I was reluctant to write any more about The BMJ and its descent into bad journalism, one of its editors amplifying antivaccine misinformation, and its publishing of outright conspiracy theories by a hack journalist, but unfortunately its editors leave me little choice. It started again when readers started emailing me links to an Open letter from The BMJ to Mark Zuckerberg, published as a Rapid Response after the utter crapfest of a “news report” by hack journalist and anti-GMO conspiracy theorist Paul Thacker. The story claimed that Ventavia Research Group, a contract research organization (CRO) hired by Pfizer to run three of its sites in Texas for its original phase 3 clinical trials of its then-experimental mRNA-based COVID-19 vaccine had all sorts of problems that, or so Thacker insinuated, called into doubt the clinical trial data behind the safety and efficacy of Pfizer’s vaccine that were used to apply for its emergency use authorization (EUA). Thacker’s main source was a “whistleblower” named Brook Jackson, who had worked at Ventavia for only two weeks. As I described, the allegations were either big nothingburgers that wouldn’t have affected the quality of the data (e.g., not appropriately using sharps containers to dispose of sharps) or were mainly insinuated and implied without actual evidence (e.g., unblinding or even falsifying clinical trial data). The article follows a familiar format for disinformation. Very definitive and serious accusations are leveled very early in the article, followed much later in the article by “facts” that do not actually substantiate such definitive and serious allegations.
I realize that Paul Thacker recently won a Special Journalism Award at Press Gazette’s tenth British Journalism Awards for three of his previous BMJ reports (but not for this one, I note), which suggest to me that the state of British journalism is dire indeed. Personally, I can only might recall Thacker as the conspiracy-mongering freelance “journalist” with a penchant for harassing scientists with frivolous Freedom of Information Act (FOIA) requests and amplifying antivaccine messages himself. I deconstructed in my usual detail everything that I found wrong in Thacker’s news story (and there was a lot), as did a number of others, some of whom I cited, some of whom I didn’t. (David Grimes and I also wrote a Rapid Response that the editors of The BMJ did not see fit to publish, which I will include later in this post, given that I didn’t include it in my last one.) Basically, current editor-in-chief Fiona Godlee and incoming editor Kamran Abbasi, current executive editor who will replace Godlee after she steps down at the end of the year, are very, very indignant about Facebook’s use of a fact checking source that—correctly, I might add—declared Thacker’s article to be misinformation.
Godlee and Abbasi complain about this early in their open letter:
The BMJ commissioned an investigative reporter to write up the story for our journal. The article was published on 2 November, following legal review, external peer review and subject to The BMJ’s usual high level editorial oversight and review.
But from November 10, readers began reporting a variety of problems when trying to share our article. Some reported being unable to share it. Many others reported having their posts flagged with a warning about “Missing context … Independent fact-checkers say this information could mislead people.” Those trying to post the article were informed by Facebook that people who repeatedly share “false information” might have their posts moved lower in Facebook’s News Feed. Group administrators where the article was shared received messages from Facebook informing them that such posts were “partly false.”
Readers were directed to a “fact check” performed by a Facebook contractor named Lead Stories.
Of course, given his long history of promoting pseudoscience and conspiracy theories, commissioning Thacker to do an article on anything related to pharma, GMOs, or vaccines was The BMJ‘s first mistake. They might as well have commissioned Andrew Wakefield to do an investigative report about the MMR vaccine and whether it causes autism. Seeing Thacker’s hack journalism in the service of a conspiracy mongering article long on insinuation and accusations but short on actual evidence to back up those accusations in the same journal that a decade ago had published Brian Deer’s excellent investigative journalism blowing the lid off of Wakefield’s MMR fraud, even going so far as to call it “Piltdown medicine” and vigorously defending against Wakefield’s libel suit over the article, was truly depressing. I even gave Godlee props once for how she handled a particularly annoying antivaxxer questioning her about The BMJ‘s reporting on Wakefield.
That history aside, apparently The BMJ‘s “usual high level editorial oversight and review” is nowhere near as high as it once was, say, eleven years ago. Let’s just put it this way. If you commission an “investigative report” by Paul Thacker about Pfizer (or any pharmaceutical company, for that matter), the results will be as preordained as they would be if you commissioned an “investigative report” by Andrew Wakefield about the MMR vaccine. If Godlee didn’t know that that’s what she would get when The BMJ hired Thacker, then she’s painfully naïve beyond belief. If she did know, then it’s hard to conclude anything other than that she was purposely putting The BMJ on the side of spreading antivaccine conspiracy theories. I don’t know what Abbasi’s role in all of this was, but the fact that he would co-sign an open letter with Godlee complaining to Facebook about its labeling of Thacker’s article as lacking context tells me that he’s part of the problem. Of course, neither of them have found enough reason to fire Peter Doshi, even after he appeared to testify at roundtable hosted by COVID-19 minimizer and antivaxxer Senator Ron Johnson, at which he spewed misinformation about COVID-19 and COVID-19 vaccines, as touted by antivax activist Robert F. Kennedy, Jr.”
Again, I deconstructed all that was wrong with Doshi’s testimony in my original article. Let’s just say that it was bad—really bad. Abbasi, you might recall, had publicly defended Doshi on Twitter with the weakest of weak defenses, “he didn’t represent The BMJ“:
Before I address Godlee and Abbasi’s embarrassing letter responding to the fact checking of Thacker’s article, which The BMJ had apparently commissioned, let’s take a look at the article that Facebook used to reach its conclusion, a Lead Stories article by Dean Miller entitled The British Medical Journal Did NOT Reveal Disqualifying And Ignored Reports Of Flaws In Pfizer COVID-19 Vaccine Trials. (Oddly enough, Abbasi and Godlee did not mention a similar deconstruction published in MedPage Today that listed the same problems with Thacker’s report (and that was also cited by Dean in his Lead Stories article), citing experts describing Thacker’s allegations as a “vague kind of hand waving”). It also cited examples of Brook Jackson retweeting and agreeing with antivaccine misinformation.
So let’s look at Godlee and Abbasi’s criticisms of the Lead Stories article, which lead them at the end to a hilariously off-base demand for a “correction”:
We find the “fact check” performed by Lead Stories to be inaccurate, incompetent and irresponsible.
— It fails to provide any assertions of fact that The BMJ article got wrong
— It has a nonsensical title: “Fact Check: The British Medical Journal Did NOT Reveal Disqualifying And Ignored Reports Of Flaws In Pfizer COVID-19 Vaccine Trials”
— The first paragraph inaccurately labels The BMJ a “news blog”
— It contains a screenshot of our article with a stamp over it stating “Flaws Reviewed,” despite the Lead Stories article not identifying anything false or untrue in The BMJ article
— It published the story on its website under a URL that contains the phrase “hoax-alert”
We have contacted Lead Stories, but they refuse to change anything about their article or actions that have led to Facebook flagging our article.
We have also contacted Facebook directly, requesting immediate removal of the “fact checking” label and any link to the Lead Stories article, thereby allowing our readers to freely share the article on your platform.
The first complaint is, while perhaps not completely false, definitely misleading. The main problems withThe BMJ story by Thacker was not the falsity of any specific claim in it, but rather how various “facts” were either poorly supported by evidence and/or woven together in a deceptive manner to suggest nefarious intent and ends where there were almost certainly none. For example, his report states that Jackson reported data falsification by Ventavia, but Thacker could only support that allegation with one throwaway line in this passage:
Documents show that problems had been going on for weeks. In a list of “action items” circulated among Ventavia leaders in early August 2020, shortly after the trial began and before Jackson’s hiring, a Ventavia executive identified three site staff members with whom to “Go over e-diary issue/falsifying data, etc.” One of them was “verbally counseled for changing data and not noting late entry,” a note indicates.
Let me cite what I wrote about this passage a month and a half ago:
I also note that, although Thacker claims that Ventavia “falsified data” in the very introduction of his report, this is the only mention of “falsifying data” that occurs anywhere else. There is no description of any specific incidents of “falsifying data” listed, only a memo mentioning three staff members, one of whom was apparently counseled about “changing data and not noting late entry.” One also notes that this memo comes from early August 2020, shortly after the trial began, which means that Ventavia intervened before that many patients could have been signed up. This is truly thin gruel to make sensationalistic claims about “falsifying data” in the introduction in such a way as to make it sound as though the practice was widespread.
Basically, we don’t know if falsification occurred, how often it occurred, or what the specific “falsifications” were. Indeed, the phrasing of the memo could have indicated that the changing of data, which could have been the correction of errors, and not noting a late entry could have been discussed in the context of how even minor violations like that without the appropriate audit trail could be interpreted as “falsifying data.” Again, we just don’t know, and Thacker doesn’t tell because he doesn’t have evidence. (If he did, you know he would have cited more than just a cherry picked quote from a memo, which is very thin gruel indeed to base a charge of “data falsification” on.)
Thacker has since tried—and failed—to bolster his allegation of data falsification by Ventavia on his Substack, and I’ll get to that before the end. First, I want to deal with the rest of Godlee and Abbasi’s complaints.
The second complaint is downright cringeworthy. The title of Miller’s article was not “nonsensical.” It said exactly what the article was about. True, it’s an arguably annoyingly click-baity title, but it’s not “nonsensical.” I’ll (sort of) give them the third complaint, that The BMJ is not a “news blog.” I say “sort of” because Lead Stories has responded, and one of its responses was:
BMJ.com objects to our use of the term “news blog.” Without a hard copy in hand, we can only responsibly work from the BMJ.com post. It is one item in a regularly updated feed of articles with the newest items at the top: a news blog, subject to the kind of after-the-fact updating not possible in a printed journal.
This one’s not worth fighting over, obviously, and it’s all a red herring anyway, a distraction on the part of Godlee and Abbasi, who take a relatively inconsequential “gotcha!” and try to use it to discredit more substantive criticisms of Thacker’s article.
The fourth response about “flaws reviewed” made me want to write Godlee and tell her that The BMJ owes me for keyboard for my new 14″ M1 MacBook Pro laptop, because I spit up the water I was drinking when I read it. Again, it’s a obvious and cringeworthy bit of misdirection that leads me to reluctantly question Godlee’s and Abbasi’s motivations. I mean, seriously, the most effective disinformation is disinformation that weaves a deceptive tale using verifiable facts. The problems with Thacker’s reporting weren’t that it reported things that were necessarily demonstrably false, but rather how Thacker wove his findings and a lot of other “facts” that were very weakly supported into a deceptive narrative suggesting that his dubious findings from three sites in the Pfizer clinical trial were generalizable to the entire clinical trial of tens of thousands of subjects. Again, one example is how Thacker outright stated as fact early in his report that data had been falsified, but the only evidence that he could muster to support that allegation was a single line from an “action plan” provided to him by his “whistleblower” Brook Jackson. He never once demonstrated that any data were actually “falsified”!
Let me just put it this way to Ms. Godlee and Mr. Abbasi. I can point out a number of articles at the conspiracy mongering antivaccine and health disinformation site Natural News that contain, strictly speaking, zero factual errors but are nonetheless completely misleading. How is it that they do not understand this, and instead retreat to haughty indignance over how a site like Lead Stories would dare question Thacker’s report without demonstrating a single error in fact? Retreating to the defense that there were no “factual errors” to defend a report like Thacker’s is the same defense someone like Mike Adams would use about one of his conspiracy articles.
And, indeed, in their response Lead Stories points out in a response that would make Orac proud:
“Missing Context” is the most accurate rating available to us in the Facebook Fact-Checking rubric: The concerns Jackson raised affect three of 153 sites, while BMJ.com’s headline makes no effort to convey how small the subset is; at no point in the article are Pfizer, Ventavia and the FDA granted the opportunity to respond to accusations of mis-, mal- and non-feasance; the whistleblower’s public statements about vaccines give vital context to her actions; and while Jackson was hired for her auditing expertise, her publicly expressed views of COVID vaccines, public health efforts and of vaccination of children are at least noteworthy. BMJ prides itself on accompanying each article with a statement of conflicts of interest. Perhaps its high-level editing could have noted their key source’s advocacy.
And that’s exactly it, the same thing that I noted when I discussed how bad Thacker’s article was. It doesn’t put its allegations into context, because if it did the allegations would then suddenly become a lot less worrisome. Basically, Lead Stories did review the flaws in Thacker’s article, it’s just that Godlee and Abbasi seem to be misrepresenting the flaws that were actually reviewed (lack of context and deceptive narrative) as meaning clear, undeniable factual errors. It’s also amusing that Godlee and Abbasi don’t know that the label placed on Thacker’s article does not inhibit the sharing of the article on Facebook at all. Anyone can share it; it’s just that the “misleading” label will be added to it—and, appropriately so, I might add. This is one of the times when Facebook actually got misinformation screening largely correct.
One can’t help but wonder if, to Godlee and Abbasi, keeping Peter Doshi on board and publishing “investigative journalism” by a hack like Thacker is all about the clicks:
While I’m at it, Grimes included our submitted Rapid Response to The BMJ that was rejected outright for publication, even as multiple Rapid Responses from antivaxxers associated with the blog Age of Autism were accepted:
Godlee and Abbasi’s final point is perhaps the most embarrassing, namely that presumably reliable sources of information are sometimes incorrectly flagged as misinformation:
There is also a wider concern that we wish to raise. We are aware that The BMJ is not the only high quality information provider to have been affected by the incompetence of Meta’s fact checking regime. To give one other example, we would highlight the treatment by Instagram (also owned by Meta) of Cochrane, the international provider of high quality systematic reviews of the medical evidence. Rather than investing a proportion of Meta’s substantial profits to help ensure the accuracy of medical information shared through social media, you have apparently delegated responsibility to people incompetent in carrying out this crucial task. Fact checking has been a staple of good journalism for decades. What has happened in this instance should be of concern to anyone who values and relies on sources such as The BMJ.
The example pointed to is the Cochrane Collaboration, which Tweeted last month:
Citing this example, as bad as it is, is misdirection, too. Basically, Godlee and Abbasi’s last argument seems to boil down to something like this: Social media fact checkers and algorithms sometimes mistakenly flag reliable sources as unreliable. Therefore, the same is true of the fact-check of Thacker’s article by Lead Stories. This is what we in the biz call a non sequitur. It does not follow from the observation how sometimes social media companies mischaracterize reliable sources as unreliable that in this instance they did the same. You have to show that they did for this article. Godlee and Abbasi most assuredly failed to do this.
Indeed, Lead Stories gets it right again:
We would like to point out that Lead Stories is not involved in enforcing said Community Guidelines and that we agree this content probably should not be blocked. We actually alerted our contact at Meta about this issue on November 10, 2021 and hope this can be resolved soon.
That is exactly how Lead Stories should have responded to this red herring.
I’ll conclude by looking at whether any new information has come to light to support Thacker’s allegations and make Godlee and Abbasi’s letter any less cringeworthy. Spoiler alert: The answer is no, although Paul Thacker did release additional documents on his Substack three weeks ago and then later as part of an attack on David Grimes. In his first article, Thacker includes a cease-and-desist letter from Brook Jackson’s attorneys to Ventavia over its statement to MedPage Today that she wasn’t involved in the clinical trials. I won’t adjudicate that particular disagreement other than to consider that I doubt a company like Ventavia would have made such a statement without having its legal department look at it first. Whatever the “truth” is; my point is more that there’s a whole lot more nothing in Thacker’s article. Nothing he adds in any way bolsters his case.
For example, he includes this as an example of the “chaos” going on at Ventavia during the clinical trial:
This appears to be cut from an email included here:
Thacker also just outright recycled another email quote that he had included in his report:
For context, here’s the whole email:
Olivia, myself and the leadership team made a decision this morning to pause our enrollment on the COVID C4591001 trial. Apparently, word travels fast in this company…and with that comes a lot of rumors. So let me help clear this up. We have been enrolling fast and furious for 7-8 weeks now. Though we have participated in high-enrolling studies before, this one had it’s own set of challenges from the beginning. Then to top it off…the enrollment period was extended for longer than we’d anticipated and that is HARD.
We realize that the more we enroll, the more chaotic and messy things get. We decided to pause our enrollment so that each of our COVID sites can not only catch up, but also make sure they are in complete compliance. I know I’ve said this several times, but I’ll say it again here…it’s not a matter of IF the FDA is coming…it’s a matter of WHEN the FDA is coming. This is the biggest clinical trial in the world…and we are atop enroller And they are coming soon. THE FDA IS COMING…SOON (like in a matter of days, if I had to make a guess).
I don’t say all of this to scare you. I say this so that you can make sure you are prepared. And this is why we have decided to pause our enrollment. We need to do our own internal audit (as a team) and make sure we are prepared.
We will be having an FDA training in the next few days (in preparation), so be on the lookout for an invite…it will be mandatory. But until then, do not forget…if they knock on your sites door…the first thing you ask for proof of who they are (they will have the documentation you need to see) and immediately notify your manager and the manager will then notify the RD (and so on). SOMs…make sure anyone who answers our doors/sits at the front desks, know to immediately ask for identification if someone says they’re from the FDA.
Let me know if you have any questions (don’t cc all, unless necessary…I don’t want the emails to get out of control with this)
The picture I get is of management trying to make sure all the i’s are dotted and t’s crossed before the FDA inspects. None of this shows that the trial was compromised or that any “data falsification” had occurred.
Then there was this email from Jackson:
Again, none of this sounds particularly nefarious, more like efforts to correct flaws in the system. Moreover, I’m not sure why Thacker thinks it’s a bad thing that Ventavia considered temporarily halting enrollment in order to make sure it had caught up on documentation and fixed problems in its systems, but he does. In any event, I perused the documents included in Thacker’s document cloud and left very much thinking that nothing in there supports his allegation of data falsification, although I always leave open the possibility that I could be wrong if anyone wants to show me where in this long email chain linked to by Thacker this is documented. I see a lot of discussion of roles of various personnel in the trial, a $500 stipend for study personnel, and questions about an unblinded CRA, and the only thing I could find about data falsification is a note that is not damning:
Go over ediary issue/falsifying data, etc. DONE-Thea also verbally counseled for changing data and not noting late entry.
If you’ve ever been involved in a clinical trial, you’ll recognize that someone likely made late entries or corrections to records but did not appropriately note that this was a late entry, which the FDA can sometimes view as data alteration. It happens occasionally, and it sounds as though it was corrected here. Truly, as in Thacker’s original article, this one little entry does a lot of work that it can’t really do. It sounds like something that happened with one employee, who was counseled, and didn’t happen again. In the scheme of things, it sounds very minor and certainly doesn’t support the impression of mass data falsification that Thacker insinuates in his article. Similarly, the issue of unblinding was never really well-supported either. I discussed that in detail in my original post; nothing here changes my assessment. Basically, there’s nothing new in Thacker’s big document dump. No wonder he never quoted more in his original BMJ article!
It’s sad to see what’s happened to The BMJ. It hired someone who is, if not outright antivax, at least antivax-adjacent who has continually spread misinformation about vaccines dating back to the H1N1 pandemic and continues to do so with COVID-19 vaccines while commissioning an awful conspiracy mongering bit of “journalism” by a known hack. Then, when criticized for it, instead of seriously looking at where they might have gone wrong, Godlee and Abbasi instead double down and throw out red herrings to try to discredit the criticism. It’s a sad way for Fiona Godlee to leave the journal, and, if this open letter to Mark Zuckerberg is any indication, I’m not reassured that things will get any better under Kamran Abbasi’s leadership.