Published by
Stanford Medicine

Autism, Stanford News

Surprise: Environment's big role in autism risk

Surprise: Environment's big role in autism risk

New research from Stanford indicates that scientists have been underestimating the influence of a child’s environment on autism risk.

The new study, led by Joachim Hallmayer, MD, appears this week in the Archives of General Psychiatry. It used pairs of twins to tease apart the relative contributions of genes and non-genetic environmental factors to autism risk. Whereas older studies said autism risk was 90% genetic, the new work estimates autism risk is 38% genes and 62% environment. In this context, “environmental” risk factors are any non-genetic contributors to a child’s disease risk from conception onwards.

The new study is the largest of its kind, is better controlled than prior research and draws on a more diverse population than has ever been examined in the past.

From a press release I wrote about the research:

“Our research shows us that we need to be studying both genetic and environmental factors as well as how they interact with each other,” Hallmayer said. “We need to explore areas of environmental risk that are shared by both twin individuals and impact the development of the child.”

To get some perspective on the implications of this study, I turned to another Stanford scientist, Antonio Hardan, MD. Hardan was not directly involved in the research, but he did discuss the study’s results with Hallmayer during the research process. Hardan also treats patients with autism at Lucile Packard Children’s Hospital. Here’s our Q&A:

What was your reaction to these findings?

The surprise is that we’re seeing less impact of genetics in autism risk compared to what we’ve believed so far. That’s the main finding from the study.

You can always say, this needs to be replicated, but I think it might be the reality that we’ve all along overestimated the true heritability of autism.

Some people might say that this study provides fresh evidence for the reputed link between vaccines and autism, even though other research has not been able to support such a link. What would you say to that claim?

There is no evidence to date to show that vaccines are linked to autism – this is supported by 10 to 20 different studies that looked at this relationship quite extensively. The way I think about it is that when you’re debating with someone, you can provide the evidence, but that may not be enough to convince them. We have evidence beyond reasonable doubt to say that vaccines and mercury exposure are not linked to autism, but some people still might not believe those findings.

Was the prior research on environmental contributors to autism too heavily slanted toward vaccines?

If you look at how many papers were published on the vaccine issue, this is manpower and financial support that was not spent wisely. Instead of doing 20 papers on vaccines, it would have been nice if we had five studies on vaccines, five studies looking at the effect of multiple birth, some studies on maternal infection during pregnancy, and so on. That’s why we should move ahead cautiously – there is this potential for overreaction that will lead to efforts that are not very wisely focused.

I know the present study is agnostic with respect to which specific environmental risk factors are important, but, based on your knowledge of other research, what environmental risk factors do you think hold promise as plausible contributors to autism?

Other people have looked at, for instance, mothers’ between-pregnancy intervals – the shorter the interval, the higher the risk of autism in the second child. There was another paper by the MIND Institute that showed that taking folate early in pregnancy is linked to decreasing autism risk and, conversely, not having prenatal care or not taking folate early in pregnancy is linked to higher risk for autism. These are the kind of things we have to be thinking of.

Some of the other environmental factors that have been suggested include low birth weight, older parental age, multiple birth or obstetric complications. These factors have been linked to other disorders such as schizophrenia and bipolar disorder. If we’re going to think about future directions in research, we should investigate these further. There are also some environmental factors that are less supported by evidence, such as allergy to gluten – we don’t want to forget those, but we should keep it balanced in terms of how we investigate different risk factors.

5 Responses to “ Surprise: Environment's big role in autism risk ”

  1. Vernon E Weckwerth Says:

    Frighteningly flawed logic…and from Stanford?

    It appears that the amount attributed to the “environment” was the unexplained residual. Do the inverse: choose any one of the “suggested” environmental factors and attribute all of the then “unexplained variance” to genetics!

    This is analytic “trap one” in the first class of multi-variate analysis. Stanford?

    (Unclear from the article what were the environmental variables included, denying any conjecture on what was really analyzed.)

  2. Rachael Says:

    Dr Andrew Wakefield’s retracted Lancet study from over a decade ago ( the one that started all the controversy about the MMR vaccine) stated that:

    “A genetic predisposition to autistic spectrum disorders is suggested by over-representation in boys and a greater concordance rate in monozygotic than in dizygotic twins. In the context of susceptibility to infection, a genetic association with autism, linked to a null allele of the complement C4B gene located in the III region of the major-histocombatibility complex, has been recorded by Warren and colleagues. C4B-gene products are crucial for the activation of the complement pathway and protection against infection; individuals inheriting one or two C4B null alleles may not handle certain viruses appropriately, possibly including attenuated strains.”

    “We identified associated gastrointestinal disease and developmental regression in a group of previously normal children, which was generally associated with environmental triggers.”

  3. Ingrid Says:

    “There is no evidence to date to show that vaccines are linked to autism – this is supported by 10 to 20 different studies that looked at this relationship quite extensively.”

    And yet… when you say “extensively” I’m curious how it could really be that extensive when a) there’s been no control group in any of those studies (ie- children who have not been exposed to vaccines) and b) there’s been not one study (not even a safety study) comparing the effect of giving multiple vaccines in one sitting vs spread apart.

    To me the only studies that have been done about autism, that I could find, are ones that look ONLY at ONE vaccine as the possible “autism culprit”…. and yet… we give 36 vaccines to children by the time they go to school. To me this is the same as doing a study on whether drinking beer makes you able/not able to drive. If you study the effects of ONE beer and find that the person can still drive… is it really proper to conclude that they could drive if they had 5, 6, 7, …or 36 beers? And the fact that this “ONE beer” study has been done 10-20 times, and each time it showed no correlation between not being able to drive and drinking that one beer… doesn’t really mean much, does it?

    A better study would be to test what’s actually done in reality – kids getting 36 shots before kindergarten. And see if those children are more likely to have autism than children who don’t get ANY shots before kindergarten. That would be a study worth looking at in terms of figuring out if there’s a link or not between autism and vaccines. As it is, still, to date, there has not been a single study like this done. And yet for some reason the medical community *insists* enough studies have been done on this topic. It’s sad.

  4. Ingrid Says:

    I’m also surprised that the medical community would seriously find these results “surprising”… I mean, do we really think that evolutionarily our genes have spontaneous modified themselves so dramatically over just ONE generation to produce an epidemic of autism? I mean, seriously – when in the history of disease did we sit around and say BAH – this epidemic must have just spontaneously come from our genetics… I find it truly bizarre that the medical community could EVER have believed it was “just” our genes causing autism. Un-frickin-believable.

  5. Polly Says:

    Despite the relentless drumbeat of propaganda from the CDC, public health authorities and the thuggish on-line goons of the medical industry, thereÂ’s a funny thing going on. The evidence of a connection between mercury exposure and autism keeps growing.

    Last year two scientists at the University of Northern Iowa, Catherine DeSoto and Robert Hitlan, published a fascinating review paper. They asked a simple question: what does the published evidence linking autism and mercury really say? To answer that question, they did a simple Pub Med search. They searched for the terms “(Autism AND Mercury) OR (Autism AND Heavy Metals)”. They found 163 articles and reviewed them. According to the authors, “Of these 163 articles, 58 were research articles with empirical data relevant to the question of a link between autism and one or more toxic heavy metals. Fifteen were offered as evidence against a link between exposure to these metals and autism. In contrast, a sum of 43 papers were supporting a link between autism and exposure to those metals.” In short, 74% of the published studies supported the theory.

    Dr. Harden says, “We have evidence beyond reasonable doubt to say that vaccines and mercury exposure are not linked to autism”.

    See what I mean!


Please read our comments policy before posting

Stanford Medicine Resources: