Balancing Epistemic Humility and Prior Knowledge

This was a textbook virus. Why didn't we act more like it?

You’ve likely heard the disease currently ravaging the planet described as a “novel coronavirus.” It was correct, of course, as it is novel. But that wasn’t all it was, and one the most frustrating aspects of this pandemic has been letting the word “novel” blind us to the rest of the phrase: coronavirus. 

This virus wasn’t something we knew nothing about. There was so much we knew, from day one, including because of SARS, the previous almost-pandemic that was also a similar coronavirus. We could have used that vital pre-information better if we had matched the requisite epistemic humility that a pandemic requires—an acknowledgement that we aren’t certain of anything—with an insistence that this situation wasn’t a blank slate. We could have utilized our prior knowledge to plan ahead, while not letting go of the uncertainty—taking thoughtful steps but without overstating our confidence.

As the Harvard epidemiologist Michael Mina writes, COVID isn’t HIV—a virus that did actually force society and health experts to rewrite the playbook. If anything, the way coronavirus has progressed has been remarkably unsurprising. As Mina wrote:

The constant drumbeat of “we do not know that yet” is tiring. We know so much about SARS-CoV-2 and Covid-19. We knew it before this virus was ever discovered! We’ve watched since January with study after study reaffirming our expectations of this virus in so many ways.

In many ways, we got lucky on this front. Take HIV for example. HIV was a new virus for which we generally did have to rewrite the textbook. But this virus is different from HIV in that it is behaving in almost all ways per the “textbook.”

Mina lists a bunch of things that we should have been prepared for, things we should have been not surprised by. Of course the disease would become a pandemic once the virus got out of China. Yes, masks would help prevent the spread—it’s a respiratory disease! Yes, antibody immunity wanes (as it does for acute respiratory viruses) relatively quickly, but we still might have immunity for some time as we know that is not the only mechanism. And so on.

But here’s the myth that has been rankling me most: the idea that the virus wasn’t going to be seasonal because…it didn’t die down in summer.

Here’s an (typical, from my recollection) example of how this was presented

The coronavirus pandemic is not affected by seasons like the flu, and – despite public pronouncements of a "second wave" as hot spots shift and case counts fluctuate in the U.S., Europe and across the world – its spread is only continuing to accelerate, according to the World Health Organization.

"The season does not seem to be affecting the transmission of this virus currently," WHO spokeswoman Margaret Harris said at a press conference on Tuesday.

Some health experts had hypothesized that the spread of the virus could slow in hotter weather, but its movement has continued during the summer. The world now has more than 16.7 million cases of the virus and over 660,000 deaths, according to Johns Hopkins University.

Here’s another (to my recollection, typical) warning not to expect seasonality:

GENEVA (Reuters) - A World Health Organization official on Tuesday described the COVID-19 pandemic as “one big wave” and warned against complacency in the northern hemisphere summer since the infection does not share influenza’s tendency to follow seasons.

Margaret Harris repeated that message in a virtual briefing in Geneva. “We are in the first wave. It’s going to be one big wave. It’s going to go up and down a bit. The best thing is to flatten it and turn it into just something lapping at your feet,” she said.

Pointing to high case numbers at the height of the U.S. summer, she urged vigilance in applying measures and warned against mass gatherings.

“People are still thinking about seasons. What we all need to get our heads around is this is a new virus and...this one is behaving differently,” she said.

To put all this in context, let’s remember that the president was often insisting that the virus would go away in warmer weather:

President Donald Trump said, “And by the way, the virus—they’re working hard—looks like by April, you know, in theory, when it gets a little warmer it miraculously goes away. Hope that’s true. … We only have 11 cases, and they’re all getting better.”

Given this context, I sympathize with people who insisted that we should not count on seasonality to alleviate the situation. In fact, this has been an ongoing challenge: if a dire forecast did not materialize because the public started behaving differently, there were claims that epidemiologists had been doomsaying. If the public received slightly better therapeutics and clinical management, reducing initial fatality rates, critics started claiming the rates had been exaggerated, that this virus was no big deal. 

Public health people have faced a frustrating conundrum: How do we work, how do we take sensible steps, when every shred of success, every shard of luck, and every improvement becomes a weapon against reality and against sensible precaution? It’s hard.

There were good reasons to assert that seasonality would not save us from a novel virus, that the virus could keep circulating in summer months, since so many people were still susceptible. But there were also a lot of reasons to warn that we could still be facing a terrible fall/winter surge. Elected officials and health leaders could have simultaneously insisted that the public needed to do everything possible to both take advantage of the relative relief offered by summer and also be appropriately aggressive over the summer combating the right targets—crowded indoor events—so that we could approach the colder weather from a much better baseline,which is  essential when facing exponential threats. We could have been in a far better position--with fewer cases, yes, but also in terms of health and psychology.. 

What would that have looked like? Holding onto a prior, reasonable assumption concerning seasonality but also not downplaying the uncertainty? Communicating the threat of a fall/winter wave—which is now happening—but also fighting both the top-down, political misinformation coming from the president and the campaign to cast doubt on every aspect of public health? How could we have encouraged the public to utilize any gains in public health without downplaying risks? 

Everything I say going forward is with the benefit of hindsight (though some of it I wrote at the time, too), but I think it’s important to think through all this, since our battle with the virus is far from over—and future threats await.

First, we should have encouraged people to take advantage of the weather. That would have meant encouraging people to take as much advantage of the outdoor opportunities as possible, rather than scolding people for going to the beach or the park. It would have meant keeping parks open while shifting all the indoor activities to outdoor environments, since we knew the virus was spread much easier inside. Even when cases were down, we should have shut down the gyms, the restaurants, the pubs—and kept them closed, with proper economic support, of course. Instead, many European countries not only opened all those establishments up, the UK even provided vouchers to encourage people to  eat in restaurants, including indoor ones! That was madness. 

Second, and crucially, an effective response would have prioritized scheduling as many of the postponed medical treatments as possible. That would have allowed us to reduce the indirect damage the pandemic will do because packed hospitals were overrun in spring, preventing the treatment of other diseases and illnesses.  Hospitals will be overrun again in many places until next spring as a result of our earlier failures. We should have taken maximal advantage of the brief period during which they could resume functioning.

I had a personal need for this type of prepared medical decision-making. This year, I needed a minor-ish surgery, a follow-up to the first one I had last year because I foolishly lacerated my flexor tendons and a nerve in one hand. It was technically a little early in the summer to get the surgery. The guidance suggested that I should have waited at least until the first-year anniversary of the initial surgery. But I had a feeling that elective surgery would get shut down again—which is indeed already happening in many places. It was fine in my case to get the procedure a little prematurely, but success rates for that particular surgery plummets greatly if delayed. I scheduled my operation for the summer. The surgery did result in remarkable improvement for my hand function—something I may never have had the chance to benefit from had I waited. 

All this felt similar to how I had pushed to schedule much of my preventative medical care in late January, early February, warily watching Wuhan, thinking that getting treatment had to happen then, lest it not happen at all.  None of these were extraordinary decisions; a pandemic was clearly headed our way. It made sense to get that eye appointment and the annual dermatology check-up out of the way. The tragedy is that this was not general guidance for the public. Similarly, hospitals were operating almost normally again in summer, which meant we should have had a national plan to push ahead with elective surgeries in order of importance: putting the ones with most implications for mortality and morbidity ahead of the rest, and doing as many as possible. I know people who were told, instead, to wait until late fall for important medical interventions. They may now never get the chance to receive treatment at all.

Third, we should have made it a priority to encourage people to recharge, psychologically, as much as possible over the summer. Instead, in a lot of places, we acted like we were still in the worst phase of the epidemic. Around the country, there were places where the virus wasn’t circulating much and the weather was suitable for outdoor activities.  But people were not encouraged to let themselves breathe a bit. Even in places where case numbers in the summer were higher, we should have properly targeted the precautions.  

Here in North Carolina, picnic areas and playgrounds—where people could gather instead of being cooped up in air-conditioned homes, which are far more dangerous--remained completely closed over the summer, wrapped up in alarmist tape. When these spaces were eventually opened, in early September, gyms and fitness centers were simultaneously opened—even though the latter should have remained closed for indoor activities as we enter the fall/winter months. And North Carolina has had relatively sensible measures compared with most states! 

I know people who tell me that they have rarely left their house since the beginning of this epidemic. They’ll now be entering the winter phase without having enjoyed the outside over the summer at all. Conversely, I also know people who spent the summer doubting the risks—because they saw all the hyper-alarmism over beaches and parks and because they kept hanging out with people and did not get infected—and may well continue acting the same way, assuming what worked for them during the better weather, at lower baseline, will keep working in colder weather. 

No doubt: some of this is Monday Morning quarterbacking. I’m not claiming that this was at all uncomplicated to navigate. There was no easy, obvious messaging path given the terrible information environment and the political set-up, especially in the United States.

But we can still learn from these failures. We can acknowledge that proper epistemic humility shouldn’t mean a complete surrender to blank-slatist assertions of complete uncertainty. This is not the first, or last, time we needed to balance the unknown with the need to act, the need to warn with the need to allow for release, and the need to use what breaks we get to increase our resilience for what we may yet face.

Programming note:

Apologies for a break in updates for Insight! In just the past week, I have wrapped up teaching and learned that my landlord is still trying to evict me (Nothing to worry yet: It’s not because I’m failing to pay rent,  but because my absentee ex-pat landlord appears to has some other plans he wants to ram through despite my lease. I’m well-lawyered and will definitely update everyone as it’s now turning into a comical drama involving courts that keep getting shut down because of COVID exposure, and perhaps a test-case for tenant rights during a pandemic). I also had to deal with a round of coronavirus testing myself ( all the tests have already come back negative but there was an unexplained minor fever that passed through the house, which gave me a chance to get reacquainted with our testing infrastructure!) And some other minor hiccups. And an election which creates some wrinkles! Back to our regular updates.


For The Atlantic, I wrote an article urging people to take seasonality and the surge into their calculations, essentially upping their cautions as we enter the fall/winter season. One of the more practical considerations for everyone is to reconsider their masks if they need to be indoors with people:

It might also be time for ordinary people to consider using higher-quality masks (N95s and KN95s)—something that public-health experts have long recommended. This is especially true for low-wage workers, a disproportionate number of whom are people of color and have to work indoors; older people, and anyone who works with them; and people with preexisting conditions that put them at higher risk. Ideally, we’d have a significant aid package, allowing businesses to remain closed and workers to stay home as much as possible, while also increasing workplace standards through better ventilation and masks. Tragically, that doesn’t seem to be in the cards.