Since the beginning of the Coronavirus pandemic there has been a myriad, a plethora, or better said a shit-ton of papers on the topic. It comes in all sorts and shapes, with varying quality and utility. Given its immediate importance there have also been a lot of discussion (and quite a bit of shouting) on their interpretation and consequences for public health policy. One prominent example is around the epidemiological concept of herd immunity, which at some point was suggested as a “strategy” to tackle the epidemics in countries like the UK and Sweden.
To be clear, I am strongly in favor of distancing measures and – in the absence of absolutely flawless test/tracing – lockdowns, It’s the only real tool available. I also strongly oppose herd immunity as a “strategy”, and think we are likely not close nor should count on it. That’s a strictly epidemiological opinion. I have my own personal opinions about the impact on the economy and society, the protests, and other non-public health aspects of the pandemic, but that’s beyond my expertise and won’t get into that.
That said, I believe there’s very little debate of actual scientific questions around the pandemic, including herd immunity. For the most part there’s the stuff we already knew, there’s straightforward application of fundamental principles, and then there’s the crazies. The crazies are people that for whatever reason, usually political, will seek out information that confirms what they wish to be true, and don’t care about (and are often not qualified for) a rational discussion of the body of knowledge available.
Now because of the crazies, many prestigious scientists/instant subcelebrities are doing basic shit so we can justify stuff we already know: avoid contact with infected people, wear masks just in case, keep track of transmission routes, incidence rates, etc. That’s not the reserach topic of mathematical epideimologists. That kind of work should probably be done by the WHO or national public health agencies like the CDC of deparments/ministries of health (or commissioned by them), if they had dedicated funding. Maybe the UK is who has something closest to this (SAGE), and maybe that’s why some early work on epidemic forecast came from there.
The British reports in late March were basic modeling, reused some code and then provided some initial estimates of intervention impact. It’s not brilliant work, but it’s what was needed at the time. I argued against unfair criticism of this work and the crazies. Also, at this point other herd immunity studies were taken out of context, while some apparently respectable researchers decided really try to get data that showed it was close, but that’s not novel science either.
At this point, however, it’s been nearly 3 months, and we need better assessments of the epidemic, so novel means whatever helps us “solve” this as fast as possible. The flip side is that there’s also the need not to feed the crazies and cause worst case scenarios (see Brazil) 9/n
That has led to things like this nonsense, and the The Lancet herd immunity assessment, which does no more than say the percentages of people positive for Coronavirus antibody is lower than the 70% expected from the most simplistic models used so far.
We published work that showed a potentially more optimistic scenario (how optimistic depends on the quantification of parameters that are still mostly unknown), and I knew it would probably be used by the crazies, so I tried to be as nuanced as possible about its context. However, it’s been really hard to get past the standard narrative/don’t feed the crazies polarization so somethings just don’t go into the discussion. Happens all the time in science, just now it’s more deadly. The bigger problem is (at least some fraction of) people are not stupid, and even if they don’t have scientific training they can see that some work gets more attention for no reason, and it’s usually what fits the preferred narrative. That undermines the credibility of science and scientists as a whole.
Another byproduct of the urgency of this situation is that it’s a great time to pad your resume with publications in unwarrantedly-named high-impact journals (Science, Nature, the Lancet, NEJM, PNAS), as long as you already have currency with the community and editors, and pat yourself in the back with how important your research is. All of that without the needed to actually tackle the most important/useful questions for the pandemic policy. That also always happened, just now it’s more fucked up.
So what are the most pressing scientific issues and debates around the mathematical modeling of Coronavirus/COVID-19? The same as before the pandemic, as long as they are also useful for guiding the best pandemic response in real time. But we are not debating those, we’re just doing the work we already did, building our resumes, and letting the power structures keep science boring and dominated by the same people. It’s just academia being the projection of the rotten structures and incentives also seen in the society in general.
-- caetano,
June 14, 2020