Skip to Main Content
Contribute Try STAT+ Today

When major decisions must be made amid high scientific uncertainty, as is the case with Covid-19, we can’t afford to silence or demonize professional colleagues with heterodox views. Even worse, we can’t allow questions of science, medicine, and public health to become captives of tribalized politics. Today, more than ever, we need vigorous academic debate.

To be clear, Americans have no obligation to take every scientist’s idea seriously. Misinformation about Covid-19 is abundant. From snake-oil cures to conspiracy theories about the origin of SARS-CoV-2, the virus that causes the disease, the internet is awash with baseless, often harmful ideas. We denounce these: Some ideas and people can and should be dismissed.

At the same time, we are concerned by a chilling attitude among some scholars and academics, who are wrongly ascribing legitimate disagreements about Covid-19 to ignorance or to questionable political or other motivations.

advertisement

A case in point involves the response to John Ioannidis, a professor of medicine at Stanford University, who was thrust into the spotlight after writing a provocative article in STAT on Covid-19. He argued in mid-March that we didn’t have enough information on the prevalence of Covid-19 and the consequences of the infection on a population basis to justify the most extreme lockdown measures which, he hypothesized, could have dangerous consequences of their own.

We have followed the dialogue about his article from fellow academics on social media, and been concerned with personal attacks and general disparaging comments. While neither of us shares all of Ioannidis’ views on Covid-19, we both believe his voice — and those of other legitimate scientists — is important to consider, even when we ultimately disagree with some of his specific analyses or predictions.

advertisement

We are two academic physicians with different career interests who sometimes disagree on substantive issues. But we share the view that vigorous debate is fundamental to the existence of universities, where individuals with different ideas who have a commitment to reason compete to persuade others based on evidence, data, and reason. Now is the time to foster —not stifle — open dialogue among academic physicians and scientists about the current pandemic and the best tactical responses to it, each of which involve enormous trade-offs and unanticipated consequences.

Since Covid-19 first emerged at the end of 2019, thousands of superb scientists have been working to answer fundamental, vital, and unprecedented questions. How fast does the virus spread if left unabated? How lethal is it? How many people have already had it? If so, are they now immune? What drugs can fight it? What can societies do to slow it? What happens when we selectively evolve and relax our public health interventions? Can we develop a vaccine to stop it? Should governments mandate universal cloth masks?

For each of these questions, there are emerging answers and we tend to share the consensus views: Without social distancing, Covid-19 would be a cataclysmic problem and millions would die. The best current estimate of infection fatality rates may be between 0.4% and 1.5%, varying substantially among age groups and populations. Some fraction of the population has already been infected by SARS-CoV-2 and cleared the virus. For reasons that aren’t yet totally clear, rates of infection have been much higher in Lombardy, Italy, and New York City than in Alaska and San Francisco. To date no drug has shown to be beneficial in randomized trials — the gold standard of medicine. And scientists agree that it will likely take 18 months or longer to develop a vaccine, if one ever succeeds. As for cloth masks, we see arguments on both sides.

At the same time, academics must be able to express a broad range of interpretations and opinions. Some argue the fatality rate will be closer to 0.2% or 0.3% when we look back on this at a distance; others believe it will approach or eclipse 1%. Some believe that nations like Sweden, which instituted social distancing but with fewer lockdown restrictions, are pursuing the wisest course — at least for that country — while others favor the strictest lockdown measures possible. We think it is important to hear, consider, and debate these views without ad hominem attacks or animus.

Covid-19 has toppled a branching chain of dominoes that will affect health and survival in myriad ways. Health care is facing unprecedented disruption. Some consequences, like missed heart attack treatment, have more immediate effects while others, like poorer health through economic damage, are no less certain but their magnitude won’t immediately become evident. It will take years, and the work of many scientists, to make sense of the full effects of Covid-19 and our responses to it.

When the dust settles, few if any scientists — no matter where they work and whatever their academic titles — will have been 100% correct about the effects of Covid-19 and our responses to it. Acknowledging this fact does not require policy paralysis by local and national governments, which must take decisive action despite uncertainty. But admitting this truth requires willingness to listen to and consider ideas, even many that most initially consider totally wrong.

A plausible objection to the argument we are making that opposing ideas need to be heard is that, by giving false equivalence to incorrect ideas, lives may be lost. Scientists who are incorrect or misguided, or who misinterpret data, might wrongly persuade others, causing more to die when salutatory actions are rejected or delayed. While we are sympathetic to this view, there are many uncertainties as to the best course of action. More lives may be lost by suppressing or ignoring alternate perspectives, some of which may at least in part ultimately prove correct.

That’s why we believe that the bar to stifling or ignoring academics who are willing to debate their alternative positions in public and in good faith must be very high. Since different states and nations are already making distinct choices, there exist many natural experiments to identify what helped, what hurt, and what in the end didn’t matter.

We believe that the bar to stifling or ignoring academics who are willing to debate their alternative positions in public and in good faith must be very high.

Society faces a risk even more toxic and deadly than Covid-19: that the conduct of science becomes indistinguishable from politics. The tensions between the two policy poles of rapidly and systematically reopening society versus maximizing sheltering in place and social isolation must not be reduced to Republican and Democratic talking points, even as many media outlets promote such simplistic narratives.

These critical decisions should be influenced by scientific insights independent of political philosophies and party affiliations. They must be freely debated in the academic world without insult or malice to those with differing views. As always, it is essential to examine and disclose conflicts of interest and salient biases, but if none are apparent or clearly demonstrated, the temptation to speculate about malignant motivations must be resisted.

At this moment of massive uncertainty, with data and analyses shifting daily, honest disagreements among academic experts with different training, scientific backgrounds, and perspectives are both unavoidable and desirable. It’s the job of policymakers, academics, and interested members of the public to consider differing point of views and decide, at each moment, the best courses of action. A minority view, even if it is ultimately mistaken, may beneficially temper excessive enthusiasm or insert needed caveats. This process, which reflects the scientific method and the culture that supports it, must be repeated tomorrow and the next day and the next.

Scientific consensus is important, but it isn’t uncommon when some of the most important voices turn out to be those of independent thinkers, like John Ioannidis, whose views were initially doubted. That’s not an argument for prematurely accepting his contestable views, but it is a sound argument for keeping him, and others like him, at the table.

Vinay Prasad is a hematologist-oncologist and associate professor of medicine at the Oregon Health and Science University and author of “Malignant: How Bad Policy and Bad Evidence Harm People with Cancer” (Johns Hopkins University Press, April 2020). Jeffrey Flier is an endocrinologist, professor of medicine, and former dean of Harvard Medical School.

  • Ok, cities states and nations shut down, the curve is flattened and then decreases, at what point can things begin to open again? If the virus is as infectious as it seems to be, there will not ever be a way to contain it, other than a vaccine, outside of restricted contact and movement for an indefinite period. There is a difference between what is medically ideal and what is possible within human societal systems. What we are doing now as a society is unsustainable economically, psychologically, and medically. We are conditioned to believe that any amount of suffering and death due to disease is unacceptable, an admirable philosophy. Yet, absurd in the face of the reality of how millions live and die.

    If eventual herd immunity is actually the only way this disease will be controlled, and from all evidence that I see, this seems likely in the long term. Our current pathway is morally indefensible.

    Starvation kills tens of millions a year and is entirely preventable, far more preventable than covid-19. Shutting down the economic system will likely cause this figure to dramatically rise, likely far in excess of those that will die to this virus. We are making decisions that will kill millions. Yes, we will save mostly the old and infirm in developed nations from a quicker death than they could normally expect but have no doubt, we are condemning millions of others to a very early death. Even developed nations will likely experience political and societal upheaval (think of the food price pressure as one variable in the rise of ME turmoil, refugee crisis in Europe, and the rise of the far right).

    • I think I get your point, if herd immunity levels are inevitable no matter what we do, then just figure out the fastest way to get there without extra deaths caused by overloading the health system. I think the answer is that herd immunity levels are not inevitable, and that something like South Korea’s testing plan is the model to look towards. Yes, there will be some reduced social activity, but with deaths minimized and with the economy able to function, even if not at peak levels. It remains to be seen if it will work, but they seem to be on a good path. I assume experts in U.S. are studying South Korean approach. See, e.g., https://www.theguardian.com/world/2020/apr/23/test-trace-contain-how-south-korea-flattened-its-coronavirus-curve , and http://www.undp.org/content/seoul_policy_center/en/home/presscenter/articles/2019/flattening-the-curve-on-covid-19.html

      Regarding starvation, yes, I expect that will be a problem in many less developed countries that have not yet had large coronavirus outbreaks. I don’t see that as an issue closely related to what we do in the U.S. Surely you’re not suggesting that in the U.S. the economy will shut down to a level so that starvation deaths will be “likely far in excess of those that will die to this virus”?

    • Our ability to mimic South Korea has long since past. The horses are well out of the barn. Yes, I was not referring to starvation in the US. The rise in prices of grain and wheat causes calamity for much of the world population that lives on the edge of starvation. People in the US will complain of bread becoming more expensive while millions around the globe like those in sub-Saharan Africa will starve. It shows people’s absolute ignorance of global economic systems when they state things such as who cares about GDP or the economy. I understand the ingroup bias people have for the strangers who they perceive as more connected to them than strangers farther away with a different label but a decision to shut down an economy that has positioned itself in the food market to be so dominant, encouraged these nations to grow cash crops rather than food crops, is a decision that will kill millions. Those lives matter as much as any others.

  • How dare you question the all-knowing scientists who fit the media and DNC narrative! Everyone should stay home indefinitely until the world ends. And people wonder how Trump got elected? People want these authoritarian elites kicked in the teeth.

    • No one is saying do nothing. What about if we just protect the susceptible? Let life resume for the rest of us.

  • Agree with much in this long article, but not what I gather is the main conclusion. Of course some views are more political than others, we cannot hide behind a denial of that. Those arguing that we are in control and can carefully restore normalcy are under a huge burden to make the argument convincingly given the politics and the horrendous situation we are in.
    So, when specious speculations about the death rate per incidence of positive tests keeps coming up, I get impatient. What are people afraid of? Dying. Not the statistics per positive test. When US deaths per million in the population begin to decrease, that is the time to ease off our restrictions. All the issues about deaths per positive tests are useful and interesting, but being manipulated if we interpret them to negate the incidence of deaths in the country incorrectly.
    Jim

  • Agreed, all opinions should be accepted regardless of their nature and analyzed for their validity. Conclusions should be logical and based on evidence that is independently validated. That is the laboratory. In real life, much is swayed by political agenda, greed, faith, despair, which is to say, it suffers from the Human condition. For the most part I believe that the decisions that took place, have slowed the progress of the virus. The social distancing, the stopping of mass movement through airlines and across borders, the isolation of potentially contaminated for 14 days. I don’t believe the use of a mask offers much protection. If you have ever sanded drywall, the evidence is clear that these masks do not seal well over time, but I suppose if it makes people feel safer, then why not. I also believe that we should have begun reopening business under the guide of the social distancing rules sooner. If I can shop for groceries with many other unknown people, then why can’t the same be said for all business. Just apply the safe rules!
    I am seeing a lot more people having a point of view on this, but even more important is that they are willing to speak out about it, which is a good thing. Freedom of speech is a pillar of the Democratic society, don’t let fear silence you. There are three kinds of people.
    1. Those that watch things happen
    2. Those that make things happen
    3. Those that sit around trying to wonder what the hell happened.
    Be a number 2!

    • “Be a number 2!” If everyone was a #2 this world would be chaotic. We need #1 and # 3 in equal numbers for this world to exist.

  • Some believe that nations like Sweden, which instituted social distancing but with fewer lockdown restrictions, are pursuing the wisest course.

    First, the US is not Sweden. Americans don’t have the civility, education and socialist attitude (that is, giving up some of your personal freedom for the common good) that Swedish have. And Swedish citizens have been voluntarily social distancing because of that. Not to mention, that the Swedish government has largely changed its stance on the matter.
    Second, all this talking about the final fatality rate is rather pointless. The final toll of these few months will out class any other “flu” season in recent history. And that, despite the fact we tried to stop it. That is, the human toll would be significantly higher if we didn’t do anything.
    Third, I haven’t heard of a single ICU medical worker describing the situation as normal, or even tough but already seen it. And, that is actually the most relevant aspect of why a lockdown was/is necessary. With inaction, the healthcare system would have just cracked after a few days, and then the bodies would have been seriously piling up.
    Fourth, there are many indications that the final count should be much worse that the current official one. Just look at Florida. By any reasonable consideration it should have ten times higher deaths than officially reported, and accidentally the local government is one of those that acted sloppily and slowly. Now, either they have been unbelievably lucky, or the swamps are filled with bodies.
    Finally, there is a cockiness circulating simply because it turned out the virus has been affecting mainly elderly people. If the virus would have been as deadly in young children, or even across all ages, in the US, there would be people shooting each other on the streets by now

    • “Additionally, CDC estimated that 151,700-575,400 people worldwide died from (H1N1)pdm09 virus infection during the first year the virus circulated.** Globally, 80 percent of (H1N1)pdm09 virus-related deaths were estimated to have occurred in people younger than 65 years of age.”
      From CDC-and we didn’t lock down nor did we have people shooting each other. Flu this year appears to have targeted kids again-over 100 have died this season-where is the media concern? Doesn’t everyone of those deaths count? You haven’t heard about ICU/medical workers say anything about when bad hotspots of the flu happen and there’s triaging in tents in hospital parking lots because the MEDIA hasn’t pushed it. Not saying CV19 isn’t bad or that the current deaths don’t matter-but remember it’s a matter of personal tunnel vision/media bias too.

    • ” Just look at Florida. By any reasonable consideration it should have ten times higher deaths than officially reported, and accidentally the local government is one of those that acted sloppily and slowly. Now, either they have been unbelievably lucky, or the swamps are filled with bodies.”

      …or formal social distancing measures are not the only factor affecting the outcome.

      Most of the US is relatively “pre-social-distanced” compared to NYC and New Jersey, where the worst results have been seen. The weather in March and April is more favorable to outdoor activities in Florida than in New York/New Jersey. There is far more travel by car vs. mass transit.

      So policies are not the only indicator of how much social distance is actually occurring.

  • I appreciate this article and the measured, fair tone the authors take towards this important issue. This is a difficult time, and I agree that it is too easy to abandon norms of discourse and too difficult to see the long-term consequences of doing so. And I certainly agree that we must avoid personal attacks.

    However, there is good reason that the community’s response here has been so forceful. That’s because some of the people associated with the study have continued to make media appearances to promote topline results for which there is no evidence on reasoning which was refuted within a few days of the preprint’s publication. These appearances – in the midst of an incredibly critical period during which we must base policy on good data – are themselves violations of standard norms. The first response of the academic community was to do what is has always done – examine the results and methods with a healthy dose of skepticism. But this did not prove sufficient to prevent a substantial segment of the population getting the incorrect impression that new evidence supports a flu-like mortality rate for COVID-19.

    The question of COVID-19 severity is an important one that is critical to determining the best way to move forward. And the authors of this piece are correct to note that there are serious trade-offs (in terms of both lives and livelihoods) of any policy decision and there still remains a tremendous amount of uncertainty.

    But resolving this question will require convincing the scientific community, not failing to do so and nevertheless trying to persuade the public.

  • The greatest need, as the authors would surely agree, is to acquire good and reliable data on which to base decisions. Antibody tests for the virus shows some promise in being able to ascertain community spread and narrow the infection fatality rate for this virus to help guide public policy.

    Unfortunately, a few states are rushing to determine and publicize community spread without releasing any details, underlying data and methodologies. New York state is using an assay that claims a specificity ranging from 93%-100%. https://on.ny.gov/2SclqMl. That represents a false positive range of 0-7%. So, when Gov Cuomo states that 15% of the state has been infected, he does not give a range. Is it 8% to 15% or some other range? What is the 95% confidence interval? How was the test validated? How many and what type of controls used? These details need to be made public so it can be reviewed by third parties. The tests in other states are even more questionable.

    The antibody test by Premier, used in the Santa Clara and Los Angeles counties, was determined by California researchers, https://covidtestingproject.org/, to have a false positive rate as high as 8% (95% Confidence Interval). The test by BioMedonics used in Florida and Massachusetts has a false positive rate as high as 21% (95% CI) https://www.dropbox.com/s/cd1628cau09288a/SARS-CoV-2_Serology_Manuscript.pdf?dl=0 (See Table 2)

    Clearly any measurements of community prevalence for the SARS-CoV-2 virus using these tests are just nonsense and just serves to mislead and give rise to many comments asserting that the mortality rate is very low because the prevalence is so high.

    According to the National Academy of Sciences “All SARS-CoV-2 serological study results should be viewed as suspect until rigorous controls are performed and performance characteristics described…most [tests] so far have not described well-standardized controls. Samples from patients with seasonal (non-SARS-CoV-2) coronavirus infections are especially important as negative controls.” See https://download.nap.edu/cart/download.cgi?record_id=25775

    • Moreover, it is worth noting that Premier Biotech was in receipt from Chinese Provincial health authorities of their evaluation of 97.3% specificity (4/150 false positives) for IgM/IgG by 3/25/20 which is in agreement with UCSF/et al at covidtestingproject.org. This was prior to Stanford releasing their preprint.
      The study results by Jiansgu Provincial CDC can be found at:
      https://imgcdn.mckesson.com/CumulusWeb/Click_and_learn/COVID19_CDC_Evaluation_Report.pdf

    • Alfred. Good to see that covidtestingproject.org evaluations of antibody tests are supported, in this case by the maker themselves. The point estimate by this group, for Premier Biotech was 97.22% (out of 108 negative samples 3 were misidentified as positive) which as you say closely correlates with the 97.33% from the maker. Given the small number of negative controls, 108 and 150 respectively, the 95% confidence interval is wide.

      Even the best performing test, according to covidtestingproject.org, Sure Biotech with no false positives out of 108 negative controls has a 95% CI ranging from 96.64%-100.00%. A test with a potential false positive rate of 3.5% is almost useless to assess virus prevalence in low to moderately infected communities. But it’s a good start and needs to be evaluated further with greater number of negative controls. Perhaps this is the test all states should be using.

      I find it horrific that New York, Massachusetts, Florida and other states are releasing prevalence numbers with inadequate information about the reliability of the tests.

  • But think of the children!

    The only rational, reasonable thing to do is whatever I feel like. Hang tight while I dig into a half-gallon of Snickerdoodle ice cream and a big bag of choc-y pretzels while wondering if the snackdown + unemployment bonus should be 6 months, 9 months, or until my bariatric scooter breaks.
    So I’m feeling pretty gross, and if you don’t feel like that too, you must be a “———!“.

  • Open the country back up June 1st. Have people who are immune compromised wear masks and everyone needs to be accountable and not go out in public if they are sick. If they are out in public sick, criminal charges should be applied.

    • Different states are using different strategies. We will be able to observe the results, and we should get a pretty good indication of which strategies were most successful.

    • Who determines who is “sick”? Am I going to get hauled off because I sneeze (covering my sneeze) because someone’s strong perfume triggered it? Am I going to have be deathly afraid to cough in public due to my asthma? It is an absolutely RIDICULOUS idea to Criminalize ILLNESS!!!!!!!

Comments are closed.