Jul 23 2008

Autism and Vaccines: Responding to Poling and Kirby

In response to my blog post on Monday, David Kirby wrote a response in the Huffington Post and Dr. Jon Poling (father of Hannah Poling) wrote an open letter to me, placed in the comment section and posted at Age of Autism. It seems only polite that I respond to their kind attention.

The primary focus of my original post (which I further developed yesterday) was that the media is focusing too much attention on what celebrities and politicians are saying about the controversy surrounding the discredited notion that vaccines are a significant cause of autism. Over the past year Jenny McCarthy (now joined by her boyfriend Jim Carrey) has become the major spokesperson for a movement that, at its core, is anti-vaccine and is dedicated to the scientific opinion that vaccines are toxic and cause autism. Recently actress Amanda Peet joined the fray, professing her belief that vaccines are safe, are not associated with autism, and that parents who do not vaccinate their children are “parasites” for depending on other parents who do. (She later apologized for that remark, calling it “divisive”.)

While I appreciate Amanda Peet’s support, I feel strongly that scientific questions should be handled by the scientific community. Celebrities are great when they support causes – but when they second guess the scientific community and decide to advocate for their own scientific conclusions, they are more likely to cause harm than good.In that post I also took David Kirby to task for criticizing Amanda Peet, saying that her position is against the “medical establishment.” Kirby used a very dubious chain of logic to come to that conclusion – relying heavily on the opinions of politicians and ambiguous cases and failing utterly to make the case that the medical establishment, or any consensus of scientific medical opinion, finds any credibility in the supposed link between autism and vaccines.

My Involvement with this Issue

Before I respond directly to Kirby and Poling I want to review my personal involvement with this controversy. Not because it is relevant (only the scientific evidence is relevant) but because my critics often try to make it relevant, usually by making false assumptions about me. So I want to just set the record straight, satisfy any curiosity, and also make this my formal disclosure of any potential conflicts.

I am an academic neurologist in the clinical educator tract at the Yale University School of Medicine. I am an adult neurologist (not a neurosurgeon) with specialties in general neurology and neuromuscular disease. My duties are primarily teaching and clinical with some clinical research thrown in.

I do not participate in autism treatment, diagnosis, or research. I am not affiliated in any way with the Yale Child Study Center or autism research at Yale. I do not speak for Yale or autism research at Yale in any regards. I do not prescribe or research vaccines. I do not work for or have any financial interest in any pharmaceutical company or any corporation that has a vested interest in the vaccine program. I have never worked for the government or any regulatory agency. I am not involved in any legal cases involving autism or vaccines.

My first involvement with this controversy was several years ago when I was invited to write a review article for the New Haven Advocate. This was an outgrowth of what I consider to be my primary academic activity – educating the public about science and science-based medicine. This is something I have been doing for over a decade, starting with the New England Skeptical Society, a non-profit educational organization I started with some colleagues to promote science and reason. My activities have evolved over the years and now include a popular science podcast (the Skeptics Guide to the Universe) and this blog.

After several months of reading and literature review, I decided that at that time (2005) the evidence did not support a link between vaccines in general, the MMR vaccine specifically, or thimerosal in vaccines with autism or any neurodegenerative disorder. I did not begin my investigation with that conclusion. My goal was to accurately reflect the current state of the scientific evidence. As an outspoken scientific skeptic, my reputation is not based upon any particular scientific conclusion. Rather it is based upon dedication to the process of science – using valid logic and fairly accounting for all available evidence.

So for me the only thing that mattered was getting the science right. If I honestly thought the evidence supported a link between vaccines and autism I would have promoted that conclusion. This would have actually worked out much better for me and my career – becoming an early advocate for a minority and highly controversial position that I thought was correct and would eventually be vindicated would have worked very well for me. As it turns out I supported the boring consensus position and was largely ignored outside of skeptical circles. Oh well.

David Kirby

In 2005, as part of my research, I read David Kirby’s book, Evidence of Harm, in which he promoted the conclusion the thimerosal probably does cause autism and that the government is essentially engaged in a cover up of this fact. This conclusion was premised in part on the belief that we are in the midst of a true autism epidemic – that true autism rates are increasing rapidly (the evidence, however, supports the conclusion that the apparent increase is due largely to an expanded diagnosis and increased surveillance). At the time I also interviewed David Kirby for several hours by phone. We both agreed that our respective positions were destined for a definitive test – for thimerosal had been removed by the beginning of 2002 from all vaccines in the routine child vaccine schedule. If thimerosal were a significant cause of autism then autism rates should plummet by 2007. If not then they would continue to rise until they plateau at the true rate. Well, it’s 2008 and autism diagnosis rates have continued to rise without a blip. This fact alone rules out thimerosal as a significant contributor to autism rates (as always – really just as a point of logic – an effect too small to be detected by the available data cannot be ruled out).

David Kirby, however, refused to admit the undeniable implications of this data. He has since moved on to other hypotheses linking vaccines and autism- just as speculative and lacking in data.

In my recent blog post I criticized Kirby for his recent Huffington Post piece – specifically that he is desperately trying to portray the public controversy over an alleged association between vaccines and autism as a legitimate scientific controversy. Yet to make his case he cites politicians and regulators – not scientists. Further he cites the fact that the IACC (the Interagency Autism Coordinating Committee) expressed interest in exploring the vaccine-autism issue. But it turns out this interest came from Lynn Redwood – a member of Safe Minds, a group promoting the idea of a link between vaccines and autism. She was included to give voice to this minority opinion. Now Kirby is using that politically-motivated inclusion to argue that scientists are taking the controversy seriously.

Kirby now takes this self-referential strategy one step further. First he writes:

I write my Huffington Post blogs in order to spark debate and commentary from other quarters about what has become — like it or not — the biggest medical controversy of our time: the potential link between vaccine ingredients and autism.

In that sense, my last piece, “Amanda Peet vs. the Medical Establishment,” has done its job.

Reaction was predictably swift and furious to this opinion essay — and that’s what blogs are: this is not news reporting.

Let me dissect this a bit. He is saying that the purpose of his blog entry was not to report or inform, but to spark debate and commentary. Does that mean we are not meant to trust the information in his blogs? Are they intended only to provoke and not inform? While I agree that blogs are largely opinion pieces, the distinction from reporting is not clean. Often blogs will report news or summarize evidence. Opinions are often based upon a factual premise – you cannot then hide from misinformation and poor facts by claiming it’s all just opinion.

Kirby is also claiming that the autism-vaccine question is the “biggest medical controversy of our time.” I disagree, but I am also not sure how to objectively measure such a thing. As a side note – notice how he wrote: “potential link between vaccine ingredients and autism” – vaccine ingredients. Without ever admitting defeat on the thimerosal issue, he has quietly shifted his claim over to “vaccine ingredients”. What ingredients are those, I wonder?

What Kirby is primarily doing here is exactly the same thing that the Discovery Institute (an anti-evolution group) tries to do with intelligent design (ID) and evolution. They promote the idea of a scientific controversy of evolution and ID where none exists. They then use propaganda and political mechanisms to create a public and political controversy. They then present the public controversy they manufactured as if it were a scientific controversy, and then scream for academic freedom to “teach the controversy.” Meanwhile they desperately claim that there must be something wrong with evolution or else why would there be such a controversy.

Kirby and the anti-vaccination crowd have created a false controversy over vaccines and autism. They then promote this controversy as if it were a legitimate scientific controversy. They then demand that their claims be investigated, that they are represented on the IACC, and they sue the government over alleged vaccine injury – and claim that the resultant controversy they manufactured is evidence for a legitimate scientific controversy and that they should therefore be taken seriously. There must be something to this controversy we manufactured because there’s a controversy – it’s nothing more than an elaborate and deceptive self-fulfilling prophesy.

Kirby is now using a strategy also familiar to the ID crowd – say something scientifically outrageous, and then use the backlash of scientific outrage to say – well at least I got them talking about it and taking the controversy seriously. Mission accomplished. It’s just more self-fulfillment.

What is lacking in the case of ID and the anti-vaccination movement is an actual scientific controversy, or reliable scientific evidence to challenge the current consensus of opinion.

For further self-reference, Kirby writes:

Now, Hannah’s father, Dr. Jon Poling, a respected neurologist and clinical assistant professor of the Department of Neurology, Medical College of Georgia, has responded to Dr. Novella, including his, “criticizing the journalism of Mr. David Kirby.”

A copy of Dr. Poling’s letter to Dr. Novella is posted at what ABC News.com called “the popular blog, Age of Autism.”

It makes for some good reading, from deep inside the medical establishment!

It is misleading for Kirby to refer to Dr. Jon Poling as being “deep inside the medical establishment.” Sure, he is a neurologist, but he is hardly a disinterested expert. He is, as Kirby points out, Hannah’s father. This means that he was involved in suing the vaccine compensation program on the claim that vaccines triggered his daughter’s neurological condition. In this case he is acting more as a member of the anti-vaccine movement than the medical establishment.

For the record, I don’t think it’s accurate to characterize me as part of the medical establishment either. As I pointed out above, I am not involved in autism research or any organization dealing with autism or vaccines. I am simply an individual academic educator distilling this information for the public. But this is a small and not very relevant point.

Also Kirby might be just being facetious, as he claimed he was in his original post about the “medical establishment.” That gives him the ability to make his point – the medical establishment is fighting amongst itself over this real controversy – without being held accountable to the details.

Dr. Jon Poling

Dr. Poling has published an open letter (his second to me) on the Age of Autism website – a site dedicated to the failed mercury hypothesis of autism. He begins:

Your assertion that the scientific question of Autism etiology belongs to the medical community rather than Hollywood Stars is correct. I also agree that Hollywood opinions are more likely to be broadcast to millions because of their position in the media. This heightened awareness is nothing but a positive thing for the million families struggling with this difficult, and all too common, disorder. Jenny McCarthy is an Autism Mom looking for answers and rattling some cages—good for her. Amanda Peet is a new mom who believes in the importance of vaccines to protect her baby—good for her too. Don’t attack the moms, listen to them.

Heightened awareness is not always a positive thing, if it’s awareness of misinformation. What Jenny McCarthy has done is spread demonstrable misinformation (like the false claim that vaccines contain antifreeze) which has served to confuse the public. Unscientific propaganda distracts from real science and real solutions. Misinformed parents, hoping for answers, have been lured by false claims of the so-called “mercury militia” to risky and likely ineffective treatments, like chelation therapy. Hysteria over vaccines has led to increasing numbers of parents choosing not to vaccinate their children – with resultant outbreaks of preventable diseases, like measles. This threatens herd immunity – which means that vulnerable populations, and even the vaccinated, are at increased risk (because vaccines are not 100% effective).

Dr. Poling is saying, essentially, that public misinformation, risky and ineffective treatments, false hope, and unnecessary outbreaks of preventable disease is all good – as long as it raises awareness.

He then tries to defend McCarthy, and criticize me, with the mommy gambit. This was tried before by RFK Jr. who tried to deflect skepticism towards the dubious claims of the mercury militia as an “attack on mothers.” McCarthy’s status as a mom, even of a child who may have autism, does not exempt her opinions from scrutiny, and does not remove from her the obligation to use her celebrity status responsibly.

He continues:

In criticizing the journalism of Mr. David Kirby, you wrote:

“He refers again to the Hannah Poling case, a girl with a mitochondrial disorder who developed a neurodegenerative disorder with “features of autism” after getting a fever from vaccines.”

Actually—Hannah has diagnoses of DSM-IV Autism (by JHU/KKI psychology) and mitochondrial disorder (by two metabolic experts). The only ‘degeneration’ that occurred (along with 6mos of total growth failure) after 18mos of NORMAL development followed vaccination and nothing else! Of course, any ‘scientist’ can obviously point out that temporal correlation in a single case never proves causation. Rule number one of pediatrics though is “LISTEN TO THE MOM.” Are 10s of thousands of autism moms over the last decade suffering from mass hysteria induced by Hollywood? Not likely.

I was quoting the summary of the case in the Autism Omnibus decision – “features of autism.” I am still not convinced that Hannah Poling has a condition on the autism spectrum, as opposed to a broader neurological disorder that has features that overlap with autism – enough to meet some of the DSM-IV criteria. The case report also listed many symptoms that are not symptoms of autism. But as even Dr. Poling admits, this is part of the problem of trying to reach any reliable conclusion from a single case.

Dr. Poling emphasizes that his daughter had “NORMAL” development until 18 months of age. Assuming this is true, that only means the she had a regressive neurological disorder – nothing exotic or unusual there. Many neurological disorders known to be genetic are regressive – children develop normally for a while and then lose ground. This by itself is not evidence for any sort of trigger, such as vaccines.

He then reiterates his mommy gambit non sequitur, this time presuming to preach to me about clinical wisdom. Listening to mothers (and all family members, for that matter) is about information gathering – not about scientific conclusions. Parents know their children and are sensitive to changes – but that does not mean they are equipped to scientifically interpret them. Any good clinician listens to patients and family members – but also knows that there is a huge tendency to over-report. This is because people do not instinctively know what information is important and what is coincidence, they have a tendency to see patterns even where they don’t exist, and they report a great deal of misleading or just unrelated information. It’s the clinician’s job to sort through that – to find the signal in the noise. Dr. Poling is telling me to listen to the noise.

He then finishes with a straw man – that the only alternative explanation is that 10’s of thousands (not sure how he came by that number or what specifically he is referring to) of moms are suffering from mass hysteria. The true alternative hypothesis – the one actually favored by the scientific consensus – is that the observation of mothers (and fathers, for that matter, but I guess fathers are not as emotionally appealing as mothers) is accurate, the mistake is in assuming causation from correlation. Autism is frequently diagnosed around the age that many vaccines are given. This does not mean that vaccines cause autism. We need controlled or epidemiological data to test that hypothesis – and the evidence strongly suggests that the correlation is coincidence – not causation.

Dr. Poling then writes (again quoting me):

“This special case – which is not a case of autism being caused by toxins in vaccines – says nothing about the broader vaccine-autism debate.”

The only thing unique about my little girl’s case is the level of medical documentation—5 to 20% of patients with ASDs have mitochondrial dysfunction. Many other cases where mitochondrial testing is WNL is because “we never looked” not because the testing would be “within normal limits.” Most mitochondrial experts will tell you that the dots of autism and mitochondrial disorders are strongly connected.

Dr. Poling is saying that his daughter’s case is not unique – that it is typical and therefore is does say something about a broader vaccine-autism connection. Hannah Poling’s history has many features that are not typical of autism – like a history of otitis media with frequent fevers, seizures, and what sounds like a rare encephalitis that probably did result from vaccines. Even if we put her mitochondrial mutation aside – this is not a typical case of autism.

Dr. Poling also makes specific claims about the association of autism and mitochondria – which is not surprising as this is now one of the favored hypotheses of the anti-vaccine crowd, now that their mercury hypothesis has failed. He does not reference his “5-20%” figure for autistic children with mitochondrial dysfunction. The highest figure I could find was 7.2%. I admit I am not familiar enough with this literature to know for sure if there is other published data showing it is more prevalent, but reviews of the literature give figures more in the 5% range.

But more important than the exact figure is what it means. Right now the answer is that no one knows. Autism is not a single disease, it is a complex of diseases with various causes. It may be that one of those causes is mitochondrial dysfunction. The anti-vaccine crowd are now seizing upon this to argue that vaccines are triggering autism in a population made susceptible by mitochondrial dysfunction. But this is pure speculation – a hypothesis, not something that can be concluded from any data. It is also possible that some forms of mitochondrial dysfunction cause autism without a trigger, or that there are a host of environmental triggers. Perhaps any fever can trigger regression, and therefore vaccines may even have a protective role by decreasing overall infections. Maybe we should be vaccinating children with mitochondrial diseases more, not less.

Much is unknown, but at present there is no evidence (beyond two isolated cases of which I am aware, Hannah Poling being one) that links vaccines to mitochondrial dysfunction and autism.

About the case he writes:

“The case was settled (not judged in Poling’s favor, but settled) because both sides realized it was a special case that could not be extrapolated to other vaccine-autism cases.”

The case was not settled, it was conceded by medical representatives of Sec HHS. We are obviously pleased with the HHS decision to concede our case, but we had NOTHING to do with the concession. This was a unilateral decision from HHS (recall that HHS is the respondent, rather than the vaccine maker, as manufacturers have blanket liability protection afforded by the Vaccine Injury Program established in 1986) I will not speculate on the obvious question—why concede? Hannah’s case was positioned to set precedent as a test case in the Omnibus Autism Proceedings for potentially thousands of other cases.

OK – I’m not a lawyer and I don’t want to get caught up in legal jargon. This is also not a normal court of law, so I am not sure if “settled” even applies to such hearings. I was referring to the fact that the case was removed as a test case – it was not decided as a test case. The HSS agreed that “compensation was appropriate” – they specifically did not concede, as Dr. Poling implies, that vaccines were actually responsible for Hannah Poling’s autism. What the HHS did not do was rule that this case sets a precedent for any other case. It seems that Dr. Poling is suggesting that the Autism Omnibus cheated – they removed a case they knew would set a precedent they apparently did not want. But wasn’t the whole point of the test cases to be test cases?

My interpretation (and that of many others) of the HHS ruling was that the Hannah Poling case is too complex and unique to be a test case, but because there is so much uncertainty they erred on the side of compensation – without conceding that vaccines caused autism in this case.

He continues:

With regard to the science of Autism, I have no argument with the assertion that a single case does not prove causation of a generalized autism-vaccine link. What the case does illustrate though is a more subtle point that many physicians cannot or do not want to comprehend (ostensibly because vaccines are too important to even question). Autism is a heterogeneous disorder defined by behavioral criteria and having multiple causes. Epidemiological studies which have not found a link between autism and aspects of vaccination do not consider the concept of autism subgroups. Indeed, in a heterogeneous disorder like Autism, subgroups may indeed be ‘vaccine-injured’ but the effect is diluted out in the larger population (improperly powered study due to inability to calculate effect size with unknown susceptible subpopulation).

At least we agree that this one case cannot prove a causation between vaccines an autism. But isn’t that the whole question? Trying to rescue the conclusion he appears to want, Dr. Poling then pulls the “you can’t handle the truth” gambit. I really love this one – yeah, if vaccines are causing harm I (as an academic neurologist with no direct ties to vaccines at all) want to just hide my head in the sand. Please, Dr. Poling, spare us your condescension. Apparently what I cannot admit is that: “Autism is a heterogeneous disorder defined by behavioral criteria and having multiple causes.” Gee, that sounds familiar – where have I encountered that concept before.

Well, a year ago I wrote in a blog entry on autism:

First it needs to be understood that ASD is likely not a single disease but rather a groups of diseases possibly with various underlying causes.

Again – I am not an autism expert. I am just distilling the consensus of opinion of experts – who seem to have no problem recognizing that autism is a group of disorders, not a single disorder. I don’t expect Dr. Poling to have read every blog entry I have ever written on autism, but if I easily gleaned that from the autism literature, why didn’t he. It seems he was just assuming whatever was convenient for him to assume about his fellow physicians.

His next statement is technically true – that studies designed to look for an effect in an entire population may not be powered to find those effects in a much smaller subgroup. But I have been reading the literature long enough to recognize this as a juicy post-hoc rationalization. In other words – this is what proponents always say after the evidence doesn’t support their contentions. It’s true – but it doesn’t mean there is an effect in a subgroup – it just means we cannot rule out effects that are smaller than the data is capable of showing. This is universally true – by itself it does not rescue data from being negative. At best it means that you can generate a new hypothesis (a subgroup effect) after the old one has been rejected (no effect in the entire group). But all you have now is an untested hypothesis – speculation, not evidence.

After some more speculation, Dr. Poling descends further into condescension:

Definition: Autism is a heterogeneous systemic disorder with primary neuropsychiatric manifestations due to complex genetic and gene-environmental interactions likely affecting synaptic plasticity early in childhood development. This new theory of Autism is rapidly replacing the ‘old guard’ dictum that Autism is a genetically predetermined developmental brain disorder of synaptic formation/pruning that is set in motion prenatally. By the ‘10 year rule of science,’ your time is about up!

Time is up for what, exactly? Is Dr. Poling assuming that I am part of some alleged “old guard” that doesn’t recognize that genes interact with their environment. Does he forget that I am a neurologist, like he (we even went to the same medical school). I treat ALS – guess what is the leading hypothesis of the causes of ALS (which, like autism, is a group of diseases with various causes) – that it is a complex interaction of genetics and environment. What about multiple sclerosis – a complex interaction of environment and genetic predisposition. I have not been in neurology long enough to remember a time when this was not the leading hypothesis as to the cause of many neurological disorders.

Genetic phenotypes interact with their environment – sure. The question is, what is the relative contribution from genes and environment, and what is the interaction. This is a question that autism researchers have asked themselves (even though they cannot comprehend such things, according to Dr. Poling). For example there have been twin studies that specifically ask what the relative contribution of genes and environment is to the risk of developing autism. The conclusion of this one study is: “Genetic structural equation modeling showed that the overlap between AQ and WB and SOC was mainly due to genetic effects.” This means when they looked at twins and non-twin siblings, shared genetics was more of a predictor of overlapping autism traits than shared environment.

This is hardly the final word, and yes this is a very complex topic that requires further research. Subgroup analysis – sure, sounds good. My point is that researchers are asking the very question Dr. Polings suggests they are not, and also that the evidence is pointing in the direction of strong genetic influences in autism. I have no problem with researchers exploring hypotheses about specific environmental triggers of specific genetic abnormalities in subgroups of autism – research away. I would not even be surprised if such research yielded fruit.

My position, rather, is that we cannot conclude from current evidence that there is a specific causal link between vaccines and autism. Further, any such potential link is likely to only be relevant to a minority of autism cases as existing evidence rules out a moderate or larger association between vaccines and autism. Finally – none of this rescues the thimerosal hypothesis as the explanation for an autism epidemic. Dr. Poling admits that they are now nibbling around the edges of autism – not proposing a possible explanation for a major cause of autism. The fact that there was no measurable effect on the incidence of autism after the removal of thimerosal rules out a significant effect for thimerosal by any mechanism.

This includes the article that Dr. Poling references on altered calcium homeostasis. He says that in the discussion the authors speculate about thimerosal as a possible trigger. This type of speculation does not derive from the data – it’s just speculation. Further, as I just pointed out, it doesn’t matter what mechanism is proposed for an effect from thimerosal, the data already rules out a significant effect.

He concludes:

Thank-you Dr. Novella and his band of skeptics for continuing the debate.

No problem, Dr. Poling. Me and my merry band of skeptics will now return to Sherwood Forest. In all seriousness – more skepticism is exactly what Dr. Poling and his side need. Skepticism is just good science.

It’s interesting that Dr. Poling finishes with Kirby’s debate gambit – as if the debate justifies it’s own existence. At least that brings this discussion full circle.


Existing evidence supports the conclusion that autism spectrum disorder – while a collection of disorders of differing causes, is dominantly genetic. There is no evidence to support vaccines or thimerosal as a significant contributor to autism. Proponents of an autism-vaccine link have been steadily retreating from negative data falsifying their claim. They are now focusing on small subgroups of autism with possible genetic and environmental factors – but at present have nothing more than speculation to implicate vaccines even in this drastically reduced role.

Meanwhile there are groups and individuals presenting to the public unsubstantiated claims that vaccines are unsafe and cause autism. These unproven claims are leading some parents to unproven and dubious therapies. It is also leading some parents to forgo immunization, threatening herd immunity and already resulting in outbreaks on preventable diseases.

David Kirby and Dr. Poling want the debate to continue – and not just in the laboratory and the clinic, among scientists who can parse the subtleties of the data – but in public where fear-mongering and misinformation are easy to spread. According to them, it’s all good.


This entry has been cross-posted at Science Based Medicine

36 responses so far