Coronavirus Sars-CoV-2/Covid-19 Megathread

[Josquius]

Things don't look good in the US...

Yet we are ending the lockdown in most states this month...

I think many people just aren't going to listen until they start getting sick, their friends and family start getting sick and dying, etc.

It's like the company I work for... nobody in management listens to anything until there's a crisis, so sometimes in order to get them to pay attention and cooperate, you just have to let a crisis happen.

Yes definitely.
A large amount of my aunt's friends were really those old "whatever, its not my problem, just let me get on with life, if its my time its my time" people...Then one of their husbands caught it. Sharp changed their tune.

[HisMajestyBOB]

I am a little surprised by the situation in the UK.  I thought they would be in same category as Germany, South Korea and New Zealand, coming out of this crisis as victors and suffering relatively few deaths.  It now looks like they are in the same boat as Italy and Spain, some of the worst performers.  I thought the UK would be good at this sort of thing, with their healthcare, technology, and good governance.

Relatively speaking. I'll trade you the GOP and Trump for the Tories and Johnson in a heartbeat.

[alfred russel]

Things don't look good in the US...

In anything comparing case numbers over that time period, it is worth also looking at the number of tests run. The US has been ramping up testing rapidly--now 7.7 million have been run--but at the beginning of the time period there were hardly any tests available. Since only a fraction of the cases are being identified, if you run more tests it stands to reason the case count will grow.

It is also notable that those graphs have the same scale. It looks like NYC is coming in ~6-7k new cases a day, while the rest of the US is ~22k. If that is an accurate reflection of how the disease is spreading, it is still spreading at a much faster rate per capita than the rest of the US.

[DGuller]

The height of the curves tells us about today, and New York metro are should be considerably higher on the y-axis due to more disproportionate under-testing there compared to the rest of US.  On the other hand, the shape of the curve hints at an R0 rate, and that tells you what the height of the curves will be in the future.  With a curve like the one on the right, opening up doesn't seem like a smart decision.

[The Brain]

The height of the curves tells us about today, and New York metro are should be considerably higher on the y-axis due to more disproportionate under-testing there compared to the rest of US.  On the other hand, the shape of the curve hints at an R0 rate, and that tells you what the height of the curves will be in the future.  With a curve like the one on the right, opening up doesn't seem like a smart decision.

Er... Trump has won the virus. Did you self-isolate from the memo?

[Legbiter]

So we've been seriously dealing with this for about 2 months now since the first confirmed case here in late February. At the moment we've had 1800 confirmed infections, 10 people have died. We've built up a fairly complete picture of the spread. The virus does not spread from children to adults like the flu and noroviruses. The biggest clusters here pop up in crowded indoor settings like sporting events and office/cubicle plantations. A few small retirement homes were hit as well. Public transport and elevators are also of concern. No known cases of outdoors spread. Household spread is observed but with proper precautions it's not as virulent like with say, nororviruses where the entire family often gets sick. Surface contamination has been greatly curtailed by ubiquitous hand sanitizer in all stores and businesses plus frequent hand washing. Mask wearing in still minimal and apart from DIYs they are still rare and precious things.

The initial response was inept to say the least and at times comical, for instance the decision to shut down border travel, etc was made for us by other governments once they started stacking up the bodies in Italy. We started testing widely for the virus a month before we had our first confirmed case using existing health care systems, a private lab rolled out massive free testing overnight on their own initiative and contact tracing teams were thrown together at the last minute. That part of the response has been and is an outstanding achievement. We had severe PPE shortages in the beginning, at one point being down to 2 days supply before public and private players managed to secure enough supplies by hook and by crook. But it was a close run thing.

The economic effects are disastrous of course but we're using the plumbing to help people and businesses that was set up in the 2008-2009 economic crash.

Overall verdict so far is I'd say we had a very incompetent start, got very lucky in the beginning but used the time fairly well. Our approach worked out to a slightly stricter Sweden but unlike them we coupled that with a massive testing/tracing/isolation effort.

[alfred russel]

With a curve like the one on the right, opening up doesn't seem like a smart decision.

DGuller--two points:

1) are you sure the modest upward trajectory of the "rest of the US" is an increase in actual cases versus just an increase in identified cases? It is so hard to tell from the graph, but if testing is up 33% from say 3-4 weeks ago, I'd expect a ~33% increase in identified cases with actual cases staying constant. In fact, it looks like testing is up about 50% over the last four weeks.

https://ourworldindata.org/grapher/full-list-daily-covid-19-tests-per-thousand?time=2020-03-08..&country=USA

2) "Opening up" can mean abandoning policies that aren't effective. We are now approaching a couple weeks since Georgia opened tattoo shops, massage parlors, etc. and the data is still coming in, but it doesn't seem like cases are spiking here.

[derspiess]

I'm pretty okay with Ohio's plan to open up in phases.  Just hope people don't screw it up.  My son's baseball season is now scheduled to start June 1 and it'd be nice to have a shot at that happening. 

[DGuller]

DGuller--two points:

1) are you sure the modest upward trajectory of the "rest of the US" is an increase in actual cases versus just an increase in identified cases? It is so hard to tell from the graph, but if testing is up 33% from say 3-4 weeks ago, I'd expect a ~33% increase in identified cases with actual cases staying constant. In fact, it looks like testing is up about 50% over the last four weeks.

https://ourworldindata.org/grapher/full-list-daily-covid-19-tests-per-thousand?time=2020-03-08..&country=USA

I'm not sure, I wish I had access to more useful numbers than confirmed cases.  One thing I am very sure about is that increased testing wouldn't have a linear effect with increased case counts like you hypothesized.  The more widespread the testing, the lower your positive test result rate.  As you expand out the testing, you will now be testing people less likely to be sick, so testing being up 33% would result in much less than 33% increase in cases. 

2) "Opening up" can mean abandoning policies that aren't effective. We are now approaching a couple weeks since Georgia opened tattoo shops, massage parlors, etc. and the data is still coming in, but it doesn't seem like cases are spiking here.

I agree with the sentiment, don't know enough about the details.  It's easy to fall into the trap of thinking that the more hardcore your policies are, the more serious you are about fighting the spread.  A more intelligent approach recognizes that your budget of getting the public's compliance is limited, and you want to spend it wisely on measures that actually work rather than spray and pray indiscriminately.

[Josquius]

It's not a US vs New York contest.

The context of that graph was that some are trying to present the overall US numbers as having flattened, however taking away New York which stands quite apart it is still in the growth phase.

An increase in testing could be a valid argument, but usually you see sharp jumps in numbers when that happens rather than gradual curves.

[alfred russel]

I'm not sure, I wish I had access to more useful numbers than confirmed cases.  One thing I am very sure about is that increased testing wouldn't have a linear effect with increased case counts like you hypothesized.  The more widespread the testing, the lower your positive test result rate.  As you expand out the testing, you will now be testing people less likely to be sick, so testing being up 33% would result in much less than 33% increase in cases. 

I know it won't be exactly linear, hence I put the "~". I can think up a scenario that would be even greater than 33% however--probable positive cases in healthy populations are not allocated tests (both due to the risk of them increasing spread by leaving home to present to PPE deficient clinics and it not serving a treatment purpose) with testing focused on screening high risk environments (such as nursing homes).

In Georgia, however, as testing has increased the percentage of positive tests did initially fall.

[HisMajestyBOB]

The White House's model says the number of deaths will hit 0 in ten days, so go ahead and open it up!
https://www.businessinsider.com/white-house-economic-adviser-hassett-model-coronavirus-deaths-zero-10-days-2020-5

The White House is reportedly relying on a "cubic model" prepared by controversial White House economic advisor Kevin Hassett that shows coronavirus deaths plunging to zero by May 15 to help guide their economic decision-making during the pandemic.

The Washington Post cited sources familiar with the model who it said it demonstrated the rate of deaths dropping through the beginning of May until hitting zero in ten days.

The "cubic model" from Hassett clashes with the assessment of public health experts who say the virus will continue infecting people and swell the US death toll for the foreseeable future. The US recorded over 1,000 deaths from the coronavirus on May 4.

A spokesperson for the Council of Economic Advisors told Business Insider the data was drawn from The New York Times database.

Jason Furman, who formerly chaired the CEA under President Obama, blasted the model Hassett designed as one cooked up to elicit results supporting the administration's drive to reopen the economy.

He said it had "no place whatsoever in any discussion of the government's response."

"The 'cubic fit' is based on an approach to epidemiology that has long been absent from any serious epidemiological discussions. It made terrible predictions back in March and April," he wrote on Twitter. "The functional form was chosen to get the result they wanted."

Another model prepared by the University of Washington's Institute for Health Metrics and Evaluation is also being used by the White House. It recently revised its death toll upward and now predicts 135,000 deaths by August 1.

The Trump administration is privately predicting the daily coronavirus death toll will double in the next month, Business Insider's Eliza Relman reported.  Some states not initially ravaged by the outbreak have begun reopening their economies, even as infections mount.

Other critics argued that an economist with an unreliable track record on issues within his own realm of expertise shouldn't wade into public health matters.

Paul Krugman, a Nobel Prize-winning economist and New York Times contributor, pointed out Hassett had made several predictions that never materialized.

Hassett, once the former Chair of the Council of Economic Advisors for the Trump administration, was in favor of the 2017 tax cuts and he argued it ramped up business investment in a Wall Street Journal op-ed in December. It didn't, per a Business Insider analysis of the law late last year.

"Hassett remains... a pillar of the modern conservative establishment, and Trump called on him to second-guess experts in epidemiology, a field in which he has no background," Krugman wrote in a New York Times op-ed that was published on Tuesday.

[mongers]

So we've been seriously dealing with this for about 2 months now since the first confirmed case here in late February. At the moment we've had 1800 confirmed infections, 10 people have died. We've built up a fairly complete picture of the spread. The virus does not spread from children to adults like the flu and noroviruses. The biggest clusters here pop up in crowded indoor settings like sporting events and office/cubicle plantations. A few small retirement homes were hit as well. Public transport and elevators are also of concern. No known cases of outdoors spread. Household spread is observed but with proper precautions it's not as virulent like with say, nororviruses where the entire family often gets sick. Surface contamination has been greatly curtailed by ubiquitous hand sanitizer in all stores and businesses plus frequent hand washing. Mask wearing in still minimal and apart from DIYs they are still rare and precious things.

The initial response was inept to say the least and at times comical, for instance the decision to shut down border travel, etc was made for us by other governments once they started stacking up the bodies in Italy. We started testing widely for the virus a month before we had our first confirmed case using existing health care systems, a private lab rolled out massive free testing overnight on their own initiative and contact tracing teams were thrown together at the last minute. That part of the response has been and is an outstanding achievement. We had severe PPE shortages in the beginning, at one point being down to 2 days supply before public and private players managed to secure enough supplies by hook and by crook. But it was a close run thing.

The economic effects are disastrous of course but we're using the plumbing to help people and businesses that was set up in the 2008-2009 economic crash.

Overall verdict so far is I'd say we had a very incompetent start, got very lucky in the beginning but used the time fairly well. Our approach worked out to a slightly stricter Sweden but unlike them we coupled that with a massive testing/tracing/isolation effort.

Thanks for the update Leggy, good to near it's now progressing well. 

Something like this is what many W.E. or US state might need to implement to get it under control, whilst the world waits for the magic bullet vaccine to appear.

[HisMajestyBOB]

Implement policies? Why would we do that? It's going away very soon!
Time to disband the task force, stop testing, and go back to normal!
https://twitter.com/WhiteHouseCEA/status/1257680258364555264?

To better visualize observed data, we also continually update a curve-fitting exercise to summarize COVID-19's observed trajectory. Particularly with irregular data, curve fitting can improve data visualization. As shown, IHME's mortality curves have matched the data fairly well.

[Tamas]

Behavioral scientists, economists, the blond mavericks of the anglo-saxon world sure know who to turn to for advice on pandemics!