Are COVID-19 deaths undercounted? Yes, especially if people with other illnesses die at home and are never tested for the virus that causes the disease. Are COVID-19 deaths overcounted? Probably also yes, especially when physicians try to compensate for the lack of testing by inferring that COVID-19 caused a death without laboratory confirmation, but also when they assume that the disease killed someone who tested positive, even if the actual cause might have been something else.
Judging from excess mortality in places hit hard by the epidemic, the first problem is bigger than the second problem, although those figures are ambiguous, may be incomplete, and so far are limited to relatively brief periods of time. In principle, the accuracy of COVID-19 death tallies is an empirical question, albeit one that may never be conclusively answered. But like virtually every other pandemic puzzle, it is also a political question, as illustrated by a recent New York Times story that charges Scott Jensen, a Minnesota family doctor and Republican state legislator, with aiding and abetting right-wing "denialists" who think all the hoopla about COVID-19 is a conspiracy cooked up by Donald Trump's enemies.
The Times piece portrays confirmation bias and motivated reasoning as problems that afflict only one side of the debate about COVID-19 control measures. In the paper's telling, lockdown skeptics, whom it conflates with "the virus 'truther' movement," let their ideology influence the way they interpret and present information, while lockdown supporters are interested only in discovering and disseminating the facts.
In covering Jensen's controversial comments about COVID-19 death tallies, the Times focuses on their political utility rather than their validity. "The claim was tailor-made for President Trump's most steadfast backers," write Matthew Rosenberg and Jim Rutenberg. Here is how they summarize Jensen's argument: "Federal guidelines are coaching doctors to mark Covid-19 as the cause of death even when it is not, inflating the pandemic's death toll."
That gloss implies that Jensen thinks federal officials are deliberately encouraging overdiagnosis. But that is not what Jensen says. Rather, he argues that some deaths may be misclassified based on guidelines from the U.S. Centers for Disease Control and Prevention (CDC) concerning "probable" or "presumed" COVID-19 cases in patients who were never tested for the virus. That is a potentially significant problem when such cases account for a substantial share of reported deaths—more than a quarter in New York City, for example.
"In cases where a definite diagnosis of COVID-19 cannot be made," the CDC says, "but it is suspected or likely (e.g., the circumstances are compelling within a reasonable degree of certainty), it is acceptable to report COVID–19 on a death certificate as 'probable' or 'presumed.' In these instances, certifiers should use their best clinical judgment in determining if a COVID–19 infection was likely. However, please note that testing for COVID–19 should be conducted whenever possible."
During an April 9 interview on Fox News, Jensen called the CDC's advice "ridiculous" and contrary to usual medical practice. "The idea that we're going to allow people to massage and sort of game the numbers is a real issue, because we're going to undermine the [public] trust," he said. "And right now, as we see politicians doing things that aren't necessarily motivated [by] fact and science, the public is going to—their trust in politicians is already wearing thin."
You can surmise from Jensen's framing that he is not a fan of lockdowns. But contrary to what Rosenberg and Rutenberg seem to think, that does not necessarily mean he is wrong.
In the case of a fragile, elderly patient with a cough and fever who happens to die during an influenza outbreak, Jensen said, "I wouldn't put influenza on the death certificate. I've never been encouraged to do so. I would put, probably, respiratory arrest [on] the top line, and the underlying cause…would be pneumonia, and under 'contributing factors,' I might well put in 'emphysema' or 'congestive heart failure.' But I would never put influenza down as the underlying cause of death. Yet that's what we're being asked to do here."
Jensen conceded that other doctors might take a less conservative approach. "Some physicians really have a bent toward public health, and they will put down 'influenza,' or whatever, because that's their preference," he said. "I try to stay very specific, very precise. If I know I've got pneumonia, that's what's going on the death certificate. I'm not going to add stuff just because it's convenient."
Even when someone dies after testing positive for the COVID-19 virus, Jensen added, it is not safe to assume the disease killed him. He asked viewers to imagine a patient who goes to the hospital with a collapsed lung after "getting hit by a bus" and tests positive for COVID-19. "They die 20 minutes later because of their collapsed lung," he said. "We're going to put that down as COVID-19? That doesn't make any sense."
While that example is fanciful, Jensen's underlying point is valid. So is his warning that deaths of people who were never tested for the virus may be erroneously attributed to it, especially in light of the CDC's guidance.
More controversially, Jensen suggested that hospitals have a financial incentive to overdiagnose COVID-19. "Any time health care intersects with dollars, it gets awkward," he said. "Right now Medicare has determined that if you have a COVID-19 admission to the hospital, you'll get paid $13,000. If that COVID-19 patient goes on a ventilator, you get $39,000—three times as much. Nobody can tell me, after 35 years in the world of medicine, that sometimes those kinds of things [don't] impact on what we do."
That claim gave rise to memes like this one, suggesting that financial incentives are driving up COVID-19 death counts:
FactCheck.org found that Jensen's estimate of Medicare reimbursements was reasonable. It noted that "the government will pay more to hospitals for COVID-19 cases in two senses: By paying an additional 20% on top of traditional Medicare rates for COVID-19 patients during the public health emergency, and by reimbursing hospitals for treating the uninsured patients with the disease (at that enhanced Medicare rate)." But it said "the fact that government programs are paying hospitals for treating patients who have COVID-19 isn't on its own representative of anything nefarious."
Jensen told FactCheck.org he did not mean to suggest that hospitals are deliberately padding their numbers. But he did imply that money tied to COVID-19 has an influence on diagnoses.
Is that plausible? "There's an implication here that hospitals are overreporting their COVID patients because they have an economic advantage [in] doing so, [which] is really an outrageous claim," Gerald Kominski, a senior fellow at the UCLA Center for Health Policy Research, told FactCheck.org. Kominski interpreted Jensen as suggesting that patients are put on ventilators when it is not medically appropriate, which "is basically saying physicians are violating their Hippocratic Oath….It would be like [performing] heart surgery on someone who doesn't need it."
But that does not seem to be what Jensen is suggesting. In an April 15 Facebook post, Jensen argued that "increasing the number of COVID-19 deaths may create an avenue for states to receive a larger portion of federal dollars." He elaborated on his argument in an April 19 video, saying: "Hospital administrators might well want to see COVID-19 attached to a discharge summary or a death certificate. Why? Because if it's a straightforward, garden-variety pneumonia that a person is admitted to the hospital for—if they're [on] Medicare—typically, the diagnosis-related group lump sum payment would be $5,000. But if it's COVID-19 pneumonia, then it's $13,000, and if that COVID-19 pneumonia patient ends up on a ventilator, it goes up to $39,000."
Under these circumstances, Jensen says, it is reasonable to think that doctors might feel pressure to diagnose COVID-19 in ambiguous cases. He emphasizes that he is not accusing doctors of lying, just recognizing that they operate within an institutional framework that can be affected by financial considerations. "I am not saying physicians in the emergency room are…gaming the system," he says. "But we do have a variety of players in the system" who may encourage doctors to note COVID-19 as a cause of death.
USA Today asked Marty Makary, a surgeon and professor of health policy and management at Johns Hopkins Bloomberg School of Public Health, about Jensen's claim. "What Scott Jensen said sounds right to me," Makary replied in an email, although he declined to elaborate.
Jensen's general take on the accuracy of COVID-19 death tallies is also more nuanced than it has been widely portrayed. "Do I think there's undercounting of COVID deaths?" he says in his video. "Yeah, for sure. A state like ours, we're probably going to undercount, because we don't have all these people who may have died of a COVID-19-related disease tested. So we may undercount." But in New York City, which last month began reporting "probable" COVID-19 deaths alongside "confirmed" COVID-19 deaths, "they sure could be overcounting….There's a lot of variety and a lot variability to what's going on here."
Instead of assessing the strength of Jensen's arguments, the Times immediately links him to the most disreputable lockdown opponents. "His assertions were picked up by Infowars, the conspiracy-oriented website founded by Alex Jones," Rosenberg and Rutenberg note in the third paragraph of their story. "They were shared by followers of Qanon, who subscribe to a web of vague, baseless theories that a secret cabal in the government is trying to take down the president."
More generally, Rosenberg and Rutenberg say, Jensen's critique is useful to Trump supporters, and you know that can't be good. "Since the outset of the crisis," they write, "elements of the right have sought to bolster the president's political standing and justify reopening the economy by questioning the death toll. Climate-change skeptics have employed techniques perfected in the fight over global warming to raise doubts about the deadliness of the virus. Others, including Mr. Trump's media allies as well as some in the anti-vaccine movement, have repurposed fringe theories about 'deep state' bureaucrats undermining the president to argue that the official numbers should not be trusted."
The implication is that anyone who wonders about the accuracy of COVID-19 death counts—even someone like Jensen, who suggests deaths may be underreported in some parts of the country and overreported in others—should be taken as seriously as someone who denies the human contribution to climate change or thinks vaccines cause autism. Likewise anyone who "raise[s] doubts about the deadliness of the virus"—an issue that remains scientifically unsettled, mainly because a dearth of testing means we do not know the true number of infections, a fact that is crucial in estimating what share of people who catch the virus will be killed by it.
The Times also looks askance at anyone who "question[s] the [epidemiological] models," which rely on assumptions about unknown variables and generate a wide range of projections that are often inconsistent with each other and with what has actually happened. "Even under the best circumstances," Rosenberg and Rutenberg concede in the 26th paragraph, "modeling how a pandemic will play out, like modeling the pace and impact of climate change, is an imperfect science. And there is indeed great uncertainty about what the death toll is now—and what it will be—given limited data about the new coronavirus and the different counting methods jurisdictions are using."
Still, Rosenberg and Rutenberg aver, "The lines of attack against the conclusions of health experts are familiar to those who have studied the climate-change denial movement." Never mind that "health experts" disagree with each other about crucial facts such as the prevalence, lethality, and transmissibility of the virus. As far as the Times is concerned, highlighting these areas of uncertainty makes you a misinformed "Covid skeptic," if not an outright "virus 'truther.'"
To clinch their case that Jensen is a crank who should be ignored, Rosenberg and Rutenberg close their story by noting that the legislator recently "plugged into a remote State Senate hearing on easing restrictions on telemedicine for addiction disorders while playing a round of golf, without a mask." While local criticism of that incident focused on whether Jensen was paying proper attention to the hearing (he says he was), the Times wants you to know he was not wearing a mask. Despite Jensen's M.D. and 35 years of practicing family medicine, readers are invited to conclude, he does not even seem to be aware that COVID-19 is contagious.
You might think that going maskless is not necessarily reckless behavior in an open outdoor space like a golf course, where the risk of transmission is low and it is easy to maintain appropriate distance from others. But if that's what you think, you had better keep your opinion to yourself—unless you feel comfortable being lumped in with the "denialists."