The Vaccine and Autism Connection: The Wakefield Study Once Again Discredited

Summer is over and as we head into the colder months, we know that the flu season is rapidly approaching. And once again, the controversy about the vaccination of children flares anew. Should children be vaccinated or not? What are the risks? Much of the consternation regarding vaccinations stems from the Wakefield work on autism and his perception on the role vaccines played in causing autism. Although his study (and theory) has been thoroughly discredited by science, it is a good time to review this calamity.

In January 2011, the British Medical Journal (BMJ) published detailed critical analyses (2011a; 2011b; 2011c) by Brian Deer, investigative reporter, of the controversial link between the MMR vaccine and autism, focusing on the original study by Andrew Wakefield, et al.(1998). These new revelations should permanently dispel the connection between the two. This new information at the same time underscores the importance of adhering to the scientific method and open and full disclosure of the scientific and research process.

The recent discoveries of what seems to be flawed and unethical research are summarized here (for full information, please read the BMJ articles by Deer). These findings can be grouped into problem with the subjects, methodology of the Wakefield study, potential researcher biases, and ethical behavior on the part of the researchers.

Through interviews and review of medical records, Deer found several inaccuracies between what Wakefield, et al. reported in their study, and information found elsewhere pertaining to the subjects. A major concern is that the participants were not randomly selected. Random selection of subjects lowers potential validity problems and allows more confidence in the results. In the Wakefield, et al. study, it is now learned that several of the participants were solicited from an organization called Justice Awareness and Basic Support (JABS). The banner on its web site states, “Welcome to the support group for vaccine-damaged children.” Although this organization publicly supports vaccinations, most of the messages imply an anti-vaccination crusade, demanding monetary damages for children injured by vaccinations and more scientific research investigating possible harm caused by vaccines. Including children solicited from JABS suggests at least two potential problems. First, these children may have been predisposed to either medical problems or parents already biased against the MMR vaccine. Secondly, this suggests that random selection by the researchers was not, in fact, done. The danger with doing this is that Wakefield could select children who were already exhibiting bowel problems. Another major problem with the Lancet study concerned participants having problems that were linked to vaccine, but in fact existed prior to getting the vaccine! For example, one child had (previous to the administration of the vaccine) already been diagnosed with regressive autism and diarrhea. Another child was described in medical reports prior to starting the MMR study as presenting “recurrent diarrhea,” “developmental delay,” “general delay,” “restricted vocabulary,” “and “….a very small deletion within the fragile X gene,” with the parent noting that she had concerns about his developmental progress when the child was only 18 months old.

Regarding the diagnosis of regressive autism, Wakefield and colleagues said that nine of the subjects clearly demonstrated signs of regressive autism after the vaccine. However, upon review of medical records, three of the nine clearly had no such symptoms. And these three didn’t even have a clinical diagnosis of autism, either at admission or discharge from the hospital after completion of the study. A recalculation of results now shows that of the 12 subjects, 6 were reported to having symptoms following the vaccine, but 6 did not – a random result, no effect of vaccines.

Deer also discovered flawed research methodologies used by Wakefield, et al. Beside selecting participants in a biased way, Wakefield and colleagues seemed to already know what findings they wanted to discover in their research. The ideal research question is assessing whether two variables are related, and have no a priori belief in the outcome. Although researchers often hypothesize a particular finding that may be discovered in the research, the researchers will take considerable steps in research protocols to protect from researcher bias. However, Wakefield and colleague (Barr) seemed to have a very clear agenda in mind. On a research application for funding for a study to follow the original Lancet research, they wrote, “The objective….is to seek evidence which will be acceptable in a court of law of the causative connection between either the mumps, measles, and rubella vaccine or the measles/rubella vaccine and certain conditions which have been reported with considerable frequency by families of children who are seeking compensation” (Barr & Wakefield,1996). Potential researcher bias towards a preferred outcome is clear here. Another problem with the design of the experiment involved poorly defined dependent variables with little apparent regard for reliability of measurement. For example, the researchers described one child as demonstrating “behavioural symptoms” as a result of the vaccine. In fact, medical records indicate that the child actually had a chest infection.

Another research problem with Wakefield’s original study concerns replication of results. Although given ample opportunity to do so, Wakefield has never attempted to replicate his findings.

Deer learned that Wakefield was not an unbiased researcher, going where the data took him. It it now known that Wakefield was employed by and working for attorneys who were building a lawsuit against vaccine manufacturers. Deer reported that Wakefield was being paid £435,000 to build a case linking the MMR vaccine to autism. Furthermore, Wakefield was planning on developing a diagnostic test that would detect measles virus in the bowels or other bodily fluids. This diagnostic procedure would be quite lucrative if Wakefield could prove that the MMR vaccine caused the “leaky gut” and developmental delay that he hypothesized.

Deer also reported potential ethical lapses on the part of the researchers. The participating children were subjected to what was considered severely intrusive medical examinations, including ileocolonoscopy, lumbar punctures, barium meal, electroencephalography and evoked potentials, and magnetic resonance imaging brain scans. Wakefield and colleagues asserted that the hospital ethic’s committee approved these tests. However, Deer found otherwise- the Ethical Practice Committee had approved data collection from investigations that were clinically indicated, which these tests were not. There were such significant problems with this original research that in 2010, the General Medical Council of England judged Wakefield and colleagues to have committed around 30 serious ethical charges, including dishonesty (4 counts) and subjecting participants to medical tests that were not justified. Wakefield is appealing these findings.

The debacle of the Wakefield study once again underscores the importance of adhering to good science and shining the light of public study and debate on the methods and results of scientific research. If good science had been used in the original study published in Lancet, the researchers wouldn’t have had a preexisting agenda for their research (i.e., to find a link between MMR and development and physical problems.) If good science had been used in the Lancet study, then the children selected for that research would have been selected randomly rather than for characteristics that might influence getting a particular result. If good science had been used in this original study, reliable and valid measurement of the critical variables would have been assured. Replication of results would have been attempted.

The Wakefield study undeniably triggered the vaccine-autism hysteria and the anti-vaccine scare across the world, resulting in a distrust in childhood vaccines that still exists today. It is a case study of how so many aspects of the scientific method were apparently violated. Deer’s investigations have shown that Wakefield developed a shoddy research design that does not stand up to the light of public scrutiny, and there is ample evidence to suggest that he was conducting an experiment to find proof that would support his previously-held belief. Good science was not used.

References