Health and Human Services Secretary Robert F. Kennedy Jr. recently said the nation’s top health agency would undertake a “massive testing and research effort” to determine the cause of autism by September 2025.
The comments by Kennedy have stirred controversy among researchers and autism experts who have expressed skepticism due to his previous promotion of discredited theories linking vaccines and autism.
M. Daniele Fallin, PhD, the James W. Curran Dean of Public Health at the Rollins School of Public Health at Emory University, spoke with WebMD about Kennedy's comments and about autism research in general.
Editor’s Note: This interview has been edited for length and clarity.
WebMD: HHS Secretary Robert F. Kennedy Jr. has directed resources to search for “the environmental” causes of autism. What do researchers know about the root causes of autism – genetic and environmental?
Fallin: What we know from the past 30 to 50 years of research is that there are certainly genetic components to the causes of autism. We see autism running in families, so that’s what we think of as inherited genetics. But we also see that people with autism have an accumulation of genetic mutations that are occurring for the first time in themselves. That's something that we call de novo mutations.
And so, from both of those lines of evidence, it’s clear that genetics plays a major role. It is also true that there is some evidence for environmental factors acting either in concert or on top of that genetic risk.
For environmental factors, the first thing that’s essential to consider is the timing window. We know from a lot of neuroscience evidence that the physiology that we think of as autism is already set in motion during earliest neuro development – either at conception or very early on in life, which would mean during pregnancy or in utero.
The next important question is: What is the scope of the term “environment”? Generally speaking, a lot of folks have said anything nongenetic is considered environment – so anything that is being experienced by the mom or the fetus. Sometimes we think of that as toxins and toxicants, which would be chemical exposures or heavy metal exposures.
And in that regard, some of those environments are things like exposure to air pollution, exposure to pesticides. There’s a lot of interest and evidence that correlates pregnancy complications like preterm birth or having diabetes during pregnancy that also is associated with that child then developing autism. But it's still unclear whether that's due to direct effects of the toxin on the fetus or on the mother or to the pregnancy complication itself.
Sometimes it’s pharmaceutical or exposures like that. Valproic acid, which is an epilepsy drug that has been considered as a risk factor for autism when taken by the mother while she’s pregnant. At one point, antidepressants were considered, but that has been shown to not be a risk factor.
WebMD: Kennedy also recently suggested autism may be preventable. Is there any evidence to support this?
Fallin: I think it’s important to acknowledge that the concept of preventing autism is controversial. Many in the neurodiverse community have a perspective that the idea of identifying a cause for the purpose of preventing autistic people is not a high-minded cause and is unwise. There are folks in the autism community that just don't like the concept of prevention in this sense because they think that it's tied to something like eugenics. [Eugenics is the largely discredited scientific theory, often associated with Nazi Germany, that the human race can be improved through selective breeding.]
There are others who are excited about the concept of preventing the disabilities and challenges that autistic people face, not preventing the people themselves. And so, if we're using prevention in that sense, then it is possible. I believe that if you understood the causes, you could prevent the challenges that people face with autism. But it's important to recognize we're not talking about preventing people themselves.
WebMD: What are your thoughts on the approach to studying environmental factors in the new HHS initiative announced by Kenedy?
Fallin: There is a field of environmental epidemiology in autism, and there are many very thoughtful researchers working to identify environmental risk factors for autism, and a lot of that is motivated by the idea that these could be modifiable risk factors that would have important value in the lives of people with autism.
That being said, that set of expertise has really pushed what the possible study designs are that could get you there. And one of the reasons that we’ve had limitations in having discovery in this area, besides just research itself, is the kind of data that one needs to ask these questions robustly. And it sounds like, from what we’ve heard from the secretary so far, that the intention is to just bring together existing data records, like in electronic health records, to ask and answer these questions.
The problem is for these environmental factors, those often don’t exist in electronic health records. If I want to understand exposure to air pollution or I want to understand exposure to pesticides or other organic phosphates, for example, that's not likely to be in my medical record.
The closest may be that you could have where I lived in my medical record, and you might be able to do some fancy connecting of my address to the general exposure in my neighborhood. But it’s very different from having a direct measurement of my exposure when I was conceiving or pregnant with my children.
So that is one of the other dangers of this design is the idea that if we wanted to study environmental factors in the right timing window, which I do think itself is a laudable goal, it doesn't sound like we're designing the right kinds of studies to do that with rigor.
WebMD: Kenedy pledged to find the cause of autism by September. Is that a reasonable goal and a reasonable timeline?
Fallin: I think saying something like we’re going to identify a cause for autism in four months’ time is dangerous for a few reasons.
The first is it ignores past science that has shown the complexity of autism and that there’s likely not a single cause, but rather multiple things that we should consider as we think about strategies that would improve the lives of people with autism or prevent these challenges or disabilities associated with autism. So, that's the first thing.
The second thing is, we are grappling with a trust in science and rigor as a country right now, and the idea that we could do really rigorous research that would come up with a definitive answer in four months is just simply not feasible and does a disservice to trying to repair the trust in science in this country.
What I mean by that is, if you were to do a rigorous study of any kind in this field, you would first gather folks who have the expertise to make sure that there’s a sound design. You would declare what the scientific question is and how you're going to design a study to answer that specific question.
This is what we do in clinical trials all the time. It’s actually required that you say up-front what the question is so that you don't overinterpret data later to answer a different question that was not the one that the study was designed for.
So you first have to have the expertise and the folks to design the study. Then, you have to carry out the study. Then, you have to take the data that was gained from that study and have experts again adjudicate and evaluate and analyze those data to make sure that multiple people looking at the same results can come up with a consensus conclusion if you're going to then declare a cause.
And then finally, if you make it that far, you’ve got to figure out what is the translatability, what is the utility of that finding in terms of driving some kind of new intervention strategy – how do you go about communicating that responsibly, etc.
And so at least those four steps – and there are probably many others – just simply cannot happen with any rigor in four months.
To promise that you would get an answer in that short of a time means that necessarily you're going to have to cut yourself short on some aspect of what I just described. You're going to have to sacrifice rigor somewhere in that process, and that’s dangerous.
WebMD: Are vaccines considered an environmental factor?
Fallin: In large scope, vaccines would be considered under things in environmental epidemiology because that is a pharmaceutical. But the challenge with vaccines and autism first is that timing issue that I mentioned – that vaccines are generally not given until after birth, and autism has already started prior to birth, so that timing window doesn't work as a causal factor.
WebMD: How did vaccines come to be linked in the public's mind with autism, and what do we know about the link between vaccines and autism?
Fallin: Unfortunately, there was a paper published in the ’90s that reported an association between vaccination and autism in a small group of kids, and that got a lot of press because it immediately spoke to a lot of folks who are vaccine skeptics across many different areas but latched on to this issue of vaccines and autism because we're talking about young kids. We're talking about leaning into parental fear, and none of us want to do anything that could harm our own children.
So it picked up steam quickly – had a lot of celebrity amplification of that particular study. Unfortunately, that study then turned out to have major errors in its scientific rigor, but also in conflict of interest. The lead author had a financial conflict of interest in doing this work, and he ultimately had his license removed. In the meantime, the co-authors retracted from the study and then ultimately the whole paper was retracted, and the journal published a retraction.
So that study has been debunked, yet the damage that was done in terms of the popular belief that the two are linked continues, and again, that's despite the fact that was in the ’90s. Now we're talking almost 30 years of refutation of that original finding, where no one has seen the same association across multiple studies, multiple designs, multiple parts of the globe.
WebMD: Does the fact that this debunked finding still lingers in certain sectors frustrate your work in public health?
Fallin: Yes. It’s frustrating for two reasons: It’s frustrating as a public health professional because we know the power and the value of vaccines, and it’s a tragedy whenever you see a vaccine-preventable death or illness. So when I’m wearing my public health hat, that is really a frustration.
And if I instead wear my autism researcher hat, it’s also a frustration because there are lots of things that we should and could still be doing in autism research to understand the causes of autism or the causes of particular experiences amongst people with autism. And those could include environmental influences – and this vaccine story has diverted both research and attention from some of those other really important avenues of research that could lead to innovations and change for the autism community.
WebMD: As a public health professional in the public eye in the current environment, do you have any reservations about speaking up at this time, whether for fear of losing federal funding, or employment, or any other reason?
Fallin: There are many researchers who feel afraid to speak out for the reasons that you said. For me personally, whenever we’re speaking truth to the scientific evidence that exists – the need for rigor, the need for answers – I’m not worried about speaking out about those things because it’s important that we share factual information for the public.
WebMD: Any final thoughts?
Fallin: It’s important to remind folks that the issue is not whether we should or shouldn’t study autism. The issue is around what is the best, scientifically rigorous approach, and does it include the concerns and interests of the autism community.
