Updated: Vaccine-autism study retracted — again

For the second time, a journal has quickly retracted a study that suggested vaccines raise the risk of autism and other neurodevelopmental disorders.

The study first raised a furor last year, prompting a Frontiers journal to quickly retract it. After it was republished in the Journal of Translational Science this month, that journal has also retracted it.

Although the titles of the two papers changed, the abstracts were nearly identical. Both studies surveyed the parents of 666 home-schooled children, 39% of whom where not vaccinated, and concluded that vaccination increased the risk of neurodevelopmental problems, particularly if children were born prematurely.

A representative of the Journal of Translational Science told usPilot comparative study on the health of vaccinated and unvaccinated 6- to 12-year-old U.S. children” has been retracted, and it will update us with an explanation.

Here’s more from the (now-retracted) abstract:

…in a final adjusted model with interaction, vaccination but not preterm birth remained associated with [neurodevelopmental disorders], while the interaction of preterm birth and vaccination was associated with a 6.6-fold increased odds of NDD (95% CI: 2.8, 15.5)…While vaccination remained significantly associated with NDD after controlling for other factors, preterm birth coupled with vaccination was associated with an apparent synergistic increase in the odds of NDD.

The journal is published by Open Access Text, which was included in the now-defunct list of possible predatory publishers, compiled by librarian Jeffrey Beall.

When the study appeared last year in Frontiers in Public Health, it caused a firestorm on Twitter, prompting Frontiers to release a public statement, noting that the study was only “provisionally accepted but not published.” It was retracted later that same week.

In 2011, first author Anthony Mawson at Jackson State University, filed a lawsuit against the Mississippi State Department of Health, in which he alleged that, after he advocated the need for more studies on vaccine safety, a state officer interfered with his then-position at the University of Mississippi Medical Center, resulting in his contract not getting renewed. The suit was dismissed the following year.

The last author of the study, Binu Jacob, is listed as a former graduate student at Jackson State.

This isn’t the first time an anti-vaccine study was republished after a hasty retraction — last February, Vaccine temporarily removed (then retracted) a study linking the vaccine for human papillomavirus (HPV) to behavioral problems in mice; in July, the paper was republished by the journal Immunologic Research, albeit with major revisions, according to one of the co-authors.

Hat tip: Timothy Caulfield

Update 5/8/17 3:43 p.m. eastern: This story was updated since it was published to reflect the fact the paper had been retracted.

Update 5/18/17 9:26 a.m. eastern: It appears a version of the paper has been reposted online. We’ve asked the journal if the paper remains retracted, and will update if they respond.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

25 thoughts on “Updated: Vaccine-autism study retracted — again”

      1. It is available through google’s cache

        What Richard Telford links to there is a second OAText paper by Mawson — another slice of salami from the same data-set — which seems to have been unpublished as well.

        The Googcache copy of the original paper (“Pilot comparative study”) is here:

  1. The journal is published by Open Access Text, which was included in the now-defunct list of possible predatory publishers, compiled by librarian Jeffrey Beall.

    Jeffrey also devoted a comment thread to the publisher.

    When the study appeared last year in Frontiers in Public Health, it caused a firestorm on Twitter
    You may be exaggerating here. Leonid Schneider tweeted his critique at 4.14 am, November 28th (and again at 4.16 am). The first of his readers responded at 7.46 am., to report that Frontiers had already taken down the Abstract for re-consideration.

    One negative tweet is hardly a firestorm.

    1. Further research reveals that as well as Dr Schneider’s critical tweet on November 28th, there was another one from Tara Smith (of Aetiology blog) the day before. So the firestorm of coordinated protest that forced Frontiers to censor the Abstract consisted of two tweets, not just one. My bad.

  2. Topic aside, what was the technical reason for the retraction of the original report in Frontiers in Public Health? Was there misrepresentation of study methods and procedure? After the first retraction and republication, was the second retraction based on the same technical reason? Or was the second retraction based on “self-plagiarism” as the cause, i.e., using the same text without reporting that it had been published and quickly retracted elsewhere after the “firestorm” of one, noted above by herr doktor bimler? I’m trying to make sense of this after reading the definition of “plagiarism” and “self-plagiarism” in Wikipedia. https://en.wikipedia.org/wiki/Plagiarism.

    1. Topic aside, what was the technical reason for the retraction of the original report in Frontiers in Public Health?

      The Frontiers position (if I recall correctly) was that the manuscript had never been fully accepted, only provisionally accepted, so the management’s decision to overrule the Editor (and the reviewers, and the whole much-vaunted paradigm-breaking Frontiers publication model) was not a retraction. No explanation required. The manuscript simply did not meet the rigorous Frontiers standards after all.

    2. The two papers seem not so much to have been retracted as to have been disappeared – rather in the manner of embarrassing Facebook posts. Normally when journals retract papers they leave the original paper in place, but append a retraction notice to it.

      A reason may never be forthcoming from the publisher, and given their reputation, I am not going to expect one.

    1. aldebert,

      Yes, but how do you obtain the reason from a trustworthy source? Brian Deer gives his opinion without documentation of Frontiers in Public Health’s reason for retracting the report. Other comments point out the predatory nature of the journal, indicating it as a bad place to publish. Yet the identified predatory journal retracted the article for some reason. I want to know whether the reason involved a misrepresentation of study methods and procedure. Or, if the identified predatory journal was reacting to external pressure from some source.

      I’m skeptical about all claims that cannot be verified. As Ronald Reagan advised, “Trust but verify.”

      Can Retraction Watch provide the verified reason for the journal’s retraction?

  3. Introducing these studies, you say that both “surveyed the parents of 666 home-schooled children, 39% of whom were not vaccinated…”

    Call me cynical, but I doubt if either figure is correct.

    This was a project controlled by a man who showered fawning, sycophantic praise on a known research cheat, Wakefield, and was promptly rewarded by anti-vaccine campaigners with money to carry out another of Wakefield’s ludicrous projects.

    The modus operandi, as usual, is to invite parents to offer detailed, retrospective clinical information about their children, in circumstances where – boxloads of other research shows – their recollections are both full of error and thick with motivation.

    It’s the purest junk, and I wouldn’t believe a word of it.

  4. The abstract makes it sound like they lumped preterm birth and vaccination together as the cause, not vaccination alone. This immediately raises questions about the methodology of the study. It also seems to rely on information from the parents of the children, not documented a medical diagnosis from a reputable physician with experience dealing with autism.

    There have been numerous studies on this topic in the years since Wakefield’s initial fraudulent study was released. Every reputable one has found no conclusive link between vaccinations and autism. Also, the claims by Wakefield and other anti-vaccine proponents claimed that it was the mercury-base preservative Thimerosal that was responsible for autism, not the vaccines themselves. The problem with this as it relates to recent claims by the anti-vaccine crowd is that Thimerosal has been almost completely discontinued as a preservative is vaccines. This is because most vaccines are now available in single dose vials that do not need nor have a preservative added. Single dose vaccine vials are used by virtually all doctors in the US for childhood vaccinations. There are still a very limited number of vaccines that are often used in multi-dose vials such as flu vaccines that are administered during seasonal vaccination campaigns. However, these are also available in single dose vials that do not contain Thimerosal and these are what is used by most doctors. The multi-dose vials are more likely to be used by places like clinics and retail pharmacies that have flu vaccinations available. What this means in general is that any study that simply looks at childhood vaccines is not even addressing the issue of Thimerosal which was the culprit that the original anti-vaccine crowd pointed to as the cause of the autism. Even with that, there have been numerous studies that have debunked the supposed link between Thimerosal and autism. Here is a link for the FDA’s current information on the topic of vaccine safety.

  5. Just read the full study (you can find in online if you look hard enough). I am not sure the official reason why it was retracted but here is my opinion ( as a specialist in research). One of the major flaws I saw was that they did not control for number of visits to doctors when comparing the two groups, despite showing in a separate analysis that the vaccinated group had overall more visits (as you might expect). Thus the “increase” in chronic conditions diagnosed in the vaccinated group may simply be due to increased chance of being detected due more visits to doctors. It also was unclear if they controlled for other important factors such as socioeconomic status, etc. That said, the study was likely retracted due to poor quality/bad science and untenable conclusions rather than political reasons. Had I been a peer-reviewer on the paper I would have recommended to reject it.

  6. Just look at the paper and the methodology.
    The authors asked the parents to give them information through the third party.
    The authors accept that it is not a randomized study.
    The authors did not check medical records of patient relying ONLY on parents’ words.
    The authors did not check whether whatever they were told is true or not.
    So, what is the science about it?

  7. “Both studies surveyed the parents of 666 home-schooled children, 39% of whom where not vaccinated….”

    This is not correct. What parents reported regarding vaccinations of their children and what vaccines the children actually had are likely to be different. Parents are not the best sources for remembering such things.

    Note also that the “study” did not bother to check to see if the participating parents even had children.

  8. It’s not clear what the problem might be with this paper, but the multiple huge differences reported would usually be “too good to be true”, making the results improbable though still possible. The suggestion made by other commentators that the principal author was associated with the discredited Wakefield, that the paper has been retracted by two journals, and that no-one has replicated the findings in the many years since the work was done, all point to likely error. If there were really such differences as indicated, one would expect there would have been some similar findings in doi:10.1136/bmj.j3862 (2017). A possible source of bias is reported in https://doi.org/10.1080/21645515.2018.1502520 (2018).

    1. Erica – Not sure if you already saw this, but the last section of this article includes this update “Update 5/18/17 9:26 a.m. eastern: It appears a version of the paper has been reposted online. We’ve asked the journal if the paper remains retracted, and will update if they respond.”

Leave a Reply to aldebert Cancel reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.