InfluentialPoints.com
Biology, images, analysis, design...
Medical Diversity Livestock Invasives
"It has long been an axiom of mine that the little things are infinitely the most important" (Sherlock Holmes)

 

 

Wakefield's initial MMR / Autism study

Prior to the mid 1980s, vaccination of children in UK against measles, mumps and rubella was performed using separate vaccines. In 1988 a combined MMR (measles mumps rubella) vaccine was introduced. Coverage was very high in the 1990s, and the number of children catching these diseases fell to an all-time low.

A box of the MMR vaccine used in UK
(Photo: InfluentialPoints )

In the Lancet medical journal, Wakefield et al. (1998)  reported gastrointestinal disease and behavioural disorders (autism) in twelve previously normal children. In most cases, onset of symptoms occurred after they received the MMR vaccine. Eight of the cases were attributed to MMR vaccination, either by the child's parents or their physician.

Whilst Wakefield et al. admitted that they had not proved an association between measles, mumps, and rubella vaccine and the syndrome described, Wakefield et al. called for further investigations into the "possible relationship between autism and the vaccine" and, at a subsequent press conference, argued that the MMR vaccine should be withdrawn.

Wakefield's paper produced a flood of letters to the journal (see for example Lee et al., 1998 ) pointing out (amongst other things) that merely observing A preceeds B, does not mean that A causes B!
 
Since Wakefield's fallacy is extremely common, you may prefer a concrete example:

Observing that few people marry after they die (most marriages occur beforehand) does not show that marriage causes death.

The MMR vaccine is given at around 12-15 months of age, and the mean age at which parents of children with autism first report concern is 18-19 months - irrespective of whether or not the children have received the MMR vaccine. Hence there will inevitably be a close temporal association for affected individuals. None of the other normally accepted criteria for demonstrating causality were met (or even discussed).

 

Was there any epidemiological evidence?

Presumably in response to the call for further investigations, Taylor et al (1999)  undertook an epidemiological study to investigate Wakefield's claim that the MMR vaccine may be causally associated with autism. They concluded that since there was no sudden change in the autism trend line after the introduction of MMR vaccination, there was no evidence for a causal association between MMR and autism. This provoked a letter to the Lancet from Wakefield (1999),  who criticized the study, and presented data himself purporting to show a delay in the increase in number of cases of autism in UK compared to the USA concurrent with the delay in introduction of MMR. For critical evaluation we reproduce his figure below:

Figure adapted from Wakefield (1999) 

Taking this graph at face value, it shows that diagnoses of autism increased rapidly from the late 1970s in both the USA and the UK. But it simply does not show a delay in UK (where the MMR vaccine was not introduced till 10 years later) as argued by Wakefield. Trying to read anything into the slightly larger increase in autism cases from 1984 to 1985 really is clutching at straws. Given the level of random error from year to year, all we can say is that from 1978 to 1985 the number of reported autism diagnoses appeared to increase (more or less) exponentially in both countries. This increase coincided with changes in criteria for diagnosing autism - to which the great majority of scientists ascribed the increase in reported cases.

In addition Cox & Kirkham (2007)  point out that it is misleading to interpret these graphs as showing time trends in diagnosis 'incidence' since they are actually derived from the age distribution of autism cases. Many assumptions have to be met for these to be interpreted as 'autism incidence'. Even if we were to accept that there was prima facie evidence for temporal association in the MMR autism trends, there are various criteria that should be met if one is going to postulate the MMR vaccine causes autism. Authors should as a matter of course go through these criteria and make a reasoned judgement. Yet again, as with the 1998 paper, this was not done - another black mark for the Lancet referees (if they existed) and the editor.

 

What was the media's role?

Some accounts of the MMR fiasco focus upon Wakefield's original 1998 paper, depite the fact that it was Wakefield's 1999 letter to the journal which really got things moving. Once the media decided to take up this story it acquired a life of its own. In just one month, early in 2002, there were hundreds of media reports - described by some as a 'feeding frenzy'. Uptake of the MMR vaccine dropped alarmingly, and health workers started warning of the probable return of measles as a serious public health problem.

The media's role in this was detailed by Hargreaves et al. (2003).  Fifty-three percent of people surveyed in UK assumed that, because the media gave both sides of the debate equal coverage, there must have been equal evidence for each. This despite the fact scientific opinion overwhelmingly rejected the claim of a link between the MMR vaccine and autism. Only 23 percent of those sampled were aware that the evidence favoured supporters of the MMR vaccine. Hargreaves et al. concluded that the use of so-called 'balanced' news reporting achieved the exact opposite, by wrongly creating the impression of an equal weight of evidence.

It is debatable in this case whether more blame should fall on the researchers (for bad science), the journal (for publishing it) or the media (for sensationalising and misreporting it). Wakefield's initial suggestion of a link on the basis of just eight cases, attributed to MMR vaccination by the child's parents or their physician, was completely unjustified. However, the fact it (and the subsequent letter) ever got into print was entirely the fault of the journal editor and its referees, and both the author and Lancet editor would appear to be responsible for this misrepresentation of results.

The effects of all this were magnified by the vast amounts of inaccurate reporting by the media so that, in the end, much of the public ended up with a highly biased fraudulent picture of the risks of the MMR vaccination. A sustained campaign in UK to reassure parents limited the damage there, but anti-vaccination campaigns thrive on the internet worldwide and have seriously affected uptake of MMR vaccination in several developing countries.

A Nigerian child with measles
(Photo: courtesy of Wikipedia/Mike Blyth )

 

Why were Wakefield's MMR papers ever published?

This story does not end there. Deer (2011a),  (2011b),  (2011c)  accused Wakefield and some of his coathors of 'data fabrication' (fraud by any other name). In most of the 12 cases the children's ailments, as described in The Lancet, differed from their hospital and GP records. Although Wakefield's research paper claimed that problems came on within days of the MMR innoculation, medical records suggest this was true in only one case. In many cases, medical concerns had been raised before the children were vaccinated. Hospital pathologists, looking for inflammatory bowel disease, reported in the majority of cases that the gut was normal, yet the Lancet paper showed them as abnormal.

We may never know how much of this fraud was perpetrated for financial gain, and how much was subconcious 'modifying' of the data to fit expectations (politely known as 'confirmation bias'). Certainly the latter is a high risk in any field of research when researchers become determined to prove their 'pet' theory. One would hope that journal editors and referees are always on the look-out for such bias. That said, Deer (2011d)  has suggested on his website that the only reason Wakefield's MMR / autism research was initially published in the Lancet was because of the powerful influence of the 'old boy network' amongst some senior medical researchers.

 

Feedback & comments

    We would be delighted to receive feedback and comments at info@influentialpoints.com

    Comments from authors quoted on this page (whether positive or negative) are displayed here, in the order received.

  • Ian Hargreaves 
    Professor of Digital Economy, Cardiff University, UK
    14 Oct 2011

    Thanks for drawing this to my attention, Bob.

    I would phrase the conclusion and implications for the news media of our work slightly differently from the way you make the point. It is not that the pursuit of "balanced news" necessarily leads to error of the kind which occurred in the MMR case, but that the goal of balance should not be deployed as a substitute for critical thinking about the quality of evidence on different sides of an argument. The remedy for an over-literal adherence to "balanced news" is self-evidently not "imbalanced news" but reporting which is founded upon sceptical inquiry with regard to all parties. In my view, that is the only reliable route to trusted journalism, as argued, for example, in my Journalism: a Very Short Introduction (Oxford University Press).

 

References

  •  Cox, A.R. & Kirkham, H. (2007). A case study of a graphical misrepresentation: Drawing the wrong conclusions about MMR vaccine. Drug Safety 30 (10), 831-836. Abstract  Full text 

  •  Deer, B. (2011a) Secrets of the MMR scare. How the case against the MMR vaccine was fixed. BMJ 342: c5347 (Published 5 January 2011). Abstract  Full text 

  •  Deer, B. (2011b) Secrets of the MMR scare. How the vaccine crisis was meant to make money. BMJ 342: c5258 (Published 11 January 2011). Abstract  Full text 

  •  Deer, B. (2011c) Secrets of the MMR scare. The Lancet's two days to bury bad news. BMJ 342: c7001 (Published 18 January 2011). Abstract  Full text 

  •  Deer, B. (2011d) Roy Pounder brokered Lancet publication of discredited MMR-autism scare study. Website accessed 24 June 2011. Full text 

  •  Godlee, F. et al. (2011). Wakefield's article linking MMR vaccine and autism was fraudulent (Editorial). BMJ 342: c7452 (Published 5 January 2011) . Abstract  Full text 

  •  Hargreaves, I. et al. (2003). Towards a Better Map: Science, the public and the media. Economic and Social Research Council, UK. Full text 

  •  Lee, J.W. et al. (1998). Autism, inflammatory bowel disease, and MMR vaccine. (Letter) Lancet 351, 905-909 (March 21). Letter  All letters 

  •  Taylor, B. et al. (1999). Autism and measles,mumps,and rubella vaccine: no epidemiological evidence for a causal association. Lancet 353, 2026-2029 (12 June). Abstract  Full text 

  •  Wakefield, A.J. et al. (1998). Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet 351, 637-641 (28 February). Abstract  Full text 

  •  Wakefield, A.J. (1999). MMR vaccination and autism. Lancet 354, 949-950 (11 September). Full text