SARS-CoV-2 Coronavirus (COVID-19)

Interesting survey of infectious disease experts, with their best estimates of coming case and death tolls:
Infectious Disease Experts Don’t Know How Bad The Coronavirus Is Going To Get, Either | FiveThirtyEight

Survey administered March 16-17, combined / modeled / posted March 18, picked up by news outlet March 20.

Projections to March 23, March 29, and end of 2020. Since today is March 23, the judging can begin.

====================================

"1.Experts ... predict 10,567 total cases (80% uncertainty interval: 7,061-24,180 cases) of COVID-19 will be reported by COVID Tracker on Monday March 23rd.

Predicted number of cases (range) Predicted probability
0–7,500 0.13
7,500 –10,000 0.29
10,000 –12,500 0.24
12,500 –15,000 0.10
15,000 –17,500 0.05
17,500 –20,000 0.04
20,000 + 0.15."

Consensus forecast (50% probability) for March 23: 10,567 cases. Actual reported cases just past midnight: 42,164

====================================

"3. Experts anticipate 19 US states will report more than 100 cases of COVID-19 within one week (80% uncertainty interval:10-36 states)"

Consensus forecast: 19 states. Actual: 36 states:

====================================

(This must have been pulled directly from authors by reporters, as I don't see it in the academic posting.)

"How many total COVID-19 cases in the U.S. will the CDC report on March 29?

The consensus forecast generated by the individual responses indicates that we should expect roughly 19,000 reported cases by March 29, with an 80 percent chance of seeing between 10,500 and 81,500 cases."

Consensus forecast: 19,000 cases
Actual: Don't know yet, because this target date is still 6 days in the future. But we blew past this consensus figure on March 20, 9 days early.

====================================

"How many people will die in the U.S. due to COVID-19 this year?"

Considering how this group so severely missed (underestimated) the cases (the 50% consensus estimate, not the high end extremes) just one and two weeks out, I don't believe it is worthwhile to include the end-of-year projection.

 

Yes, vitamin C is quickly spilled into the urine. But it does an amazing job before leaving. Like curing Covid-19 in the case of the Chinese usage mentioned above. Small amounts clear in a matter of a few hours, so effective therapy requires either huge amounts or regular dosing on the order of every few hours. But any amount takes a load off your immune system.

The issue of how much to take is similar to the question of how much water does it take to put out a fire. Dose, dose, dose. 100 mg isn't going to do much with a 10,000 mg cold, let alone a 100,000 mg one. And 10,000 mg that is used up / pissed out in 4 hours isn't much use for the next 20 hours.

The idea of even massive IV doses seems archaic in this day and age. I'd like to see something like an insulin pump, controlled by a sensor/microprocessor that maintains a target blood level continuously.

High dose vitamin C has been around for over 80 years. It's a medical miracle, but a political failure. A generation of doctors have been trained to believe that it's good for scurvy, but not much else. A few doctors, particularly those who follow the Orthomolecular Society, somehow have gone past the negative training to learn how to use it. But they've generally had to take a low profile to avoid punishment.

A friend just told me that IV C has been approved by the FDA for over a year. News to me. Perhaps the formal study with Covid-19 patients in WuHan will pry loose some recognition. Zero deaths and shorter hospital stays. Sounds useful to me...

 

My wife and a neighbor went to a small town Walmart to pick up a grocery order yesterday. The parking lot was packed and the employees said the crowds and sales were worse than before Christmas! People are ignoring the health advice. Do there need to be mass arrests and prison??

 

Well, they are self selecting to be culled, they are why you are self isolating.

 

Overnight I received and processed WU's. Folding at Home is really hammered from all the new interest in solving Corvid-19. I'm throwing almost 500 cores at it.

 

I understand the roads were pretty crowded with traffic too.

 

i wouldn't go near a box store now, unless the supermarkets were empty

 

Would argue this was not a problem of poor epidemiological forecasts per se but of very limited and unrepresentative data due to severe lack of testing. The problem in the U.S., particularly during this time interval, was that we had very limited counting of actual cases in the beginning that has only recently (last few days) started to improve:



Calculated Risk: March 23 Update: US COVID-19 Tests per Day #TestAndTrace

Even in the most recent available data, since ~15% of tests are positive, this high percentage continues to indicate limited testing.

We are getting closer to adequate testing to make accurate forecasts, but they will need to double the latest daily testing numbers to make accurate forecasts and for test and trace measures.

 

Yes, they have 20x as many volunteers as two weeks ago. they continue to be hammered. But importantly they are up to 12x more results.

 

Actually, they're leaning the otha way with the prisons...

If this thing gets serious enough there IS a playbook that will empty the streets and stores, but that might be another case of the cure being worse than the sickness.

 

nature ... thinning out the bad chromosomes day after day - in so many different ways.
.

 

Yes, jails are being unloaded of the lower level offenders because of lack of space to keep them properly separated, and lack of health care capacity to handle the inevitable mass illness.

Courts will decide later, after this plague has been controlled, whether to give the released inmates credit fortheir time out or whether they should return to complete the sentences inside. I thought the existing law generally required their sentence clocks to keep running. And since jails handle the shorter sentences (vs prisons get the longer ones), most of these released jail inmates should run out their detention clocks before this plague is resolved.

 

Including those who believe the press headlines. (I know, untrustworthy source, but there are other similar)

Arizona man dies after taking coronavirus 'cure' Trump touted with false claims | World news | The Guardian

 

Yeah, pretty grim when that's "good news". Hopefully beginning of a downward trend.

 

For the layman, what does that mean? What are you doing, and what hardware involved? Thanks.

 

Except that tracing is being abandoned, we are already beyond the point where it would be useful for containment. That horse is out of the barn, that ship has sailed.

Even much testing is being restricted, discouraged even for folks who are already ill, unless the results would make a difference in treatment path. Preemptive testing is consuming valuable resources that are now better applied elsewhere.

 

To clarify, this is to say once we get handle on our U.S. situation and enter current China/South Korea mode, it will help mitigate backwash of cases from other areas.

At the moment this unfortunately remains true. As an internal medicine hospitalist physician, I am confronted with this dilemma on a nearly daily basis. We are still being asked (told) to continue to only order these tests by our higher ups and ID colleagues essentially as you describe. Yet we had the opportunity to be weeks ahead on the testing curve and should have the capacity for massive population testing already, but here we are.

 

The Johns Hopkins, dynamic map is pretty good.

Bob Wilson

 

Background: Some of you may have heard of the SETI@home project way back when it started in 1999 along with Astropulse and others funded by the National Science Foundation and run/operated by Berkeley. These were projects with massive amounts of collected data to be processed. David Gedye devised and implemented a plan to make a "virtual supercomputer" composed of voluntary Internet connected home/business computers. This became known as distributed computing.

In distributed computing, each computer's "physical" CPU chip depending on its age and size may have only one "core" or multiples. Remember the "Pentium" processor? It had either 2 or 4 cores. IE 2 or 4 processors on the single CPU chip. The old 386 or 486 chip, just one core. The machine I'm on right now has a Intel Core i9 10980XE chip which has 18 cores in it's CPU chip. With the distributed computing Folding at Home client, I can devote one or all of those cores to the project it's working on at the time.

Some projects lend themselves to the type of processing that your GPU (graphics card processor) is better suited to work on. Some high end gaming machines might have up to 3 graphics cards installed. You can also let Folding at Home use your graphics card as well.

Any PC can do this. You can set the level as to home much of your PC's power it uses.

I have access to a large number of currently idle PC's and servers. That's where most of the 500 cores are coming from.

Here's a list of the current Folding at Home projects.

 

Much shorter URL here:
Johns Hopkins Coronavirus Resource Center