The idea that Andy Wakefield is a fraud is the quick-and-dirty way to dismiss anyone with vaccine safety concerns. I was reminded of that on a couple of fronts this week. An Age of Autism reader who wrote Sen. Dianne Feinstein of California to support vaccine choice just received this reply:
“I understand that many parents are also concerned that vaccines may cause autism. This claim was published in 1998, in an article in the Lancet, a British medical journal. The researcher who authored the article was later found to have deliberately falsified data to produce a fraudulent link ..."
The source of the "fraud" claim, the British Medical Journal, decided to remind everyone of that 2012 report this week. In a "Dear Colleagues" letter, BMJ Clinical Director for North America Carolyn Wong Simpkins wrote that in the current measles outbreak, "we are seeing the sad consequences of parents opting out of these [vaccine] benefits. But do you remember the origins of the rumors attempting to connect the MMR vaccine with autism? It began with a research paper—later retracted—from investigators at a London medical school, but soon spread fear, guilt, and now the resurgence of a nearly eradicated infectious disease across the globe.
"In 2011, The BMJ published an in-depth, three-part investigation that described the problems with data corruption and bias in the original paper. As we move forward, and encourage parents to vaccinate their children, I think it’s important to revisit this history and remember the detrimental effects that fraudulent data can have on the health of the global population, and the importance of championing transparency, integrity, and scientific literacy."
May I be permitted to interject a minor quibble here? There was no fraud! I first made that case in a 10-part series in 2012, and am recapitulating it here in the hope that someone of Feinstein's stature might decide to look more closely before they repeat the "fraud" canard yet again.
The BMJ began its 2011 attack on Wakefield's "elaborate fraud" by claiming he altered every single one of 12 children's anonymous case histories to create a phony link between the MMR vaccine and autism. In five cases, it said, signs of autism actually began before the shot was even given.
If true, yep, that's "deliberately falsified data ... a fraudulent link" -- in a word, good old-fashioned fraud. But let's meet the claim at its strongest point and see if it holds together. That is the story of Child 11 in the case series. In the BMJ, author Brian Deer claimed Child 11's symptoms couldn't possibly have been caused by the MMR shot because they appeared “too soon” -- a full two months before the shot. Deer said the father himself spotted the "anomaly" and was deeply upset about Wakefield's deception. Wow. Gotcha! An "elaborate fraud" indeed.
But none of that is true.
Like Deer, I was able to identify the 12 Lancet families, and I set about contacting them in the months after the BMJ series was published; I eventually spoke to more of them than Deer did both in the U.S. and England, where I spent a week taking trains from Wales to Bath. I met Father 11 -- the only American case -- closest to home, at a Peet’s Coffee shop in an affluent, picture-perfect Southern California enclave.
We sat outside in the mid-60s sunshine he jokingly called “a little frosty.” A wealthy businessman who lives in a gated community nearby, he wore a light jacket emblazoned with “Cal,” for the University of California at Berkeley where he got an engineering degree. He carried a thin file folder and a spiral notebook.
In this laid-back setting, it was hard to grasp the role he and his family have played in one of the major medical controversies of our time, one that unfolded in a foggy city 6,000 miles to the east.
This father is Deer’s best witness among the parents of the 12 children described in the Lancet paper – in fact, his only one, the lone parent who is hostile to Wakefield, not just a little frosty, but coldly angry. His anonymous comments to Deer in the BMJ seemed to fully support its January 5, 2011, cover story: “Secrets of the MMR Scare: How the Case Against the MMR Was Fixed.”
“My investigation of the MMR issue exposed the frauds behind Wakefield’s research,” wrote Deer. Child 11 was among those “whose parents apparently blamed MMR,” but in truth “Child 11’s case must have been a disappointment. Records show his behavioural symptoms began too soon.” [Italics in original] Deer quoted from a Royal Free Hospital discharge summary: “His developmental milestones were normal until 13 months of age. In the period 13-18 months he developed slow speech patterns and repetitive hand movements. Over this period his parents remarked on his slow gradual deterioration.”
Deer summarized: “That put the symptom two months earlier than reported in the Lancet, and a month before the boy had MMR. And this was not the only anomaly to catch the father’s eye. …”
This "fraud" by Wakefield -- blaming the vaccine for symptoms that began two months before the shot -- was the coup de grace for serious consideration of a link between vaccines and autism. Wakefield was “convicted of fraud,” wrote Time magazine in an article titled “The Dangers of the Antivaccine Movement.” An editorial in The New York Times, titled "Autism Fraud," noted Britain’s General Medical Council had already stripped Wakefield of his medical license, and the Lancet retracted the paper: “Now the British Medical Journal has taken the extraordinary step of publishing a lengthy report by Brian Deer, the British investigative journalist who first brought the paper’s flaws to light — and has put its own reputation on the line by endorsing his findings.”
Indeed it did.
“Clear evidence of falsification of data should now close the door on this damaging vaccine scare,” Editor in Chief Fiona Godlee wrote. She said “there is no doubt it was Wakefield” who was responsible for the “elaborate fraud,” despite having 12 co-authors.
Hold the door, please. I was about to learn that Deer’s explosive claim about Child 11 – Exhibit A in this alleged hoax -- was false.
The father opened the file folder – guarding the papers against a fickle coastal breeze -- and showed me a letter he had written on January 1, 1997, to “Dr. Andrew Wakefield, Royal Free Hospital, London, England.”
"My son [name deleted] at age 15 months, was immunized with the Merck MMR vaccine and became ill for the next several months,” the letter began.
“As his pediatric records indicate he came down with a viral infection, and shortly thereafter viral pneumonia. His condition slowly deteriorated over time, and was diagnosed as being autistic on his birthday at age 3. The onset of his autistic behavior began around 18 months. … He was diagnosed as moderate to severe, with no speech, no eye contact, and cognitive function at 6 months overall.”
Multiple specialists in the United States confirmed the autism diagnosis, the letter added, as well as their suspicions of the MMR vaccine as the cause. Further workups in California also revealed “indeterminant inflammatory bowel disease” -- the dual syndrome Wakefield was then investigating at the Royal Free. That was why the father wanted the hospital’s pediatric gastroenterologists to evaluate his child.
So – first came the shot, then the symptoms. The father’s account, and medical records created before he got anywhere near Wakefield, could not be clearer. But didn’t he tell Brian Deer exactly the opposite, as recounted in the opening of the BMJ cover story? And didn’t a hospital record confirm that?
No. And no.
Though you’d never know it, the father was actually disputing a much more minor "anomaly" -- how long after the shot specific symptoms occurred -- that didn't bear on the essential truth of the paper. There was no question of which came first -- it was the shot. In fact, the father did directly blame the MMR for causing his son’s illnesses and autistic regression – a fact that appears to have escaped Deer’s notice, or at least acknowledgement.
Yes, the father was angry at Wakefield. Yes, he disagreed with other points, some of them unrelated to the content of the Lancet article. But no – he did not say that the symptoms came before the shot. He gave Deer no basis to claim that the shot came "too soon" to have triggered the symptoms.
No basis, in short, for a claim of fraud.
But what about the the discharge document that was the sole basis for Deer's claim? It was simply wrong, one of thousands of pieces of paper generated by many medical personnel in a complicated medical case stretching over many years; perhaps the “13-18 months” was a typo for “15-18,” since that is what the father had reported all along. Regardless, the father says he never told Deer that the symptoms came first, and there is no evidence to the contrary. Deer apparently did not bother to check that one piece of paper against the large volume of other evidence, or to confirm it with the father, or to make sure that his own claim that symptoms began “a month before the boy had MMR” coincided with any actual chronology.
As far as I can tell, no one on the planet -- no doctor, no parent, no document – has ever said Child 11 was anything but healthy and developing normally before the MMR. No one, that is, but Brian Deer in the BMJ. And here we see Deer at work: Because Wakefield was a fraud – because Deer said so – any discrepancies between data in the Lancet paper and any other source was proof against Wakefield. One document says 13-18 months for the period of regression? That was evidence enough that Wakefield “used bogus data … to manufacture a link” between the MMR and autism.
And so it went in the other 11 cases -- no evidence at all of the "fraud" Deer and the BMJ alleged; in fact, plenty of evidence of no fraud at all. The claim was really a selective, manipulative, deceptive -- one might say fraudulent -- exercise that no one bothered to subject to a case-by-case critique.
To my surprise as we sat outside in Southern California, Father 11 told me he hadn’t read the BMJ article, and he declined my offer to quote from it or have him read it during our visit. He would rather lay out the sequence in his own words, he told me.
That turned out to be a useful approach.
His son had been completely healthy and developing normally, he said, until the MMR shot at 15 months triggered a downhill progression.
“I very much believe it,” he said about the relationship of the shot to the symptoms: The measles component of the vaccine triggered an immune deficiency that produced the cascade of devastating physical and mental problems. This, in fact, was Wakefield’s provisional hypothesis.
When I showed Father 11 what Deer had written about the shot-and-symptoms sequence, he said, emphatically, “That’s not correct.”
A few days later, after he read the BMJ piece, the father sent Deer and myself an email.
“Mr. Deer’s article makes me appear irrational for continuing to believe that the MMR caused difficulties which predated its administration, but until the incorrect dates in the discharge summary were pointed out to me this week, I failed to realize that the discharge summary was inaccurate.”
The father, still seething at Wakefield for other slights, imagined or not, wrote that this was an honest mistake on Deer’s part.
“Based on the incorrect discharge summary I shared with him, Mr. Deer reasonably inferred that my son’s autistic symptoms predated his receipt of the MMR vaccination, which they did not.”
I’m no engineer, but neither is Father 11 a journalist. As someone familiar with the norms of my profession, I had rather a different reaction. I found it hard to see how Deer -- who interviewed the father in person twice, once in California and once in London, corresponded by email, and must have heard the same story I did — could get something so important so wrong. The number of times he used the father’s quotes to misleading effect – making him appear to angrily assert that the symptoms preceded the shot – was too high; the way he did it seemed too artful.
Yet when I met him, Father 11 was as straightforward and precise as you might expect from a successful engineer. By the time I finished my Peet’s, I had no doubt about the chronology or the documentation.
Besides, the imperative to get the facts correct – and to correct them promptly and prominently if called for – is implacable. Even more than a hurried newspaper account, there were not supposed to be any mistakes in Deer’s work in the august British Medical Journal. They said so themselves.
Yet this rather fundamental mistake, if you can call it that, in Deer's reporting -- which went to the very heart of the claim that Wakefield committed an "elaborate fraud" -- did not perturb the BMJ. Nor did the other cases I dissected that pointed to, at most, issues of interpretation and analysis in the paper, most of it supplied not by Wakefield -- a researcher -- but by the clinicians he worked with.
This is a single example of many factual misrepresentations by Deer and the BMJ. In 11 out of the 12 cases, the case histories were actually taken not by Wakefield but the senior clinician and author in the paper Prof John Walker-Smith who, unlike Wakefield, was funded to appeal and completely exonerated in the English High Court in March 2012. The judge found no evidence of misreporting of data or inappropriate medical procedures. The two histopathologist co-authors defended the clinical findings of the paper in the face of BMJ's allegations (here and here) and in November 2011, after the intervention of whistleblowing scientist David Lewis, Deer and BMJ's editor Fiona Godlee along with their expert advisor Prof Ingvar Bjarnason were forced into an embarrassing climb down in the columns of Nature.
But they've stood by their fatally flawed "fraud" claim, and that has caused irreparable harm. “The BMJ stands by the article by Brian Deer and the linked editorial published on 5 January,” Godlee wrote in February 2011 in response to e-mails critical of its reporting by readers at Age of Autism.
“The article, which was subjected to peer review and editorial checking, was based on enquiries carried out over some seven years, involving, among other things, interviews with parents of children enrolled in Andrew Wakefield's research. Four such parents are quoted in the article. As made clear in the article, the core data on which the findings were based were evidenced, except in the case of one child, by the transcript of a General Medical Council fitness to practise hearing which sat between July 2007 and May 2010.”
That “one child” – the exception for which no independent evidence existed -- was Child 11. No one could check what that father said, or so it seemed. I imagine Deer was counting on that. Now Deer and the BMJ (which has partnered with MMR manufacturer Merck on other projects) just go on repeating their many long-ago disproven allegations in the cynical expectation that no one will be able to do anything about it.
Dan Olmsted is Editor of Age of Autism.