Monday, August 25, 2014

Andrew Wakefield Tortures History

I've written before about how the anti-vaccine movement regularly gets the science wrong (see, for example, here and here). Most often, it involves how they interpret (or, rather, misinterpret) studies that they think support their view that vaccines are the root of all evil, causing all manner of disorders. They regularly misrepresent and torture the scientific literature to fit their agenda. But occasionally, it involves actually attempting to do science themselves, the latest of which being a study [Edited to Add: this study has been fully retracted by the journal.] recently published by Vaccine Injury Compensation Program petitioner, Brian Hooker, PhD and funded by the anti-vaccine group Focus Autism. This study has been examined by an epidemiologist, another epidemiologist, and a cancer researcher. The players have also been examined by a dyslexia consultant and autism advocate (see also here). This study, in addition to being of questionable validity, has spawned quite the brouhaha among anti-vaccine activists. In short, Hooker took a dataset that was analyzed by a group of CDC investigators. Their finding was that age at first vaccination with MMR did not appear to [edited to clarify and correct] affect the risk of be associated with autism [ETA: cases (those with autism) were no more or less likely to have been vaccinated with MMR than controls before 18, 24 and 36 months; the study design could not establish causation, only correlation]. The results of their study were published as a case-control study. Hooker used the same dataset to create a cohort study and found that there was no increased risk of autism in any of the groups (i.e., MMR does not cause autism) except one: African American males, for which he puportedly found a 3.36-fold increased risk of autism when they were vaccinated between 24 and 31 months. Read the two posts by those epidemiologists for why his analysis is suspect.

I'm not going to go into the science behind the studies, because it relies very heavily on statistics, and I'm not a statistician or epidemiologist. As an aside, neither is Dr. Hooker. Instead, I want to focus on some statements made by one Andrew Wakefield, the British doctor who committed scientific fraud, resulting in the full retraction of his 1998 case series study on MMR and the stripping of his medical license. Wakefield boldly added himself to the list of not only torturing science, but now adds to his accomplishments torturing history and ethics (granted, we already knew he was ethically challenged). You see, in his videos (here and here) about the alleged "whistleblower", William Thompson,Wakefield compared the purported "cover-up" to the Tuskegee syphilis debacle. It's a false comparison used simply to inflame people and claim the race card.

Wakefield and Hooker allege that the CDC found data that the MMR increases the risk of autism in African American males when given between 24 and 36 months of age and that they purposefully added arbitrary criteria to suppress this information from their final published results. Wakefield says that this purported cover-up of the CDC suppressing this information from the public is similar to the Tuskegee cover-up. But there is just one problem with this.

There was no cover-up in the Tuskegee syphilis experiment.

In fact, just about everything in Wakefield's comparison is wrong, except for the fact that both studies have a race component. Let's take a look at what happened in Tuskegee and how it does (or doesn't) relate to the DeStefano study, rather than the impression being given by Wakefield and his fawning acolytes. Overall, Wakefield does actually give an accurate description of what happened at Tuskegee, but the conclusions he draws from that, and the comparisons he makes to DeStefano et al. are completely off-base.

Back in the 1920s, researchers began to look at the clinical course of syphilis to find out what exactly it does and how quickly it does it. Syphilis is a bacterial disease transmitted through sexual contact. It is a slowly progressing, easily curable disease that, if left untreated, can cause some very serious complications, such as problems with muscle control, paralysis, blindness, dementia, and death, as it attacks the internal organs. Back in the early 1900s, not a great deal was known about the disease, and the only treatments available were, at best, mildly effective. It was also thought that the disease progressed differently based upon the patient's race.

In the Tuskegee experiment, which started in 1932, researchers at the Tuskegee Institute in Macon County, Alabama, in collaboration with the Public Health Service (the precursor to today's CDC), started a study to examine the effects of syphilis in poor, rural black men. This population was particularly vulnerable to exploitation, since not only did they have poor resources and a lack of education, but rampant racism caused many to view them as less equal than Caucasians.

When the study began, there weren't really any major ethical issues with the study. There were no known cures for the disease and the subjects were given free health care and checkups, , free meals, as well as funeral insurance. It was mainly an observational study, so risks were minimal. The biggest ethical issues at the start were that subjects were not told the purpose of the research, and the incentives are what would now be considered "undue influence". They were simply told the researchers were examining "bad blood", a term to describe a wide range of ailments. However, more ethical problems started to crop up. The researchers prevented subjects from getting treatment elsewhere, and they never informed subjects they were infected with syphilis, leading them to spread the infection to their wives and, consequently, children, who were born with congenital syphilis.

The biggest ethical lapse arose in the 1940s. In 1945, penicillin was shown to cure syphilis when given in a timely manner. This information was never disclosed to participants in the Tuskegee study. Furthermore, even after "Rapid Treatment Centers" were set up by the Public Health Service in 1947, the men participating in the study were prevented from seeking treatment by keeping them in the dark about not only the existence of a cure, but even that they had syphilis in the first place. The subjects were even given placebos and told they were being treated. This went on into the 1960s, when researchers rotated onto the study stated their concerns, and on up to 1972, when a researcher at the Public Health Service, Peter Buxtun, leaked information to the media about the study, after his attempts to lodge internal complaints over the previous 6 years failed. PHS and the study investigators justified their conduct by saying that the study was not complete and that the information to be gained was of such importance to the advancement of medical knowledge that the subjects had to remain in the dark and untreated. They also believed, as did many scientists of the era, that the subjects could not possibly have grasped the complexities involved in scientific and medical research, and so could never have possibly given anything resembling informed consent.

Buxtun's leak to the press resulted in a public uproar, the closing of the study and the begin of reparations to the men and their families. In all, by the end of the study, 28 subjects had died from syphilis, 100 died of disease-related complications, 40 of their wives had been infected, and 19 of their children were born with congenital syphilis. It was one of the worst ethical failures in human subjects research in the history of the United States, and it led to the Belmont Report (establishing the research principles of respect for persons [informed consent], beneficence [balance of the risks and benefits], and justice [equitable subject selection]), as well as the creation of the Office for Human Research Protections (OHRP).

Finally, all of the information from the study was published and publicly available. PHS did not hide anything from the public, even though they did lie to the study participants.

To sum up the Tuskegee study:
  • subjects were misinformed about the study purpose, their diagnosis and the availability of treatment
  • subjects were denied proper treatment when it became available, putting themselves and their families at increased risk for no benefit
  • subjects bore an undue burden compared to the ultimate benefit from the knowledge gained
  • none of the study information was covered up or suppressed
Comparing that to the DeStefano study, we'll assume, just for argument's sake, that Hooker and Wakefield are correct about information being suppressed. Study subjects were not misled about the purpose of the study. Being a retrospective case-control study, there was no treatment that could have been given or changed during the course of the study. No subjects involved in the study were put at increased risk, since DeStefano et al. were looking at medical histories, not conducting treatment. Subjects were equitably selected. No preferences were given for one population over another. There were a total of 1,188 controls of white/other/unknown and 636 black. Of those, 46% (552) of white/other were excluded from the subgroup analysis due to lack of a birth certificate, while only 30% (252) of black were excluded. Similarly, from a total of 394 white/other/unknown cases and 230 black, 45% (176) of the white/other were excluded and only 30% (93) of the black were excluded due to lack of a birth certificate. So, other than the purported claim that DeStefano hid information, there are no other similarities between Tuskegee and the 2004 study.

What about information suppression and cover-up? In Tuskegee, information was kept from the study participants, but the data collected were freely available to the public, and the results of the study were published. In DeStefano, the data set has always been available to researchers (in fact, CDC even has instructions on how to access the data on their web site), and the methods DeStefano et al. used and why they were used were disclosed in their paper. In fact, even Hooker's findings of an increased risk of autism among African American males appear to be due to a misapplication of statistics, as noted by the two epidemiologists linked at the beginning of this post.

At the end of the day, Hooker's findings are, at best, questionable, and Andrew Wakefield's comparison to Tuskegee are not only off the mark, but incredibly insensitive to the real suffering and tragedy that was Tuskegee. Anti-vaccine activists have latched onto the perceived, though completely inaccurate, racial disparity in a dishonest attempt to inflame public opinion against the CDC and vaccinations. Fortunately, they have found no mainstream voice in the media to repeat their nonsense. If they do, MMR vaccination rates are likely to plummet, and this year's historically large outbreak will seem like a mere blip.

I'll end with this musing: William Thompson is supposed to have "blown the whistle" and provided information to Brian Hooker that was supposedly covered up. Yet Hooker used the same data set that DeStefano et al. used, so it's unclear exactly what information Thompson provided that was not already available from the CDC. We do not know exactly what Thompson said; all we have are a few carefully edited sound bites from Wakefield.

Addendum: As Dorit notes in the comments below, there is also Wakefield's equating the CDC to Hitler, Stalin and Pol Pot. Actually, he says the CDC is worse than those dictators. His video, including all those awful comparisons, has been lapped up and regurgitated by anti-vaccine activists. Even assuming that Hooker's analysis is correct, what these people are saying is that autism is worse than death, that a slightly increased risk (keep in mind Hooker's confidence interval for African American males vaccinated between 25 and 36 (or 31) months is very broad, with the low end coming awfully close to 1 (i.e., no effect)) of autism is worse than the irrevocable death of their children.
[Edited to Add (Aug. 28, 2014): Here is a timeline through about 1pm today, courtesy of Liz Ditz (thanks for the amazing work on this, Liz!).]

Posts discussing Hooker's allegations, excluding anti-vaccine sources

Posts discussing Hooker's allegations from anti-vaccine sources

Other links

1 comment:

  1. Wakefield's misuse of history also covers his comparison of the researchers to Hitler, Stalin and Pot Pol.


You need a Google, LiveJournal, WordPress, TypePad, AIM or OpenID account to comment. If you need one of those, click on the appropriate link. Spam comments will be deleted. Comments on posts older than 30 days will go into moderation, due to the activities of commercial spammers.