Narrowly because it's actually unfair point by point, as you say.
And slightly dishonest, because Scott does in fact entertain ideas that are scandalous to an expected arrival from the NYT. (Though not, I would say, for the reasons they might assume.)
But also more broadly, it doesn't take the opportunity to make intelligible to that arrival his insider's view of these related-but-distinct phenomena (his blog, the Rationalists, "Grey Tribe", the counter-techlash). Which the article predictably muddles.
Perhaps saddest to me, by responding defensively (however understandably!), he reinforces the emphasis in the piece on this sort of culture war stuff. Rather than perhaps channeling this new interest towards the meta-science and history and psychiatry and the like.
Besides close monitoring of people on O2 to adjust amounts and intubate if that time comes, there are potentially other components of supportive care that might alter outcomes, like IV anti-virals (e.g. remdesivir) or anti-coagulants, e.g. heparin. Some anti-coagulants can be given at home, but the strong doses being discussed also come with high risk of bleeding, benefitting from bed rest and close observation. For a good slice of the sickest COVID patients, hospitalization is helpful and hard to replicate at home, even with our limited therapies.
That said, oxygen and heparin at home could be good for a lot of patients. Improvements in prognostication may help distinguish that group in the future.
It's based on 282 deaths among the 320 ventilated patients who either died or were discharged.
But 831 patients were still on ventilators!
It's a snapshot taken too soon.
Estimates of mortality on vents very widely still, partly because of real underlying variation in practices or population, partly because the data is really messy.
Any vent strategy being debated on Twitter (early! late! APRV!) is also being discussed by working pulmonary critical care docs. Judging those discussions or the variations in practice as an outsider is hard. But relative silence on Twitter doesn't equal mindless orthodoxy.
An outright cure at a year strikes me as unlikely, given the record with anti-virals.
But improvements in supportive care (anticoagulants?) and lucky repurposing might win meaningful reductions in morbidity (some permanent) and mortality within months.
I'm not sure how to weight that possibility. (Metaculus has settled on a likely ~5% relative risk reduction for chloroquine, which is no game changer.) But it seems underweighted in general in discussions about different paths to herd immunity.
That framing implies that all the relevant actors have committed to strategies that they'll see through to the end.
Learning from what's happening in Sweden should help guide decisions elsewhere.
The process of comparison is messy and politicized and in some sense always potentially premature, but I don't think for all that it should be forsworn.
My understanding is that there are variable lags up the chain, and some states don't report deaths for months. NYC as it happens is a relatively prompt reporter. Not sure if there is any public information locally before it's reported upwards, e.g. if you could send volunteers to county health offices.
This was discussed fairly widely recently when people were interpreting recent decline in year over year deaths as evidence that COVID responses were saving lives (e.g. less driving). But it was almost all just typical reporting lag.
Sampling uncertainties aside, this is super sensitive to mis-estimating the false positive rate. Which they seem fairly likely to have done, using an estimated test specificity of 100% (with error bars) in one branch of their analysis after getting 30/30 negatives against pre-COVID serum samples. Too small a sample to get that confident in an atypically high specificity, especially if that collection of samples was disproportionately low antibody (e.g. summer blood rather than post cold/flu season blood).
That's to my mind the most consequential misestimation of the test characteristics, but Balaji Srinivasan details more:
Narrowly because it's actually unfair point by point, as you say.
And slightly dishonest, because Scott does in fact entertain ideas that are scandalous to an expected arrival from the NYT. (Though not, I would say, for the reasons they might assume.)
But also more broadly, it doesn't take the opportunity to make intelligible to that arrival his insider's view of these related-but-distinct phenomena (his blog, the Rationalists, "Grey Tribe", the counter-techlash). Which the article predictably muddles.
Perhaps saddest to me, by responding defensively (however understandably!), he reinforces the emphasis in the piece on this sort of culture war stuff. Rather than perhaps channeling this new interest towards the meta-science and history and psychiatry and the like.