How the 1998 Lancet MMR paper started a vaccine scare

A 1998 paper on 12 children set off a long-running vaccine scare in Britain. Vaccination rates fell and measles returned. It took 12 years for the journal to fully retract the paper, and the claim kept a following after the retraction.

October 31, 2026 · 4 min read · TruthRanking

The first source

On February 28, 1998, The Lancet published a paper by Andrew Wakefield and 12 co-authors at the Royal Free Hospital in London. It described 12 children referred to the hospital with bowel symptoms and a loss of developmental skills. The parents of eight of the 12 linked the start of those problems to the MMR vaccine, the combined shot for measles, mumps and rubella. The paper itself stated that it had not proved an association between the vaccine and the syndrome it described.

The hospital held a press conference to mark publication. An Open University account, citing the journalist Brian Deer, says Wakefield told reporters he wanted the three vaccines given separately. A 2010 report in the Canadian Medical Association Journal says media coverage was driven by speeches and appearances in which he recommended single vaccines. The single-shot advice came from those public remarks.

How it spread

The expert response came fast. In March 1998, at the request of the Chief Medical Officer, the UK Medical Research Council convened a group of 37 experts. They found no current evidence of a link and advised no change to vaccination policy, according to the Open University. Coverage grew anyway. A Cardiff University study, reported by the BMJ in 2003, counted 561 media reports on MMR over seven months of 2002.

Vaccination followed. Health Protection Agency figures show MMR uptake peaked at 92% in 1995 and 1996, then fell below 80% by 2003, and was as low as 60% in some areas. In 2003, 442 measles cases were reported, three times the 1996 figure.

In 2004, an investigation by Deer for the Sunday Times reported that the research had been backed by £55,000 from lawyers. The Lancet said it regretted that the funding and related litigation had not been disclosed to its editors, but it did not withdraw the paper. Ten of the paper's 13 authors later renounced its conclusions, the Associated Press reported. In December 2006, Deer reported that Legal Services Commission figures showed Wakefield had been paid £435,643 in fees, plus £3,910 in expenses, from the legal aid fund. His work for the lawyers began two years before the paper was published, Deer wrote.

What it became

By 2008, measles was again considered endemic in England and Wales, the AP reported. The UK Health Protection Agency linked a large outbreak in 2008 and 2009 to the fall in MMR uptake, according to the CMAJ.

The UK General Medical Council's hearing into Wakefield and two colleagues began on July 16, 2007. On January 28, 2010, its panel issued findings that Wakefield had acted unethically and with "callous disregard" for the children, as the CMAJ reported. The BMJ reported that the panel found him guilty of dishonesty and irresponsibility. On February 2, 2010, The Lancet fully retracted the paper, saying parts of it were incorrect. On May 24, 2010, the panel found Wakefield guilty of serious professional misconduct and ordered his name erased from the medical register.

In January 2011, the BMJ began a three-part series by Deer, based on seven years of inquiries. Deer compared the paper with the children's hospital records. He reported that five of the 12 children had documented developmental problems before vaccination, although the paper described them as normal until the shot. The journal's editors checked his findings against the medical council's six-million-word hearing transcript and called the paper "an elaborate fraud." The editors noted that Wakefield has repeatedly denied doing anything wrong.

Large studies kept testing the question. A Danish cohort study published in 2019 in the Annals of Internal Medicine followed 657,461 children born from 1999 to 2010. Of them, 6,517 were diagnosed with autism. The researchers, who compared vaccinated children with 31,619 unvaccinated ones, found no association between MMR and autism.

What the record shows

The paper is often remembered as proof that MMR causes autism. Its published data were narrower: parent reports about eight children, in a paper that said it had not proved an association. The causal message traveled mainly through the press conference and later appearances.

The retraction came in stages. In 2004 the journal kept the paper while regretting the undisclosed funding. Most of the authors then renounced its conclusions. The full retraction came in February 2010, 12 years after publication.

The findings also came from different bodies and say different things. The medical council judged the conduct of the research and of individual doctors. The fraud finding is the BMJ editors', built on Deer's comparison with records. The panel also found co-author John Walker-Smith guilty of serious professional misconduct. On March 7, 2012, a High Court judge quashed that finding and his erasure, citing "inadequate and superficial reasoning" by the panel. A third doctor, Simon Murch, was found not guilty of serious professional misconduct. Those outcomes concern individual doctors. The paper remains retracted.

The medical question has its own answer, separate from the fraud finding. A report on 12 children can raise a question. Settling it takes large population studies, like the Danish one.

How to trace a claim like this

1. Read the first source, not the launch. The paper and the press conference said different things. Compare the first source with the press version before rating the claim.
2. Check what each number counts. Twelve children is a sample size. Eight is a count of parent reports. The 561 media reports of 2002 count coverage. None of these measures whether a vaccine causes a condition.
3. Separate coverage from corroboration. Hundreds of articles repeating one study add up to one study. See republication vs corroboration.
4. Check what happened to the source later. Look for retractions, corrections, funding disclosures and disciplinary rulings, and note their dates. This paper stayed in the journal for six years after its funding was questioned.
5. Weigh study size and design. A description of 12 patients and a cohort of 657,461 children are different kinds of evidence. Our methodology explains how we weigh them.