The vaccine-autism link: bad information for scared parents

Two articles published in the New York Times caught my interest this month (3 Rulings Find No Link to Vaccines
and Autism
and Court Says Vaccine Not to Blame for Autism).  The articles described the dismissal of court cases where parents were suing vaccine manufacturers for giving their children autism.  Some parents claimed that a preservative containing mercury, Thimersol, caused autism. Others claimed that the Measles, Mumps, and Rubella (MMR) vaccine can contribute to causing either autism or gastrointestinal dysfunction (and gastrointenstinal dysnfunction has links to autism).  This made me curious – where did the whole MMR-causes-autism hypothesis come from?  What science is there on either sides of this debate?  And where does the science stand now? If you just want the summary, here you go: there is no conclusive evidence that MMR causes autism.  There is lots of evidence that it in fact DOES NOT cause autism.  Read on for more information on some of the original science to arm yourself against the shucksters who perpetrate this myth to prey on desperate parents.  While this is not an exhaustive review of every paper published on this topic, I tried to pick representative papers and looked at only those articles published in peer-reviewed journals (a basic and important standard for scientific research).I started with the original 1998 paper published in the British medical journal The Lancet that many anti-MMR vacciners cite, titled “Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children“.  The paper was a little hard to read with the giant red ‘Retracted’ printed across every page, but I muddled through.  They looked at children with ‘regressive development disorder’ and chronic enterocolitis, a gastrointestinal dysfunction.  At the beginning of the paper, the authors stated that parents of the autistic children noted that the symptoms of autism showed up around when the children received the MMR vaccine.  They also noted that all of the children had abnormal gastrointestinal symptoms.  This was a sample of 12 children, so was a preliminary investigation, not the end to a long series of related studies.  They use most of the discussion for potential links between gastrointestinal symptoms and autism, but they do touch on the MMR-autism hypothesis.  They say that following encephilitis brought on by measles, some children develop psychosis.  They also stated that other studies had found that parents reported autism symptoms around the time their children were received either the MMR or measles vaccine.  However, they specifically stated that

“We did not prove an association between measles, mumps, and rubella vaccine and the syndrome described.”

but further investigations were needed to investigate this charge.  If I understand the history of this allegation correctly, people started pointing the finger at MMR for the following reason: the MMR vaccine came in to wide use late in the 1990s (depending on the country) and this corresponds with a reported increase in the number of children diagnosed with autism.  This was supported by the parent-reported onset of autism symptoms around the time their child received the MMR vaccine.  Also, studies around this time reported finding measles virus DNA in bowel biopsies, potentially tying in to the whole gastrointestinal disorder-autism hypothesis.

So, the jury is still out in 1998. The hypothesis did not stand up against further scrutiny in 1999, when another paper was published in The Lancet titled “Autism and measles, mumps, and rubella vaccine: no epidemiological evidence for a causal association.” They looked at British children born since 1979 that were diagnosed with autism.  They did find that the rate of autism is increasing but importantly they did not find that there was a sudden jump in the number of children with autism that corresponded to the introduction of the MMR vaccine.  The increase in autism has yet to be determined.  But hypotheses are not proved (or disproved) based on the outcomes of one study – rigorous science requires experiments to be repeated in different populations and with different researchers, so the case was still not closed.

Another prominent epidemiological study to investigate this purported relationship was published in the New England Journal of Medicine in 2002 titled “A Population-Based Study of Measles, Mumps, and Rubella Vaccination and Autism.”   This study looked at Danish children who either did or did not receive the MMR vaccine from 1991-1998.  This covered over 500,000 children, and they found that both vaccinated and unvaccinated children had the same risk for autism.  They also noted that the average age for MMR vaccination was 1 year 5 months, while the average age for diagnosis with autism was 4 years 3 months.  This finding doesn’t support earlier claims that children were diagnosed with autism close to the time they received the vaccine.

The debate really heated up with a series of letters to the editor in The Lancet where authors of the original paper from 1998 take opposing viewpoints and are obviously very upset.  I kind of liked seeing this spark in the scientific literature – from my experiences in organic chemistry people rarely come out as straighforwardly to call someone else’s BS.

Simon Murch fired the first volley with his plea for people to stop using the 1998 paper as a reason to not vaccinate their children.  He laments the fact that vaccination rates in the UK have fallen to 61% for MMR and that children are getting sick and dying as a result.  He points to the fact that the hypothesis that gastrointestinal disorders may be related to autistm has been holding up to further scrutiny, but no causal relationship between MMR and autism has even been proven.  I kind of feel bad for this guy- he published some preliminary researched that has ballooned out of control, and feels partially responsible for the rise in measles because of the hype and non-science around this paper.  Unfortunately reining in this tiger has turned out to be nearly impossible.  David Thrower responds that they studies that showed no link between MMR and autism are ALL flawed (sounds like a conspiracy) and that upon further examination

“exposes each one as flawed, with unsupportedassertions, questionable hypotheses, and overstated outcomes.”

Talk about the kettle calling the pot black.  I don’t really follow his logic after this: he states that there is evidence to support the link between gastrointestinal disorders and autism (which is true) and somehow this supports the MMR-autism connection.  This is strange.  He cites studies support the gastro-autism link and somehow concludes that this supports the MMR-autism connection.  This doesn’t compute.  I do feel sorry for him, as he ends his letter with the statement that he has an autistic child.

Simon Murch’s co-author of the 1998 paper, Peter Harvey, joins the exchange of letters to the editor in The Lancet.  He makes a very similar argument to David Thrower, somehow intuiting that gasto-autism links prove MMR-autism links.  He provides evidence to support this hypothesis: children with gastrointestinal disorders AND autism have traces of measles virus in their gastrointestinal system.  Simon Murch responds to defend himself against Murch’s and Thrower’s attacks.  He responds that reports of measles virus in autistic children’s guts are preliminary and require much more evidence.  He concedes that epidemiological studies (such as ones examining links between autism and MMR) can be difficult to perform and should be evaluated carefully, but he puts it nicely:

“However, the concordance of all studies represents an impressive body of data. Thrower’s suggested tactic, in which every study that shows an unwanted outcomeis destruction-tested from a hostile viewpoint, is essentially that used for years by the tobacco industry.”

Translation: there is a increasing body of work to disprove the MMR-autism connection, and it is bodies of work taken together that prove or disprove hypothesis – not cherry picking.

Later in 2004, 10 of the 12 authors of the original 1998 Lancet piece partially retracted the 1998 paper.  Even though no causal link between MMR and autism was hypothesized in the 1998 paper (in fact they explicitly stated that “We did not prove an association between measles, mumps, and rubella vaccine and the syndrome described.”) they felt responsible for the public health maelstrom that resulted from the MMR-autism hypothesis, and wanted to formally state they did not support this.

Evidence against the MMR-autism hypothesis continue to mount. All of these studies were carried out with large cohorts of children.

2006: “Is there a ‘regressive phenotype’ of Autism Spectrum Disorder associated with the measles-mumps-rubella vaccine? A CPEA Study.”  found no evidence that onset of autistic symptoms was related to measles-mumps-rubella vaccination.
2007: “MMR-vaccine and regression in autism spectrum disorders: negative results presented from Japan.”  This study was great because MMR was only administered in Japan from 1989 to 1993. They found that during the period of MMR usage there was no increase in the number of children diagnosed with autism.
2008: “A sampling of the literature includes the following: Lack of Association between Measles Virus Vaccine and Autism with Enteropathy: A Case-Control Study” studied whether MMR vaccination or presence of measles virus in the GI of children corresponds with diagnosis of autism.  They found that healthy children and those with autism had the same levels of measles virus present in their GI.  They also found no relationship between when the children were given the MMR vaccine with onset of GI symptoms or autism.

I had hoped this conspiracy theory would be finally put to bed in 2010 when the original 1998 paper was formally and completely retracted from the journal The Lancet.  Not only did they say that several elements of the 1998 paper were incorrect, but that the investigation itself had never been properly approved by an ethics committee (an important part to any scientific research involving humans, and that several of the authors of the paper were sanctioned for acting dishonestly.  Unfortunately, as one person was quoted in the New York Times article,

“…it’s very hard to unscare people after you’ve scared them…”

Given the number of pseudo-science websites you can find by googling words like MMR, vaccine, safety, and autism, it looks like it will be a long time before the good science penetrates through all of the misinformation.