by Brian Deer
But why leave out the “integral and fundamental” from a journal read mostly by doctors? For some, this knowledge might have aided patient care: raising the profile of an important issue for developmentally challenged children, including those without language to express it.
Yet, for specialists, it might also have raised questions, doubts, about the validity of the so-called “syndrome.”
“Constipation is the exact opposite of what you would normally be looking for,” Booth explained, “in a patient with inflammatory bowel disease.”
Such a disease, of course, is what Wakefield claimed. It was integral and fundamental to Barr’s lawsuit. And if mild inflammatory cell changes were reported in constipated children, it “raises the question,” Booth told the panel, “of whether the constipation was responsible.”
Constipation? Not MMR? Booth’s analysis was far from speculation. Fecal stagnation behind intestinal obstructions, as well as abrasion of the intestines’ surface epithelium (just one cell deep, in both the large and small bowels), had long been linked with inflammation. Indeed, nine years before the hearing, at the meeting at the Royal College of Surgeons, the Scottish gastroenterologist, Anne Ferguson, who’d asked about JABS, had also spoken of constipation and “little ulcers.”
This would surely have been known to some of the paper’s thirteen authors. But, to be fair, only one man wrote it. Strictly speaking, most of his colleagues weren’t entitled to authorship, falling foul of stringent criteria. According to the Uniform Requirements for journal manuscripts, “substantial contributions” to multiple aspects were expected—which was true for hardly anyone but Wakefield.
None of the authors was eminent. Three were trainees. “In 1996–1998, you needed very little involvement to be named as an author,” confirmed one of them, a junior pathologist, Andrew Anthony, in a statement. And the consultants were barely better placed. “I did not write the histology sections,” Dhillon said in a reply to a story by me. Murch (who says he never saw the final version before printing) told me that “enterocolitis” was a “dreadful term” for what was found, while the swollen glands were “hugely oversold.”
Indeed, when other authors sought amendments to the paper, before publication, it was left to Wakefield whether he took any notice. Mark Berelowitz, the child psychiatrist who’d spoken in the Atrium, not only said that he didn’t know which of the children was which but that he didn’t agree with Wakefield’s descriptions of autism. It was “not a behavioural disorder,” he told the panel, “and it was not clear that it was a regressive disorder.”
Even Walker-Smith had accepted Wakefield’s claims based on a secondhand data compilation. He said that Wakefield brought “all the clinical and laboratory details” to his office one day, prepared in a “sort of master chart.”
“We all rely on trust,” the Australian professor told the panel, on his twenty-fourth day in the witness chair, when asked about the descriptions in the paper. “I trusted Dr. Wakefield.”
Whoaah. “I am sorry?” asked the black-dressed Sally Smith, from her position at the far end of the rectangle.
“We all rely on trust,” the now-seventy-year-old pediatrician repeated, quietly. “Yes, and I trusted Dr. Wakefield.”
“You trusted him in what context?”
“Just in general.”
It was an emotional moment, on a bright August day. I think all of us knew what he meant. But there was more on the horizon regarding the pathology, from the next defendant in the chair. This was Murch, who would speak of an astonishing event—which Davies, Walker-Smith, and Wakefield all testified as having no recollection of happening.
Giving evidence on day 113, Murch spoke of a meeting between the paper’s authors roughly three months before publication. Wakefield had circulated the latest version, he said, and several doctors, including Murch himself, Walker-Smith, and Davies, plus possibly two juniors and a few others in the department, gathered in a histopathology seminar room to look at the slides again.
Would the others forget this? Or did Murch imagine it? I can’t gauge which would be the more incredible. This was around ten months after the last child left Malcolm Ward; about three months since a media storm over MMR had erupted, thanks to stories placed in the magazine Pulse; Wakefield and his mentor, Roy Pounder, had met with management; the Royal Free was gearing up for the Atrium event. And yet it seemed that, at least, someone influential still agonized over the accuracy of Table 1.
“I do have a very good recollection of the meeting,” Murch told the panel, chaired by Surendra Kumar, a family doctor, flanked by Stephen Webster, a gerontologist, Parimala Moodley, a psychiatrist, Wendy Golding, an educationalist, and Sylvia Dean, a former local government chief executive. “I think the reason was initially that Dr. Davies had seen the draft of the paper and just wondered whether the description of the histology perhaps oversold it.”
Think about that. Real or dreamt, this was remarkable testimony. If the descriptions were “oversold,” what would be left to publish? Would the authors unnerve The Lancet by correcting Table 1? Would Wakefield tell the nation that he made a mistake? Everyone in the seminar room had résumés to think of. Would any junior raise a hand of objection?
Nobody present needed a PhD in psychology to know what Walker-Smith wanted. “Prof,” as they called the top doctor in the room, who sat on committees with professors of medicine and surgery, wasn’t only set for a tasty academic credit (which the university submitted to the Research Assessment Exercise), but he’d been putting all the children (apart from the American) on powerful Crohn’s anti-inflammatory drugs that carried black box warnings to prescribers.
Wakefield, likewise. He was being paid at hourly rates to advance his “syndrome” in Barr’s lawsuit. If he didn’t find fault with the solicitor’s target product, his lavish personal fees would stop. He’d filed for two patents claiming single measles shots. And, with Pounder, he’d already announced to the world, in Pulse, evidence that “confirm our suspicions.”
Murch told the panel that the debate over the biopsies was led by the pathologists, Davies and Dhillon—who must have known their specialty’s consensus. Mild elevations of inflammatory cells—as noted here and there in Davies’s reports, and similarly scored on Dhillon’s grading sheets—were frequent, normal findings in healthy guts, and shouldn’t be reported as colitis.
“A common error is to diagnose ‘mild chronic nonspecific colitis’ in biopsies of normal colon because of the normal population of mononuclear cells,” explains, for example, a landmark guide from the period, published in November 1989 in the American Journal of Surgical Pathology. (And I collect a small stack of papers saying the same.) “As a rule of thumb, the diagnosis should not be made unless there is evidence of injury to the colonic epithelium.”
Davies, nevertheless, took an authorship credit, and Murch told the panel that the reporting was accurate. “All the pathologists present when the slides were reviewed,” he said, “agreed that the wording was reasonable.”
So, yes please, they’d have a paper with their names on it in The Lancet, the world’s number two general medical journal.
TWENTY-FIVE
We Can Reveal
As what Wakefield would later call a “war” over vaccines spread from Britain to the United States, I began to get invitations to give talks. That meant PowerPoint, Microsoft’s slide-maker, and weekends spent sizing and pasting.
From early days, I devised a less than lecture-like color scheme: orange and yellow on a solid black background, with a gambling theme in the artwork. The left bottom corner featured a card-fan motif: a hand with five aces, including two spades. And an opening run of photographs featured Wakefield with his base: an ever-increasing number of women.
One . . . four . . . five . . . nine . . . Then stock images of hundreds . . . and thousands.
At the start, my carbon footprint remained comme
ndably small. Which was probably just as well. After revealing Wakefield’s secret legal deal, the money he billed at hourly rates, his planned vaccines and products meant to make his fortune, how his own lab had failed to find measles virus genomes, and that he’d refused to perform a gold-standard study, what I had was too exhausting for a forty-minute presentation. It was more like a shopping list of conflicts and outrages than a hard news “We can reveal.”
But all that changed on Sunday, February 8, 2009: five years after my first MMR splash. By now, the hearing on Euston Road was temporarily adjourned. And Wakefield was back on page 1.
I was at Heathrow Airport for a dawn flight to Detroit, Michigan, when I picked up the paper and spread it on the floor to read my latest installment. Around me, Terminal 5 was all but deserted as I gazed down at the front and scanned a couple of paragraphs, with my fists under my chin and my elbows on my knees.
MMR doctor fixed data on autism
It was about time, too. We’d probably waited too long. But we needed to be sure to get it right.
The doctor who sparked the scare over the safety of the MMR vaccine for children changed and misreported results in his research, creating the appearance of a possible link with autism, a Sunday Times investigation has found.
Confidential medical documents and interviews with witnesses have established that Andrew Wakefield manipulated patients’ data, which triggered fears that the MMR triple vaccine to protect against measles, mumps and rubella was linked to the condition.
I wasn’t sure about “the condition,” but newspapers were team efforts. What followed, inside, was more so. I licked a middle finger, reached down, and turned pages, until I came to a spread at pages 6 and 7: three thousand words, across sixteen columns, plus two panels of further information. Reversed, in white lettering—upper case, boldface—on a gray background strip across the top of both pages, ran the headline:
HIDDEN RECORDS SHOW MMR TRUTH
Then a photograph of me, and an introductory standfirst:
A Sunday Times investigation has found that altered data was behind the decade-long scare over vaccination
Neat.
Three color pictures ran from left to right: the inevitable baby crying after being stuck with a needle; the actors Jenny McCarthy and her then boyfriend Jim Carrey, waving their arms in “Green Our Vaccines” T-shirts; then Wakefield grinning, exposing a dental overbite, outside the hearing on Euston Road.
“Key Dates in the Crisis,” one panel was headed. The other: “How the Scare Led to the Return of Measles.”
I’d seen this coming. I’d gotten the interview with Ms. Two, which I’d not yet reported. I had the stack of poorly redacted reports from the pilot study database. I’d gotten expert submissions, filed in Richard Barr’s lawsuit. I had transcripts of hearings before senior judges, countless documents obtained under the Freedom of Information Act. And, pursuant to court order, I’d read the kids’ records at my lawyer’s office on the day Wakefield dropped his claim for defamation.
Some of this stuff was legally sealed. And there were still gaps to fill in the evidence. Like a piano with missing keys, the recurring pattern was obvious. But I still couldn’t play the tune.
The Australian professor came to my aid, however. His change-of-story strategy with the General Medical Council, to argue that the scopings were all pure patient care, meant his lawyers—and everyone around the rectangle of tables—rehearsed, over and over, sometimes a dozen times, every note and letter in those stacks of cardboard crates. Those three-inch-thick binders, stuffed with secret records, flopped and flopped far more than I needed.
Seated by the door, filling notebook after notebook, I not only had cases where normal pathology was reported in The Lancet as inflammatory bowel disease. I had kids, tabulated with diagnoses of regressive autism, who’d not been diagnosed with autism. I had boys who’d been listed with the first symptoms within days, whose records revealed nothing for months. And I had patients who’d been the subject of doctors’ concerns before they received MMR.
I counted not one case where I could reconcile the records with what was published, to such effect, in the journal.
To me, it was pretty much what you might expect if you picked a bunch of patients from anti-vaccine groups’ lists for a project supervised by provincial lawyers. And while our page 1 headline was shocking enough, when individual children’s cases were drilled down through the records—histories, diagnoses, and claimed temporal links—the audacity of what had happened became staggering.
Each anomaly, in itself, seemed technical. Dry. A scratchy note here. A biopsy report there. But from this web of discrepancies had emerged a case series that had fooled The Lancet’s editors, peer reviewers, and readers, provoked a storm of publicity that ensured public funding for Barr’s class action, and so created the engine for what would become a global crisis of confidence in vaccines.
One example was the case of the only girl among the dozen: a three-year-old, not only a JABS referral from the same town as Child Four, but a patient at the same general practitioners’ office. “She was reported in the journal as having suffered a brain injury ‘two weeks’ after MMR,” ran my story, on page 6.
Her medical records did not support this. Before she was admitted, she had been seen by local specialists, and her GP told the Royal Free of “significant concerns about her development some months before she had her MMR.”
A six-year-old boy’s colon, reported in The Lancet as diseased, was revealed to be unremarkable in his records:
He was reported in the journal to be suffering from regressive autism and bowel disease: specifically “acute and chronic nonspecific colitis”. The boy’s hospital discharge summary, however, said there was nothing untoward in his biopsy.
Incredibly, two of the twelve (including Child Six, whose poorly redacted report I’d read at my lawyer’s office) weren’t merely similar cases, they were brothers. And, along with another boy, introduced to the research by the brothers’ mother, they hadn’t been diagnosed with autism. Indeed one, Child Seven, aged nearly three, was discharged from the hospital with a letter stating, “He is not thought to have features of autism.”
There was so much stuff, we couldn’t get it all in. Even the three thousand words were mangled. But pick-up was good, especially in the United States, where USA Today, Newsweek, the Los Angeles Times, the Chicago Tribune, and a string of other outlets ran reports.
And now I got more invitations to give talks on what I’d learned, where I could really dig into the detail.
My trip to Detroit let me do just that, starting on the day after we published. I’d been invited to give a week of lectures, seminars, and “grand rounds” presentations on the snowbound campus of the University of Michigan, Ann Arbor, where I unveiled my first-ever PowerPoint. I’d yet to get the colors right, and my formats were amateur: for the most part crude bullet points.
Wakefield’s Assignment from the Lawyer
1. Establish a time link between MMR and disorders (historically 14 days)
2. Find a fingerprint of damage
3. Propose a mechanism for this damage
The Lancet Paper Delivers on the Assignment
1. 8/12 “MMR,” with 14 days maximum to “behavioural symptoms”
2. A “new syndrome” of regressive autism and bowel disease
3. Proposal that the ultimate culprit may be measles virus
I also framed a Q&A quiz, which I would use in my talks ever after. Sharp observers might have noticed (although, if they did, they never said) an apparent contradiction in my narrative. If, as I reported, the children were recruited and brought to Hampstead to construct a case against MMR, why did The Lancet only report eight with the link? Surely the parents of all twelve would stake a claim?
“So why didn’t all the kids’ families name MMR?” I asked during my main event in Mic
higan: the Susan B. Meister Lecture in Child Health Policy.
I let that question hang, in the hope of raised hands. Then I answered myself, “They did.”
By now, the binders had been remorselessly scoured. The kids’ records had been read into mine. And it was clear that when the parents gave their accounts at the Royal Free, eleven children’s mothers or fathers blamed the shot. The one remaining family blamed a “viral infection” (first thought to be rubella, but then deemed to be measles), and even they later changed their story to finger MMR, after being visited at home by a lawyer.
Thus three allegations were withheld from The Lancet. Hence, eleven, not eight of twelve.
So now, when I asked the audience why the figures might be different, hands rose like spring-loaded umbrellas. The commonest opinion was that eleven, or twelve, would have given the game away. The cohort’s parents would be exposed for who they really were: a group of marshaled, prescreened complainants against the vaccine, not a bowel clinic’s typical work.
Much later, Wakefield raised a different explanation, however, citing a criterion not stated in the paper. “We reported on those eight who had made the link at the time of their child’s deterioration and excluded those who made the link later,” he said, in a 148-page affidavit, with underlining.
If a child’s parents made the association later in time and not on their own because, for example, they had read about the issue in a newspaper article, their association was not included. To have included those who only came to this position more recently and “secondhand” would have clearly biased the Lancet paper’s reporting.
He came up with that account when he was suing me again: this time, unsuccessfully, in Texas. At face value, it might seem to make some kind of sense, even if not revealed, as it should have been. However, while the maximum “interval from exposure to first behavioural symptom” for the reported eight was claimed to be fourteen days, it was between one and three months for those he omitted. So, including them would have blown his temporal link. But he was at least acknowledging that parents’ stories might be wrong—an admission I never heard from him elsewhere.