Renowned journal retracts controversial autism/vaccine paper


by Kayt Sukel

February 10, 2010

On Feb. 2, renowned medical journal The Lancet retracted a controversial 1998 paper that hypothesized a link between the measles, mumps, and rubella (MMR) vaccine and autism. Citing ethics violations by the lead author, Andrew Wakefield, the journal’s decision set off a global media firestorm. Many scientists argue the paper’s retraction definitely debunks the premise that vaccines play any role in the development of autism. But advocacy groups, as well as many parents of children who have autism, argue that Wakefield’s work still has merit—and questions about why autism rates have risen so dramatically over the past few decades still need to be answered.

The retracted study

Wakefield was the lead researcher on the paper, “Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive development disorder in children,” published in the Feb. 28, 1998, issue of The Lancet. It was a series of 12 case studies of children who were referred to a pediatric gastroenterology unit with both gastro-intestinal issues and autistic symptoms.

“It wasn’t a study, per se,” says Paul Offit, the director of the Vaccine Education Center at the Children’s Hospital of Philadelphia. “He looked at children believed to have intestinal abnormalities and hypothesized that maybe after they received their MMR vaccine, they developed the intestinal problems and that may have led to the autism. He raised a hypothesis but, in that paper, he didn’t study a thing.”

Soon after the paper was published, the General Medical Council (GMC), the organization responsible for regulation of doctors and proper medical practice standards in the United Kingdom, received a complaint about the work.

“The GMC can look into a doctor’s registration when any concerns are raised by a member of the public, a newspaper, a journal, or even an institution,” says Lisa Crack, a spokesperson for the organization. The GMC investigation of the complaint culminated in a panel hearing that lasted 148 days. According to Crack, it’s the longest hearing in GMC’s history. The panel found Wakefield did not receive proper approvals from a governing research ethics committee and that he was biased in his selection of participants.

Based on the GMC’s findings, editors of The Lancet decided to retract the paper, releasing the following statement:

"It has become clear that several elements of the 1998 paper by Wakefield et al. are incorrect, contrary to the findings of an earlier investigation. In particular, the claims in the original paper that children were 'consecutively referred' and that investigations were 'approved' by the local ethics committee have been proven to be false. Therefore we fully retract this paper from the published record."

Tony Kirby, press officer for the journal, said there would be no further comment on this decision, but did say such a decision is not unprecedented.The Lancet has retracted more than a dozen papers in its history, including one in October 2009.

Precedents notwithstanding, Offit argues The Lancet’s decision sends a strong message to the research community. “Bad science gets published all the time. Rarely does a journal retract a paper, they usually just let the bad science die in the face of irreproducibility,” he says. “But in this case, not only was the paper incorrect, it was fraudulently incorrect.” 

Three separate hypotheses

Irva Hertz-Picciotto, an environmental epidemiologist who has been studying autism, says that there are three specific hypotheses regarding vaccines and autism that need to be examined separately.

“Wakefield’s hypothesis about the MMR vaccine is a different question than thimerasol or other preservatives in vaccines. And then there’s a third hypothesis that perhaps we’re giving too many antigens at once and tiny babies’ immune systems can’t handle it,” she says. “Each need to be looked at in their own right.”

Hertz-Picciotto says most rigorously conducted studies have not reported any associations between vaccines and autism, with regard to any of the vaccine hypotheses. But the studies are not as definitive as they could be. Offit agrees that a vaccinated/non-vaccinated study would be important but it would be very difficult to conduct.

“You’d have to have a large enough sample of children completely vaccinated and an equal number completely unvaccinated and then hope, likely beyond reason, that both groups are the same in terms of healthcare-seeking behaviors and socio-economic background,” he says. “The big problem is that parents who choose to not give any vaccines to their children usually have very different healthcare-seeking patterns than those who fully vaccinate. There are too many variables to potentially control for.” 

But even without that kind of study, Offit contends that enough time and resources have been put into the vaccine question.

“Almost all studies regarding autism suggest that development occurs very early on in life, probably in utero,” he says. “It just doesn’t make sense for vaccines to be involved. They are the best studied of all environmental factors and though I can’t say with certainty that there’s no link, they’ve had a pretty good look.”

Offit thinks that the fear of vaccines that studies like Wakefield’s founded has been a very sad chapter in science for the health of children. “Carl Sagan once said that extra claims need to be backed by extra evidence—and given what we know about the benefits of vaccines, the risks in avoiding them is tremendous. We’ve seen children hospitalized and even dying from the measles,” he says. “The evidence just isn’t there. It’s time to start focusing on other factors.”

Hertz-Picciotto agrees that it’s time to broaden the investigation to other factors. “There are so many suspect causal factors in autism that this huge focus on vaccines may be keeping us from doing important work on potentially hundreds of others,” she says. “Autism is not a simple story and we’re not going to find one single factor. We’re more likely to find a minefield of environmental factors than some single smoking gun.”