Issue 40: In praise of armchair epidemiology

The battle over expertise is one of the hottest fronts of the current culture war. But knowing R0 from R2D2 isn't a sign you've succumbed to the Dunning-Kruger Effect, it's a basic act of citizenship

Today’s briefing is by Andrew Potter (AP), who is an associate professor in the Max Bell School of Public Policy at McGill University.

__

I. A FABLE

Back in early March I took my car to the garage around the corner to get the winter tires switched out for the summer set. Around midday the mechanic sent me a text saying my front brake pads were shot and needed replacing. He also sent a snapshot of the spent brake pads as proof. 

But here’s the problem: I have no idea what new brake pads look like compared to worn out ones. Looking at the photo, I wouldn’t have bet much money on my ability to actually point out which part of it was the brake pad. I don’t know anything about how cars work. 

So I texted him back and asked him whether it was absolutely necessary to replace the pads. “Yes,” came the response. “It’s dangerous to drive with them like this.” It wasn’t that I think the guy was trying to screw me (though for all I know he was), but I wished I understood what decision I was being asked to make, and the reasons behind it. And I wished I knew enough about the problem to understand the risks and tradeoffs that would be involved in not following his advice. Just how badly worn were the pads? And how dangerous was it to drive with them like that, given that I probably wasn’t going to be doing more than city driving for the foreseeable future? 

I had no idea. So I took the expert’s advice on blind trust, got new brake pads, and what was supposed to be a $50 tire change ended up costing me $500. Not for the first time, I found myself wishing that I was something of an armchair car mechanic.

II. A PANDEMIC

In the early weeks of the lockdown, a couple of widely circulated pieces lamented the sudden surge in what they called “armchair epidemiology”. The first piece, a Medium post by a Stanford post-doc named Noah Haber, was a satirical piece entitled “Flatten the Curve of Armchair Epidemiology”. The second, more sober reflection was published in Slate and was written by the science journalist Tim Requarth. But it, too, was a plea to “Stop the Epidemic of Armchair Epidemiology.”

The COVID-19 pandemic has definitely brought a lot of amateurs into the mix. Anyone with a capacity for numbers -- especially mathematicians, statisticians, and economists --  has jumped into the fray, producing their own charts and graphs and making sweeping generalizations about the virus, its likely propagation through the population, and the merits of our public health and political responses. And there is no question that a lot of this stuff has been overconfident, ill-informed, and misguided -- what Requarth calls “superficially convincing but flawed epidemiological analyses of the ongoing global pandemic.”

But then he goes on to complain about amateurs who

 draw conclusions that contradict accepted public health advice. Other casual assessments of the numbers are being used to make uselessly terrifying cases for containing the virus’s spread. Wherever you look, people have become suspiciously comfortable with concepts like R0 and CFRs, which they use to argue in favor of or against widespread social distancing. 

A few things about this passage made me pause. To begin with, my reading of how things have gone during the pandemic is that the one group that has spent the most time contradicting accepted public health advice are… the public health experts. Changing your mind is not a crime, but no one would accuse them of having been consistent.

Similarly, the most uselessly terrifying case I’ve seen for containing the virus’s spread was not some some back-alley Medium post, but the impeccably-credentialed analysis produced by the researchers at the Imperial College London, who predicted 40 million deaths worldwide if the virus were to spread unchecked.

And finally, what on Earth is “suspicious” about someone being comfortable with concepts like R0 and the case fatality rate?

In addition to R0 and CFR, here are a few other ideas or concepts with which I’ve personally become suspiciously familiar, if not comfortable, over the past three months: comorbidity; variolation; human challenge trials; virology; vaccines; PCR tests; serology; transmission heterogeneity; N95; PPE; self-isolation; rhesus monkeys. (Masks, though, I’ve avoided. Life is short.)

These are all things I knew almost nothing about before the pandemic. Do I know enough about any of it to qualify as an expert? Of course not. But nor have I come down with a feverish case of what Haber calls DKE19, (DKE= “Dunning-Kruger Effect,” so hahaha geddit. Requarth invokes the DKE as well; someday I’m going to write something about the professional class’s obsession with throwing “Dunning-Kruger Effect!” at anyone who steps on their lawn, but that’s another axe for the grindstone). 

So I’m not about to make any sweeping claims about how contagious or deadly COVID-19 really is, and I don’t have any grand opinions on how the lockdowns should be rolled back (except on this one issue)

But what this knowledge has given me are some basic tools for understanding, evaluating, and judging what we are being told by the public health experts and the politicians. That is, I’ve become an armchair epidemiologist, and you should too, for at least three reasons: 

First, it will give you a better sense of what it would look like if someone was pulling a con. I don’t think any of Canada’s public health experts are the epidemiologist equivalent to a crooked mechanic, but I don’t think all of them are entirely free of self-interest or ulterior motives either. It’s easier to see those motivations at work if you understand the stakes. 

Second, even if it’s not a con, it helps to understand not just what you are being asked to do (wash your hands for 20 seconds, self isolate for three months, stay away from grandparents, etc), but why.  That is, it helps if you understand the principles and the reasoning behind these requests or orders. It also helps you understand why these requests or orders might change.  

Third, an armchair epidemiologist is better placed to understand the risks that go along with certain behaviours, and the tradeoffs involved in non-compliance. It’s the difference between blindly accepting a mechanic who says “it’s dangerous to drive on these brakes”, or understanding that you probably shouldn’t go on a road trap with these pads, but a summer of trips to the grocery store are probably safe. 

We live in a world where expertise is one of the most highly contested concepts in our public discourse. But if the right can be accused of encouraging the demise of expertise, the left has responded by turning it into a cult. The right treats science as just fodder for a game that is just politics all the way down, while the left fantasises about getting politics out of the act of governing entirely. 

Good armchair epidemiology cleaves to neither side. Instead, it is nothing more, and also nothing less, than an act of citizenship, of taking ownership and responsibility for your own health and safety and that of the people around you. Hope you have a comfy chair, this could take a while. (AP)

___

Policy for Pandemics is produced and edited by Andrew Potter and co-edited by Charlotte Reboul and Paisley Sim (bios here) If you have any feedback or would like to contribute to this newsletter, please send an email to andrew2.potter@mcgill.ca