Fake Charity Nerd Girl

resist microstructural oppression

Posts tagged #coronavirus cw

worldoptimization

No matter what changes the coming years bring,
As the leaves of our youth fade from green,
I am sure we will always remember the spring
When we fought about COVID-19.

We all became experts on MERS and on SARS,
Debated the hammer and dance.
Quoted from studies that found CFRs,
Weighing sample size, bias, and chance.

For others the year has brought lockdowns in vain,
Or loved ones in hospital beds.
We happened to live somewhere it was contained
So we argued on Twitter instead.

And they’ll bury the dead when it’s over at last,
As the streets fill with laughter and song.
Over time our disputes will fade into the past,
But we’ll never admit we were wrong.

worldoptimization
worldoptimization

ok dumb question what is the best current research on long-term effects of COVID?

I haven’t really been following COVID discourse much for several months but like … we have enough data by now to know like what percentage of people by age group, etc are experiencing various symptoms six months later, right?

googling has so far failed to turn up anything concrete

worldoptimization

ok, reading through this article:

  • a study of people who were hospitalized in Italy showed that 87% had at least one symptom 2 months later. this is just people who were hospitalized though, which accounts for 15% of diagnosed cases, and a lot lower in younger people
  • another study in Ireland found half of people having fatigue 10 weeks after being infected, but I think this has similar selection effects (it looks like participants were recruited from people who were attending a post-COVID follow-up appointment?)
  • the COVID Symptom Tracker app in the UK found that 2% of people were still experiencing any symptoms after 90 days. that seems pretty different from the above two numbers! I’m not sure how representative of a sample their userbase is, but they do have 4m users which is a lot. I can imagine it being biased toward young people, but at least that tells us that the rate of long COVID is pretty low in young people.

looking back at a spreadsheet I made back in March, my guess was that DALY costs from long-term effects of COVID would be 2x DALY costs from deaths. if the 2% number is real, then I feel like 2x is a reasonable upper bound, and probably significantly too high; either way, seems like much lower than would justify the costs of lockdown.

@correct-conflagration

But see: https://mobile.twitter.com/ahandvanish/status/1313973286364229638. tl;dr the app may be significantly underreporting because it’s clunky & people get tired of reporting symptoms

worldoptimization

hm ok, I’m having a hard time confirming this but it sounds like the 2% number isn’t “98% reported that they were symptom free after 3 months” but “2% were still bothering to put their symptoms in an app every day after 3 months,” which seems pretty different :/

this spreadsheet (from the twitter you linked) has what seems like a reasonable compilation of studies on long-term COVID symptoms and the studies seem … mostly not that helpful?

out of the 21 studies listed, only 6 seem to be on potentially representative samples (the rest are on hospitalized patients, or recruited from support groups, etc). out of those, two are < 1 month, and one was a sample of fifteen people. looking at the remaining ones:

  • one was a news article in French citing a researcher in Switzerland who said that one in three patients was still having symptoms two months later. didn’t get any more details than that
  • one was a study from Wuhan of 153 patients, finding that 11% still had symptoms after 3 months. seems ok though not very big, not really controlled, and they warn there might be bias in who responded to their survey
  • the most promising one looks like a preprint posted a few days ago (!) it looks at 233 people in the US who tested positive for COVID, comparing to a large group of the general population. it finds that 24% of COVID patients have at least one symptom after 90 days, compared to 6% of the control group reporting one of the same symptoms lasting for 90 days. there were only a few symptoms that were significantly more prevalent in the COVID group than the control group (loss of smell and taste, headache, and rapid heartbeat) though probably the sample size isn’t big enough to say that much about individual symptoms

so yeah, idk. this definitely updates me higher on the number of people who still have symptoms after 3 months: seems like 20% is a better estimate than 2%. mostly though, the state of the evidence seems surprisingly bad given that COVID’s been around for a while.

worldoptimization
worldoptimization

ok dumb question what is the best current research on long-term effects of COVID?

I haven’t really been following COVID discourse much for several months but like … we have enough data by now to know like what percentage of people by age group, etc are experiencing various symptoms six months later, right?

googling has so far failed to turn up anything concrete

worldoptimization

ok, reading through this article:

  • a study of people who were hospitalized in Italy showed that 87% had at least one symptom 2 months later. this is just people who were hospitalized though, which accounts for 15% of diagnosed cases, and a lot lower in younger people
  • another study in Ireland found half of people having fatigue 10 weeks after being infected, but I think this has similar selection effects (it looks like participants were recruited from people who were attending a post-COVID follow-up appointment?)
  • the COVID Symptom Tracker app in the UK found that 2% of people were still experiencing any symptoms after 90 days. that seems pretty different from the above two numbers! I’m not sure how representative of a sample their userbase is, but they do have 4m users which is a lot. I can imagine it being biased toward young people, but at least that tells us that the rate of long COVID is pretty low in young people.

looking back at a spreadsheet I made back in March, my guess was that DALY costs from long-term effects of COVID would be 2x DALY costs from deaths. if the 2% number is real, then I feel like 2x is a reasonable upper bound, and probably significantly too high; either way, seems like much lower than would justify the costs of lockdown.

worldoptimization

ok dumb question what is the best current research on long-term effects of COVID?

I haven’t really been following COVID discourse much for several months but like … we have enough data by now to know like what percentage of people by age group, etc are experiencing various symptoms six months later, right?

googling has so far failed to turn up anything concrete

worldoptimization
slatestarscratchpad

Somebody poke holes in this argument for me:

Suppose US lockdown will last one year before a vaccine or something else solves coronavirus.

And suppose US lockdown will save 1,000,000 Americans who would have died without lockdown. And each death costs 10 DALYs, because they’re mostly old people with only 10 years left to live. And there’s an equal amount of non-death disability, so total 20 million DALYs lost. I’m not 100% sure of any of these numbers but I think they’re the right order of magnitude and if anything skewed towards being overestimates.

Lockdown affects 300 million Americans. So it’s net negative if it costs them more than 1/15th DALY each, ie if one year of lockdown is less than ~94% as good as a year not on lockdown.

But it seems like a year on lockdown is less than 94% as good as a year not on lockdown. Therefore lockdown is net negative.

worldoptimization

yeah I think this is basically right

worldoptimization
worldoptimization

My company tries to hire the best person we can find for any given job. And it just so happens that a lot of the best people turn out to not be American. Add to that the fact that we’re running a global business marketing to customers all over the world where local expertise can be useful, and you end up with employees from all over: US, Canada, Australia, Europe, Japan, Hong Kong, mainland China, etc.

And all we want is a place in the world where we can all stay for an extended period of time and work together, and it seems crazy that it’s so hard! Immigration laws are weird.

worldoptimization

Like things that have happened in the four months since I moved to Hong Kong:

  • walking to work and passing smashed and burned-out storefronts from the previous night’s riots
  • my coworkers getting tear gassed
  • having to get my temperature checked every time I want to enter my apartment building
  • apparently there’s a toilet paper shortage now

and yet Hong Kong is still just clearly a better place for us to be running a business than the US, on many dimensions. I’m impressed how badly the US has fucked this up, and surprised by what proportion of world-changing startups are nonetheless headquartered there.

worldoptimization

weird to think it was only a few short months ago when rioting and coronavirus were considered disadvantages of Hong Kong relative to the US

worldoptimization

It’s sort of a weird shift in perspective to read the latest SSC post and see the list of people praised for getting things right: all Westerners, many of whom started tweeting about it in mid-February or so, all seemingly distinguished by their ability to do cost-benefit analyses and reasoning under uncertainty.

When like, you know who else got it right? The entire city of Hong Kong. Back in mid-January, before there was a single case in Hong Kong, my Hong Kong coworkers were already warning me to avoid crowds and wear a mask when I went outside. The city effectively went on lockdown as soon as the first cases were reported, and it wasn’t the actions of a few informed people in the government, either (the widely reviled chief executive was in Davos at the time, and it took a strike by hospital workers to get the government to take steps like “restrict travel from mainland China”).

Idk, it’s obviously not that everyone in Hong Kong has rationality superpowers. I guess they just have a cultural script of “potential deadly pandemic -> start taking precautionary measures” which … sounds pretty reasonable in retrospect but apparently most of the world just doesn’t have.

This is the most cliche takeaway ever from a white person living in Asia but I think it has pushed me a bit towards appreciating the ways in which we can learn from other cultures, etc. Like, sometimes I think that the internet has connected the whole world into one big unified cultural blob–and then I see things from Chinese social media, which has a fuckton of people and its own totally different (and incomprehensible, to me) set of memes and aesthetics and hot-button issues. Hong Kong social media overlaps with Western social media a bit–they’ll both comment on articles in the New York Times or whatever–but it also has its own set of issues and views, and it’s impressive how with all this information traveling at the speed of light it still manages to stay about a month ahead of the discourse on my facebook.

worldoptimization

further updates in “worldoptimization is annoyed about Efficient Market discourse”:

https://www.facebook.com/robert.wiblin/posts/884145899415

this is like, the opposite of complaining about Predictit markets on Bloomberg being wrong. 

the problem there is that there’s just not enough money to make it clearly worth worrying about. that’s not a problem if you think you can predict the S&P 500! if you can do that you can make arbitrary amounts of money, limited only by your capital and your risk tolerance. sure, maybe you, random internet user, don’t have immediate access to a ton of capital. but if you’re smart and motivated, you can find ways to get it.

not just to pick on this post, but I feel like a common reaction I’m seeing is “lol, guess markets aren’t so efficient after all.” when the right reaction is somewhere between being pissed that you missed out on this opportunity (if you did), and being excited because this is an update toward you and other smart people who share your values being able to make lots of money easily

worldoptimization

I haven’t been following it that closely so I’m still kinda confused about how the coronavirus fits into the culture wars

my vague impression is that:

  • social justice-y people are being contrarian by saying that it’s not a big deal
  • tech/libertarian people are being metacontrarian by saying that it is a big deal and we should all be freaking out about it

I’m friends with more of the latter so seeing more of the latter discourse. And even though I think people think we should be freaking out, I’m not really sure what that means–is it jut that we should acknowledge that this is an important societal issue? Or do they think that it’s worth my time for me to take actions that decrease my personal risk of coronavirus?