problems of the child were activated by that particular stimulant. Perhaps in the future we will be able to have our cake and eat it too, as children who test with this genetic problem can avoid vaccines and other possible toxins, and children who don't will be vaccinated and still ensure a critical mass of non-infectable kids.

Jon Best

As a person with Autism I deem that any news on this is better than what the doctors told my parents in 1986. I am 25 years old and still living with the consequences of there discision to put me under their rulling of Autism. I do want to say that I am at least not Autistic as some other children because I have a high learning and greatly, miraclous, normal life. While the doctors told my parents that I will never read, write, do math, function in a socitiy as a normal person and which also means not able to drive, walk, talk, and communicate as a person to another person. Yes I do have some symptoms of it but bless God that he gave me a way out of it through video games, is this a shock? I have heard that a year ago on a tv program that made a reliasm of having about a room full of autistic children and put half of them with video games and half of them not and half of the children who did had better hand and eye reflexas than the ones that did not, reading, and God (which is my explanation of everything).


anne borden

Some pregnant women are refusing the H1N1 vaccine, thus risking their unborn children's lives -- and their own.

We KNOW that swine flu can kill a pregnant woman and her unborn child. The autism link, by contrast, is disproven or unsubstantiated.

What does it say about our culture that parents are taking significant known risks (H1N1, pneumonia) out of fear based on a dubious claim (autism link)?

Gary Pears

As a parent of a 19 year old with what we think of as Autism, I would like this topic to be processed by the Freakonomic engine. What can we deduce by looking at the drugs that have been accepted as being harmful and their related spikes, such as thalidomide and then looking at MMR in the same way. I think that MMR 1 may spike differently to MMR2.


I just want to know what "incentives" the government had for making it impossible for parents to sue drug companies for supposedly vaccine-related injuries inflicted on their children. That's all. Hope you guys can help me make a decision soon.