The surprise came in what Ioannidis had to say. As many public health experts and government officials were urging people to stay home to avoid infection, he speculated that the coronavirus might be less dangerous than assumed. News media were overhyping the disease. The greater risk lay not in covid-19 but in overzealous lockdowns to prevent its spread.

“We’re falling into a trap of sensationalism,” Ioannidis told the documentary filmmakers interviewing him remotely on March 23 as he sat in a studio at Stanford. “We have gone into a complete panic state.”

The video would be viewed more than a half-million times on YouTube before it disappeared. Six weeks after it was uploaded, the footage of Ioannidis was removed by YouTube, which said the interview with one of the world’s foremost epidemiologists had violated its policies on covid-19 misinformation. Ioannidis — whose talks for TEDx, at Google and in university lecture halls have circulated online for years — said he was stunned. But he was hardly silenced.

At a time when President Trump was openly at war with his own administration’s medical experts, Ioannidis’s doubts about the wisdom of lockdowns became part of the rancorous debate about how the country should respond to the threat of covid-19. His arguments in a string of appearances on Fox News, CNN and other news networks were seized on by right-wing firebrands seeking to discredit public-health officials and reopen the economy. It was a remarkable turn for Ioannidis, a longtime evangelist for science-based health policies who has argued for zealous gun-control measures and the abolition of the tobacco industry.

Ioannidis, 55, insists he is doing what he has always done: following the data and sometimes contending with the head winds of conventional wisdom or popular opinion. He says governments should focus on protecting the sick and elderly from infection while keeping businesses and schools open for the less vulnerable.

“There is a lethal virus circulating out there. We all have responsibility to do our best to contain it as much as possible. It’s not a joke. It’s not a conspiracy. It’s not fake,” he told The Washington Post. “But we don’t panic. We don’t destroy our world. We don’t freeze everything.”

But as the pandemic enters its deadliest phase, Ioannidis is losing the argument over how to combat covid-19. Among epidemiologists, consensus now exists that it was inaction, not overreaction, that helped create the worst public health crisis in a century. The uncontrolled spread of the virus has led to overrun ICUs in South Dakota and makeshift morgues in Texas. States and countries are locking down in a bid to preserve lives as vaccines start to roll out. Even Sweden, which resisted tough restrictions through the spring, is now reversing course to avert catastrophe.

Ioannidis still has defenders, who say he offers a cautionary tale of how partisanship has poisoned scientific discourse about the coronavirus — and in the case of the YouTube video, led to its outright suppression.

But his critics say the Stanford doctor is violating the principles of intellectual rigor he has spent much of his career espousing — refusing to admit his mistaken judgments and recklessly lending a scientific imprimatur to forces that defy public-health directives for irrational reasons.

Steven Goodman, an epidemiologist who co-directs a Stanford research institute with Ioannidis, said he worries about the role his longtime colleague and some other scientists have played in the highly politicized dispute over the coronavirus response.

“Debates among scientists about the evidence are healthy. But if conducted in public the rules change,” Goodman said. “They can confuse people and undermine the consistent messaging needed for public health. Politicians can also misuse these debates to undermine public health policies they don’t like. The result? Our complete failure to contain covid-19.”

Doubt is a cardinal virtue in the sciences, which advance through skeptics’ willingness to question the experts. But it can be disastrous in public health, where lives depend on people’s willingness to trust those same experts. And in 2020 America — in the era of QAnon, covid denial, bleach ingestion and disputed mail ballots — some accuse Ioannidis, the famous skeptic, of becoming a much more dangerous man than he realizes.

“I’m so excited about this next interview,” Fox News host Laura Ingraham said on April 21. Her upcoming guest, she said, would explain that the coronavirus’s prevalence in the population “may be 55 times higher than previously thought by the, quote, ‘experts,’” — here Ingraham raised her hands in a pair of mocking air quotes — “meaning the true fatality of the virus is somewhere below that of seasonal influenza.”

“Professor,” she continued, “it’s great to have you on tonight.”

Ioannidis appeared on-screen and began to guardedly explain the results of new research suggesting the virus might be less deadly than previously estimated (though not, as Ingraham claimed, less deadly than the flu). He said more studies were needed.

“Science is the best thing that can happen to humans,” he said. “And if we have good data, we can use science to really act decisively …”

Ingraham cut him off, asserting that the virus and the deaths it was causing were “largely relegated to a few key areas.” She asked whether it was responsible for Centers for Disease Control and Prevention Director Robert Redfield to warn of another wave of deaths when winter arrived.

“All predictions and all forecasts, they can be made,” Ioannidis said. “Personally, I would avoid making them, putting a lot of trust in them, because we just don’t know.”

“Professor, I wish I could read your entire CV, because it is so impressive,” she said.

Ioannidis was born in New York City and raised in Athens, the only son of Greek physicians. He was a child math prodigy, valedictorian at one of Greece’s premier high schools and first in his class at the medical school of the National University of Athens. He also cultivated interests in literature and classical music. To this day, he teaches a course at Stanford in Greek poetry and moonlights as an opera librettist.

He attended Harvard, and then Tufts, for advanced training in the treatment of infectious diseases, before spending two years at the National Institutes of Health working under Anthony S. Fauci. Ioannidis helped run the clinical trials that would lead to breakthrough drugs for the long-term treatment of HIV, and he worked at a clinic for HIV-positive patients at Johns Hopkins University.

Along the way, he began to harbor doubts that some of the world’s foremost medical experts had evidence to back up their advice.

“I was going through an increasing crisis of belief in our tools,” he recalled.

That crisis culminated in 2005 with a pair of papers that reverberated through the scientific world. “Why Most Published Research Findings Are False” demonstrated the many ways in w
hich researchers’ sloppiness, personal bias, careerism, financial incentives and thirst for publicity could lead them to dubious findings. A companion paper showed that such problems were common even among prominent studies published by top journals.

His writings made Ioannidis a star in the burgeoning field of evidence-based medicine, which seeks to test experts’ clinical judgments against hard data. In 2010 he was recruited to Stanford, where he would go on to co-found and co-direct the Meta-Research Innovation Center at Stanford (METRICS).

Ioannidis became a kind of beat cop in the world of high-stakes epidemiological research, turning a pitiless eye on subjects from nutritional guidelines to antidepressants. For all his willingness to eviscerate other scientists’ work, he always insisted on decorum in academic debates, and hardly cut an intimidating figure in the flesh. Fond of white summer suits, Ioannidis rode an adult-size tricycle to work from his home in Menlo Park. Students remembered him as unfailingly kind. He never got a Twitter account.

And he always insisted, as he tried to do on “The Ingraham Angle,” that his life’s work was to improve science, not discredit or discard it.

At a time when the United States had logged fewer than 150 confirmed covid-19 deaths, he urged more research to determine how deadly the coronavirus was and how far it had spread. Without that data, he argued, countries risked overreacting with draconian policies that could do more harm than good — “like an elephant being attacked by a house cat,” as he put it. “Frustrated and trying to avoid the cat, the elephant accidentally jumps off a cliff and dies.”

Despite warning about the hazards of forecasting with incomplete data, Ioannidis ventured a prediction of his own. Based on mortality figures from the Diamond Princess cruise ship outbreak, he wrote that the virus might claim only 10,000 lives in the United States. (Later he would seek to portray this figure as his “lower bound” estimate.)

In a sense, this was classic Ioannidis, assessing a poorly understood medical intervention — the halting of the world’s economy — and demanding evidence of its effectiveness and risks. But his arguments diverged from those of experts urging rapid quarantine measures to slow infections. Marc Lipsitch, an epidemiology professor at the Harvard T.H. Chan School of Public Health, recalled his bafflement at Ioannidis’s essay and said it was already clear that strict social distancing measures were necessary to avoid overflowing hospitals.

“We had enough evidence to see that uncontrolled spread was very dangerous,” Lipsitch said. “The idea that we should just sort of sit by and gather data calmly struck me as incredibly naive.”

But the backlash over the Stat essay was mild compared with what would ensue when Ioannidis and his colleagues set out to gather the data he said was needed. On April 17, a group led by Stanford researchers published a preliminary version of a study — the first of its kind in the United States — that tested a sample of Santa Clara County residents for antibodies to the virus. Based on their results, they estimated that the county probably had more than 50 times the 1,000 cases that had been officially reported.

The finding, if correct, meant some influential estimates of the virus’s lethality had been exaggerated. The Stanford paper estimated a mortality rate among the infected of 0.2 percent or less — well below the 0.9 percent figure used in a widely reported Imperial College London study that helped shape lockdown policies. Ioannidis was among more than a dozen authors whose names appeared atop the preprint, but he became one of its go-to publicists, appearing on CNN, Fox and elsewhere.

Problems with the study emerged almost immediately.

It seemed that every criticism Ioannidis once hurled at the work of sloppy scientists was now turned back upon him. Statisticians attacked the study’s methodology. Medical experts questioned its antibody tests. BuzzFeed News revealed that JetBlue Airways founder David Neeleman — an outspoken advocate of reopening the economy — had helped fund the research, and that Neeleman had offered additional funding to a Stanford scientist whose efforts to ensure the test kits’ reliability were holding up publication.

“What a weird turn to see John Ioannidis pushing one of sloppiest studies in the deluge of Covid-19 papers,” Alex Rubinsteyn, an assistant professor of computational medicine and genetics at the University of North Carolina School of Medicine, wrote on Twitter. “If he weren’t an author I would expect [the study] to show up in one of his talks as a particularly potent cocktail of bad research practices.”

Administrators at the Stanford School of Medicine launched a fact-finding review, which uncovered “no evidence that any of the study funders influenced the design, execution, or reporting of the study,” according to documents reviewed by The Washington Post. But the review did conclude that funders’ involvement could “suggest an apparent conflict of interest in the public mind,” and it faulted the authors for shortcomings in other areas, such as the Facebook ads they used to recruit test subjects and insufficient oversight of the volunteers who collected blood samples.

Ioannidis and his co-authors corrected and revised some calculations in the paper after it was made public, a not uncommon practice for studies posted online before peer review. But they stood by their results, which would prove to be within the range of figures observed in other states and countries. (The infection fatality rate differs from the case fatality rate, which measures mortality only among confirmed cases. Because cases have been undercounted at every stage of the pandemic, the infection fatality rate is usually the lower of the two numbers.)

“That’s exactly the way that science should work,” Ioannidis declared.

He said he received many messages from supporters who praised his courage in bucking the public-health consensus. But criticism also continued, and escalated. A meme accusing Ioannidis of eating his own feces was sent to Stanford email accounts. Someone started a rumor online that his 86-year-old mother had died of covid-19, and when friends began calling her apartment in Athens to ask about the funeral, she suffered a life-threatening hypertensive episode.

“I don’t think I’ve ever felt so miserable,” Ioannidis said.

Even some who did not embrace all of Ioannidis’s views were appalled at his treatment. Jeffrey Flier, former dean of Harvard Medical School, supports social distancing and favored the spring’s lockdowns. He is harshly critical of Trump, but also troubled by what the Ioannidis episode showed about scientists’ intolerance for facts that might be seen as buttressing the Trump administration’s policies.

“I don’t think anything I’ve ever seen in biomedical science has been so affected by politics,” Flier said. “This is the force field that has taken over.”

Late in October, Ioannidis joined a forum with scientists from across the world, including his Stanford graduate students, over Zoom. Their guest speaker was Atle Fretheim, research director at the Norwegian Institute of Public Health. Fretheim described how he had tried to
launch a randomized controlled trial — the gold standard in medical experiments — to determine whether closing schools reduced coronavirus outbreaks.

The project had failed. In March, Norwegian government officials were unwilling to keep schools open. In May, as the virus waned, they refused to keep schools closed. At one point, Fretheim related, he had granted an interview to a TV journalist in Norway. “Of course, it came out as, ‘Crazy researcher wants to experiment with children,’ ” he said.

Ioannidis watched amid the on-screen mosaic of faces. He had spent the summer and fall in Europe, shuttling between Berlin — where he had founded a sister branch of the Stanford meta-research center in 2019 — and Athens. In Greece he had been swimming regularly at Vouliagmeni Beach, and he was deeply tanned. But it was in Berlin that he felt happiest.

Over the summer Germany seemed to have tamed the virus, and many aspects of life had returned to something resembling normalcy. Wearing a mask, Ioannidis was able to meet with other scientists at his office and even give in-person lectures to socially distanced audiences. He attended operas and concerts, sitting through performances of Mozart and Schumann in the eerily depopulated Berliner Philharmonie music hall.

“I feel that I am defending my world when I am there,” he wrote in an email to The Post.

With the fall, however, threats to that world reemerged. Deaths and hospitalizations took off in Europe as the virus surged. The U.S. death toll kept rising, on its way to passing 300,000 by mid-December. In many large American cities, bars were open and schools were closed. Millions of Americans still refused to wear masks, and the president continued to insist that the virus was on the verge of “going away.”

In October, a group of scientists published the Great Barrington Declaration, which proposed hastening herd immunity to the virus by encouraging the young and healthy to resume normal life while increasing protections for the vulnerable. Jay Bhattacharya, a Stanford medical school professor who partnered with Ioannidis on the Santa Clara antibody study, was one of its organizers.

The declaration was quickly embraced by Trump administration officials, but not by Ioannidis. He refused to sign, saying questions of scientific fact should not be settled by petition.

But by then, Ioannidis had already made himself heard. He had appeared at least 18 times on major cable news networks, repeatedly questioning the severity of the pandemic. He acknowledged that his statements might have been exploited for political purposes but said the alternative — silencing his doubts, or withholding them from certain audiences — is dangerous as well.

“I think it’s very scary when we try to suppress scientific thinking and investigation, whatever the reason might be,” he said. “Conveying more certainty than there is does not save lives.”

But as the days grew colder and the death toll mounted, many public-health officials concluded that uncertainty was not saving lives, either. Desperate to beat back the virus, governments around the world were again closing businesses and imposing stay-at-home orders.

The predictions of a second wave Ioannidis had told Fox News viewers not to trust had proved right. The 10,000 American lives he said might be lost to the pandemic were vanishing almost every week. Germany began to quarantine visitors from Athens, preventing Ioannidis from returning to his research center in Berlin. So last month he boarded a flight back to California, where record-setting hospitalizations are now prompting a new round of lockdowns.