Thread by @asymmetricinfo

Seeing a lot of this circulating on the right, so let me explain why folks are worried even though it is not literally true that every ICU bed in the country is occupied at the moment.

https://twitter.com/boriquagato/status/1335956092833161218

#1, the big worry is ICU space, not hospital beds, and as you can see from this very thread, ICU utilization is running well above hospital utilization generally.

#2 The constraint on ICUs isn't beds, it's staff. ICU beds are (relatively) easy to build. They're not much good if the only people you have to staff them are the cafeteria workers.

#3 It's true that ICUs can flex to deal with high utilization. But to do so, they have to:

1) Stretch existing workers to do more (potentially compromising care)
2) Recruit workers from other specialties (potentially compromising care)
2) Hire additional temporary workers

Hiring temps is the best strategy. The problem is, it's a good strategy that's hard to implement when a staggering fraction of the nation's hospitals are all having the same problems, requiring exactly the same skills from the same shrinking pool of workers, at the same time.

#4 This is all further complicated by the fact that this is a novel coronavirus, so your workers--the very people you're expecting to help you flex above normal capacity--are falling sick in droves. And staying sick, because ...

#5 One of the qualities that sets covid-19 apart from the flu is its staying power. Of course, some people just shrug it off (my Dad never even ran a fever!). But people who get sick are often out for weeks.

Of course, the patients also stay sick longer, which brings us to ...

#6 The beds don't turn as fast as ICUs are used to. People who get covid-19 can stay very ill for a long time.

That's very worrying because we might be 3 weeks from peak infections, & over a month from peak hospitalizations--falling as beds are still tied up w/Thanksgiving cases

#7 Especially a problem because--contra the frequent conservative assertions--covid-19 doesn't pay nearly as well as other stuff that hospitals could be doing with those beds, so on top of operations overload they're also having a fiscal crisis as elective surgeries crater.

#8 This is a constant theme in this pandemic: the lag sucks. Everything happens with a substantial lag; hospital admins are often worrying, not about the problems they have now, but the ones they'll have in 3 weeks that are already baked into the cake.

When covid cases are growing, you can temporarily score political points by pointing to things not being that bad right at this instant, if you ignore the lag--the hospitalizations and deaths that are now inevitable but haven't yet happened. But this doesn't make any sense.

If you are talking about a rapidly growing epidemic, and your analysis focuses only on the instant, without accounting for significant lags and quasi-exponential growth, then you are not making a serious critique; you are engaged in a very stupid personal hobby.

Now, of course, I am not saying that all the hospitals will topple--many won't, regardless, and possibly transmission has peaked. Epidemic dynamics are hard to model that finely.

But hospital admins should worry that it's still growing, because that's sure what the graphs suggest. Nor are journalists dumb for reporting that hospitals are under severe strain, because they absolutely are, as you can ascertain by talking to them.

#9 Exhaustion is another major worry. It's one thing to surge for three weeks. It's another to surge for three months. Staff get exhausted, they make mistakes, patients die, they themselves are more likely to make mistakes with PPE and get sick--or run out of supplies.

This is something that hospitals have to plan for, and it's never been more challenging, because these aren't regional emergencies, they're national emergencies.

An emergency too many people wish away by staring hard at isolated statistics & saying "I don't see a problem."

There are reasonable arguments about how bad this will be, whether behavior changes will drive caseloads lower or Xmas/NYE will supercharge them, etc.

There are not reasonable arguments that doctors, administrators and journalists are all lying & everything is secretly awesome.

Now I remain agnostic as to whether the next few months will be seared into American memory as the worst medical disaster of all time, or some more modest level of disaster. But these just aren't the right statistics with which to address that question.