Semi-OT: Coronavirus (COVID-19) Discussion Thread

Moderators: Clav, Domejandro, ken6199, bisme37, Dirk, KingDavid, cupcakesnake, bwgood77, zimpy27, infinite11285

bidde
Senior
Posts: 556
And1: 455
Joined: Mar 13, 2020
 

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#601 » by bidde » Sat Apr 18, 2020 9:43 am

zimpy27 wrote:
Spoiler:
bidde wrote:
HotRocks34 wrote:Just a brief wrap-up on something touched upon a few pages ago in this thread. I'm putting this here to shoot down the more radical interpretation of events that was being considered. This is a new report on the investigation of the origins of the virus.

https://www.foxnews.com/politics/us-officials-investigation-coronavirus-wuhan-lab



So, the bioweapon hypothesis (and, apparently, the "altered virus structure" hypothesis) is now firmly returned to the conspiracy theory corner. It's official.


.
.
.
.


Interesting study out of Stanford (Covid antibody testing in California):
https://www.medrxiv.org/content/10.1101/2020.04.14.20062463v1.full.pdf

Summary:
- Number of actual infected persons may be 50-85 times higher than known
- CFR (mortality rate) of Covid (at least in California) may be around .12% to .20%


This data is similar to what a German study found (0.37% CFR):
https://reason.com/2020/04/09/preliminary-german-study-shows-a-covid-19-infection-fatality-rate-of-about-0-4-percent/

It's also not so far off of the current figures in Iceland. Iceland is the most tested nation per capita in the world on the virus. That makes them possibly the best "not antibody researched" study of what the Covid CFR may be.

Iceland ----------> 9 deaths/1754 cases = 0.51% CFR

A virus with a .1% CFR will kill 1 in 1000 people who catch it. A virus with a .5% CFR will kill 5 in 1000 (1 in 200) people who catch it.

As many have suspected, there are likely a lot more people who have, or have had, the virus in the USA than we know about. The current "known" CFR of the virus in the USA is around 4.6% (30,449 deaths from 666,573 cases). Data from here:
https://covidtracking.com/data

Picking a number between 0.1% and 0.5%, let's just say that the virus CFR is 0.3%. That would make the virus about 15 times less deadly than it currently appears to be based on the known numbers. That's a welcome thought.

Of course, the CFR for the virus is a lot higher for certain populations (elderly, those with per-existing conditions).


Comparisons:
https://en.wikipedia.org/wiki/Pandemic_severity_index#Guidelines
https://en.wikipedia.org/wiki/1957%E2%80%9358_influenza_pandemic
https://www.medicinenet.com/script/main/art.asp?articlekey=208914

Average flu ----------> 0.1% CFR (or less; kills roughly 291,000 to 646,000 globally per year)
1957 flu --------------> 0.3% CFR (in the UK; killed 1-2 million people globally)
Spanish flu ----------> 2.0%+ CFR or higher (killed 17-100 million globally)


People usually differentiate between CFR (case fatality rate) and IFR (infection fatality rate). CFR uses confirmed cases as the denominator, while IFR uses all infected. So you can't really be wrong about the CFR. IFR is the more interesting and difficult question, but scientists believed it to be a lot lower than some of those CFRs for a while. For example the Imperial College Study (the one that predicted 2.2 million deaths in the US without intervention) was published a month ago and assumed an IFR of 0.9% and a study focused on estimating IFR published in late March found a 0.4 %- 1.3% confidence interval for IFR.

As far as the Stanford Study goes, ideally you would want a random sample of a population. But this is how they found their participants:
We recruited participants by placing targeted advertisements on Facebook aimed at residents of Santa Clara County.


The problem with an approach like this is that you might introduce a bias, for example you end up getting more people that are worried about having covid, because they had symptoms. That could lead to the prevalence being overestimated and the IFR being underestimated.

Also, I was kind of shocked to see those CIs on the sensitivity of the used test:
Our estimates of sensitivity based on the manufacturer’s and locally tested data were 91.8% (using the lower estimate based on IgM, 95 CI 83.8-96.6%) and 67.6% (95 CI 50.2-82.0%), respectively. Similarly, our estimates of specificity are 99.5% (95
CI 98.1-99.9%) and 100% (95 CI 90.5-100%).


At this point we can be pretty certain that the IFR is above 0.2% just by looking at the fatalities in NYC and some of the hard hit regions in Italy.


Imperial College Study:https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf
Estimates of the severity of coronavirus disease 2019: a model-based analysis: https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(20)30243-7/fulltext


As we discussed before, 0.2% is what the Iceland data shows as the fatality rate but that is an ideal situation, SCC is similar base don this study of antibody tests.

I wouldn't be certain IFR is over 0.2% but IFR will be different in different cities based on population density, demographics, pollution, etc.

I definitely agree on demographics and I get what you mean with pollution, but how is population density playing a role?
Fairview4Life
RealGM
Posts: 70,153
And1: 34,002
Joined: Jul 25, 2005
     

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#602 » by Fairview4Life » Sat Apr 18, 2020 12:05 pm

Interesting info on the powers the WHO doesn’t have over member states. Which makes some sense but obviously doesn’t help when people lie a lot.

Read on Twitter
9. Similarly, IF THOU HAST SPENT the entire offseason predicting that thy team will stink, thou shalt not gloat, nor even be happy, shouldst thou turn out to be correct. Realistic analysis is fine, but be a fan first, a smug smarty-pants second.
basketballRob
RealGM
Posts: 37,063
And1: 14,900
Joined: May 05, 2014
     

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#603 » by basketballRob » Sat Apr 18, 2020 12:30 pm

NY 567 wrote:If the infection rate is 50+ times higher than reported, that means the death rate is very low. If that's the case, I think we need to start opening up the country before our economy totally collapses.
I think they're hiring at hospitals right now. Did you try there?

Sent from my SM-G950U using RealGM mobile app
User avatar
BadMofoPimp
RealGM
Posts: 49,057
And1: 12,513
Joined: Oct 12, 2003
Location: In the Paint

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#604 » by BadMofoPimp » Sat Apr 18, 2020 12:40 pm

TMU wrote:
BadMofoPimp wrote:
TMU wrote:
I'd re-consider. Flu-like symptoms for a day or two are common after the flu vaccination.


I have caught the flu one time in about 35 years. How often have you gotten the vaccine?


Ah, it now makes more sense why you don't want to get it. I believe in flu vaccination and I get it every year as I'm in health care. A few years ago, I developed a generalized rash that lasted several hours after getting a vaccine - which is pretty rare. I'm still not sure what happened then, but I've had no other issues since.


If I was in Health Care, I would probably get it like you. The one time I got it was when I was visiting someone in a hospital!
Image

Provin Ya'll Wrong!!!
User avatar
Gold Dragon
Assistant Coach
Posts: 4,025
And1: 4,607
Joined: Jul 02, 2006
Location: Oz
 

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#605 » by Gold Dragon » Sat Apr 18, 2020 2:23 pm

I don't know if this has been posted yet but I just discovered this and found it very helpful to visualize and compare how countries are or are not managing covid19.



Trajectory of confirmed cases
User avatar
Nuntius
Retired Mod
Retired Mod
Posts: 23,172
And1: 23,616
Joined: Feb 28, 2012
   

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#606 » by Nuntius » Sat Apr 18, 2020 3:17 pm

Fairview4Life wrote:Interesting info on the powers the WHO doesn’t have over member states. Which makes some sense but obviously doesn’t help when people lie a lot.

Read on Twitter


Gold Dragon wrote:I don't know if this has been posted yet but I just discovered this and found it very helpful to visualize and compare how countries are or are not managing covid19.



Trajectory of confirmed cases


I want to thank both of you for your posts. Both are great resources.
"No wolf shall keep his secrets, no bird shall dance the skyline
And I am left with nothing but an oath that gleams like a sword
To bathe in the blood of man
Mankind..."

She Painted Fire Across the Skyline, Part 3
- Agalloch
User avatar
zimpy27
Forum Mod
Forum Mod
Posts: 45,621
And1: 43,867
Joined: Jul 13, 2014

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#607 » by zimpy27 » Sat Apr 18, 2020 4:27 pm

bidde wrote:
zimpy27 wrote:
Spoiler:
bidde wrote:
People usually differentiate between CFR (case fatality rate) and IFR (infection fatality rate). CFR uses confirmed cases as the denominator, while IFR uses all infected. So you can't really be wrong about the CFR. IFR is the more interesting and difficult question, but scientists believed it to be a lot lower than some of those CFRs for a while. For example the Imperial College Study (the one that predicted 2.2 million deaths in the US without intervention) was published a month ago and assumed an IFR of 0.9% and a study focused on estimating IFR published in late March found a 0.4 %- 1.3% confidence interval for IFR.

As far as the Stanford Study goes, ideally you would want a random sample of a population. But this is how they found their participants:


The problem with an approach like this is that you might introduce a bias, for example you end up getting more people that are worried about having covid, because they had symptoms. That could lead to the prevalence being overestimated and the IFR being underestimated.

Also, I was kind of shocked to see those CIs on the sensitivity of the used test:


At this point we can be pretty certain that the IFR is above 0.2% just by looking at the fatalities in NYC and some of the hard hit regions in Italy.


Imperial College Study:https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf
Estimates of the severity of coronavirus disease 2019: a model-based analysis: https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(20)30243-7/fulltext


As we discussed before, 0.2% is what the Iceland data shows as the fatality rate but that is an ideal situation, SCC is similar base don this study of antibody tests.

I wouldn't be certain IFR is over 0.2% but IFR will be different in different cities based on population density, demographics, pollution, etc.

I definitely agree on demographics and I get what you mean with pollution, but how is population density playing a role?


Population density does a couple things.
1) It spreads the virus faster which overruns hospitals faster and reduces level of care.
2) The reason it spreads faster is because there is more virus per cubic foot of space (the same reason why denser populations have more air pollution). That not only increases chance of catching it but increases the chance of a higher amount of viral load upon infection.
3) This is just an untested hypothesis, the chance of breathing virus into lower respiratory tract rather than depositing in upper respiratory tract might be higher in a denser population.
"Let's play some basketball!" - Fergie
User avatar
zimpy27
Forum Mod
Forum Mod
Posts: 45,621
And1: 43,867
Joined: Jul 13, 2014

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#608 » by zimpy27 » Sat Apr 18, 2020 4:39 pm

Promezclan wrote:
zimpy27 wrote:
bidde wrote:
People usually differentiate between CFR (case fatality rate) and IFR (infection fatality rate). CFR uses confirmed cases as the denominator, while IFR uses all infected. So you can't really be wrong about the CFR. IFR is the more interesting and difficult question, but scientists believed it to be a lot lower than some of those CFRs for a while. For example the Imperial College Study (the one that predicted 2.2 million deaths in the US without intervention) was published a month ago and assumed an IFR of 0.9% and a study focused on estimating IFR published in late March found a 0.4 %- 1.3% confidence interval for IFR.

As far as the Stanford Study goes, ideally you would want a random sample of a population. But this is how they found their participants:


The problem with an approach like this is that you might introduce a bias, for example you end up getting more people that are worried about having covid, because they had symptoms. That could lead to the prevalence being overestimated and the IFR being underestimated.

Also, I was kind of shocked to see those CIs on the sensitivity of the used test:


At this point we can be pretty certain that the IFR is above 0.2% just by looking at the fatalities in NYC and some of the hard hit regions in Italy.


Imperial College Study:https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf
Estimates of the severity of coronavirus disease 2019: a model-based analysis: https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(20)30243-7/fulltext


As we discussed before, 0.2% is what the Iceland data shows as the fatality rate but that is an ideal situation, SCC is similar base don this study of antibody tests.

I wouldn't be certain IFR is over 0.2% but IFR will be different in different cities based on population density, demographics, pollution, etc.

Scariest is Ecuador, which also shows that the numerator is inaccurate, not just the denominator.
Image
Around 6,000 people, or close to 0.2% of the entire population of Guayas (population 3.6 million) were killed by coronavirus in only 2 weeks, and that's with a very young population. So leaving it unchecked and hoping it will turn out like the flu just won't work.

In the chart, we have death totals by region per month, and for the first two weeks of April. Half of Zone 8/Guayas's death toll in January would be 970 - instead 6700 were killed in the first two weeks of April, an excess of more than 5,700 deaths. The other less-affected regions serve as perfect controls (for seasonality or deaths of despair from seeing the stock market drop), so it's really undeniable that coronavirus was the cause. In almost all zones (besides Zone 5, which is next to Guayas), the total for those two weeks in April is actually well under half of the totals for January or February (despite having some coronavirus deaths of their own), since social distancing stops traffic accidents and the spread of other diseases like flu.



This virus isn't the flu, that's important to stress for a number of reasons. Our population has no immunity against it, this virus attacks in a different way, this virus kills in a different age proportion to the flu.

These all mean that this virus isn't necessarily killing the same people that would die from the flu. So it is additive.

It's killing more and sends many more to hospital for life-saving treatment that may otherwise die.

The mortality rate can increase rapidly based on the health care system in the country. That's an important factor I didn't list above. Probably the most important.

But just about every health system is built on efficiency and built to handle the disease predicted based on the year before. No hospital system is prepared for this virus. There aren't enough doctors and nurses anywhere to maintain a standard level of care for this virus.

In many ways I'm glad the mortality rate has been presented at around 3%. Scares more people, makes them more likely to adhere to lockdowns. It would be too difficult to say it was 0.3% but we still need to be in lockdown otherwise that could jump much higher.
"Let's play some basketball!" - Fergie
NY 567
Lead Assistant
Posts: 5,237
And1: 7,438
Joined: Dec 18, 2016

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#609 » by NY 567 » Sat Apr 18, 2020 4:56 pm

TMU wrote:For those who are basically saying that they don't believe in the flu vaccine therefore they won't be getting the Covid-19 vaccine, are you insinuating that the Covid-19 is just another variant of the seasonal flu? Still in denial of how potent this virus is?


Is it really that potent though? If it's 50+ times as widespread as reported, that means it's not that much more deadly than the flu. The main issue is how easily it spreads and how it could infect the compromised+elderly and overwhelm our hospitals. But at this point, it's probably spread throughout the country, the death toll isn't too tremendous and a lot of people already have antibodies for it. I still think we need to be on guard for those who are in the high risk bracket, but now that the fog of war is starting to clear, it doesn't seem as deadly as it once did. I don't know if it's worth it to wait until a vaccine comes along, that could be a year away, and closing our country down for that long will cause more chaos than this virus will imo
Triples333
Assistant Coach
Posts: 3,786
And1: 3,672
Joined: Sep 05, 2016

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#610 » by Triples333 » Sat Apr 18, 2020 5:14 pm

Sweden has been a very interesting test case I have been monitoring. They are recommending the immune deficient and elderly stay home when possible (many do not), and for those that can work from home do, but they never shut down any business (restaurants, etc are all running as usual) and apparently masks are not prevalent in any sense. Schools for all kids 16 and under remain open. In a nation of >10 million they currently have ~13K official cases and 1,511 deaths. Their curve does appear to be flattening and there are thoughts of a herd immunity already beginning to develop. It is a topic of huge national debate within their country. Their smaller Scandinavian neighbors did practice massive shutdowns/social distancing and do have predictably lesser death rates (but if a herd immunity is indeed being formed in Sweden, some argue they are simply getting ahead of it). Again, very interesting to monitor.
User avatar
TMU
Forum Mod - Rockets
Forum Mod - Rockets
Posts: 30,188
And1: 10,413
Joined: Jan 02, 2005
Location: O.R.
       

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#611 » by TMU » Sat Apr 18, 2020 5:45 pm

NY 567 wrote:
TMU wrote:For those who are basically saying that they don't believe in the flu vaccine therefore they won't be getting the Covid-19 vaccine, are you insinuating that the Covid-19 is just another variant of the seasonal flu? Still in denial of how potent this virus is?


Is it really that potent though? If it's 50+ times as widespread as reported, that means it's not that much more deadly than the flu. The main issue is how easily it spreads and how it could infect the compromised+elderly and overwhelm our hospitals. But at this point, it's probably spread throughout the country, the death toll isn't too tremendous and a lot of people already have antibodies for it. I still think we need to be on guard for those who are in the high risk bracket, but now that the fog of war is starting to clear, it doesn't seem as deadly as it once did. I don't know if it's worth it to wait until a vaccine comes along, that could be a year away, and closing our country down for that long will cause more chaos than this virus will imo


I would say it is given the number of incidence and how quickly the disease can progress in certain patients. Unlike the flu we still don’t know a whole lot about the Covid-19, and I think it would be better to err on being more cautious before prematurely re-opening the country. When I say that I don’t mean that states should wait until a vaccine becomes approved by the FDA. States could potentially reopen with more people being tested with potentially isolating those who are positive. This ensures a degree of safety for society while a vaccine or treatment is developed. Right now there’s still too much handwaving without any proper execution in place.
HotRocks34
RealGM
Posts: 17,198
And1: 21,129
Joined: Jun 23, 2007

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#612 » by HotRocks34 » Sat Apr 18, 2020 5:49 pm

Good news from New York.

Based upon Cuomo's press conference today (his numbers are different than the ones De Blasio is using for New York City), there were 540 fatalities from the virus yesterday in the state (that number will likely go up later once "probable deaths" are added).

That is the lowest daily figure since April 2 (16 days ago).

If the trends hold, New York may finally be through the worst.
Jokic 31/21/22
Luka & Oscar = 5 x 27/8/8
The Brodie = All-out energy
nikster
RealGM
Posts: 14,436
And1: 12,937
Joined: Sep 08, 2013

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#613 » by nikster » Sat Apr 18, 2020 6:15 pm

lakerz12 wrote:
nikster wrote:
steger_3434 wrote:The point of this was to flatten the curve, not make people not get it. The graph shows the exact same amount of people getting it just spread out over more time so the healthcare system isn’t overrun. Besides a couple hot spots the hospitals are ok. I went to a larger hospital in Milwaukee this past Saturday and it was the slowest I’ve ever seen it. What we did worked. We flattened the curve. Now open up society and continue with life while being more cautious


Sent from my iPhone using RealGM Forums

US was getting about 2000 deaths daily the past 5 days, yesterday was a slight drop but still above 1500. Its a bit premature to think the peak is passed overall (and some US states that got cases later are yet to peak)


If you trust this model then deaths did indeed peak 3 days ago, as far as USA total: https://covid19.healthdata.org/united-states-of-america

But if that model turns out to be correct we don't know. But it is appearing that deaths are on the decline now.

This website also shows daily new cases and new deaths leveling and dropping: https://www.worldometers.info/coronavirus/country/us/

Of course yes some individual states may peak slightly later.

Only 5 days later but looks like model was incorrect. we will see how close they are but US had 6000 deaths on the 14th and has been above 2000 deaths per day since then.
nikster
RealGM
Posts: 14,436
And1: 12,937
Joined: Sep 08, 2013

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#614 » by nikster » Sat Apr 18, 2020 6:20 pm

NY 567 wrote:
TMU wrote:For those who are basically saying that they don't believe in the flu vaccine therefore they won't be getting the Covid-19 vaccine, are you insinuating that the Covid-19 is just another variant of the seasonal flu? Still in denial of how potent this virus is?


Is it really that potent though? If it's 50+ times as widespread as reported, that means it's not that much more deadly than the flu. The main issue is how easily it spreads and how it could infect the compromised+elderly and overwhelm our hospitals. But at this point, it's probably spread throughout the country, the death toll isn't too tremendous and a lot of people already have antibodies for it. I still think we need to be on guard for those who are in the high risk bracket, but now that the fog of war is starting to clear, it doesn't seem as deadly as it once did. I don't know if it's worth it to wait until a vaccine comes along, that could be a year away, and closing our country down for that long will cause more chaos than this virus will imo

it would be nice if it really was that wide spread but that seems just like hopeful thinking at this point

Look at countries that are doing tons of testing as a proportion of the population. Positive results are in 10% of patients or less. Its 2% in south korea for example.
EH15
General Manager
Posts: 8,099
And1: 7,157
Joined: Jan 15, 2008

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#615 » by EH15 » Sat Apr 18, 2020 6:26 pm

nikster wrote:
lakerz12 wrote:
nikster wrote:US was getting about 2000 deaths daily the past 5 days, yesterday was a slight drop but still above 1500. Its a bit premature to think the peak is passed overall (and some US states that got cases later are yet to peak)


If you trust this model then deaths did indeed peak 3 days ago, as far as USA total: https://covid19.healthdata.org/united-states-of-america

But if that model turns out to be correct we don't know. But it is appearing that deaths are on the decline now.

This website also shows daily new cases and new deaths leveling and dropping: https://www.worldometers.info/coronavirus/country/us/

Of course yes some individual states may peak slightly later.

Only 5 days later but looks like model was incorrect. we will see how close they are but US had 6000 deaths on the 14th and has been above 2000 deaths per day since then.

They have not had 6000 in a day. It was probable deaths added to NY's total.

The high was yesterday of around ~2500 so far.
nikster
RealGM
Posts: 14,436
And1: 12,937
Joined: Sep 08, 2013

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#616 » by nikster » Sat Apr 18, 2020 6:38 pm

EH15 wrote:
nikster wrote:
lakerz12 wrote:
If you trust this model then deaths did indeed peak 3 days ago, as far as USA total: https://covid19.healthdata.org/united-states-of-america

But if that model turns out to be correct we don't know. But it is appearing that deaths are on the decline now.

This website also shows daily new cases and new deaths leveling and dropping: https://www.worldometers.info/coronavirus/country/us/

Of course yes some individual states may peak slightly later.

Only 5 days later but looks like model was incorrect. we will see how close they are but US had 6000 deaths on the 14th and has been above 2000 deaths per day since then.

They have not had 6000 in a day. It was probable deaths added to NY's total.

The high was yesterday of around ~2500 so far.

so we are just to completely ignore probable deaths because we dont have the testing capacity to confirm? Probable deaths might be an overestimate of deaths, but using only confirmed cases is definitely underestimate

either way, the model was wrong about the peak
EH15
General Manager
Posts: 8,099
And1: 7,157
Joined: Jan 15, 2008

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#617 » by EH15 » Sat Apr 18, 2020 6:40 pm

nikster wrote:
EH15 wrote:
nikster wrote:Only 5 days later but looks like model was incorrect. we will see how close they are but US had 6000 deaths on the 14th and has been above 2000 deaths per day since then.

They have not had 6000 in a day. It was probable deaths added to NY's total.

The high was yesterday of around ~2500 so far.

so we are just to completely ignore probable deaths because we dont have the testing capacity to confirm? Probable deaths might be an overestimate of deaths, but using only confirmed cases is definitely underestimate

either way, the model was wrong about the peak

I don't think you understand. They added deaths from the previous month all in a single day to that tally. It wasn't deaths from a single 24 hour period.

Similar to how China added a bunch of deaths a couple days ago.
User avatar
Kabookalu
RealGM
Posts: 63,103
And1: 70,115
Joined: Aug 18, 2006
Location: Long Beach, California

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#618 » by Kabookalu » Sat Apr 18, 2020 6:42 pm

nikster wrote:
EH15 wrote:
nikster wrote:Only 5 days later but looks like model was incorrect. we will see how close they are but US had 6000 deaths on the 14th and has been above 2000 deaths per day since then.

They have not had 6000 in a day. It was probable deaths added to NY's total.

The high was yesterday of around ~2500 so far.

so we are just to completely ignore probable deaths because we dont have the testing capacity to confirm? Probable deaths might be an overestimate of deaths, but using only confirmed cases is definitely underestimate

either way, the model was wrong about the peak


I don't have the tweet at hand, but I saw a model that showed confirmed covid19 deaths and overall deaths in New York compared from a month to month, year to year basis. The overall deaths in New York is astronomically higher than it usually is, even when compared to 9/11.
Read on Twitter
bidde
Senior
Posts: 556
And1: 455
Joined: Mar 13, 2020
 

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#619 » by bidde » Sat Apr 18, 2020 6:42 pm

nikster wrote:
EH15 wrote:
nikster wrote:Only 5 days later but looks like model was incorrect. we will see how close they are but US had 6000 deaths on the 14th and has been above 2000 deaths per day since then.

They have not had 6000 in a day. It was probable deaths added to NY's total.

The high was yesterday of around ~2500 so far.

so we are just to completely ignore probable deaths because we dont have the testing capacity to confirm? Probable deaths might be an overestimate of deaths, but using only confirmed cases is definitely underestimate

either way, the model was wrong about the peak


It's not about not counting them, it's about a change in reporting that resulted in a months worth of cases being attributed to a single day.

On April 14, New York City reported 3,778 additional deaths that have occurred since March 11 and have been classified as "probable," defined as follows: “decedent [...] had no known positive laboratory test for SARS-CoV-2 (COVID-19) but the death certificate lists as a cause of death “COVID-19” or an equivalent"

As with similar instances in the past (with other countries), we have added the additional data on the day it was reported. If and when the historical distribution is provided, we will make the needed adjustments.


https://www.worldometers.info/coronavirus/us-data/
nikster
RealGM
Posts: 14,436
And1: 12,937
Joined: Sep 08, 2013

Re: Semi-OT: Coronavirus (COVID-19) Discussion Thread 

Post#620 » by nikster » Sat Apr 18, 2020 6:43 pm

EH15 wrote:
nikster wrote:
EH15 wrote:They have not had 6000 in a day. It was probable deaths added to NY's total.

The high was yesterday of around ~2500 so far.

so we are just to completely ignore probable deaths because we dont have the testing capacity to confirm? Probable deaths might be an overestimate of deaths, but using only confirmed cases is definitely underestimate

either way, the model was wrong about the peak

I don't think you understand. They added deaths from the previous month all in a single day to that tally. It wasn't deaths from a single 24 hour period.

Similar to how China added a bunch of deaths a couple days ago.

ahhh i misunderstood. thanks. i figured it would have included additional days but didnt realize it was the full month

Return to The General Board