The dumpster diving VAERS preprint debacle: How did we get here?

Last week, I wrote about a study published as a preprint that I referred to as a “dumpster dive” into the Vaccine Adverse Events Reporting System (VAERS) database. Adding to the criticism were two guest bloggers, Dr. Dan Freedman and pediatric cardiologist Dr. Frank Han, as well as the newest regular blogger, Dr. Jonathan Howard. Given the copious coverage we’ve provided to this one bad preprint, you might wonder, what’s left to say? Quite a bit unfortunately, albeit not so much about the science anymore, given that, between Twitter and SBM bloggers and guest bloggers, the study has been thoroughly deconstructed, with lengthy explanations of why the study’s design was so poor and its conclusions not justified by its data and analysis. Rather, I’ll be discussing what’s been revealed since this preprint was published and the reaction of those supporting its authors, not to mention that one of the authors, Josh Stevenson, had an undisclosed conflict of interest in that he is associated with a COVID-19 minimizing group opposing masks, lockdowns, and vaccine mandates.

There’s a regular segment on The Amber Ruffin Show called “How Did We Get Here?” Now seems as good a time as any for a postmortem on this incident thus far that asks, “How did we get here?” So how did we get here, where normally-respectable physicians teamed up with a COVID-minimizing crank to dumpster dive in the VAERS database and produce a study that seemed almost intended to spread fear of COVID-19 vaccines and discourage mandates for adolescents, particularly adolescent boys?

How we got here, a recap

Before I look into the fallout that helped reveal how we got here, a brief recap is likely required for most readers, who can click on the links above and in this section if they desire more detailed information. Let me just say now that what surprised me more than anything else about the preprint VAERS study is that doctors with training in epidemiology could misunderstand VAERS so badly that they unwittingly carried out the same sort of awful, incompetent VAERS analysis that antivaxxers have been doing for a couple of decades, all in order to produce evidence that suggests that the risk of myocarditis in adolescents due to COVID-19 vaccines makes the risk/benefit equation of the vaccine unfavorable in this age group. VAERS, of course, is a 30-year-old database to which anyone can literally report anything that might be an adverse event after vaccination.

As I like to say, the greatest strength of VAERS is that anyone can report anything to it, but simultaneously the greatest weakness of VAERS is that anyone can report anything to it. This open nature makes VAERS useful as an early reporting system, a “canary in the coal mine”, if you will, that can find safety signals if there are a large number of reports of a particular adverse event after a given vaccine, but to determine if such a safety signal indicates actual correlation that might indicate causation requires other, more rigorously maintained, active surveillance systems such as the Vaccine Safety Datalink (VSD). That’s why, at best, VAERS can be hypothesis generating, not hypothesis testing, and it is not suitable for estimating the incidence of an adverse event after vaccination. Indeed, as I said last time, were I a peer reviewer of that study, I’d have been tempted to say that unadjudicated VAERS reports (CDC officials try to verify the reports to determine if they are real and to find more medical information before using them in any analysis) cannot, by the very nature of the VAERS database, be used to do the sort of analysis that its authors (Tracy Beth Høeg, John Mandrola, Allison Krug, and Josh Stevenson) were trying to do, much less to compare the incidence of myocarditis after COVID-19 vaccination to hospitalizations due to COVID-19 itself, and that any attempt to use it that way is folly and incompetent science of the worst sort.

Indeed, the open nature of VAERS is why antivaxxers love VAERS so much. Not only can trial lawyers and antivaxxers seeking to sue vaccine manufacturers or to bring claims before the Vaccine Court game VAERS by encouraging reports of “vaccine-induced autism”, but, as a passive reporting system, VAERS has certain biases, including, paradoxically, both underreporting and overreporting of cases depending on the specific adverse event. Longtime regular readers of SBM and my not-so-secret other blog know that what I like to call “dumpster diving” into VAERS is a common technique used by antivaxxers to try to find spurious associations between autism, sudden infant death syndrome, autoimmune diseases, premature ovarian failure and infertility, and basically any other adverse outcome that antivaxxers want to attribute to vaccines. Indeed, the first time I ever encountered this technique was in 2005, and, as I—and others familiar with how antivaxxers have misused VAERS before had been warning before any COVID-19 emergency use authorization (EUA)—had been warning, antivaxxers continue to use VAERS to given the impression that COVID-19 vaccines are causing a “holocaust” of death and depopulation and numerous other harms. Again, the fact that Høeg et al either didn’t know or didn’t care about the limitations of VAERS and how unsuitable it is for the sort of analysis that they carried out, coupled with their inclusion of a COVID-19 minimizing crank, is very troubling.

How we got here, part two: Josh Stevenson

Regular readers might recall that I had no idea who Josh Stevenson was when I wrote my deconstruction of the dumpster diving preprint. Later in the day, after my post had gone live, I added an addendum including a bit of what I had learned about Josh Stevenson, namely that he had listed only Truth in Data, LLC as his affiliation for the preprint and that there was a huge omission, namely that he was a major contributor to Rational Ground, a COVID-19 minimizing website opposed to most public health interventions to mitigate COVID-19, where he is described as:

Josh is a data visualization expert who focuses on creating easy to understand charts and dashboards with data. He is based out of Nashville, where he provides analysis to support local advocacy groups for in-person learning and other rational, data-driven covid policies. His background is in computer systems engineering & consulting, and his Bachelor’s degree is in Audio Engineering.

Yes, this is exactly the sort of expertise one would want to include in a team analyzing unadjudicated VAERS reports, right?

Meanwhile a coauthor of mine, Duke University Professor of Global Health & Public Policy Gavin Yamey pointed out to me:

How did we get here? Why was Josh Stevenson an author on this preprint? What is Rational Ground, anyway? And how is Rational Ground’s grift so expensive?

It didn’t take much of a perusal of Rational Ground to find a lot of articles railing about “lockdowns” and opposing mask mandates. Of course, there are rational risk-benefit analyses and economic considerations about the use of “lockdowns” to slow the spread of COVID-19, but Rational Ground is clearly not about that. Rather, it’s about fear mongering. Indeed, sampling of the articles opposing “lockdowns” (which, at least in western democracies are far from complete “lockdowns”) blames them for a veritable panoply of evils, more tuberculosis, deaths from cancer, heart disease, suicides, starvation, childhood deaths, more cases of HIV, child abuse, domestic/sexual abuse, opioid deaths, and much, much more. It’s not just “lockdowns,” either. Basically, Rational Ground paints an entirely dire picture of any sort of public health intervention to slow the spread of COVID-19, particularly mask mandates, for which they have an entire page full of articles opposing masks, falsely claiming that masks don’t have any effect on the spread of COVID-19, and a whole lot of other misinformation about COVID-19, ranging from semi-responsible to pure denialist.

Unsurprisingly, Rational Ground enthusiastically endorsed the Great Barrington Declaration not long after it was published. The Great Barrington Declaration, as you might recall, was a document published nearly a year ago (two months before there was even a vaccine as the Pfizer vaccine wasn’t issued an emergency use approval until two months later!) by a trio of scientists, Dr. Sunetra Gupta of the University of Oxford, Dr. Martin Kulldorff of Harvard and Dr. Jay Bhattacharya of Stanford, that called for the United States and the United Kingdom to end their lockdowns and promotes allowing the virus to spread among young people in order to build herd immunity. At the time, I described it as eugenics, given that its authors advocated “focused protection” of those “vulnerable” to severe disease and death due to COVID-19 (the elderly, those with chronic health conditions that predispose to severe disease, etc.), ignoring that it is impossible to “focus” protection of these people from an infectious disease if that disease is spreading unchecked through the very population with whom they live and who take care of them. In addition, the Great Barrington Declaration was instigated by a right wing libertarian think tank, the American Institute of Economic Research (AIER), which widely promoted it, while its signatories were able to appear on conservative media to promote their message gain access to the highest levels of government in the UK and US, with one of them prominently defending Florida Governor Rick DeSantis and his opposition to mask and vaccine mandates:

This is not the first time billionaires aligned with industry have funded proponents of “herd immunity.” Gupta, along with Harvard University’s Martin Kulldorff and Stanford University’s Jay Bhattacharya, wrote the Great Barrington Declaration (GBD), which, in essence, argues that covid-19 should be allowed to spread unchecked through the young and healthy, while keeping those at high risk safe through “focused protection,” which is never clearly defined. This declaration was sponsored by the American Institute for Economic Research (AIER), a libertarian, climate-denialist, free market think tank that receives “a large bulk of its funding from its own investment activities, not least in fossil fuels, energy utilities, tobacco, technology and consumer goods.” The AIER’s American Investment Services Inc. runs a private fund that is valued at $284,492,000, with holdings in a wide range of fossil fuel companies (e.g. Chevron, ExxonMobil) and in the tobacco giant Philip Morris International. The AIER is also part of “a network of organizations funded by Charles Koch—a right-wing billionaire known for promoting climate change denial and opposing regulations on business” and who opposes public health measures to curb the spread of covid-19.

And:

In October 2020, Gupta, Kulldorff, and Bhattacharya met with two of US President Donald Trump’s senior health officials, Health and Human Services Secretary Alex Azar and Scott Atlas. Atlas was at the time on leave from his fellowship at the Hoover Institution, a conservative think tank affiliated with Stanford University. The meeting reportedly led the administration to eagerly embrace the GBD. Nor did the GBD authors limit their efforts to national governments. For example, in March 2021 Florida Governor Ron DeSantis hosted a video roundtable with Atlas, Gupta, Kulldorff, and Bhattacharya, where they expressed opposition to masks, testing and tracing, physical distancing, and mass vaccination. YouTube removed the video “because it included content that contradicts the consensus of local and global health authorities regarding the efficacy of masks to prevent the spread of Covid-19.” GBD authors, predictably, cried, “Censorship!” Bhattacharya continues to advise Governor DeSantis on Florida’s covid-19 policies, including providing legal testimony in support of DeSantis’s ban on mask mandates in public schools.

True, Rational Ground is a relatively small fish in a big pond of groups promoting COVID-19 misinformation, but it’s a nasty one. In particular, Justin Hart is a particularly vile conspiracy theorist spreading the silliest antimask disinformation:

As well as the “casedemic” conspiracy theory and election fraud conspiracy theories:

Naturally, when it was pointed out that Josh Stevenson was a coauthor of the preprint and those of us who had been unfamiliar with his activism, a lot of us wondered: How did we get here?

So people asked the lead author, Dr. Høeg, who showed up in the comments of Dr. Freedman’s guest post to deny any knowledge of his association:

Even more interestingly, before the publication of the preprint, Mr. Stevenson appeared to have gone through an intentional effort to hide his association with Rational Ground, at least on his social media profiles (he was still listed on the website):

All of this left me wondering: Were the other three coauthors at least sympathetic to the aims of Rational Ground, or were they “useful idiots” for the antimask crowd? I must admit that it strained credulity (at least mine) that Dr. Høeg did not know Josh Stevenson’s background. Apparently, they go back at least to February, but did Dr. Høeg truly not know about Stevenson’s activism minimizing COVID-19 and opposing masks and public health mandates? Take a look:

You be the judge! Maybe Stevenson really did hide his activism. Maybe Dr. Høeg and her coauthors were willfully blind. I don’t know. I just know that it looks…bad, particularly given how there seemed to be a network of media and social media outlets poised to publicize it as evidence that we should not vaccinate, for example:

Meanwhile, COVID-19 minimizer and antivaxxer Rep. Marjorie Taylor Greene Tweeted out:

As did “inventer of mRNA vaccines” Robert Malone:

Interesting indeed.

How we got here, part three: The Medscape report

About a week after the preprint was published, Medscape published a news report about it that was most…revealing. It was by Alicia Ault and titled “COVID Vaccine Preprint Study Prompts Twitter Outrage“. I will admit that I very much hated the headline, as it simply made the incident sound as though it was just another Twitter kerfuffle, but fortunately the article was better and revealed some interesting things. First of all, Drs. Mandrola and Høeg appeared to be backtracking a bit:

Two of the authors, John Mandrola, MD, a cardiac electrophysiologist who is also a columnist for Medscape, and Tracy Beth Hoeg, MD, PhD, an epidemiologist and sports medicine specialist, told Medscape Medical Newsthat their estimates are not definitive, owing to the nature of the VAERS database.

“I want to emphasize that our signal is hypothesis-generating,” said Mandrola. “There’s obviously more research that needs to be done,” he said.

“I don’t think it should be used to establish a for-certain rate,” said Hoeg, about the study. “It’s not a perfect way of establishing what the rate of cardiac adverse events was, but it gives you an estimate, and generally with VAERS, it’s a significant underestimate.”

That last part, of course, is utter nonsense and simply reinforces my impression that none of the authors truly understand VAERS. Yes, it is true that VAERS is a passive reporting system, meaning that it relies on doctors, healthcare workers, and relatives to submit reports, which can lead to underreporting. Certainly, for minor adverse events, the underreporting is likely to be significant. After all, who’s likely to bother reporting, for example, a sore arm, brief fever and chills, or a headache to VAERS? Not very many. In contrast, though, severe adverse events, such as death (or myocarditis, particularly myocarditis requiring hospitalization) are far more likely to be reported—or even overreported, which is why the VAERS team at the CDC doesn’t use unadjudicated VAERS data to estimate the incidence of adverse events after vaccination. They try to adjudicate the data by contacting the person filing the report to verify whether the incident actually happened and trying to obtain clinical data. Then they look at other databases more suitable to the task of determining correlation, frequency, and potentially inferring causation, like the Vaccine Safety Datalink (VSD). Worse, as we documented on SBM and others did on Twitter, there were a fair number of reports used in the preprint analysis that were…sketchy…at best.

Now here’s what really got my skeptical antennae twitching:

Both Hoeg and Mandrola said their analysis showed enough of a signal that it warranted a rush to publish. “We felt that it was super time-sensitive,” Mandrola said.

“Super time-sensitive”? Why? Why was it “super time-sensitive” that Høeg et al publish their results, some of which we were discussing nearly three months ago in balancing the risks of vaccination against the risk of the virus? Particularly curious is that a common defense of this preprint from its authors and others is that the frequency of myocarditis after vaccination of adolescents was within the range of published results, which is true (albeit at the high end of the range). So, again, what was the rush?

According to Medscape:

Hoeg told Medscape Medical News that the paper went through peer-review at three journals, but was rejected by all three, for reasons that were not made clear.

She and the other authors incorporated the reviewers’ feedback at each turn and included all of their suggestions in the paper that was ultimately uploaded to medRxiv, said Hoeg.

They decided to put it out as a preprint after the US Food and Drug Administration (FDA) issued its data and then a warning on June 25 about myocarditis with use of the Pfizer vaccine in children 12 to 15 years of age.

That was over two months before the preprint was submitted. The authors’ story is just not lining up. Again, what was the rush? The Medscape story wonders, too:

Harlan Krumholz, MD, SM, the Harold H. Hines, Jr. professor of medicine and public health at Yale University, New Haven, Connecticut, which oversees medRxiv, tweeted, “Do you get that the discussion about the preprint is exactly the purpose of #preprints. So that way when someone claims something, you can look at the source and experts can comment.”

But Ziaeian tweeted back, “Preprints like this one can be weaponized to stir anti-vaccine lies and damage public health.”

That was Boback Ziaeian, MD, PhD, assistant professor of medicine at the David Geffen School of Medicine at UCLA, Los Angeles. Krumholz replied:

It is true that anything can be (and is being) weaponized, but that doesn’t mean that outlets like medRxiv should make it so easy. From my perspective, Krumholz was being either disingenuous or utterly oblivious, take your pick.

The Medscape article then finished with:

Both Mandrola and Hoeg said they welcomed critiques, but that they felt blindsided by the vehemence of some of the Twitter debate.

“Some of the vitriol was surprising,” Mandrola said. “I kind of have this naïve notion that people would assume that we’re not bad people,” he added.

However, Mandrola is known on Twitter for sometimes being highly critical of other researchers’ work, referring to some studies as “howlers,” and has in the past called out others for citing those papers.

Hoeg said she found critiques about weaknesses in the methods to be helpful. But she said many tweets were “attacking us as people, or not really attacking anything about our study, but just attacking the finding,” which does not help anyone “figure out what we should do about the safety signal or how we can research it further.”

The Tweet:

My goodness, Dr. Mandrola sounded almost like…me, given that I characterize his preprint as a “howler” of a study.

This brings us to the lamentations of certain doctors in the same ideological tent with respect to COVID-19 of how very, very mean critics have supposedly been. Indeed, I found it very interesting how I was so thoroughly criticized for supposedly relying on ad hominems in my post last week, when in fact the original version of the post barely mentioned the authors, other than to lament that they should know better than to have done such a crappy analysis. In fact, I bent over backwards not to engage in ad hominem attacks. Then I learned that Josh Stevenson was a coauthor of the paper, was informed who he was, and then had to add more about him and start asking the other coauthors why they would team up with someone like him.

Enter Dr. Vinay Prasad.

How we got here, part four: Dr. Vinay Prasad and “medical tribalism”

As I noted a week ago, one of the fans of this preprint was Dr. Vinay Prasad. You remember Dr. Prasad, don’t you? First, he was a big fan of the preprint, calling it a “bombshell” by a “dream team” of authors, which made me wonder if he knew who the heck Josh Stevenson was, leading to Dr. Høeg thanking him profusely:

You might also recall that Dr. Prasad, someone whom I like to refer to as a rising star in oncology, was someone whose work, prepandemic, met with some approval here at SBM. Then, unfortunately, for whatever reason, Dr Prasad decided that the sort of work we do here at SBM and other doctors and scientists do to combat scientific and medical misinformation was the equivalent of Michael Jordan or LeBron James “dunking on a 7′ hoop”, leading to retorts by Steve Novella and myself, who expressed nothing but disdain for our efforts, seemingly thinking it beneath him to address alternative medicine, antivaccine pseudoscience, and the like. Ironically, if there’s anything that the COVID-19 pandemic has shown us, it’s just how dangerous the sort of medical misinformation that we at SBM have long combatted actually is.

In any event, Dr. Prasad used his outlet at Medscape to weigh in on the preprint and the reaction to it in an article titled “Vaccine Tribalism Is Poisoning Progress on COVID Science“. Oh, dear. It turns out that Dr. Prasad, in an utterly unsurprising twist, views critics of the preprint as the problem, not those who wrote the preprint, and reveled in the publicity. He begins with an astounding contortion of “bothsidesism”:

As I look across social media, I am concerned. Clearly there is a small, but vocal minority of people who are critical — often irrationally so — of vaccination. They are wrong. But, what I see more and more, among the ranks of physicians and other professionals, is a growing minority on the other end of the spectrum. These are people who are quick to label legitimate scientific dialog as “anti-vax” or “dangerous misinformation.” In many cases, the speakers have little knowledge of the issues themselves. They couple this condemnation with a strong sense that they are “morally” correct, working to purge the world of dangerous anti-vax thinking. Ironically, they are further polarizing an already polarized debate, and worse, they are simply wrong. These are real and live issues. Intelligent scientists have to discuss these policy implications openly. The stakes could not be higher.

As we continue to investigate unanswered questions during this unprecedented pandemic, the tribalism and polarization must end. Simply raising questions about the vaccines isn’t anti-vax, so let’s not label it as such. Instead, we need to welcome new questions and invite ongoing, open discussions from medical professionals. If not, we risk poisoning progress.

How “reasonable.” How “measured.” How utterly ignorant of the history of the antivaccine movement. Even more unsurprisingly, after citing some studies, Dr. Prasad pivots to point to the reaction to the preprint as an example of exactly what he’s talking about:

But on social media, the calls against the authors came vociferously, passionately, and very nearly constantly. Many were skeptical of the study being based on VAERS data, which is self-reported. And, I agree that generally VAERS data is unsuited to draw conclusions about the frequency of adverse events. But there are several unique features here: the authors reviewed all cases by hand, using the expertise of a cardiologist; many cases had features that were challenging for alternative explanations; and finally, the authors acknowledged many limitations in the paper and provided the primary source data for others to code differently in a convenient app. All that said, I am sure there are persistent errors in the paper, both of inclusion and exclusion, and others may wish to make different external comparisons. The paper is not perfect, but it offers something, even if it’s not the end of the story.

It is indeed highly amusing to me to observe how quickly the eminent Dr. Prasad went so quickly from praising a “bombshell” paper by a “dream team” of investigators to describing it as “not perfect” and likely still containing “persistent errors” that lead others to want to make “different external comparisons”. (One wonders how rapidly that Tweet of his will go down the memory hole.) Of course, critics of the preprint pointed out how none of the “unique features” cited by Dr. Prasad redeems the analysis in the least.

None of that stopped the so-much-more-rational-than-us Dr. Prasad from characterizing this as “medical tribalism” that poisons everything:

The “tribe” of med-Twitter that had existed before was fractured. Many felt that dogged an unyielding devotion to vaccines (i.e., two mRNA doses has to be good for everyone, everyone must get a booster ASAP) was synonymous with being pro-vaccine, and anti-pandemic. One driving factor has been a new entrant into previously academic debates: the public. As medical and research discussions have increasingly become part of the mainstream during COVID-19, many doctors have been labeled as anti-vaxxers or COVID-deniers because the public will take their questioning of the science and treat it as fact. While I understand the desire to prevent this, we can’t neglect the fact that an ongoing and careful reconsideration of harms and benefits is often warranted in this complex and constantly changing pandemic. What works for an 80-year-old might not be right for a 16-year-old boy. Log fold risk by age is not intuitive, not fully processed, nor understood. A tribe that developed a policy platform over years was unable to handle a scientific moment that twisted and fractured the usual rules, especially as the public entered the debates.

The funny thing is, it’s not the pro-vax “tribe” that lacks nuance. Sure, there are occasional members who do. Vaccine advocates are human, and there will always be a range of behaviors, from civil to overly vociferous. Overall, though, it is the vaccine advocate side that has been discussing how to promote vaccination and problems with various vaccines with far more “nuance” than Dr. Prasad or the authors of this study.

In fact, I’d go so far as to say that it’s not the “nuance” brought by Dr. Prasad, the authors of this preprint, and others that vaccine advocates criticize. It’s the antivaccine talking points that these physicians and scientists parrot, I’ll assume unknowingly. Indeed, Dr. Prasad’s entire article is basically a retread of an antivaccine talking point frequently inadvertently parroted by people who view themselves as so very “reasonable” and “nuanced” that I’ve dealt with more times than I can remember, portraying vaccine advocates as unreasonable, frenzied, self-righteous zealots who engage in “vaccine exceptionalism” and “bully” parents to vaccinate and turn well baby visits into “battlefields“. Meanwhile, if these “nuance-lovers” mention it at all, they only mention in passing how “tribal” those whose talking points they sometimes parrot go. Indeed, antivaxers are quite enamored of Holocaust analogies, either with vaccines causing a Holocaust or laws requiring children to be vaccinated before they can attend school being likened to Nazi-ism.

It is also very much amazing to me how utterly clueless and oblivious Dr. Prasad is to how the work of these scientists whom he praises is used. In fairness, it is definitely true that it is impossible to control how one’s work is used once it’s published, but let’s step back a minute and look at the whole picture. The preprint study’s authors admitted that their manuscript was rejected by three different journals. In the Medscape story, Dr. Høeg claims it was for “unclear reasons,” but I have a hard time believing that, unless the editors of all three journals “triaged” their manuscript, declining to send it out for peer review because they deemed unsuitable. If any or all of the journals sent the manuscript out for peer review, the authors would have received detailed critiques from at least two peer reviewers from each journal. The authors know that. Dr. Prasad knows that. In fact, the authors even admitted it in the Medscape report. Høeg outright stated that she and her coauthors had incorporated critiques from peer reviewers into the version of the manuscript that they published as a preprint! (I can’t help but observe that, if this version has been “improved” by addressing reviewer comments, I’d hate to see the first version of the manuscript.) In light of that history, publishing the manuscript as a preprint reeks of bypassing peer review to put the study’s conclusions before the general public, having failed to produce an article that could pass peer review.

But none of that stops Dr. Prasad from parroting another favorite antivaccine talking point, again I’ll assume unknowingly:

Many of these commenters need to be honest and appraise their own skill set. Are you capable and do you have experience quantifying rare harms? Are you certain that the federal agencies tasked with this appraisal — the same agencies that have made colossal errors — are not making errors here? For most people on Twitter, the best way to be a science communicator is to sit this one out. When used correctly, science is the greatest lighthouse for human endeavors, but if misused, science has led to some of the greatest errors in our history.

Demonizing people interested in better characterizing vaccine safety signals is not pro-vax; it is pro-ignorance and anti-science. Vaccine tribalism is poison.

I’d ask Dr. Prasad the same questions. What he’s engaging in is pure credentialism. Of course, on paper I appear not to have the “credentials” to criticize Høeg et al. However, that leaves out the 16 years of practical experience I have deconstructing dumpster diving VAERS studies carried out by those with an antivaccine agenda, experience that armed me with the knowledge to know immediately that Høeg et al were doing the same thing and falling into the same trap of using VAERS data inappropriately to estimate the frequency of an adverse event. Dr. Prasad, for all his “credentials,” was utterly unaware of that history, what VAERS is, and how it has been misused. How else could he have so risibly characterized the study a “bombshell” by a “dream team of authors”?

How we got here, conclusion

The publication of this preprint, the reaction to it, and the defense of its authors by a certain contingent of medical social media combine to create a depressing affair. How did we get here? Allow me to speculate. First, we got here because of an unholy alliance between antivaxxers, the resistance to public health interventions, conspiracy theorists like QAnon believers, and political actors backed by industry to promote COVID-19 misinformation in order to stoke resistance to masks, “lockdowns,” and now even vaccine mandates. Am I saying that Dr. Prasad and Høeg et al. are conspiracy theorists? Of course not. What I am saying is that there is a contingent in medicine (most of medicine) that has been utterly unaware and unconcerned about medical misinformation, some (like Dr. Prasad) even going so far as to express contempt for efforts to address this misinformation as beneath them.

This is not a new problem, and one “good” effect of the pandemic is that it’s forced a number of my colleagues who had been oblivious to the power of quackery, antivaccine conspiracy theories, and health misinformation to have their noses rubbed in it to the extent that they have to admit that I had a point all those year. Unfortunately, there remains a contingent of doctors who view themselves as so much more “reasonable” than those awful vaccine advocates, as so much more capable of “nuance,” to whom part of the message of conspiracy theorists resonates because they themselves are resistant to, for instance, government health mandates and view themselves as apostles of evidence and science. Dr. Prasad, for instance, was well known before the pandemic for (correctly) railing against weak science, and yet that didn’t stop him from initially characterizing Høeg et al. as a “bombshell” by a “dream team”. I know I keep repeating that, but it can’t be repeated too often, given Dr. Prasad’s long-expressed contempt for “debunking” antivaccine pseudoscience.

This can lead to really bizarre incidents, such as a formerly reasonable thoracic surgeon not minding Robert F. Kennedy, Jr. representing his writing as part of Children’s Health Defense or Monica Gandhi liking Tweets by rabid antivaxxers like Rob Schneider because antivaxxers share their disdain of big pharma:

We saw the same phenomenon with Peter Gøtzsche before the pandemic, a phenomenon that has become depressingly common during the pandemic, with embarrassing results.

Meanwhile, I can’t help but point out that at least two of us at SBM are not as zealous as Dr. Prasad seems to think. Indeed, back in January, when there were a few insinuating that he was antivaccine, Jonathan Howard and I both pushed back and defended him, in spite of our strong disagreements with him on vaccines and other topics:

I rather suspect that these doctors’ self-image as being so much more “reasonable” and “nuanced” than everyone else, combined with the ego massage that they get from expressing “contrarian” views about COVID-19, is what leads them astray. Tribalism can certainly be a problem, particularly when one “tribe” doesn’t recognize that it is just as liable to fall prey to “tribalism” as any other human being and believes its “tribe” to be so much more “reasonable” and “nuanced” than any other. The result is often anything but reasonable or nuanced, and that appears to be what has happened in the sad debacle of the VAERS preprint, in which contrarians team up with a COVID-19-minimizing activist to produce an poorly done dumpster dive into VAERS that was immediately seized on by antivaxxers and antimaskers as evidence that vaccinating adolescents against COVID-19 is more dangerous than the disease.