Coronavirus

[ohmman]

Just a quick note to those who are in a secure financial situation right now. If you can manage it, move your planned annual donations up and donate now instead of (or, if possible, in addition to) doing it at the end of the year. Nonprofits are struggling, especially those which serve an already economically insecure segment of the population. Having cash on hand today is really important.

Food banks, food runners, and education foundations that are trying to provide food to schoolchildren who typically qualify for free or discounted breakfast/lunch are at the top of my list, but there are obviously a myriad of others.

I recognize not everyone feels secure enough financially to shift things around, but if you do, it's something to consider.

 

[EVNow]

Why is France having more deaths than the US with less cases or why is France not posting new or active case data?

View attachment 527528

France looks incongruous to me looking at the table sorted by total deaths.

Serious question or rhetorical ?

Aggregating by country (esp. large countries) with different outbreak starting points, density & testing regimen gives incongruous data.

For eg. NY is looking worse than the worst French province.

 

[jhm]

Here is a link to an interview of a New York doctor who has treated 699 COVID-19 patients with hydroxychloroquine/Azithromycin plus zinc and has had 100% success. Rudy Giuliani did the interview. The doctor says the reason for his high success is that he starts the treatments quickly enough to keep the patient from going into the very dangerous battle within the lungs known as ARDS. He says once a patient gets ARDS it's about a 50/50 chance of survival. Personally, I think 100% success ratio is a game changer and the secret is to start the treatment soon enough. The ventilators are essential for treating ARDS and the drugs can keep a patient from reaching that point. This is huge.

Oh so Giuliani is a medical researcher these days? I would rather lick public doorknobs than get "info" from Giuliani.

 

[EinSV]

An interesting new modeling effort from Imperial College London. I haven't reviewed in detail but a few things that are likely to get some attention.

Caveats: prepublication, not peer reviewed, large error bars on everything.

1. Estimated lives saved so far through interventions in 11 major European countries: 59,000

2. They estimate orders of magnitude fewer infections detected than true infections due to mild and asymptomatic infections and limited testing capacity ("attack rate" is "proportion of the population infected to date" (through 3/28)):

3. This chart is really interesting. As noted above, they estimate 4.9% of the European population has been infected (as of 3/28), including ~10% in Spain and ~15% in Italy (huge error bars). But still not close to herd immunity, which they estimate requires 50-75%

4. They conclude it is not clear if existing measures in place across Europe will be sufficient to begin shrinking growth:

"Overall, we cannot yet conclude whether current interventions are sufficient to driveRt below 1 (posterior probability of being less than 1.0 is 44% on average across the countries). We are also unable to conclude whether interventions may be different between countries or over time."

https://www.imperial.ac.uk/media/im...urope-estimates-and-NPI-impact-30-03-2020.pdf

 

[jhm]

Quick question: Is the graph normalized by population? Or is this total number of cases independent of population?

I prefer per capita charts. The pandemic is widespread at this point, so it is not like clusters are limited geographically. So the more interesting questions are what fraction of the pop will ultimately become a case or die.

This chart will be updated in about 4 hours.

 

[bkp_duke]

An interesting new modeling effort from Imperial College London. I haven't reviewed in detail but a few things that are likely to get some attention.

Caveats: prepublication, not peer reviewed, large error bars on everything.

1. Estimated lives saved so far through interventions in 11 major European countries: 59,000

2. They estimate orders of magnitude fewer infections detected than true infections due to mild and asymptomatic infections and limited testing capacity:
View attachment 527553

3. This chart is really interesting. As noted above, they estimate 4.9% of the European population has been infected (as of 3/28), including ~10% in Spain and ~15% in Italy (huge error bars). But still not close to herd immunity, which they estimate requires 50-75%

View attachment 527552

4. They conclude it is not clear if existing measures in place across Europe will be sufficient to begin shrinking growth:

"Overall, we cannot yet conclude whether current interventions are sufficient to driveRt below 1 (posterior probability of being less than 1.0 is 44% on average across the countries). We are also unable to conclude whether interventions may be different between countries or over time."

https://www.imperial.ac.uk/media/im...urope-estimates-and-NPI-impact-30-03-2020.pdf

So back of the napkin math . . . if there is any validity to these numbers.

Taking Italy since they are about 10% through their population being infected. To date that's about 11,500 deaths.

To get to herd immunity levels, 60% immune in the population, you are looking at a potential 6X increase in the number of deaths, but lets say they get caught up and stop overwhelming their hospital system, and cut that in half. 34,500 deaths with those assumptions, till herd immunity is reached. That's about 575 deaths per 1 million.

If that translates to the US - we see where the administration got their 200,000 deaths number . . . and gives us some insight into their (potential modeling).

/ramblings of a man stuck in front of a computer and bored

 

[EVNow]

I made a post in the TSLA Daily Trading Charts thread that touches heavily upon COVID-19.

Here's the link.
{Edit: Error in link has been corrected}

If you'd like to comment on that post, please do so in the coronavirus thread, not the daily trading charts thread. Thx!

Here is a new paper (pre-publish) - but with control groups.

Efficacy of hydroxychloroquine in patients with COVID-19: results of a randomized clinical trial : COVID19

Key findings: For the 62 COVID-19 patients, 46.8% (29 of 62) were male and 53.2% (33 of 62) were female, the mean age was 44.7 (15.3) years. No difference in the age and sex distribution between the control group and the HCQ group. But for TTCR, the body temperature recovery time and the cough remission time were significantly shortened in the HCQ treatment group. Besides, a larger proportion of patients with improved pneumonia in the HCQ treatment group (80.6%, 25 of 32) compared with the control group (54.8%, 17 of 32). Notably, all 4 patients progressed to severe illness that occurred in the control group. However, there were 2 patients with mild adverse reactions in the HCQ treatment group.

Significance: Among patients with COVID-19, the use of HCQ could significantly shorten TTCR and promote the absorption of pneumonia.​

 

[TheTalkingMule]

So back of the napkin math . . . if there is any validity to these numbers.

Taking Italy since they are about 10% through their population being infected. To date that's about 11,500 deaths.

To get to herd immunity levels, 60% immune in the population, you are looking at a potential 6X increase in the number of deaths, but lets say they get caught up and stop overwhelming their hospital system, and cut that in half. 34,500 deaths with those assumptions, till herd immunity is reached. That's about 575 deaths per 1 million.

If that translates to the US - we see where the administration got their 200,000 deaths number . . . and gives us some insight into their (potential modeling).

/ramblings of a man stuck in front of a computer and bored

It's like we're making the same assumptions all over again, based on the same source no less.

So if 10% of Italy is infected, we need 6x more people die before getting to herd immunity? Nevermind that all the most at-risk citizens have already contracted the virus and died.

If this new Imperial College report is where they're getting the 100-200k deaths scenario in the US I'll be even more disappointed in our leadership. Didn't think that was humanly possible.

Testing, testing, testing. It's practically April and we have no widespread testing plan.

 

[Daniel in SD]

large error bars on everything.

You ain't kidding!

So back of the napkin math . . . if there is any validity to these numbers.

Taking Italy since they are about 10% through their population being infected. To date that's about 11,500 deaths.

To get to herd immunity levels, 60% immune in the population, you are looking at a potential 6X increase in the number of deaths, but lets say they get caught up and stop overwhelming their hospital system, and cut that in half. 34,500 deaths with those assumptions, till herd immunity is reached. That's about 575 deaths per 1 million.

If that translates to the US - we see where the administration got their 200,000 deaths number . . . and gives us some insight into their (potential modeling).

/ramblings of a man stuck in front of a computer and bored

That is extremely wishful thinking. Deaths can lag infections by a month (~3 weeks from symptoms to death). Look at Korea. Also I think 60% is extremely optimistic. (EDIT: they use 3.87 for R0 in the paper before interventions, I think that would be 74%)

 

[traxila]

Here is a new paper (pre-publish) - but with control groups.

Efficacy of hydroxychloroquine in patients with COVID-19: results of a randomized clinical trial : COVID19

Key findings: For the 62 COVID-19 patients, 46.8% (29 of 62) were male and 53.2% (33 of 62) were female, the mean age was 44.7 (15.3) years. No difference in the age and sex distribution between the control group and the HCQ group. But for TTCR, the body temperature recovery time and the cough remission time were significantly shortened in the HCQ treatment group. Besides, a larger proportion of patients with improved pneumonia in the HCQ treatment group (80.6%, 25 of 32) compared with the control group (54.8%, 17 of 32). Notably, all 4 patients progressed to severe illness that occurred in the control group. However, there were 2 patients with mild adverse reactions in the HCQ treatment group.

Significance: Among patients with COVID-19, the use of HCQ could significantly shorten TTCR and promote the absorption of pneumonia.​

This is definitely something. We should have a couple more studies soon.

 

[traxila]

Oh so Giuliani is a medical researcher these days? I would rather lick public doorknobs than get "info" from Giuliani.

I despise Giuliani. From before 9/11.

This isn't a study, but what I saw in the interview was high risk patients (above 60 or preexisting conditions) being treated across the board at first sign of disease with HCQ, AZT and zinc sulfate. Yes, call it observational or anecdotal, but this information cannot just be ignored if you have a sample size above 500 and absolutely no progression to ARDS in a patient population that if really infected should be hitting the hospital at a 20% clip. These cases and numbers need to be confirmed. If this is what it says it is, should not simply be dismissed.

 

[SageBrush]

News from Denmark: So far, so good.

Denmark has a growth rate way below some other countries but I'm not seeing bending. Below are the last 10 days

 

[jhm]

It's like we're making the same assumptions all over again, based on the same source no less.

So if 10% of Italy is infected, we need 6x more people die before getting to herd immunity? Nevermind that all the most at-risk citizens have already contracted the virus and died.

If this new Imperial College report is where they're getting the 100-200k deaths scenario in the US I'll be even more disappointed in our leadership. Didn't think that was humanly possible.

Testing, testing, testing. It's practically April and we have no widespread testing plan.

It could be that 90% of the most at-risk have not yet been exposed to the virus. If herd immunity means knocking off the most frail first, what's the value in it?

 

[EVNow]

An interesting new modeling effort from Imperial College London. I haven't reviewed in detail but a few things that are likely to get some attention.

Caveats: prepublication, not peer reviewed, large error bars on everything.

1. Estimated lives saved so far through interventions in 11 major European countries: 59,000

2. They estimate orders of magnitude fewer infections detected than true infections due to mild and asymptomatic infections and limited testing capacity ("attack rate" is "proportion of the population infected to date" (through 3/28)):
View attachment 527553

3. This chart is really interesting. As noted above, they estimate 4.9% of the European population has been infected (as of 3/28), including ~10% in Spain and ~15% in Italy (huge error bars). But still not close to herd immunity, which they estimate requires 50-75%

View attachment 527552

4. They conclude it is not clear if existing measures in place across Europe will be sufficient to begin shrinking growth:

"Overall, we cannot yet conclude whether current interventions are sufficient to driveRt below 1 (posterior probability of being less than 1.0 is 44% on average across the countries). We are also unable to conclude whether interventions may be different between countries or over time."

https://www.imperial.ac.uk/media/im...urope-estimates-and-NPI-impact-30-03-2020.pdf

Note that the infection rate is back-calculated based on fatalities.

Josh Michaud on Twitter

The authors use reported deaths to back-calculate an estimated attack rate in each country through March 28.​

 

[kyzhk]

I despise Giuliani. From before 9/11.

This isn't a study, but what I saw in the interview was high risk patients (above 60 or preexisting conditions) being treated across the board at first sign of disease with HCQ, AZT and zinc sulfate. Yes, call it observational or anecdotal, but this information cannot just be ignored if you have a sample size above 500 and absolutely no progression to ARDS in a patient population that if really infected should be hitting the hospital at a 20% clip. These cases and numbers need to be confirmed. If this is what it says it is, should not simply be dismissed.

Not sure if its been done yet - one way to quickly confirm efficacy of hydroxychloroquine would be to check the proportion (or lack thereof) of patients who have lupus / arthritis with confirmed covid19 and hospitalization rate. A significant difference in the distribution of lupus/arthritis patients among the covid19 positive patient compared to general population should confirm efficacy rather easily given the number of confirmed covid19 cases we already have.

 

[JRP3]

Medium is an open blogging platform that anyone can post to. The author here cites no scientific study, just conjecture. I can post a hundred SeekingAlpha articles with charts and graphs proving Tesla is doomed - did you read this article with the same skepticism that you would them? If not, it's worth asking yourself why.

The entire country of South Korea believes that fans left on overnight kill people. There are plenty of blog posts to support this, and fans in South Korea are legally required to have timers that automatically switch them off to protect people. There is no science to back this either.

Odd then that all the medical professionals are begging for masks, they seem to think masks make a difference. The author did in fact support his premise with scientifically valid reasoning. I also have enough understanding of basic physics to know that additional layers between you and particles in the air reduce the chance of ingestion. Feel free to ignore reality if you wish.

 

[SpaceNuts]

Data through Mar 30

Least effective:
US
Turkey

Best Benders
Switzerland
Italy

View attachment 527444

Thanks for your daily update. This is great information. May I ask how do you get your raw data?

 

[EVNow]

Odd then that all the medical professionals are begging for masks, they seem to think masks make a difference. The author did in fact support his premise with scientifically valid reasoning. I also have enough understanding of basic physics to know that additional layers between you and particles in the air reduce the chance of ingestion. Feel free to ignore reality if you wish.

I always assumed the downplaying the importance of masks was to stop any run on them and hoarding - so it is available to healthcare workers.

Didn't really matter - as the masks are nowhere to be found and healthcare workers don't have enough of them.

Europe & US spent 2 months - basically doing nothing (end of December to end of Feb) to prepare for the pandemic. Inexcusable.

ps : 1918.

 

[SageBrush]

Thanks for your daily update. This is great information. May I ask how do you get your raw data?

CSSEGISandData/COVID-19

IIRC, the source data is from the JHU database. I turn the table into a CSV by clicking on the 'raw' button, and then copy/paste into a Google spreadsheet. PM me if you would like access to my Google SS since it has some housecleaning built in.

The data used to also break the USA down by state but in the past ~ week they are only reporting a cumulative value. I was sad to see the change.

 

[TheTalkingMule]

It could be that 90% of the most at-risk have not yet been exposed to the virus. If herd immunity means knocking off the most frail first, what's the value in it?

The new report starts with the assumption real infection is an order of magnitude greater than identified cases. For Italy, that means it's everywhere since identified cases are so widespread. The frailest of folks have literally zero immune response, so one sneeze can take out a whole auditorium whereas a room of 28 year olds might all be fine.

Obviously I'm talking out of my ass, but I think it's a rational line of thinking.

I've seen cases near me where 85+ year olds haven't even left a controlled environment for 3-4 weeks then test positive out of the blue. Nursing home residents contracting it from staff, etc. And we're still just now seeing significant case growth in PA.

I think it's everywhere to some degree and certainly is in Italy.