Coronavirus

[FrankSG]

I just got this very troubling SMS from a cell provider in Singapore called Singtel:

Dear customer, be protected with free COVID-19 coverage with $1,000 payout upon diagnosis. Get free 30-day insurance by NTUC Income with any top-up of $20 or more. Up to S$5,000 coverage for COVID-10, accidents, hospitalisation and job loss. Policy T&Cs apply. Info: singtel.com/prepaidinsurance. TO claim call NTUC Income 6788 6616

You want to give people $1,000 if they get infected with a disease? That's the stupidest thing I've ever heard. If students donate blood for $50, I bet some wouldn't mind getting infected with a for them most likely very mild virus for $1,000.

 

[Causalien]

...


The number of infected in the U.S. is estimated to be above 100,000 already, which is at least 1 but possibly 3 orders magnitude higher than any new infections arriving via borders.

It's a classic "the horse already left the barn" situation, and the exponential curve of the early outbreak is pretty much inescapable for the next 10 days - and that's while there's still not nearly enough testing in the U.S., where officially there's still only 2,800 known and reported cases.

Man. It's a sad day when your estimate is more pessimistic than mine.

 

[AlanSubie4Life]

he number of infected in the U.S. is estimated to be above 100,000 already, which is at least 1 but possibly 3 orders magnitude higher than any new infections arriving via borders.

I haven't seen an estimate that high - Trevor Bedford estimates between 10k and 40k. And says he could be off by a factor of two either direction. (I suspect it is more to the low end of that range currently - it just feels right. But maybe it is as high as 50k. Not sure.)

Anyway, regarding the arrivals from Europe, certainly agree this is a silly policy; it has minimal impact at the moment on additional cases (Bedford analyzed it). The super-spreading events introduced by these huge lines (which definitely will have some infected people in them) will have minimal additional effect. In addition, just a reminder that all of these people in these airport lines are under a mandatory self-quarantine for 14 days in the US after arrival. Hopefully most of them take it seriously, so that that self-quarantine acts to quell the effects of these super-spreading events that occurred today in the airports.

I think Trump should have just mandated strict 14-day self-quarantines for all travelers coming from Europe, and not bothered to screen at the airports or even apply the ban to non-resident aliens. That would have been a policy that would have nibbled around the edges of the problem. But obviously it should have been a side policy, an afterthought - not the main focus; the main focus is stopping the spread of the thousands of cases already in the country.

Long term, we'll have to do quick EFFECTIVE screening at airports, to avoid the mandatory quarantine after arrival, but they're not equipped to do any such useful screening right now. They can do questionnaires or whatever and leave it at that for now.

 

[adiggs]

My current primary measure for US virus progress is the actual number of people we're testing on a daily basis. Not the politician's talk about how many tests we can do - the actual number completed each day. If we're using the 10-40k estimates, and it takes testing 10 people to find 1, then we've got a need for 100-400k tests to be done.

Preferably over the next week. So that's 14-56k people tested per day - anybody hear of testing at that scale going on around here?

I'm thinking we're still blind - completely blind. And it's looking like we'll stay blind until we can stop counting people tested/day, and start counting hospitalizations / day. Presumably hospitalizations / day will be a pretty complete and accurate number.

 

[AlanSubie4Life]

Preferably over the next week. So that's 14-56k people tested per day - anybody hear of testing at that scale going on around here?

I'm thinking we're still blind - completely blind. And it's looking like we'll stay blind until we can stop counting people tested/day, and start counting hospitalizations / day. Presumably hospitalizations / day will be a pretty complete and accurate number.

We are certainly still blind.

The testing is absolutely key long term - it's the only way to exit this situation as far as I can tell. Extensive, rapid testing - hundreds of millions of tests need to be available, with capacity to test a million people a day.

So obviously that capability needs to be ramped up quickly over the next month or two.

But right now, we have to get extremely serious about social distancing and avoiding any further spread EVERYWHERE, and especially critically do so NOW in the places that have large numbers/clusters of cases. If that distancing and spread reduction is done diligently, we don't NEED to immediately test everyone with the virus right now. They simply need to stay at home and get better, or get worse, and be hospitalized, without propagating to anyone else (and obviously family members would have to remain quarantined if there was any suspicious case in the household).

That's the only way to stop the increase in cases - and it looks like we have just a few days to do so in New York, and possibly Seattle as well. What we're doing now isn't good enough. Have to get serious. I had a long conversation with my parents tonight about their church in Washington County, OR, and how they're having a service tomorrow. I told them to not go. I could not convince my dad - he wants to run the sound system there for them. They're going to be limiting contact and making changes, but frankly that's likely not good enough. We have to stop all gatherings.

But yes, long term, we need lots of testing. That needs to be ramped over the next 2-4 weeks so it is ready on the downward side of the new case curve, so people can get back to work and doing their regular things - with extensive testing available for anyone who could conceivably be a risk for restarting the epidemic. It's going to be mandatory to have this ability to test, to be able to stamp out new embers or remaining embers.

 

[aubreymcfato]

Just as has been repeatedly posted in the Coronavirus thread, no matter how manny times one posts that graph, it remains not a graph of the mortality rate. The denominator is not "the number of people who have the disease", it's "the number of people who've gotten tested and the test came back positive", which is an entirely different number. Dividing deaths by "people who tested possible" is a meaningless figure.

I understand, but I've posted this chart on the other thread to show the difference of rate.
I don't really care about the numbers, but that chart, to me, shows how and overstressed health system works much worse and this means much more fatalities. Then mods moved the post.

In the other thread, I feel people are downplaying the C19 threat (to Tesla¹, to their lifes). IMHO Musk is downplaying it too. I hope to be proved wrong. But I won't discuss anymore with people praising Trump and shitting on European public health.

¹ This doesn't mean I'm selling all my shares. I should've done it weeks ago, but I'm a very bad investor and I don't know to time the market, to me it's just a matter of luck. I just unloaded a few shares to have some dry power in case it dips below and I can buy a bigger number of shares.

 

[AlanSubie4Life]

Looks like America is on course to be #1! We have all the best people! 'Murica! Looks like just 11 days to hit the situation in Italy - but unclear whether we'll have the testing capacity to even keep up with that growth and track it accurately, though (this is a plot of detected cases, not actual cases, of course). You'd need to do roughly 100k tests daily nationwide to detect the 10k additional cases detected in the US on the 10th-11th day from now (though maybe the positive rate of the tests will increase from the current 10% at that point, due to the prevalence of the disease, so we'd need fewer tests than 100k). Right now we don't have the capacity for that, in any case - we can do more like 30k tests daily nationwide I believe.

*I can't quite square the 33% daily increase in this plot with the Trevor Bedford 5-day doubling estimate (which would be 12.5% daily increase). That's a big difference, but assuming this plot data is accurate (pretty sure it is), it looks like the actual rate of increase is 33% per day typically without decent containment measures.

 

[Matias]

I won't discuss the Elon tweets or mail any more: if I'm overracting, in a few weeks I'll be here apologizing for my mistake. I'd be happy to.

I'll just leave you a sobering graph, showing how one of the best European health systems fares when things go really south (the red one is Lombardy fatality rate):

Do you have data about the absolute death rate at the moment?
I’m thinking, how much it has rised. I’m thinking about the possibility, that it might be possible, that at the moment deaths could be more easily categorised as being coronavirus deaths.

It is in many cases a bit arbitrary to name just one cause of death, if person has many conditions.

I’m just speculating. And by no means, I don’t mean that the situation would not be extremely bad. I’m just thinking, how bad..

 

[Matias]

Edit: As luck would have it, a number of posts discussed hydroxychroroquine while I typed this. The links are worth looking at though.
-----------

I'm surprised how little discussion I see on hydroxychloroquine (Plaquenil) with regards to Coronavirus. The drug has been around awhile for malaria and arthritis.]

Yes, it has been around for decades for those conditions, at is considered a bit obsolete for both of them. And it is very cheap.

 

[Fact Checking]

Preferably over the next week. So that's 14-56k people tested per day - anybody hear of testing at that scale going on around here?

Yes, these are actual figures from Andy Slavitt (health official responsible for ACA/Medicaid/Medicare under Obama).

7 days ago:

Andy Slavitt on Twitter

"NEW: Because we can’t get a straight answer from the Trump Administration, I have checked with lab companies. The best estimate is it will be 8 weeks before we have all the nationwide testing we need."
​

4 days ago:

Andy Slavitt on Twitter

"BREAKING: The US is failing to produce even small number of testing kits for COVID-19 extraction despite Trump & Pence statements. Labs and states are worried. Expect next to no availability to continue for weeks. It is vital to self-quarantine."
​

2 days ago:

Andy Slavitt on Twitter

"We can at best do 10,000 tests/day. We should be able to do millions."
​

Earlier today:

Andy Slavitt on Twitter

"I talked to several on the WH team who went from 0 to 50 at lightning speed. Why they were at 0 weeks into an outbreak when South Korea was 10 days before is a topic for another time (e.g., November) but the labs, mobile testing & supply will be straightened out."

"But, testing in a few weeks will be like a sports team scouting college players after the draft. It will help us, but it won’t diagnose & prevent if we are following the Italy curve in Slide 3."

"I need to close with three messages: 1/ stay socially isolated for real. Not for you, but for everyone. And if someone minimizes this all, patiently listen & explain. People can’t be bullied."

"Second, root for the government to succeed here whether you like them or not. The stakes are higher than any most of us have ever experienced: wars, 9/11, whatever. Let’s have an election with this behind us having pulled together. This does spread from Republican to Democrat."​

He is politically opinionated, but discount his professional opinion based on his politics at your own risk.

 

[Matias]

Is that a positive ... uh ... good thing?

 

[Fact Checking]

I haven't seen an estimate that high - Trevor Bedford estimates between 10k and 40k. And says he could be off by a factor of two either direction. (I suspect it is more to the low end of that range currently - it just feels right. But maybe it is as high as 50k. Not sure.)

The 100k infected in the U.S. is purely my estimate and it might be too high, but there was a two months long active suppression of testing and a disinformation campaign by the Trump administration, which created Wuhan levels of oblivion in at least half of the population.

Here on TMC there's one member whose wife is a health worker in California and was infected and was tested Covid-19 positive, he came down within hours with very similar symptoms - yet despite asking repeatedly he was refused a test. He's not included in the U.S. statistics.

If in our very small TMC circles we have evidence of under-testing then it's probably very widespread.

I got to the 100k infected in the U.S. based on taking the known 57 U.S. fatalities, taking a 3x factor in assuming that not all respiratory fatalities in the U.S. got tested for Covid-19 due to the suppression effort, taking into account the 3 days delay in testing which with a 40% early exponential growth rate adds a factor of 2.7x, and using the China-mainland, South-Korea and Diamond Princess fatality rate to a long term whole-population fatality rate with the relatively young demographics of the U.S. of ~0.5%, which gives the size of the infected population in the U.S. at around 92,000.

But it's a very crude estimate and I could be wrong by a factor of 5 in either direction - and 20k sounds plausible too. Let's hope it's the lower estimates.

Note that due to the speed of the early exponential if it's as low as 20k today, it will probably be 100k in 4-5 days.

 

[Matias]

BTW, Trump's test is negative

That’s aladeen news!

 

[ev-enthusiast]

No panic, just the new numbers from Italy:

Deaths (deceduti) is for Italy, in the Lombardy region where most infections are registered it is above 8%.
From a medical point of view the outbreak in Europe is lagging behind a couple of days behind the Asian outbreaks. And some other regions like the US lag behind a couple of days behind Europe. Other countries lag behind even more days and will have the outbreak later. It takes some time for the virus to spread from country to country and from continent to contient.

 

[Matias]

Correct - it is the CFR (Case Fatality Rate). Where Case is defined as someone ill enough to present for healthcare and a health provider decides to test them.

We will never know the true IFR (Infection Fatality Rate) for COVID-19, although as more data comes in from around the world we will get a reasonable extrapolation.

I think with the serum antibody tests we will know it afterwards. Of course that data comes after the epidemic has burned out and has more academic value than actually helping with the current situation.

In medical school I heard a joke about the serum antibody tests; they’re good, except when you get the results, the patient has already been either cured or dead..

 

[renim]

https://www.medrxiv.org/content/10.1101/2020.03.09.20033217v1.full.pdf

' Both viruses showed markedly longer viability on stainless steel and 118 polypropylene: the median half-life estimate for HCoV-19 was roughly 13 hours on steel and 16 hours on polypropylene'

i expect the severity of the covid-19 pandemic to have very local effects.
the use of public transport being one of the top contributors. so as far as USA is concerned, high public transports locations have greater scope for community transmission before response can be co-ordinated.

 

[avoigt]

I haven't seen an estimate that high - Trevor Bedford estimates between 10k and 40k. And says he could be off by a factor of two either direction. (I suspect it is more to the low end of that range currently - it just feels right. But maybe it is as high as 50k. Not sure.)

Anyway, regarding the arrivals from Europe, certainly agree this is a silly policy; it has minimal impact at the moment on additional cases (Bedford analyzed it). The super-spreading events introduced by these huge lines (which definitely will have some infected people in them) will have minimal additional effect. In addition, just a reminder that all of these people in these airport lines are under a mandatory self-quarantine for 14 days in the US after arrival. Hopefully most of them take it seriously, so that that self-quarantine acts to quell the effects of these super-spreading events that occurred today in the airports.

I think Trump should have just mandated strict 14-day self-quarantines for all travelers coming from Europe, and not bothered to screen at the airports or even apply the ban to non-resident aliens. That would have been a policy that would have nibbled around the edges of the problem. But obviously it should have been a side policy, an afterthought - not the main focus; the main focus is stopping the spread of the thousands of cases already in the country.

Long term, we'll have to do quick EFFECTIVE screening at airports, to avoid the mandatory quarantine after arrival, but they're not equipped to do any such useful screening right now. They can do questionnaires or whatever and leave it at that for now.

Due to the German Robert Koch Institute they expect the true cases to be by a factor of 4 - 11 times higher which is a really huge number.

However if that many more people are infected they either are in an early status or do have a very mild outcome and don't need hospitalization.

The German fatality rate did fall to 0.17%. today That down trend has been predicted from RKI before.

Closing borders does help with a narrative that the infection is coming from outside which is really nonsense. Many European countries are closing borders to a certain extend but Germany does not for good reasons. Restricting smaller areas if you have a large amount of infected peoples somewhere is much more effective.

China should be a copy and paste example how to fight the virus best but politicians still use the virus for their narrative and with that personal interest.

 

[sparcs]

which will only cause lot’s of people to be together in the same building. Not what you need to limit an epidemic.

24 hour stores so there wouldn't be much traffic in those overnight hours anyways. Best time to restock shelves and sanitize is when fewer customers would naturally be there.

 

[SageBrush]

We have to stop all gatherings.

Masks

 

[SageBrush]

China should be a copy and paste example how to fight the virus best

Pretty much, although S. Korea and Singapore are also very successful models.

The truly amazing thing is that those countries have densely packed populations so despite a HUGE built-in disadvantage they tamped down the spread.