The Ioannidis Affair: A Tale of Major Scientific Overreaction

Editor’s note: This article was originally published on November 30, 2020 with a number of errors and misleading claims. First, it should have been labeled “Opinion,” but was not. Second, the authors’ bylines were omitted. Third, the authors failed to note that they have collaborated in the past with both John Ioannidis and Vinay Prasad, who are discussed in this essay, and also in this accompanying story. This, we now understand, was also the case with a similar opinion piece by the same authors in Undark magazine in June. Fourth, the authors did not disclose that there were other problematic issues raised about the design of a study co-authored by John Ioannidis, most notably how the study authors recruited study participants and how independent faculty at Stanford said that they were unable to verify the accuracy of their test.

Other specific errors or omissions are noted with asterisks in the text below. Scientific American sincerely regrets all of these errors.

In March, when John Ioannidis published an opinion essay in STAT, the reaction was swift and brutal. Ioannidis estimated that deaths in the U.S. from COVID-19 could potentially be as low as 10,000—or they could approach levels not seen since the flu pandemic of 1918. Anything was possible, he wrote, and he pleaded for better science in order to make informed decisions. Fellow researchers latched on to his low figure, accusing him of “horrible science” and of missing the point by calling for more data when “coffins of … victims are accumulating.” Richard Ebright, a professor of chemistry and chemical biology at Rutgers University, tweeted that Ioannidis’ STAT essay was “content-free, logic-free drivel.” James P. Reichmann, a student of public health at Georgia State University, wrote that Ioannidis’ “contrarian position” could cause “irreparable harm” and cost “thousands of human lives.”  Detractors accused him of recommending that the nation “do nothing” in response to the virus.

But that was not Ioannidis’ position. In the flood of public shaming, his central focus was ignored: estimates of COVID-19 mortality were all over the map, and without testing a “random sample of a population” and repeat testing “at regular time intervals to estimate the incidence of new infections,” the real answers were unknown. Data, not guesswork, he suggested, should guide public health decisions about interventions such as quarantine, travel bans, work and school closures, travel bans and physical distancing, which had their own risks of harm.

As attacks on Ioannidis mounted, two prominent physicians rose to his defense. Vinay Prasad, associate professor of medicine at the University of California, San Francisco, and Jeffrey Flier, former dean of Harvard Medical School, published an article entitled, “Scientists who express different views on Covid-19 should be heard, not demonized.”

But far worse was to come for Ioannidis. BuzzFeed,* a left-leaning media outlet, implied that a financial conflict of interest had distorted his findings. The BuzzFeed article advanced a trifecta of misleading claims: in addition to having a financial conflict,** the author suggested that his science was poorly done and that he had a right-wing political agenda.

The charges were wrong on all accounts.

First, the science: the charge arose out of a study to determine the percentage of people in Santa Clara County who were infected with COVID-19. Critics complained that the low rate determined in the study, on which Ioannidis was a co-author, could have statistically been all false positives.*** In other words, they asserted that in reality, there may have been zero individuals infected with COVID-19. The absurdity of their statistical argument, couched in a fair deal of math, eluded most observers.**** Second, critics accused Ioannidis of being a right-wing Trump supporter on the basis of his having appeared on a number of Fox news shows and of writing to President Trump of his concerns about the lack of evidence regarding the efficacy of lockdowns. However, unreported by BuzzFeed, Ioannidis also appeared far more often on CNN, which Trump disparages as a “fake news” outlet, and on the left-leaning BBC.

But the real below-the-belt charge came with the implication that Ioannidis was in league with David Neeleman, the founder and former CEO of JetBlue,***** who gave $5,000 to an anonymized fund at Stanford to support the Santa Clara study. Ioannidis told us he received “zero dollars” for the study*, and his lack of any financial incentive was further underscored by a “fact-finding” investigation conducted by an external law firm for Stanford. The investigation found “no evidence that any of the Study funders influenced the design, execution, or reporting of the Study.” Nor did the investigators “identify any conflict of interest of the faculty principals.” Stanford never released the results of its fact finding, because no researcher had been formally accused, and Ioannidis continues to be attacked. ******

* The original version of this piece misspelled “BuzzFeed” as “Buzzfeed.”

**BuzzFeed, which was not given a chance to respond to this piece in advance, disputes this claim, pointing out that “we know that Neeleman donated $5,000 to the study, he himself acknowledged that the study authors knew about it, and another co-author on the study, biotech investor Andrew Bogan, thanked him for his financial support in one of Neeleman’s many emails to the Santa Clara study authors during the course of the study.”

***The original version of this piece incorrectly referred to “false negatives” instead of “false positives

****BuzzFeed points out that “what the many critics of the study actually said was that, due to the low prevalence of COVID-19 at the time the study was conducted, the small number of individuals included in the study, and what was known about the specificity of the test, the researchers could not rule out the possibility that the positive test results they got could have all been false positives. In other words, they couldn’t disprove the null hypothesis that all the positive results were false. This is not an “absurd” statistical argument; this is a basic principle underlying such testing

*****The original version of this piece incorrectly identified Neeleman as the CEO of JetBlue; he is the founder of the airline and has not served as the company’s CEO since 2007.

******The authors of this piece failed to mention that the BuzzFeed story was based on a whistleblower complaint filed to Stanford by someone involved in the research. This complaint spurred the investigation cited later in the piece. The whistleblower complaint clearly stated, “Concern that the authors were affected by a severe conflict of interest is unavoidable.”