Thread by @BillHanage

I’ve no objection to working with economists and indeed I have done so during the pandemic! That’s just one of the reasons this article seems to be addressing a straw man. It also misses something really important that shouldn’t be forgotten 1/n nytimes.com/2021/11/19/opi…

Few serious people argue that there are not trade offs between costs and benefits of different elements of pandemic response, but there is only so much a combined model of the sort proposed in the article can do. And at the start of a pandemic it could be actually damaging 2/n

The more complicated a model, the more information you need to make it work. At the start of a pandemic this is typically in very short supply. Remember the arguments about the infection fatality rate that lasted nearly a year? 3/n

That’s exactly the sort of information you need, and it will be uncertain for ages. Exponentials are not kind to those who wait - let alone those who delay action while waiting for certainty about how bad something is exactly, rather than realising it is bad enough 4/n

At a later point in a pandemic there is much more room for these sorts of discussions. And indeed I’d encourage them to facilitate a long term strategy for pandemic management (which I’ve been advocating for since summer 2020). That’s not what we got… 5/n

Instead of a long term plan, as soon as case counts start to drop a rotating panel of usual suspects (most without experience in ID epi) start jumping up and down declaring ‘herd immunity’ is here and the pandemic is over. They don’t seem overly fazed by having been so wrong 6/n

And then something changes (seasons, contact patterns, variants) and cases climb again. None of this is rocket science, and as we’ve been saying since the start it’s a marathon not a sprint. There’s another important thing to remember 7/n

The situation changes over time. The consequences of increasing cases in places with high vaccination rates (esp in older folks) are utterly different from those without such protection. Similarly places vary in the quantity and quality of available healthcare resources 8/n

And so I am anxious that an overly complicated model will only produce endless discussion over how it should be parameterised. And a place for policy makers to hide from hard decisions 9/n

And of course epidemiologists and economists should talk to each other.

Indeed, they already do

10/end

https://twitter.com/CT_Bergstrom/status/1462519368256950272