National disgrace
In this blog’s now 15-year-history, at Waterloo and then MIT and now UT Austin, I’ve tried to make it clear that I blog always as Scott, never as Dr. Aaronson of Such-and-Such Institution. (God knows I’ve written a few things that a prudent dean might prefer that I hadn’t—though if I couldn’t honestly say that, in what sense would I even enjoy “academic freedom”?) Today, though, for only about the second time, I’m also writing as a professor motivated by a duty of care toward his students.
A week ago, most of my grad students were in the Bay Area for a workshop; they then returned and spent a week hanging around the CS building like normal. Yesterday I learned that at least one of those students developed symptoms consistent with covid19. Of course, it’s much more likely to be a boring cold or flu—but still, in any sane regime, just to be certain, such a person would promptly get tested.
After quarantining himself, my student called the “24/7 covid19 hotline” listed in an email from the university’s president, but found no one answering the phone over the weekend. Yesterday he finally got through—only to be told, flatly, that he couldn’t be tested due to insufficient capacity. When I heard this, I asked my department chair and dean to look into the matter, and received confirmation that yeah, it sucks, but this is the situation.
If it’s true that, as I’ve read, the same story is currently playing itself out all over the country, then this presumably isn’t the fault of anyone in UT’s health service or the city of Austin. Rather, as they say in the movies, it goes all the way to the top, to the CDC director and ultimately the president—or rather, to the festering wound that now sits where the top used to be.
Speaking of movies, over the weekend Dana and I watched Contagion, as apparently many people are now doing. I confess that I’d missed it when it came out in 2011. I think it’s a cinematic masterpiece. It freely violates many of the rules of movie narrative: characters are neither done in by their own hubris, nor saved by their faith or by being A-list stars. But Contagion is also more than a glorified public service announcement about the importance of washing your hands. It wants to show you the reality of the human world of its characters, and also the reality of a virus, and how the two realities affect each other despite obeying utterly different logic. It will show a scene that’s important to the charaters for human reasons, and then it will show you the same scene again, except this time making you focus on whose hand touched which surface in which order.
But for all its excellence and now-obvious prescience, there are two respects in which Contagion failed to predict the reality of 2020. The first is just a lucky throw of the RNA dice: namely, that the real coronavirus is perhaps an order of magnitude less fatal than the movie virus, and for some unknown reason it spares children. But the second difference is terrifying. All the public health authorities in the movie are ultra-empowered and competent. They do badass things like injecting themselves with experimental vaccines. If they stumble, it’s only in deeply understandable ways that any of us might (e.g., warning their own loved ones to evacuate a city before warning the public).
In other words, when the scriptwriters, writing their disaster movie, tried to imagine the worst, they failed to imagine a US government that would essentially abandon the public, by
(1) botching a simple test that dozens of other countries performed without issue,
(2) preventing anyone else from performing their own tests, and then
(3) turning around and using the lack of positive test results to justify its own inaction.
They failed to imagine a CDC that might as well not exist for all it would do in its hour of need: one that didn’t even bother to update its website on weekends, and stopped publishing data once the data became too embarrassing. The scriptwriters did imagine a troll gleefully spreading lies about the virus online, endangering anyone who listened to him. They failed to imagine a universe where that troll was the president.
“I mean, don’t get me wrong,” they told me. “Trump is a racist con artist, a demagogue, the precise thing that Adams and Hamilton and Franklin tried to engineer our republic to avoid. Just, don’t get so depressed about it all the time! Moaning about how we’re trapped in a freakishly horrible branch of the wavefunction, blah blah. I mean look on the bright side! What an incredible run of luck we’ve had, that we elected a president with the mental horizons of a sadistic toddler, and yet in three years he hasn’t caused even one apocalypse. You’re alive and healthy, your loved ones are alive and healthy. It could be a lot worse!”
The above, I suspect, is a sentiment that will now forever date any writing containing it to January 2020 or earlier.
Follow
Comment #1 March 10th, 2020 at 12:24 pm
The average person spends an entire lifetime just trying to grow up and quit being such a jerk all the time and then they find themselves faced with this raging 2-year-old in the nation’s highest office.
It’s enough to make you question what it was all worth …
Comment #2 March 10th, 2020 at 12:28 pm
I’m continually amazed at how much you have to offer mathematically. Your readership comes here for mathematics. When, instead, they find politics, it looks (to me, at least) a bit disingenuous, as if you’re trying to take advantage of your mathematical platform to advance your political opinions. It dilutes the effectiveness of your blog. The problem is a significant portion of your non-academic readership has differing political opinions from you, nor do they regard you as a political expert or authority. I certainly don’t. If someone were to mention the name of Scott Aaronson, I would immediately think, “Oh, the computer scientist at UTA who works in complexity theory.” I would not immediately think, “Oh, the well-known political theorist.”
If you want to say political things, why not start a different blog? Or use a different platform?
Comment #3 March 10th, 2020 at 1:09 pm
My sister is in public health and has worked with the CDC quite a bit. She is not a fan of Trump (to say the least). But she tells me it is unfair that he is being blamed for this, because (she says) the CDC was a complete, dysfunctional mess already long before he showed up, and if anything, his fault has been to leave it too much alone.
Take it for whatever it’s worth. You don’t know my sister or me, so there’s no reason you should credit what I say that she says (whereas I trust her a good deal on such things).
Comment #4 March 10th, 2020 at 1:14 pm
Adrian #2: Did you read the post? It wasn’t about “politics” in the conventional sense of “some faraway tribal thing that you gripe about but that doesn’t directly affect your life.” My PhD student is now in quarantine. My other students and postdocs, who were close to him all week, don’t know whether they should quarantine themselves as well. I don’t know if we should continue our quantum information group meetings after spring break. And the central reason for the uncertainty is that no one can get tested, and the reason for the latter is that the federal government no longer has a functioning nervous system for responding to emergencies, but only a Twitter simulacrum of one.
More broadly, though, this blog has never been only about CS. It wasn’t back in 2005, it’s not today, and its title doesn’t promise that. It’s always been about whatever was bothering me most the day I was blogging. Who told you otherwise?
Comment #5 March 10th, 2020 at 1:16 pm
Re: Keister, look up! The header says “The Blog of Scott Aaronson”. It turns out that people are bigger than what you want to extract from them. Maybe *you* come to this blog for mathematics—speak for yourself. If you think that’s all that happens here, you haven’t been reading it for very long.
Comment #6 March 10th, 2020 at 1:19 pm
There is more detail in the March 5 article
“Why the CDC botched its coronavirus testing” in the MIT technology review:
https://www.technologyreview.com/s/615323/why-the-cdc-botched-its-coronavirus-testing/
Comment #7 March 10th, 2020 at 1:25 pm
I am sick and fucking tired of the nonstop Trump bashing. Look, I am well aware that Trump is a buffoon and a national embarrassment, but every bad thing in the entire world is not Trump’s fault. If the CDC has remained incompetent under the 3+ years of the Trump administration, then let me point out that it was very likely every bit as incompetent for the entire *EIGHT* YEARS of the Obama administration — he merely lucked out that this didn’t happen then. OK, everybody hates Trump, I get it. But the non-stop multiple articles that have been in the papers literally EVERY SINGLE DAY since this clown was elected are out of control.
Enough already! This is not news!
Comment #8 March 10th, 2020 at 1:42 pm
Joe #7: Trump deserves all the bashing since he proclaimed himself as a dictator (confirmed by impeachment trial) and constantly sucking attention of the people around the world.
Comment #9 March 10th, 2020 at 1:52 pm
Edward #3 and Joe #7: OK, fair point. By heaping well-deserved blame on the baboons at the top, I certainly didn’t mean to let the CDC off the hook. I can easily believe that the CDC was incompetent and dysfunctional even before Trump, although his installing Redfield (who seems like a yes-man ideologue) as director can’t possibly have helped. But the sort of breathtaking abdication of responsibility that we’ve seen could only happen via multiple layers of incompetence, the primary and secondary and tertiary fail-safes all failing. All it would’ve taken was a president who would pay attention for a few minutes, learn the relevant facts, think about all the lives at stake, and then like a normal person say “this is completely insane! I’ll just instruct the FDA to set aside its ban on independent tests, and then go on TV and urge anyone with symptoms to get tested yesterday!”
Comment #10 March 10th, 2020 at 2:16 pm
I watched the link below the other day, and the situation isn’t as clear when it comes to responsibility.
(that channel is a bit pro-Trump because he’s being tough on China, and their main mission is exposing the CCP).
The other thing is that the so-called “testing kits” are really only a swab with a little vial to preserve the sample. They’re working on ramping those up asap.
But after that you need to ship the samples to a lab, where they replicate the RNA and detect the ones that match the virus.
The real problem according to Governor Cuomo here in NY is that the CDC refuses (at least as of 2 days ago) to let the states contract private labs in order to expend the testing.
Comment #11 March 10th, 2020 at 2:37 pm
I’ve always had a disdain for this modern “wrong expertise fallacy”, e.g. the idea that because someone specializes in complexity theory they aren’t well qualified to discuss politics. It’s a fact that some people are lifelong lovers of learning and critical thought with exceptional reading comprehension and recall, and people with those traits are destined to become experts on a wide range of topics. Long ago I naively hoped that earning a PhD would give me some general authority in intellectual discussions, but I’ve since learned that the ones who are determined to dismiss opinions that contradict their beliefs will do so by any means, with pigeonholing experts as one of their first lines of defense.
Comment #12 March 10th, 2020 at 2:52 pm
Scott, to add to the voices of the other commenters, consider for a second — would there be a course of events in this scenario where you would write approvingly of the actions of the president? A month ago, for example he suggested travel restrictions to/from China (a smart thing in hindsight). The entire media jumped at him for being a racist demagogue, suggesting isolation when only free movement of people can contribute to the solution.
I changed my opinion of him and his opposition, when he first suggested large-scale military withdrawal from the middle east, and many left-leaning people attacked him (some with an anti-war protest history going back to the Vietnam war).
Comment #13 March 10th, 2020 at 3:00 pm
Elizabeth #11: It’s interesting to imagine what must go through the minds of these complainers. “For crying out loud, Einstein and Russell and Bohr and Feynman and Hawking and Chomsky were all vastly more important scientists and thinkers than this guy is, and yet none of them ever shared their opinions about anything else besides … oh no, wait, they did.” 😀
Comment #14 March 10th, 2020 at 3:17 pm
Elizabeth @11: It’s not that I regard Scott Aaronson as unqualified to discuss politics. I merely don’t regard him as an expert in politics. If you think I was saying he is unqualified to discuss politics, you are committing the strawman fallacy. All I was saying is that the value of this blog, for me, is the mathematics, not the politics. There is a difference between someone being an “expert” – an authority who could reasonably be quoted to support your position, and someone merely presenting their own opinions. Moreover, if you are saying that I am “determined to dismiss opinions that contradict [my] beliefs … by any means” then you don’t know me very well. That statement looks very much like the hasty generalization fallacy, to me.
It has been a standard position by logicians throughout the centuries (so, definitely not just a modern idea) that the “appeal to authority” needs to be an appeal to a _relevant_ authority. So appealing to Michael Jordan to sell toothpaste in an advertisement is committing this fallacy. However, appealing to Michael Jordan to sell a particular kind of basketball shoe might not be. It is a fact that you simply can’t assume that because someone is an expert in one area, that therefore that same person is an expert in a different area. It takes a _very long time_ to become an expert in something, to the point of qualifying as an authority. Someone might indeed be well-read in many fields, but it still takes more than that to become an authority.
Comment #15 March 10th, 2020 at 3:19 pm
I called our doctor’s office in Austin, and asked what the protocol was to get a test. She said only in the ER, only if you have respiratory distress. She added other conditions:
Fever
Contact with some known covid case
Recent travel to certain countries
She didn’t seem clear about whether those were ands / ors between those conditions. She finally settle for “or”.
Comment #16 March 10th, 2020 at 3:26 pm
Note that your students are very likely going to be okay.
Those are the death rates by age (percentage of deaths among the ones who got infected, with visible signs):
80+ years old 14.8%
70-79 years old 8.0%
60-69 years old 3.6%
50-59 years old 1.3%
40-49 years old 0.4%
30-39 years old 0.2%
20-29 years old 0.2%
10-19 years old 0.2%
0-9 years old no fatalities
The USA has 50 million ppl over 65, and that’s the main population at risk.
The issue here isn’t about personal safety, but realizing that we all have a duty to limit the spread to others who are at risk.
So the point of testing is really about identifying hotspots and containing the spread, or at least slowing it down as much as possible. We need to avoid overwhelming the hospitals with people at risk who could be in ICUs for weeks.
“The first is just a lucky throw of the RNA dice: namely, that the real coronavirus is perhaps an order of magnitude less fatal than the movie virus, and for some unknown reason it spares children.”
If a virus is too lethal, it won’t spread as much, there’s a sweet spot.
The novel coronavirus is a dangerous mix of being more serious than the flu, with longer incubation time, and transmits very easily.
Also as of 2 days ago the NY gov and NYC mayor were saying that the virus can only survive a few minutes out of a host.
Today they’re saying there’s evidence that the virus could survive for as long as 12 hours on surfaces…
Going back the the politics, it’s undeniable and unsurprising that no matter what Trump does or says, it’s going to be weaponized against him almost instantly.
For many, it doesn’t even matter when he does the right thing, because they think he’s not being genuine, or caring, or he’s just being lucky in spite of his own stupidity, or in the end they just can’t stand him.
Everyone else who’s a bit more realistic/cynical about what *any* president can and can’t do in the end, are getting pretty sick and tired of that constant weaponization.
E.g. when the outbreak became obvious in February, Trump blocked all travel from China. The response from China and the WHO (going out of their way to praise China) was that it was overkill.
China was saying that the US should look at their own flu death toll compared to the flu and coronavirus numbers in China (they happen to under-report flu cases by orders of magnitude because they reclassify them based on existing pre-conditions, which we don’t do in the USA).
And the US media as a whole jumped on the “look at the flu number!” bandwagon that Trump was overreacting by banning travel.
All the focus on Trump missed the point that China is over and over the source of outbreaks you would only expect out of third world countries (the main reason is that the rich minority in China is lobbying to keep “wet markets” alive, where they have access to endangered wild species in order to cure their libido issues).
Now, Trump and many governors (NY) and many mayors (NYC) are saying “let’s not panic, let’s look at the flu numbers for perspective”, and of course that’s again being turned against Trump.
Comment #17 March 10th, 2020 at 3:34 pm
Steve #12: Racking my brain, I can think of two (discretionary) things Trump did as president that I supported:
(1) Rescinding the 2011 “Dear Colleague letter,” which eliminated due process rights for college students accused of sexual offenses.
(2) Complaining about toilets that you need to flush like 20 times before they work, and that therefore waste water rather than saving it.
There were a couple other things, like resisting Huawei or targeting Soleimani, where Trump made a non-obvious call and I definitely don’t know enough to say that it was wrong.
Why isn’t there more? Shouldn’t even a stopped clock have been right more than a few times in 3+ years? Of course, there might be more that I didn’t hear about. But I think there is a force that makes Trump worse than a random number generator in the Oval Office, that consistently biases him in the direction of horribleness. Namely, he seems to follow a heuristic of always choosing the action that will bring the most despair to the acknowledged experts in a given topic, since the experts’ despair is a central source of joy for his base. This is a problem if you think, like I do, that acknowledged experts, speaking about empirically measurable phenomena of which they have expertise, are going to be your best source of guidance at least ~90% of the time.
In the current crisis, for Trump to earn my praise would’ve been as easy as
(1) telling the truth about the magnitude of what’s going to hit us—that this could easily kill millions of Americans, that it’s not just a bad flu,
(2) ordering testing as widely as possible as quickly as possible, and
(3) putting people of demonstrated competence in charge (he could even indulge in one of his favorite activities by firing his CDC director).
If he did anything like the above, though, he wouldn’t be what he is.
Comment #18 March 10th, 2020 at 3:56 pm
Scott #17
I would add helping significantly with very positive prison reforms.
This was an interesting podcast about it
https://samharris.org/podcasts/176-knowledge-redemption/
Sam Harris hates Trump (to say the least), but his guest (Lynn Novick, who co-directed most of Ken Burns documentaries) said that the Trump administration definitely deserves some credit.
Comment #19 March 10th, 2020 at 4:09 pm
Adrian #14: I’m much more radical than you, in that I reject the entire concept of “authorities” in anything. (To invoke the authority of Carl Sagan 🙂 : “in science there are no authorities; at most there are experts.”)
Mostly, for me there are just people whose thinking makes more or less sense to me. And then there are people whose thinking makes sense to me about topic X but not topic Y—as with the brilliant mathematician who refuses to accept quantum mechanics, or the woo-woo mystic with amazing advice about how to relax. The correlation across topics is strong but very far from perfect. Usually, though, if I admire someone about X, I at least want to know what they think about Y even if I end up disagreeing.
Comment #20 March 10th, 2020 at 4:38 pm
Don’t forget to vote
Comment #21 March 10th, 2020 at 5:03 pm
Over the past few days I’ve come across a lot of crazy takes from seemingly intelligent people:
– COVID-19 does not actually exist at all, is driven solely by false positives on tests and psychosomatic illness, i.e. worried people with the common cold
– COVID-19 is not any more deadly than the common cold, high mortality rate driven solely by massive false negatives on tests
– COVID-19 is Democrat/Chinese hoax to crash the US economy (zero cases actually exist in China)
– COVID-19 is completely uncontrollable (100 million cases already exist in China but are being covered up) so we shouldn’t bother trying to control it
– COVID-19 is bioweapon (made by /insert nefarious group/) directed at the US/EU/Hong Kong, which was released in China just to misdirect online sleuths
– COVID-19 overloaded hospitals in China and Italy are just the results of statistical fluctuations during flu season, everybody there just has ordinary flu
– COVID-19 is real and exponentially growing, but it’ll shut off by itself before anything _really_ bad happens
Sometimes people even believe more than one of these at once, despite them being clearly contradictory.
The only thing in common between people with these takes is Trump support. It’s honestly depressing how a little intelligence plus a little motivated reasoning can lead people so far off track. This is why people suggest deferral to experts on everything… clear thinking is just an extremely rare ability.
Comment #22 March 10th, 2020 at 5:24 pm
I’m a bit disappointed, not because this post is political, but because in the way of many political posts, it describes the problem in a top-level, shallow way, and therefore I don’t feel like I learned much, which is unusual for this blog.
Like, I already knew that at Stanford and in Washington state there are independent testing facilities, and the Gates foundation is starting some testing, and that for some reason CDC testing was botched the first time around and things are getting moving very late. I’ve also read that the tests used in China are very bad and often give false negatives until given multiple times, though I’m not sure how credible that is. It seems like higher standards might be a reason for getting moving late? I would like to learn more. How about you?
Comment #23 March 10th, 2020 at 5:36 pm
Throwaway #21
Your list seems to miss a few items.
– COVID-19 is essentially the flu blown out of proportion by racists.
– COVID-19 isn’t all that dangerous and the CCP mass quarantines, violent and illiberal, just show you how important is western democracy.
– COVID-19 isn’t all that dangerous and the calls\ decisions in a number of places (USA, Israel, Singapore) to enact forced quarantines for incoming international visitors are right-wing hysteria.
– COVID-19 is everybody’s problem, ergo open borders are actually a solution rather than a barrier for its spreading.
– COVID-19 just illustrates how you must have government regulation over drug-and-vaccine-producing companies.
The only thing in common between people with these takes is not Trump support. It’s honestly depressing how a little intelligence plus a little motivated reasoning can lead people so far off track. This is why people can sometimes be skeptical about deferral to “experts” on everything… clear thinking is just an extremely rare ability. But wishing to label your side as the ones possessing this ability is extremely common.
Full disclosure: Trump hater just like the rest of us. But not willing to ignore the role government oversight (pun intended) under Obama played in leaving the CDC and FDA not only incompetent but actively harmful. Nor willing to ignore the deep stupidity of too many Democrats believing too many of the takes above.
Comment #24 March 10th, 2020 at 5:59 pm
Brian #22: Honestly, one reason for writing this post was my hope that it might come to the attention of someone, somewhere, who might know who UT Austin students can call right now to get tested. Do you know? What about for other universities (besides Stanford)? Please share! I’m not a public health expert or journalist; I’m a theoretical computer scientist and a connoisseur of horribleness.
A second reason for writing the post was simply to have a record to show my grandkids, that as this preventable human catastrophe unfolded I discharged my moral duty to call it what it was.
A third reason was to work through my guilt about having recruited students and postdocs to my group in the US, when the US (as it turns out) is considerably less equipped to safeguard their lives than their home countries would have been.
Some people don’t like blog posts that cast blame. But I’m more worried about a different question: one month from now, when the entire US looks like northern Italy does today, with all the ICUs overwhelmed and forced to triage—where, in the furthest reaches of the moral universe, will we possibly find a quantity of blame sufficient for those whose incompetence let this happen, those who wasted the months that the US had to prepare?
Comment #25 March 10th, 2020 at 6:15 pm
Thanks for reporting from the trenches. I just had my hopes up after reading about South Korea’s containment through testing and tracing. Well, that hope is gone after reading this.
The sad thing is what people are arguing about: politics. And what leadership is most worried about: stock market. Instead of testing everyone for coronavirus, and containment, they offer: tax cuts until election! bailout for shale companies a day after oil price dropped!
What will it take to restore sanity? 5 congressmen are in self-quarantine right now, will someone finally speak up, admit that we are in a pandemic, that we are not well prepared, but that there is a way out.
I mean, Warren’s already said this, and had a plan (free testing, government supported paid leave). Neither Biden, nor Sanders have an official coronavirus position. This is the true test of leadership, you’d think it’d be a no-brainer to demonstrate what kind of leader you are.
Comment #26 March 10th, 2020 at 7:54 pm
Adrian #14 –
“All I was saying is that the value of this blog, *for* *me*, is the mathematics, not the politics”
I added my own emphasis to the most important words in your comment.
Similarly, the value of http://blogaboutpiesandcake.com/, for me, is the pie posts, not the cake posts. I hate cake. Can you make that guy stop with the cake all the time?
In all seriousness, I thought that there was a pretty well-established tradition of personal blogs that they’re about whatever the blogger wants to blog about. That’s one of the features that distinguish blogs from, say, academic journals, which tend to put a rather prescriptive stake in the ground with regard to topic area.
With regard to Shtetl-Optimized in particular, I’m in the camp of being generally interested in what Scott thinks about various topics.
Comment #27 March 10th, 2020 at 9:13 pm
In Spring 2018 Trump fired the leaders of and disbanded the Global Health Security and Biodefense directorate of the National Security Council, which was the Executive Branch group charged with coordinating response to a pandemic. It had been set up by Obama in 2016 after the Ebola crisis. https://www.snopes.com/fact-check/trump-fire-pandemic-team/
He also “cut funding for the CDC, forcing the CDC to cancel its efforts to help countries prevent infectious-disease threats from becoming epidemics in 39 of 49 countries in 2018. Among the countries abandoned? China.” (Same source; Snopes makes the quibble that he did not cut funding, he did not continue to fund it after the previous allocation expired.)
He appointed a CDC head who has proposed preposterous approaches toward public health in line with his right wing and fundamentalist Christian beliefs: https://www.motherjones.com/politics/2020/03/trumps-cdc-director-has-a-history-of-controversial-opinions-on-controlling-viruses/. He also put Pence as the head of the coronavirus response, who also completely botched his handling of HIV public health work as governor of Indiana, https://www.politico.com/news/magazine/2020/03/02/how-mike-pence-made-indianas-hiv-outbreak-worse-118648. Obama in contrast appointed someone with real scientific expertise: https://en.wikipedia.org/wiki/Tom_Frieden
Those who want to say ‘this is all Obama’s fault’ had better reckon with the above facts.
Comment #28 March 10th, 2020 at 9:51 pm
Meh. I have trouble taking the anti-Trump spin on this seriously given the system-wide failure we are witnessing. It’s not just Trump, it’s the democrats, the media, local officials, and most public health experts and officials. Nearly everyone wasn’t taking this sufficiently seriously until two weeks ago, and most of them still aren’t.
If anything, I see this as another symptom of the underlying civilizational rot that gave rise to Trump in the first place. I just hope we can get things together on this before too many people die.
Comment #29 March 10th, 2020 at 10:09 pm
Jonathan #28: Scientific expertise matters. Trump doesn’t believe in it and dismantled the teams and leadership that should have taken charge of this, that had the expertise to know from the beginning how seriously to take this and would have started from the beginning to coordinate effective responses. Not believing in science or expertise or facts matters when nature presents a crisis. It is not a general societal failure.
Comment #30 March 10th, 2020 at 10:33 pm
A back of the envelope calculation from a friend: the US currently counts 990 confirmed cases and 29 deaths. The death rate from the disease seems to be around 1% and it takes about 3 weeks to die from it. So those 29 contracted the disease at least 3 weeks ago. Suppose they all contracted it 3 weeks ago. Then they would be around 1% of the number with the disease at that time, meaning around 2900 cases 3 weeks ago. The doubling time of the infections has been said to be about 6 days, you could imagine it varies with region according to sparsity of the population, but let’s posit 1 week. So that means right now there should be around 8700 cases, of which only a little over 10%, 990, have been detected. That also means, without very effective public health measures to slow things, which this administration does not seem interested in, in 4 weeks we’ll have around 140,000 cases and if it keeps going we’ll have crossed 1 million by 7 weeks, around April 28. Around 10% of cases need ICU treatment.
What’s our hospital capacity? “According to the AHA 2015 annual survey, the United States had 4862 acute care registered hospitals; 2814 of these had at least 10 acute care beds and at least 1 ICU bed. These hospitals had a total of 540,668 staffed beds and 94,837 ICU beds (14.3% ICU beds/total beds) in 5229 ICUs.” Hospital and ICU’s both normally have about 65% occupancy, so around a third of those beds might be available for coronavirus infections. Say, 35,000 ICU beds, so when we cross 350,000, in around 5.3 weeks, we run out of ICU beds.
Reducing the above numbers doesn’t buy you a lot of time. For example, if the death rate is 3%, add Log[3]/Log[2]=1.5 weeks.
To slow this thing down, in addition to shutting down many or most venues and situations in which groups of people mix, you have to do enough testing to identify a high percentage of the cases, quarantine those you find, go out and test all of their contacts, quarantine all of those who test positive, and so on. That is what an effective public health response would do, and what the effective responses in China and other Asian countries have done. The CDC isn’t even at the point where we can run enough tests to find most of the infected.
Comment #31 March 10th, 2020 at 10:37 pm
Scott,
it is scary to hear about your student’s experience, and I hope things turn out OK.
Covid-19 started to spread in earnest during the peak of the hiring season in math; it is awful to contemplate what choices grad students and postdocs must need to make in these times. Go for that conference or interview, and risk either quarantine/isolation/infection; or stay put, and miss out on time-critical opportunities. In the grand scheme of things this may be a small concern, but it’s a year’s worth of sparse academic opportunities for people who have worked really, really hard.
Comment #32 March 10th, 2020 at 10:38 pm
My reaction to this post, before reading any of the comments, was, why aren’t there more posts like this from more people? Isn’t it obvious that the Emperor is naked? Why don’t more people notice it, or if they do, speak up? Thank you, thank you, thank you.
And yet, there are people who apparently read math and science blogs who will defend Trump. It seems like I’ve known what kind of person Trump is most of my life: serial adultery, racist acts, went bankrupt building a casino, and you only had to watch two episodes of “The Apprentice” to see he was also a blowhard who surrounds himself with yes-men. Since then I’ve found out even worse things about him.
So if people are tired of hearing how bad a person they have as President, maybe they should have voted for somebody better, which is a very low bar. Until Trump is out of office he’ll be doing something that deserves censure almost every day, and get far less than he deserves.
Comment #33 March 10th, 2020 at 11:43 pm
Ken #30: Thank you—that’s one of the clearest analyses I’ve seen of how this catastrophe is going to unfold.
The crazy part is that, in the total absence of credible guidance from the CDC or any other arm of government, we’ll each just need to study the statistics, reconcile conflicting projections, and make our own decisions about when it’s time to get ourselves or (especially) our at-risk relatives into quarantine.
Comment #34 March 11th, 2020 at 12:57 am
Hi Scott!
I have decided to stop lurking on your blog for this. My friend and I have spent the past week or so basically only thinking about COVID-19. Here is a summary of our thinking as of 3/8: https://docs.google.com/document/d/1T0j2NP2ZMFwJfrfAiBJlro-ivaYLI6Spi3oIdKg52sQ/edit?usp=sharing
[Note that a lot of the information about number of cases in the US is outdated, but the methods used to make estimates should still be sound].
Some relevant excerpts:
TL;DR: COVID continues to spread virtually unchecked in the vast majority of the world. Countries like China and Singapore demonstrate that decisive governments can contain the spread. The US government is far from decisive. Taking precautions slows the spread, potentially saving many lives. This suggests that you should take precautions, even if you are a low-risk individual that won’t personally be harmed that much by COVID.
…
Due to the lack of testing in the US, it is pretty important to be able to estimate the true number of local cases. Here are 3 ways that I’ve been doing it (note that having multiple ways to estimate is important because each of them has different flaws – in theory, the average of 3 estimations using different methods should be more accurate than any single estimate):
1. Doubling time estimate. The mean time between getting infected and becoming symptomatic is 5 days [1]. Eyeballing graphs, the time from infection to confirmed infection is about 10 days [2]. The doubling time is about 5 days [3]. This means that there are about 2^3 = 8 times as many cases in the area as there are reported cases, adjusting for specific information about how many people you think are being tested.
In the US, LiveScience [4] gives 564 infected, so this would imply that about 4.5k are currently infected in the US.
2. Death rate estimate. The true death rate of COVID is about 0.5%-1% [5]. If the local death rate in your area, which can be calculated as # dead / (# dead + # recovered), is substantially larger than this, that probably means there are many unconfirmed cases around.
Confusingly, none of the data about the US makes sense. As of 3/9, there are roughly 22 dead [4], but JHU [6] only counts 8 recoveries? This suggests that the death rate is 22/30 = 73%, so the number of confirmed cases in the US is 73 times lower than the number of actual cases. LiveScience [4] gives 564 infected, so this means 40k infected? These numbers are probably wrong because most of the deaths are from the outbreak in Washington, which infected a lot of very high-risk individuals. I am continually frustrated at the lack of available data in the US.
3. Death number estimate. If the true death rate of COVID is about 0.5%-1% [5] and 1 person has died, by statistical law there were, on average, 100-200 people that got COVID around the same time as the person who died (with some adjustment based on the how at-risk that person was).
The US has 22 dead [4], so there are 2k-4k infected. This estimate roughly lines up with the first estimate, which is slightly reassuring.
[1] https://onlinelibrary.wiley.com/doi/full/10.1111/tmi.13383
[2] https://www.who.int/docs/default-source/coronaviruse/who-china-joint-mission-on-covid-19-final-report.pdf
[3] https://en.wikipedia.org/wiki/Template:2019%E2%80%9320_coronavirus_outbreak_data/WHO_situation_reports
[4] https://www.livescience.com/coronavirus-updates-united-states.html
[5] https://cmmid.github.io/topics/covid19/severity/diamond_cruise_cfr_estimates.html?fbclid=IwAR0RxX5CfdYC3pyXuDTKxm8uKvKmHCZyZDex7mrG6Sn874y6trni8XnD8pg
[6] https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6
Comment #35 March 11th, 2020 at 1:37 am
Hi Scott, I’m writing from a lockdown in North Italy. All previous attempts to contain the virus were totally completely useless here. Even after three days of total lockdown we still see no reduction from the previous exponential growth (due to the delay between infection and symptoms). Our government is traditionally very incompetent, but they did the right thing for once. Our public health system works great and doctors are doing heroics right now. It’s sad to see the downfall of USA (my home for many years). Good luck from the deepest of my heart.
Comment #36 March 11th, 2020 at 2:32 am
Scott:
Serious question: Are you trapped in the US? If you value freedom, here’s a list of fourteen countries that are freer than the US. I’m sure you also value other things, like proximity to family, friends, and people interested in quantum information theory. But maybe you can persuade some of them to come here to the freer world, too.
fred #16: Mortality rates might not be that bad. Still serious enough to be worth paying attention to, especially for those in the potential-US-president age-bracket, but low enough to be at least a little less of an excuse for Scott to despair.
Comment #37 March 11th, 2020 at 3:19 am
Honestly, Scott, what’s happening in Switzerland, France, Germany is much worse. The politics and the media are depicting as demagogic/populist/fearmongerers the one who complain about the very things that you complain about. Very few academics dare to raise their voices.
It’s a much easier task to (rightfully) complain about Trump policies while being in academia than to complain about the same policies in a country run by articulate “experts”. Actually, if you could take a few minutes to look at what is happening in Europe (outside Italy) and acknowledge on this blog that the policies given the situation are basically the same or worse, it would help a lot. Like really a lot. I’m desperately hoping that our universities close.
Best,
Clément
Comment #38 March 11th, 2020 at 4:00 am
Ken #30: If the doubling time is one week, and there were 2900 cases 3 weeks ago, then today there are 23200 cases (2900*2*2*2), not 8700. There will be 185,600 cases just 3 weeks from now.
Applying the same template to Italy:
Italy, on February 28th counted 888 confirmed cases and 21 deaths. The death rate from the disease seems to be around 1% and it takes about 3 weeks to die from it. So those 21 contracted the disease on February 7th or earlier. Suppose they all contracted it on February 7th. Then they would be around 1% of the number with the disease at that time, meaning around 2100 cases 3 weeks ago. The doubling time of the infections has been said to be about 6 days, you could imagine it varies with region according to sparsity of the population, but let’s posit 1 week. So that means on February 28th there should be around 16800 cases, of which only 5.3%, 888, had been detected. That also means, without very effective public health measures to slow things, one week later on March 7th Italy would have had around 33,600 (5,883 detected, 17%) cases and if it keeps going it’ll have crossed 1 million 5 weeks after that, around April 10.
Predicted cumulative deaths on March 7th: 42. Actual: 233
Comment #39 March 11th, 2020 at 7:04 am
The issue isn’t just the stock market, it’s the economy in general.
It doesn’t take long for the supply chains to start to collapse (they’re already straining with the production slowdown in China).
Scott #33
your local governor and mayor aren’t giving daily news conferences?
Comment #40 March 11th, 2020 at 7:25 am
Before anyone of us starts to feel a bit too sorry or worried for themselves:
Regardless of politics, let’s keep in mind that the vast majority of Americans don’t have the possibility to flee to Israel or take an arbitrarily long sabbatical. They can’t keep their kids at home away from school or even work from home because they have to show up at work no matter what, to keep the system going and to keep making a living.
Medical workers also have no choice but be around the infected all day long, weeks after weeks.
Comment #41 March 11th, 2020 at 7:42 am
Hi, from Italy (Lombardy).
Here, the doubling time is estimated to be more like 2.5 – 4 days.
Furthermore, even in Italy, with its public health system, testing is far less common than it should. In order to test you, they require that you have BOTH symptoms AND a proven transmission link. But the transmission link cannot really be proven without technology such as implemented in South Korea. I know of at least one person who developed symptoms and called the dedicated regional services: criteria at the time were that you had to have symptoms, and close contact with at least one person which “tested positive” (or better, which you knew was tested positive). “Close contact” did not include working in the same room, but the same DESK was required, for example. One 8-hour train trip with clearly symptomatic people near to you was not a cause of concern. And so on.
The government did almost nothing to prepare for widespread testing, and even now the authorities are downplaying the usefulness of testing, though it seems to be well established that widespread testing and good usage of the testing results to follow the transmission chains (thanks to technology, GPS tracking…) provides very good results without the need of shutting down the economy entirely. We are testing less than China and South Korea with respect to the population, and far less than China and South Korea with respect to the number of cases (even if we have still less cases than China, we have more cases per capita due to our smaller population).
Even in Italy the different amount of testing in the different local outbreaks made a difference – the outbreak in Veneto was contained more effectively.
So what you have to do is not to push for “draconian” measures of containment, which governments will happily follow, but push EXTREMELY for widespread testing. The containment measures are very easy to implement – a law is enough, and the price will be paid by the population. Testing requires preparation and laboratories, which cost time and whose price will be paid by the federal or state balance sheets.
Comment #42 March 11th, 2020 at 9:12 am
Dear Scott, it seems that pro-Trump bots (or paid trolls) have taken an interest in the comment section of your blog. On one side this should make you feel at least a bit proud, if nothing it’s a good indicator that you’re doing something Good. On the other side, oh boy, I’m SO happy I live in Europe 🙂
Take care, be strong, the virus threat will pass sooner or later.
Comment #43 March 11th, 2020 at 10:02 am
Rogan just interviewed Michael Osterholm, a public health scientist and a biosecurity and infectious disease expert.
According to him, washing your hands and not touching your face is alright, but the reality is that the main vector of infection is through breathing (not even coughing or sneezing):
(43 min mark)
https://youtu.be/E3URhJx0NSw?t=2579
Comment #44 March 11th, 2020 at 11:06 am
From snopes.com: “The Trump administration fired the U.S. pandemic response team in 2018 to cut costs.” So, yes, Trump is due a big chunk of the blame. He has also been working very hard to downplay the illness, even though there’s plenty of evidence from China, Italy, and the rest of the world that this is going to be a big problem. There’s definitely no magic here in the US that will protect us. It would be ok if he was simply trying to avoid panic but there’s all the evidence in the world that he and his cronies believe what they are telling people.
Comment #45 March 11th, 2020 at 11:14 am
Scott said in #24
“…where, in the furthest reaches of the moral universe, will we possibly find a quantity of blame sufficient for those whose incompetence let this happen, those who wasted the months that the US had to prepare?”
Out of context this could really be about a massive terrorist attack killing 3,000 people, a massive tsunami followed by a horrific triple meltdown nuclear accident, or uncontrollable gigantic forest fires due to global warming.
It’s human nature that we just can’t be thinking at once about the near infinite list of all the bad shit that can/will eventually happen.
Unsurprisingly, even politicians avoid it – it’s not secret that we’ve been badly prepared for decades on many many things, but even focusing on a single issue is challenging… which one to pick and campaign for?
Had Andrew Yang made noise over the risks of viral outbreaks instead of the dangers of AI automation on the economy, he could be our president by cheer luck!
Comment #46 March 11th, 2020 at 11:46 am
I am fasinated by the fact that no one in this thread has touch the idea that the U.S. should ramp up face mask production, while the face mask supply has been a focal point in the asian countries.
Comment #47 March 11th, 2020 at 12:05 pm
Re-reading what Scott wrote at the top:
“Yesterday I learned that at least one of those students developed symptoms consistent with covid19. Of course, it’s much more likely to be a boring cold or flu—but still, in any sane regime, just to be certain, such a person would promptly get tested.”
the advice is that everyone should deal with it as if it’s covid19, so to stay isolated in any case, testing won’t really change anything:
– either he has the coronavirus, with mild symptoms, in which case he should avoid others, so as not to pass it.
– or he has a “regular” flu, with mild symptoms, in which case he should avoid others as well, because he wouldn’t wanna pass the flu to others or get the coronavirus on top of the flu.
And if he really starts feeling very sick, he should go to the ER, and they’ll decide if he should get tested.
Others who’ve been in contact should also avoid contact.
We’re not gonna be testing everyone who has a cough or a sneeze. We’ve all been told to practice social distancing (unless your job is about dealing with people face to face).
Comment #48 March 11th, 2020 at 12:15 pm
Apparently, if Trump does something wrong, any complaints about him will be met with one of three responses by his troll army:
1) Fake news!
2) Keep crying, liberals. Prepare for four (eight?!?) more years! MAGA KAG 2020!!!
3) Your attacks on Trump are so disappointing 🙁 This is a matter of politics 🙁 Any attacks on Trump are necessarily politically motivated by the leftist media, are not factually based, and by parroting them you are therefore being partisan 🙁
This tier 3 faux-disappointment is meant to appeal to very confused observers. When a person or group is complaining about something repeatedly (even with good reason!), it’s easy to act disappointed at how unreasonable their repeated complaints are (even if they actually are reasonable!) and appear reasonable to an observing third party who isn’t sure of what the facts are. It’s an effective rhetorical tool. I am not surprised the method is being employed.
Comment #49 March 11th, 2020 at 12:21 pm
@Scott, You would have blamed Trump no matter what. Even if the handling of coronavirus was superb in the US you definitely wouldn’t have praised him for it. E.g. the US economy is doing very well under Trump. The liberal establishment doesn’t celebrate that but claims it’s all luck. Then how can you claim that the response to coronavirus is due to his incompetence but not due to chance?
Many countries botched their initial response to coronavirus, including China, Iran, and Italy. Trump isn’t in power in any of them. Also the fumbling here may be due to state or local incompetence. The CDC is manned by career bureaucrats, how is Trump responsible for their failing? Looks like whenever something bad happens you just close your eyes and start yelling Trump.
Comment #50 March 11th, 2020 at 12:21 pm
Clément #37: Ok, let’s have a look at what is happening in Germany that is related to one topic that Scott is adressing in this post, the official communication of the federal administration regarding the Corona pandemic. Today’s press conference with Merkel (chancellor), Spahn (minister of health) and Wieler (director of the Robert-Koch-Institute):
https://www.youtube.com/watch?v=7HXrhgrCj24
(Obviously in German).
TL/DR: This is a serious crisis, up to 70 % of the population could become infected. We will act according to the advice that experts give us, but consider that our knowledge is incomplete and might change in the future. We will provide one additional billion Euros of budget for a crisis response. It is important for everyone to understand, especially young people, that even though about 80% of cases have no or mild symptoms, that everyone needs to act responsibly in order to not endanger others. It matters what we do, because we need to at least slow down the infection rate, to gain time, to avoid a collapse of our health system.
Feel free to take this as a base line in comparison to the communication tactics of other governments 🙂
Comment #51 March 11th, 2020 at 12:47 pm
Competence matters. Surprise, surprise, when you hollow out an organization for years by disdaining expertise and competence in favor of promoting incompetent lickspittle toadies, this WILL HAVE a significant deleterious effect on the overall competence of the organization. Somehow we see this is a given and obvious in private industry, but we think the government is different?
If 4 years ago the CEO of Apple was replaced by an incompetent man child who made it widely known among the engineering ranks that competence and expertise had no particular value for him, but rather you saw lots of incompetent yes-men elevated into positions of decision making… would anyone be at all surprised that the iPhone model put out at the end of those four years would be utter shit compared to what came before?
Why on earth do people expect anything different from the government I just don’t get. For YEARS Republicans have made very clear they disdain competence in government. Hell, competence in government makes there WHOLE PROJECT more difficult! They are in the business of convincing people that government can do no real good so you might as well elect people that will defund it and give the taxes back to the people. That’s there whole schtick! All Trump did was add a layer of animosity to it by targeting vulnerable groups… effectively changing the message to, “The only good thing a government can do is give tax cuts back to the people AND punish the uppity brown people to make sure they know there place!”
So lo and behold when a crisis comes that necessitates coordinated action from highly competent people with good rational thinking and problem solving skills in government… this is what you get. This really should not be a surprise.
As for all those saying it is the CDC at fault… hey, remember Ebola and H1N1? Millions of tests were produced in record time and there was absolutely no shortage. The administrations message was coordinated and effective. The CDC was orders of magnitude more *competent* in the last administration. Why? Because the last administration actually *gave a shit* about competence and about science and rational thinking and problem solving.
It wasn’t replete with lickspittle toadies at the top whose only real skill is in managing the emotional tantrums of the 2 year old orange blusterfuss at the top of the piling heap of shit that is left of the federal executive branch of this country.
Comment #52 March 11th, 2020 at 12:53 pm
Trump literally said he’d prefer if we just leave the folks on the cruise ship because it would make his numbers look better. That’s what we have at the top folks.
And his payroll tax cut to zero that is being universally panned on the left and the right? Guess how long he wanted that to last… to the end of November. Hmmm…. what could be happening in November that would make the President choose that particular date…
Come on folks. This guy gives less than two shits about any of us. He isn’t declaring a nationwide state of emergency because that’d help the “snake” of a Governor in Washington State who had the unmitigated GALL to question the administrations actions. Meanwhile, had no problem declaring a nationwide *non-emergency* over the brown people coming over the border last year.
Trump does not care about any of us. And deep down we all know it.
Comment #53 March 11th, 2020 at 1:11 pm
Trump has said some stupid things on the subject, but this wasn’t his doing:
How Government Red Tape Stymied Testing and Made the Coronavirus Epidemic Worse:
FDA and CDC bureaucrats stopped private and academic diagnostic tests from being deployed.
https://reason.com/2020/03/11/how-government-red-tape-stymied-testing-and-made-the-coronavirus-epidemic-worse/
Comment #54 March 11th, 2020 at 2:15 pm
AdamT #52
“Trump literally said he’d prefer if we just leave the folks on the cruise ship because it would make his numbers look better. That’s what we have at the top folks.”
The NHK constantly remind their audience each time they report on the latest coronavirus count for Japan that 700 of them come from the Diamond Princess.
Comment #55 March 11th, 2020 at 2:18 pm
Tim Makarios #36:
Serious question: Are you trapped in the US? If you value freedom, here’s a list of fourteen countries that are freer than the US.
I’ve lived in Canada for two years and in Israel for one year, and I’ve seriously considered living in both of them for longer. And certainly if civilization collapses in the US while it remains fine in one of those countries, I won’t hesitate to revisit that option if it’s still available! 🙂 Or the UK, or Switzerland, or the Netherlands, or Australia, or someplace else. Having said that, while it now seems clear that the US will have nearly the worst coronavirus response of any “advanced” country, it’s not yet obvious (at least to me) how well or poorly these others will do.
Comment #56 March 11th, 2020 at 2:21 pm
It absolutely is Trump’s fault. As all the most credentialed R’s will tell you he is the head of the Unitary Executive and therefore ultimately responsible for the actions of inactions of his executive branch. He has hallowed out the executive of competent expertise in exchange for personal and party loyalists. He has purged “the swamp” of scientifically literate, evidence based, rational thinkers, and problem solvers in favor of people who will flatter and pacify his whims and those of his supporters.
We’ve watched for four years as this man child has taken credit for any conceivably good thing that has taken place under his watch, but now we are supposed to lay off any blame for the abject failure of his government in a literally life and death worldwide crisis? Get serious.
Comment #57 March 11th, 2020 at 2:22 pm
Anon #46
I’ve pointed out a few times that the main vector of contamination is through breathing.
But every official or expert in the US downplays the usefulness of masks because you can’t find any… the US simply doesn’t have any significant local production for masks and most other basic medical/health supplies, they’ve all been outsourced to China and India.
Comment #58 March 11th, 2020 at 2:28 pm
fred #40: Indeed, while I’ve felt many strong emotions about the coronavirus, and while self-pity is not entirely alien to me, I haven’t felt a twinge of self-pity about this. I’ll probably be fine and my wife and kids will probably be fine. Mostly I’m worried about my parents, about other older people who are close to me, and about the entire world.
Comment #59 March 11th, 2020 at 2:54 pm
Tim Makarios #36:
Lol, in recent history, and on every possible sides, many in the intellectual elite have always considered whatever country they live in as a mere support system for propping their exceptional brain.
And, when things suddenly became a bit too uncomfortable, those privileged few have had no qualms leaving behind the unwashed and deplorable masses that have been toiling away at making their world go round.
Comment #60 March 11th, 2020 at 2:57 pm
fred #43:
According to [Michael Osterholm], washing your hands and not touching your face is alright, but the reality is that the main vector of infection is through breathing (not even coughing or sneezing)
This seems like a pretty crucial question. If he’s right, then all the advice about frequent hand-washing and not touching your face is a giant distraction—once you’re in the same room with infected people, you’ve just about already lost (modulo the quality of the ventilation in that room).
Does anyone else have insight to share about this? Or a link to an academic debate, or to a view that explicitly challenges Osterholm’s?
Comment #61 March 11th, 2020 at 2:58 pm
He does not care about any of us and never has:
“ White House told federal health agency to classify coronavirus deliberations – sources”
https://mobile.reuters.com/article/amp/idUSKBN20Y2LM
If you tried to imagine an administration who actively tried to make this pandemic worse you’d have a hard time topping what Trump’s administration has actually been doing.
Comment #62 March 11th, 2020 at 3:02 pm
fred #45: In the passage you quoted, I wasn’t even talking about the lack of preparations before we knew about the coronavirus (though Trump did gut funding for pandemic preparedness relative to what it had been in the Obama era, and his hiring freeze led to something like 800 unfilled positions at the CDC). I was talking only about the lack of preparations after we knew about the coronavirus (!).
Comment #63 March 11th, 2020 at 4:01 pm
A freshly released and really interesting discussion going into details with an infectious disease specialist at the Johns Hopkins:
https://samharris.org/podcasts/191-early-thoughts-pandemic/
He says the lethality rate seems to be “only” 0.6%.
But he expects 50% of the US population to catch it within 18 months, so 800,000 deaths.
Comment #64 March 11th, 2020 at 4:20 pm
For reference, roughly 6,200,000 americans die every 18 months (in order to replace a population of 327 million over 78.7 years).
So 800,000 deaths would be a bump of 13%.
Comment #65 March 11th, 2020 at 5:59 pm
I just had a duh moment while reading an Atlantic article. The coronavirus tests are not screening tests! Like, didn’t we all learn about Bayesian probability, sensitivity, and the dangers of false positives and false negatives? And then, when I started reading about Coronavirus testing in the news, I forgot all about that.
This seems like a pretty clear explanation:
> Expectations are tempered; a similar promise from Vice President Mike Pence of 1.5 million tests by the end of last week did not come to pass. But even when these tests eventually are available, some limitations will have to be realized. Among them, these are diagnostic tests, not screening tests—a distinction that should shape expectations about the role doctors will play in helping manage this viral disease.
> The difference comes down to a metric known as sensitivity of the test: how many people who have the virus will indeed test positive. No medical test is perfect. Some are too sensitive, meaning that the result may say you’re infected when you’re actually not. Others aren’t sensitive enough, meaning they don’t detect something that is actually there.
> The latter is the model for a diagnostic test. These tests can help to confirm that a sick person has the virus; but they can’t always tell you that a person does not. When people come into a clinic or hospital with severe flu-like symptoms, a positive test for the new coronavirus can seal the diagnosis. Screening mildly ill people for the presence of the virus is, however, a different challenge.
> “The problem in a scenario like this is false negatives,” says Albert Ko, the chair of epidemiology of microbial diseases at the Yale School of Public Health. If you wanted to use a test to, for example, help you decide whether an elementary-school teacher can go back to work without infecting his whole class, you really need a test that will almost never miss the virus.
> “The sensitivity can be less than 100 percent and still be very useful,” Ko says, in many cases. But as that number falls, so does the usefulness of any given result. In China, the sensitivity of tests has been reported to be as low as 30 to 60 percent—meaning roughly half of the people who actually had the virus had negative test results. Using repeated testing was found to increase the sensitivity to 71 percent. But that means a negative test still couldn’t fully reassure someone like the teacher that he definitely doesn’t have the virus. At that level of sensitivity, Ko says, “if you’re especially risk-averse, do you just say: ‘If you have a cold, stay home’?”
https://www.theatlantic.com/health/archive/2020/03/where-do-you-go-if-you-get-coronavirus/607759/
Comment #66 March 11th, 2020 at 6:24 pm
@fred #63
This death estimate, and others, are concerned only with the direct deaths caused by the pandemic (please correct me if I’m wrong). Many, many others are likely to be caused indirectly (collapsing healthcare system, drug shortages, sickness-related accidents – what not).
Though to be fair, some deaths are going to be prevented. Fewer car accidents, fewer flu deaths (arguably – people will be more isolated and more careful. Though also less supported by the healthcare system once sick). Many of the corona deaths will be among the very elderly, some of whom would have died anyway in the next 18 months. But this is getting morbid(der?). All in all, an actual estimate seems to be very hard to make, but is quite likely to be much worse than “merely” the direct impact of the virus.
Comment #67 March 11th, 2020 at 6:26 pm
I think it’s gonna soon dawn on every country that there just won’t be any easy way out of this.
Angela Merkel is being very honest: Germany expects 70% of the population to eventually get infected.
There’s just no way to avoid that eventually, all we can hope is to slow down the spread without overloading the hospitals.
The news reports that China have things under control, but it’s not really the case in the sense that things haven’t gone back to normal… they haven’t reopened factories yet, and most cities are still practicing aggressive social distancing. It’s as if they’re standing one one foot while juggling three balls at once… they can’t keep it up forever, and once they (or any other country) relaxes things a bit, the virus will just propagate again from some random part of the world.
Realistically, how long before India, Africa, South America become new massive sources? And those countries don’t have top health care.
No single country can isolate itself from the rest of the world for more than a few months.
The effect on the economy is going to be massive.
Comment #68 March 11th, 2020 at 7:44 pm
Thanks for this sobering thread. Some very good information articulately presented.
Trump will get the blame for this, and after gutting Science in this country, most deservedly.
Comment #69 March 11th, 2020 at 8:20 pm
Here’s a rather pessimistic thought! A major slowdown of economic activity caused by the virus leads to a significant reduction in the burning of fossil fuels. As a result all the phosphates in the atmosphere, which reflect a fair amount of sunlight, fall out of the sky quite rapidly whilst the CO2 levels remain constant thereby triggering a sudden and dangerous rise in average global temperature and then….well….let’s hope I’m wrong!
Comment #70 March 11th, 2020 at 10:20 pm
Scott #60, see the WHO report:
https://www.who.int/docs/default-source/coronaviruse/who-china-joint-mission-on-covid-19-final-report.pdf
“Routes of transmission
COVID-19 is transmitted via droplets and fomites during close unprotected contact between
an infector and infectee. Airborne spread has not been reported for COVID-19 and it is not
believed to be a major driver of transmission based on available evidence; however, it can
be envisaged if certain aerosol-generating procedures are conducted in health care facilities.
Fecal shedding has been demonstrated from some patients, and viable virus has been
identified in a limited number of case reports. However, the fecal-oral route does not
appear to be a driver of COVID-19 transmission; its role and significance for COVID-19
remains to be determined. Viral shedding is discussed in the Technical Findings (Annex C).”
Droplet transmission (but not airborne) is somewhat standard for respiratory tract infections, including also the flu. Note that we eject droplets when speaking, coughing, and sneezing, but usually they don’t travel more than a few yards before falling to the floor.
Hand washing for flu and cold prevention has beed studied, and seems at least somewhat effective:
https://onlinelibrary.wiley.com/doi/full/10.1111/j.1365-3156.2006.01568.x
Perhaps washing hands even more would be even more effective.
The efficacy of masks is less clear. It’s harder to study because people don’t usually walk around in face masks to prevent the flu. Maybe there are some asian studies about this, since masks are more common there.
Comment #71 March 12th, 2020 at 1:01 am
Scott #55: It may be too soon to tell how well New Zealand will cope, but so far signs are promising:
* We’ve imported the virus three times, I think, the first of which was confirmed on the 28th of February.
* Testing is happening: “Yesterday around the country, at least 100 tests were being done, but all our tests yesterday were negative.”
* Those tests have confirmed 5 cases so far, and a further 2 cases are suspected but not confirmed.
* There’s capacity for more testing: “At the moment our labs can do over 500 tests in a day if they have to, and that capacity can be scaled up if needs be.” For context, 500 tests a day is equivalent to testing 1% of the population every 100 days.
* People with even mild symptoms are being encouraged to stay at home.
However, I have seen some express doubts as to the intensive care capacity in New Zealand hospitals, so I guess the push to slow the spread of disease makes sense, to ensure that the peak demand for intensive care is as low and late as possible.
Links:
* https://www.rnz.co.nz/news/national/411572/no-more-cases-of-covid-19-in-nz-for-fifth-straight-day
* https://www.newshub.co.nz/home/new-zealand/2020/03/coronavirus-ministry-of-health-urging-anyone-who-is-slightly-sick-to-stay-home.html
fred #59:
I wouldn’t exclusively invite the intellectual elite to leave the US for the freer world; I would very much like New Zealand to say to the US (with apologies to Emma Lazarus):
But sadly, even here in the freest country in the world, that freedom is still essentially an aristocratic freedom, primarily granted on the basis of accidents of birth.
But don’t worry; even if Scott’s exceptional brain does flee future dysfunction in the US, it’ll still be liable for US taxes — the US being very unusual in that it taxes income earned by its citizens who live and work elsewhere.
Comment #72 March 12th, 2020 at 2:14 am
LGS #70:
I found this review on google scholar:
https://www.bmj.com/content/336/7635/77
It suggested that hand washing and surgical mask both being somewhat effective, and N95 being very effective. (And gowns are effective… probably because those data are pulled from health workers.)
The data about masks are mostly pulled from 2003-2004 SARS epidemic without controls and some of them on health workers, so interpret at your discretion.
Somewhat interestingly, this article have two negative response on the website.
https://www.bmj.com/content/336/7635/77/rapid-responses
One of them has competing interest from the pharmaceutical industry, and its argument makes no sense to me. (Different respiratory viruses have different ways of transmission -> so washing kid’s hands is useful for the contact ones -> ??? so we actually have no evidence on face mask ??? )
The other one seems more reasonable and supported by citations. However, its strongest argument is this sentence: “surgical mask… However, there is no evidence that these types of masks are effective in protecting healthcare workers from SARS (Derrick & Gomersall 2005.)”
Well, if you actually look into Derrick & Gomersall 2005
https://www.ncbi.nlm.nih.gov/pubmed/15749326
What the paper do is a filtration test on surgical masks. They found that 5 layers of surgical mask only filtered out 12/13 of the particles, so it’s not as good as a N95. The authors didn’t claim that surgical masks are as good as nothing.
Comment #73 March 12th, 2020 at 6:39 am
A good part of the efficiency of masks is that it reduces contact between your hand and your mouth and nose, which in many cases is involuntary. Adding glasses to your outfit would be even better.
Comment #74 March 12th, 2020 at 8:16 am
[…] The Blog of Scott AaronsonIf you take just one piece of information from this blog:Quantum computers would not solve hard search problemsinstantaneously by simply trying all the possible solutions at once. « National disgrace […]
Comment #75 March 12th, 2020 at 8:30 am
Tim Makarios #71
Doesn’t New Zealand (like Canada and Australia) have an immigration based on a point system?
I.e. the more skill/assets, the easier you get accepted with open arms?
So much for taking the tired and the poor…
The US doesn’t have a point system (yet), but you’re saying it would be wrong for the US to adopt it?
Maybe you’re right!
Let’s look at Australia: rich party officers from China are dropping by, investing one or two million $, they get citizenship, and he country is slowly turning into a province of China, under heavy influence from the CCP (they control all Chinese media in Australia, using soft power to push Xi’s propaganda, even buying out Australian politicians…).
Canada isn’t much better, the Huawei dynasty all have/had Canadian citizenship.
PS: the US has tax agreements with most countries, it’s not like US citizens working abroad gets double-taxes… then, you’re no longer required to pay any US tax once you renounce US citizenship.
Comment #76 March 12th, 2020 at 8:50 am
Anon #72
N95 is very effective because they seal very well.
Other facial masks are looser, there’s a chance you could breathe bad air from the gaps. But even so, they’re good to wear in case you don’t want to infect others (most of your own breath and cough particles are captured)… so once everyone is wearing them, they’re pretty effective too: the virus has to pass through two consecutive masks to contaminate (much better than nothing).
Another source of contamination is through the mucus of the eyes, but the probability is lower, so you should wear some goggles only if you’re interacting with lots of ppl or if someone in your household is sick (I’ve seen footage from China where a wife was mildly sick with fever, isolated to the bedroom, and the husband was taking care of her just fine without getting contaminated).
Comment #77 March 12th, 2020 at 10:08 am
fred #59
I think many among “the unwashed and deplorable masses” would aslo very much like to emigrate, had they the same opportunities as really talented people.
Comment #78 March 12th, 2020 at 12:07 pm
Fred: Thanks for the age / mortality table. Do you have a reference for that? Do you have an update?
Everyone: Ars Technica has a updated daily Q&A on the pandemic. People at Ars ALWAYS know what they are talking about.
Comment #79 March 12th, 2020 at 1:39 pm
Raoul:
I took it from here, the source being WHO and china research (whatever it’s worth)
https://www.worldometers.info/coronavirus/coronavirus-age-sex-demographics/
Comment #80 March 12th, 2020 at 5:47 pm
There is a question about how much the CDC screwed up, how much that was due to funding, vs bureaucratic rigidity, vs orders from the top. (And note that in the Seattle case the WA officials also objected.) But regardless, it should not have been a single point of failure. Ultimately, this was going to have to be decentralized and the States were going to have to handle testing. The State of Texas is testing now, but only in Dallas, and in highly constrained numbers (and I think that is typical of the states). The states should have seen this coming and ramped up way earlier. They should have defied the FDA ban. At the very least, they should have ramped up production in advance of approval.
Should individuals have done more defiant testing? I’m not sure. Demonstrating that the virus had arrived had value early on. Demonstrating that it arrived in Austin today has much less value. Most of the value of testing is in contact tracing, which is difficult to do as an individual.
Comment #81 March 12th, 2020 at 6:32 pm
Ken #30 & Scott #33: I have to say, for a group of (presumably) highly educated scientists, most of whom have surely had at least once course in statistics, I’m rather appalled that at least one of you hasn’t commented on the obvious fallacies (yes, there’s 2) in Ken’s friend’s “back of the envelope” calculation. And the first fallacy should be obvious even to any lay person following the data: of those 29 initial deaths, 18 occurred in that nursing home; these are elderly folks in poor health, and their chance of dying wasn’t 1% — it was probably closer to 50%. So right off the bat, the calculation is based on bad data. But even the statistical portion of the calculation is wrong; in fact, it’s almost the reverse of the classic gambler’s fallacy (ie, if you throw 10 heads in a row, the next throw is more likely to be tails). You’re saying if I take a random coin, and guarantee the next throw to be tails, and it is, then the previous 10 throws must have been heads. The 1% number is simply the odds, not a guarantee. If you take 100 random people with the virus, after 3 weeks, it’s very possible they could all be alive, or 5 or 10 could die, and it’s even possible (albeit unlikely) that half of them could be dead (especially if all 100 were in a nursing home). It is simply statistically false to say, if one person died, the initial sample must have been 100.
Comment #82 March 12th, 2020 at 7:32 pm
Ken #30 (continuation of previous post, hit submit too fast!): So Ken, I would be willing to place a wager with your friend, that his prediction of 1,000,000 cases in this country is probably off by at least an order of magnitude. In fact, any layperson without a knowledge of statistics, but with a little common sense, who has been following the data, should be willing to place the same wager. And here is my reasoning: the number of cases in China appears to have leveled off at roughly 80,000. China has a population of roughly 4 to 5 times that of the US. So just based on that, it seems hard to believe that the US, with less than 25% of the population, would, for some reason, wind up with over 10 times the number of cases. He’s effectively saying the situation here is going to be 40-50 times worse than China — pretty hard to believe, even making some unlikely worst case assumptions. My only caveat would be not entirely believing the data coming out of China.
Comment #83 March 12th, 2020 at 7:52 pm
Joe #81: Good point about 18 of the 29 deaths being from the one nursing home. Suppose we discount them altogether and say there were 11 deaths. So all our numbers of infected on given days should be multiplied by 11/29. But, Michiel #38 pointed out I made a dumb thinko, for the initial 3 weeks I multiplied by 3 instead of 2^3=8. So all the numbers should also be multiplied by 8/3. Turns out fortuitously the two errors cancel: (8/3)*(11/29)=88/87 ~1.
Re your 2nd point that the number of deaths will only be 1% in the mean, but in reality we will be picking from some distribution about that mean: I know. That’s why it’s a back of the envelope calculation, trying to get some rough orders of magnitude. It would be easy to calculate 95% confidence intervals, which wouldn’t change anything very much because the exponential growth sort of sweeps away any numerical factors, multiplying the initial number by x just adds or subtracts log(x)/log(2) doubling times to get to the same number of infected.
A more impactful error is that a one week doubling time is probably far too long. A friend of mine for the WHO data on number infected vs time finds that the number is multiplied by about 1.25/day, consistently across countries. That means the doubling time is log(2)/log(1.25)=3.1 days. So a week multiplies by a factor of more than 4, say 4. That means it will only take 1/2 as many weeks to get to any given level as I had originally calculated
Michael#38 I didn’t understand how you got 42 deaths, should have been 1% of 33,600 infections, ie 336? Which isn’t too bad compared to the actual, 236. But if we add the above corrections we have to multiply by 16 for 4 weeks at factors of 4 instead of by 2 and by 11/29 for initial dress, giving
Comment #84 March 12th, 2020 at 7:56 pm
(sorry, on my phone, last one accidently posted before done)
continuing: giving around 2000, of by an order of magnitude, or around a 1 2/3 weeks. Well, these estimates aren’t an exact science. The main point is that with exponential growth you move beyond your capacity to treat very quickly.
Comment #85 March 13th, 2020 at 2:49 am
Raoul Ohio #78: See also https://slate.com/technology/2020/03/coronavirus-mortality-rate-lower-than-we-think.html
fred #75:
That sounds about right, though I don’t know the details.
That’s why I said “I would very much like New Zealand to say…” (emphasis added); New Zealand is not currently saying that, either literally or metaphorically. (See also “But sadly, even here in the freest country in the world…”)
I don’t think I said much at all about what it would be right or wrong for the US to do from here, but my line “You clearly do not want them any more” was written in reference to things like the recent decade or so (perhaps more?) of the US incarcerating people en masse for pursuing happiness without a licence.
Regarding the US double-taxing its expatirates, I’m no expert, but I think there is a fair bit of double-taxing going on. And I think the tax paid to the government of the expatriate’s new home is only a tax credit for US tax purposes, so I don’t think you’d be able to pay less in taxes by moving to a country with a lower tax rate. (And I’m certain you wouldn’t be able to get away with less tax paperwork complexity, though this is drifting a bit from whether Scott’s exceptional brain would be totally abandoning its support of the “unwashed and deplorable masses” who are prevented from leaving the US and joining him not only by the lack of another country that might welcome them, but also by the US’s own tax system.)
As for renouncing US citizenship, according to Adam Gering’s answer here, even that is taxed!
Comment #86 March 13th, 2020 at 2:42 pm
Scott,
Here is some perspective.
All of the data and publications that I’ve read to date — from medical journals such as JAMA, as well as more layperson’s publications like STAT News (www.statnews.com) — suggest that out of all confirmed cases of COVID-19 throughout the world, ~80% only suffer mild symptoms similar to the flu or the common cold (fever, cough), ~15% suffer more severe symptoms, including pneumonia, and ~5% suffer critical symptoms.
Of course those numbers also depend crucially on the age of those infected and/or their overall health prior to infection. So wouldn’t it be fair to say that someone who is < 60 years of age and in good health more than likely recover after being infected with the novel coronavirus causing COVID-19?
As for the tragedy befalling Italy, I also wonder if that country is experiencing a particularly challenging outbreak due to the high numbers of the elderly as a percentage of the population (I have read before that Italy has the second highest percentage of the elderly as a proportion of the population after Japan) and the fact that the elderly (at least from what I understand) tend to live in multi-generational homes compared to other countries, which I suspect will greatly increasing their risk of getting infected.
Comment #87 March 14th, 2020 at 12:21 am
They aren’t testing seriously in Vancouver either. I got back from Seattle and developed symptoms consistent with coronavirus 3 days later. I called the hotline and (after hanging up on me once) they said that they would only test if I was in severe respiratory distress or passing out, or if I had known contact with a confirmed coronavirus case. Seattle, they said, was not an area of concern. It also didn’t matter that I live with two older people. If I really wanted to, I could go to the urgent care centre and sit in the waiting area with everyone else. Don’t worry, a mask would be provided. Wait times are 4-5 hours. Given that I’m young and generally not dying, I didn’t do that, because it’d be more likely to just infect others than do any good, so I’m just isolating myself from the people I live with. But the case numbers are way under reported. I think they must be legitimately running low on test kits, otherwise this makes no sense at all.
Comment #88 March 30th, 2020 at 11:25 pm
The problem RE: Trump is that a simple, dumb, right-wing solution would work for a variety of human problems. Would the performance of a more sophisticated liberal solution be better? Would the cost in e.g. freedoms or other human factors be lower? Yes, but only if it’s properly administered. The child exploitation scandals in those English cities could have been handled by one of several simple, dumb, right-wing solutions. Most of those could not have solved a similar scandal by e.g. the Catholic Church, but the Church has institutional power, and that makes it easier for them to get away with things. (It could also have been handled by a simple, dumb application of more liberal principles – some of which have become unfashionable.)
So every time someone criticizes Trump, a right-wing reader will think, in the back of their minds, that e.g. blocking all travel from China earlier would have been effective. …and they would be (at least partially) right! A number of early Trump supporters called this before more mainstream people and late Trump supporters did. (There is a certain sense in which Trump is not, himself, a Trumpist.)
The other problem is that political actors have worn down the brand of “expertise” itself. For instance, the criticisms that “worrying about the virus is racist” from establishment channels… If a regular person sees that and doesn’t realize that it’s not a matter of experts, but rather gatekeeping institutions, or just doesn’t have the time to sort through who is and is not a legitimate expert – why would they vote for a government of experts rather than an emotionally-satisfying government, if they have no guarantee that they’ll even get real experts? This sort of thing has been going on for a while now and IMO it has a truly staggering cost. (That cost includes, for the people that hate Trump, the election of Donald Trump.)
Comment #89 August 6th, 2020 at 5:32 am
I agree with you