The journal Pediatrics recently published another “vaping is a gateway to smoking” article by Benjamin W. Chaffee, Shannon Lea Watkins and Stanton A. Glantz. As previously reported, it was shown to be fatally flawed by Brad Rodu and Nantaporn Plurphanswat. Rodu immediately called for the paper’s retraction in an letter-to-the-editor online comment. Two blog posts by Rodu and a Science Lesson here further explained why the paper was fatally flawed. But rather than seriously consider retracting the paper, as they should have, the editors of Pediatrics attempted a series of coverups.
The story begins when Rodu and Plurphanswat sent their comment exposing the fatal flaw in the Glantz paper on March 15, shortly after the paper became available online. The editors of the journal refused to allow the comment to appear for more than two weeks, until Chaffee et al. submitted their response to it. This fact was hidden by the editors, who eventually posted the letter (available at the current comments page, at least for the moment) with the original submission date, as if they had allowed it to appear then. They also made the table in the comment, the key to the content, almost unreadable, and they refused to fix it when Rodu asked. These actions alone were inappropriate, but not the greatest crimes against science. What followed was much worse.
The Chaffee reply led with an ad hominem attack on Rodu, which included actionable libel (a false assertion that he was hiding financial conflicts of interest). It is important to keep in mind that events demonstrate that the Pediatrics editors were micromanaging the comments on this paper. This means they actively reviewed and agreed to publish these attacks. Chaffee’s only substantive claim was a very awkward and amateurish version of the claim that was debunked in the previous Science Lesson, accusing Rodu of employing a “statistical trick.” Rodu protested the ad hominem and defamation to the editors. This should have resulted in an apology, or at least a published statement retracting those claims. Instead, the editors tried to memory-hole the attacks that they had previously approved.
The editors apparently asked Chaffee to provide a new comment that did not include those attacks. They then simply replaced the old defamatory comment with the new one. Fortunately, thanks to the “Wayback Machine” site, an archival copy of the original defamation is still available. What is worse, the editors swapped out the comment without even changing the date. Thus, they are attempting to claim that the new non-defamatory version of the comment was there all along, and that they never approved the original abuse.
The Daily Vaper presented a detailed list of the facts and inferences that appear in this article to the three Pediatrics editors involved. They were encouraged to dispute any of them they did not believe to be correct. They did not dispute any of them.
During the dispute over the comments, the editors informed Rodu that they were commissioning an independent “expert” review of the dispute. Clearly there was no genuinely expert review, because that would have concluded that Rodu was right and Chaffee’s comment was nonsense. The editors did not post the review to the comments thread, as they should have, and refused a request by The Daily Vaper to see a copy of it. One might conclude that the review never actually existed, but for the changes between the memory-holed version of the Chaffee comment and the replacement version. In the original version, they lashed out and floundered for words, while in the new version they miraculously created something that superficially resembles a scientific argument. Their argument is still wrong, but it the new version gives the illusion of understanding. Presumably the “expert review” was used by the editors to coach Chaffee on how to make his comment seem less floundering.
This suggests that the purpose of “independent review” was to provide Chaffee and Glantz a kind of public defender, someone to step in and replace their agitated and abusive denial with a seemingly rational argument. Of course there is a big difference: Someone accused of a crime is entitled to an attorney even if guilty and should not be punished because they simply do not understand how to navigate the legal system. But “researchers” who cannot offer a prima facie defense of their methods without assistance deserve a retraction for that alone. Editors who simultaneously act as judge and attorney for the defense, and who also cannot understand the methods, have no business pretending they produce a scholarly journal.
Meanwhile, a team led by Clive Bates submitted another comment, endorsing the Rodu analysis and calling for the paper to be retracted. The editors refused to post it, citing some vague ongoing review of the paper. When Bates pointed out that this offered no genuine justification for censoring their comment, the editors relented and posted it. Then, without warning or explanation, they deleted it. (It too is available via Wayback Machine archive.) It is clear that the editors of Pediatrics are trying to scrub the record.
In response to inquiries, the Pediatrics editors offered only one substantive claim. Editor-in-chief Lewis R. First and assistant editor Mark Neuman insisted, without further elaboration, that they were adhering the the Committee on Publication Ethics guidelines for journals. This is roughly equivalent to someone saying “we obeyed all applicable laws” in a press conference: First, it is inevitably not true. The COPE guidelines, like criminal law, is such a sprawling collection of statements that everyone violates some of them every day. More important, despite sounding official and legalistic, this is really meaningless cheap talk. The guidelines are often vague and depend on matters of fact, meaning that even for the violations that are quite apparent, the editors can technically deny them. There are extensive guidelines about transparency, but none of them say specifically that you should not attempt to memory-hole a defamatory comment you approved and put a fake date on the replacement version. If confronted with the standard that they should retract fundamentally misleading articles, they can claim that the inaccuracy of the paper has not been definitively demonstrated. It has, but there is no method for adjudicating matters of fact.
The most obvious explanation for the behavior of the Pediatrics editors is that they are embarrassed that they published a fatally flawed paper, and are trying to hide that fact. But it also reasonable to suspect that they like the flaw. Pediatrics is second among journals only to “Tobacco Control” as a home for anti-tobacco junk science. The average scientific quality of articles about tobacco product use in Pediatrics is actually worse than those in “Tobacco Control.” Thus it is apparent that the editors actively support anti-tobacco junk science. This would suggest that they do not merely want to avoid embarrassment, but that they actively want to promulgate the false conclusions of the paper they published.
The proliferation of anti-vaping junk science, and anti-tobacco junk science more generally, begins with the “researchers” who create it, supported by government funders like the FDA and NIH. But it is journal editors who willingly publish the obvious nonsense. In this case, the nonsense was not obvious; attentive honest researchers had to figure out the hidden error the authors made (probably intentionally). But instead of responding properly to this revelation, the editors not only redoubled their commitment to having published it, but went to great lengths to try to hide the problem from their readers. Once again, we have evidence that the phrase “according to a peer-reviewed public health article” means little more than “I read this on the internet.”
[Disclosure: The author was in ongoing communication about this matter with Rodu and Bates, and offered some suggestions regarding their statistical analysis and communications with the Pediatrics editors.]