In addition to being full of undisclosed conflicts of interest and unreasonably invasive and traumatizing for the autistic kids studied, Andrew Wakefield's now retracted 1998 study linking vaccines to autism was also fraudulent. At this point it's unreasonable to suppose that he made these errors by accident or that he was framed through some grand conspiracy. Enough evidence has come to light to show that Andrew Wakefield faked up nearly all of the details of that study. All of this information came out as a result of reporter Brian Deer's investigation, Wakefield's misguided attempts to sue Deer, and an investigation by the General Medical Council regarding whether Wakefield should maintain a license to practice medicine (narrator: Wakefield should not practice medicine and in fact his license to practice was revoked).
In what follows I review all of the ways that we now know Wakefield's retracted study was fraudulent/erroneous:
1. Autism symptoms did not suddenly appear within 14 days of receiving the MMR vaccine
Andrew Wakefield had a problem: Autism and vaccines both happen in early childhood so one would expect them to co-occur by chance. In order to connect autism to a specific vaccine, he had to show that the symptoms started to appear right away following a vaccine-related illness.
So, he faked up the histories of all of his subjects so that it would look like symptoms of autism began to appear less than 14 days after they got the vaccine. In the retracted 1998 paper, Wakefield claimed that the children in the study started manifesting symptoms of autism an average of 6.3 days after the MMR injection (range 1 to 14 days). In fact, most of the children's symptoms started to appear either long before or long after getting the MMR vaccine. And some didn't really have symptoms of autism at all but had some unrelated developmental delay.
Below I outline discrepancies between what was falsely reported in the retracted paper and what reporter, Brian Deer was able to dig up:
Subject | Onset of first symptoms of autism (according to parents and doctors) | Onset of first symptoms of autism according to Wakefield et al. (1998; retracted) |
Child 1 | Either 3 months before or 6 months after vaccination (conflicting accounts) | 1 week after MMR |
Child 2 | 2 to 6 months after MMR | 2 weeks after MMR |
Child 3 | Unclear - First visit to doctor was 15 months after MMR | 48 hours after MMR |
Child 4 | Before MMR | Immediately after MMR |
Child 5 | 5 months before MMR | 2 months after MMR |
Child 6 | | 1 week after MMR |
Child 7 | | 24 hours after MMR |
Child 8 | "Some months" before MMR | 2 weeks after MMR |
Child 9 | 2 to 4 months after MMR | 1 week after MMR |
Child 10 | 4 months after MMR | 24 hours after MMR |
Child 11 | 1 month before MMR | 1 week after MMR |
Child 12 | | Less than one month after MMR |
In perhaps the clearest sign of fraud, Andrew Wakefield circulated an earlier draft of the paper 6 months prior to publication that listed different times of onset with a maximum of 56 days after receiving the MMR vaccine.
Again, Wakefield needed to show a temporal connection between autism and the MMR jab in order to make his case. Knowing that the temporal connection didn't exist in reality renders the entire paper useless by itself. Furthermore, why would we trust someone who would just fake up all of this data? What else has he faked up?
2. No evidence that the children had "regressive" autism
Another problem is that there was really no evidence that the children had "regressive" autism. Only one child showed a decline. All of the other children were delayed from the start, which is consistent with conventional autism, a condition people are born with. Many of the children were never diagnosed with autism at all and the symptoms of the children ranged from severely autistic to just acting out a bit.
3. Subjects' digestive system disorders were invented or altered in the reporting
Subjects' digestive system disorders (if any) differed from the reporting. In the retracted study, Wakefield and colleagues claimed that 11 out of 12 children had "non-specific colitis." There are several problems with this claim:
The pathology reports of 8 out of 11 children came back normal during the original study (though their results were represented in the study as being pathological and chronic).
Most of the children were severely constipated, a fact omitted from Wakefield's report because it didn't line up well with the "colitis" claim
At best, evidence of mild inflammation was found in the children's bowels. This did not constitute "colitis."
Because the colitis finding was fictionalized, there is no evidence here at all linking autistic tendencies to any kind of gut disorder.
Conclusions
Andrew Wakefield's retracted 1998 paper was a work of fiction. In order to help sue the vaccine industry, he invented and distorted all of the facts of each of the children's cases. (1) the children did not develop autism symptoms within 14 days of receiving the MMR shot, (2) there was no evidence that the children regressed (developed normally then declined) except in one case, and (3) the children also didn't have colitis. The entire MMR vaccine-colitis-autism claim was entirely built on a foundation of falsified data.
About the author
Aaron Charlton, PhD is a science and medical blogger and entrepreneur. He writes for Away Clinic and other medical clients. He also maintains a website called OpenMKT.org that is aimed at improving transparency and quality of scientific research within the field of marketing. He is sometimes quoted by the media on matters of scientific integrity.