View previous topic :: View next topic |
Author |
Message |
GlennMacGrady
Joined: 03 Jan 2005 Posts: 8289 Location: Heisenberg
Back to top |
Posted: 03/12/20 12:38 am ::: Coronavirus Deaths, Death Rates, Testing, Other Stats |
Reply |
|
In this thread I will calculate from time to time the current death rates from coronavirus worldwide and in various countries. The data will be derived from the Johns Hopkins coronavirus website.
I have a tentative hypothesis that the worldwide death rate is currently distorted upwards by the huge numbers in China, which had grossly insufficient hospital space and likely has a medical system closer to third world than first world, and by Italy, which also has a shortage of hospital space.
The format will be Deaths/Confirmed Cases = Death Rate
As a reference point, the death rate for seasonal flu in the U.S. is about 0.10%.
____________________
Worldwide: 4,630/126,136 = 3.67%
(Editorial comment: Note that the worldwide deaths from CV are only ~20% of the >20,000 deaths from this season's flu in the U.S alone.)
China: 3,172/80,932 = 3.91%
Italy: 827/12,462 = 6.63%
Iran: 354/9,000 = 3.93%
S. Korea: 60/7,755 = 0.77%
France: 48/2,284 = 2.10%
Spain: 55/2,277 = 2.41%
Germany: 3/1,966 = 0.15%
U.S.: 38/1312 = 2.89%
Cruise Ship: 7/696 = 1.00%
Switzerland: 4/652 = 0.61%
Japan: 15/639 = 2.34%
Norway: 0/629 = 0.00%
Denmark: 0/516 = 0.00%
Netherlands: 5/503 = 0.99%
Sweden: 1/500 = 0.20%
U.K.: 8/459 = 1.74%
Belgium: 3/314 = 0.95%
Australia: 3/128 = 2.34%
Canada: 1/117 = 0.85%
Qatar, Austria, Bahrain, Singapore, Malaysia = 0.00%
Last edited by GlennMacGrady on 03/23/20 11:26 am; edited 3 times in total |
|
tfan
Joined: 31 May 2010 Posts: 9816
Back to top |
Posted: 03/12/20 12:50 am ::: Re: Coronavirus Death Rates |
Reply |
|
GlennMacGrady wrote: |
(Editorial comment: Note that the worldwide deaths from CV are only ~20% of the >20,000 deaths from this season's flu in the U.S |
But also note how long the flu has been in circulation and how far it has traveled and to what circulation level, versus COVID-19.
Every time I see those high flu death rates I try to recall hearing about someone I knew or someone famous dying from the flu. Nothing comes to mind, but that doesn’t mean I haven’t forgotten it. I do hear about pneumonia doing people in, particularly older people who are in the hospital for some other issue, but maybe the flu sets up the body to get pneumonia and is considered the cause.
|
|
GlennMacGrady
Joined: 03 Jan 2005 Posts: 8289 Location: Heisenberg
Back to top |
Posted: 03/12/20 12:52 am ::: |
Reply |
|
Looking at the data I just calculated gives me fodder to speculate about the ban that Trump announced yesterday on travelers from Europe, except the U.K.
Methinks the medical gurus advised Trump that the really dangerous "seed source" is Italy. They may not have wanted to single out Italy publicly for diplomatic reasons, and they may thought it conservative to widen the ban to all of contiguous Europe, which is easily accessed by travelers from Italy.
The U.K. is further away on islands and doesn't have that many confirmed cases yet. |
|
tfan
Joined: 31 May 2010 Posts: 9816
Back to top |
Posted: 03/12/20 12:56 am ::: |
Reply |
|
GlennMacGrady wrote: |
The U.K. is further away on islands and doesn't have that many confirmed cases yet. |
They will likely see a jump since so many hurried home recently to avoid the lockdown of Northern Italy. And the government let them come right on in with no contact. Although the health agency asked them to self-quarantine for 14 days which is good, and something that I don’t believe the USA government was doing.
Last edited by tfan on 03/12/20 1:01 am; edited 1 time in total |
|
GlennMacGrady
Joined: 03 Jan 2005 Posts: 8289 Location: Heisenberg
Back to top |
Posted: 03/12/20 1:00 am ::: Re: Coronavirus Death Rates |
Reply |
|
tfan wrote: |
maybe the flu sets up the body to get pneumonia and is considered the cause. |
Indeed it does, especially in elderly patients. The CDC's weekly FLUVIEW site has a section that combines flu-pneumonia deaths. I'll bet pneumonia is what ultimately kills lots of the elderly CV patients.
|
|
GlennMacGrady
Joined: 03 Jan 2005 Posts: 8289 Location: Heisenberg
Back to top |
Posted: 03/12/20 1:51 am ::: |
Reply |
|
If we subtract Washington state's 30/373 from the U.S. totals in the OP, the remaining 49 states of America have a death rate of: 8/939 = 0.85%.
Those data surely call for cancelling the currently ongoing Iditarod sled dog race in Alaska, or at least forbidding spectators at the finish line. (Sorry, my bad angels come out late at night, along with my mice.) |
|
FrozenLVFan
Joined: 08 Jul 2014 Posts: 3519
Back to top |
|
justintyme
Joined: 08 Jul 2012 Posts: 8407 Location: Northfield, MN
Back to top |
Posted: 03/12/20 9:42 am ::: |
Reply |
|
It will be interesting to see what happens with these numbers over time.
One thing to consider is the countries that have the highest death rates suffered from hospital collapse. If you run out of beds, the numbers are going to skyrocket, but also with a pandemic that is a major concern and a real potential to happen just about everywhere.
The other thing is these numbers are all still relatively small sample sizes. From what we know so far, COVID-19 has a huge variance in its death rate based upon age and other health factors. For instance, kids under 14 are basically at 0 risk of death while people over 80 have a death rate of almost 15%.
So who is getting sick in these initial cases in each country will likely cause major swings in the numbers. If most of the cases are in younger healthier individuals the rate will be significantly lower, while if it hits older people with other health issues it will skew upwards. These general numbers won't tell us the whole story until we know for sure that the cases tracked are actually a representative sample of the entire population.
If someone can find it, a list that tracks country by country rates broken down by age and health would be more informative at this point, imo.
_________________ ↑↑↓↓←→←→BA
|
|
PUmatty
Joined: 10 Nov 2004 Posts: 16393 Location: Chicago
Back to top |
Posted: 03/12/20 10:43 am ::: |
Reply |
|
Glenn, fascinating. I didn't realize you had a degree in epidemiology. You've been holding out on us all of this time.
|
|
FrozenLVFan
Joined: 08 Jul 2014 Posts: 3519
Back to top |
Posted: 03/12/20 12:21 pm ::: |
Reply |
|
justintyme wrote: |
One thing to consider is the countries that have the highest death rates suffered from hospital collapse. If you run out of beds, the numbers are going to skyrocket, but also with a pandemic that is a major concern and a real potential to happen just about everywhere.
|
If hospitals are overwhelmed, you'll also need to consider people who die from heart attacks, trauma, etc because there are no resources available to care for them. The furor over the cost of healthcare has ensured that we have no excess resources, whether it be beds, staff, or equipment, in hospitals especially during flu season. Every bed taken by a coronavirus patient means someone else won't receive care.
|
|
GlennMacGrady
Joined: 03 Jan 2005 Posts: 8289 Location: Heisenberg
Back to top |
Posted: 03/12/20 1:50 pm ::: |
Reply |
|
All I did in the OP was apply simple division to data on the Johns Hopkins site, which I've not seen any news organization do. Plus, I research for what seems to be authentic and interesting data on the internet.
I'm also interested in the infection and death rates in each of the U.S. states, but don't feel like doing 50 calculations that are changing daily. But to summarize, there have been U.S. deaths in only five states:
30 - Washington (most clustered in one nursing home of elderly)
4 - California
2 - Florida
1 - New Jersey
1 - South Dakota
Here's a current chart on the infections and deaths in the 50 states:
Next, although testing is very early, the following chart suggests the U.S. infections are already approximating the bell curve predicted by Farr's Law of Epidemics:
|
|
Luuuc #NATC
Joined: 10 Feb 2005 Posts: 21999
Back to top |
Posted: 03/12/20 5:36 pm ::: |
Reply |
|
GlennMacGrady wrote: |
All I did in the OP was apply simple division to data on the Johns Hopkins site, which I've not seen any news organization do. |
Probably because it would be basically irresponsible to publish such a flawed calculation.
Very few countries in the world have been testing comprehensively enough for their number of confirmed cases to be accurate.
_________________ Thanks for calling. I wait all night for calls like these.
|
|
5thmantheme
Joined: 11 Apr 2016 Posts: 540
Back to top |
|
jammerbirdi
Joined: 23 Sep 2004 Posts: 21046
Back to top |
Posted: 03/12/20 7:16 pm ::: |
Reply |
|
Again. Not a scientist or a doctor and numerically illiterate but I just think there’s something going on in this country with the death rate. There are by now so many people in this country who have been exposed to coronavirus. So now a government body in Ohio has this estimation. I don’t doubt that for a minute. I would guesstimate that a million or more in my state have been exposed. Compared to what I believe and what Ohio is estimating is happening there... I’m asserting that people in this country are simply not dying of coronavirus. Maybe no more or less even than the flu.
Tell me how I’m wrong._________________ Every woman who has ever been presented with a career/sex quid pro quo in the entertainment industry should come forward and simply say, “Me, too.” - jammer The New York Times 10/10/17 |
|
GlennMacGrady
Joined: 03 Jan 2005 Posts: 8289 Location: Heisenberg
Back to top |
Posted: 03/12/20 7:20 pm ::: |
Reply |
|
Luuuc wrote: |
GlennMacGrady wrote: |
All I did in the OP was apply simple division to data on the Johns Hopkins site, which I've not seen any news organization do. |
Probably because it would be basically irresponsible to publish such a flawed calculation.
Very few countries in the world have been testing comprehensively enough for their number of confirmed cases to be accurate. |
So, therefore, it would be irresponsible for any medical, academic or governmental body to publish data on confirmed cases. Or to make predictions of spread based on those data. Or to publish or testify before Congress, or opine in the media, about current or projected infection or fatality rates.
Ridiculous.
Sorry, but the only irresponsible comment is yours. |
|
jammerbirdi
Joined: 23 Sep 2004 Posts: 21046
Back to top |
Posted: 03/12/20 7:29 pm ::: |
Reply |
|
And honestly, nor will there ever be testing comprehensive enough to allow us to do anything more than guess at the exact number of people who have been exposed. What percentage of them/us get sick, etc.
And the media being responsible?
You would need to take things to the old days where they did a tuberculin skin test. On EVERYONE. Which they once did in this country. And that's never going to happen._________________ Every woman who has ever been presented with a career/sex quid pro quo in the entertainment industry should come forward and simply say, “Me, too.” - jammer The New York Times 10/10/17 |
|
Luuuc #NATC
Joined: 10 Feb 2005 Posts: 21999
Back to top |
Posted: 03/12/20 7:40 pm ::: |
Reply |
|
GlennMacGrady wrote: |
Luuuc wrote: |
GlennMacGrady wrote: |
All I did in the OP was apply simple division to data on the Johns Hopkins site, which I've not seen any news organization do. |
Probably because it would be basically irresponsible to publish such a flawed calculation.
Very few countries in the world have been testing comprehensively enough for their number of confirmed cases to be accurate. |
So, therefore, it would be irresponsible for any medical, academic or governmental body to publish data on confirmed cases. Or to make predictions of spread based on those data. Or to publish or testify before Congress, or opine in the media, about current or projected infection or fatality rates.
Ridiculous.
Sorry, but the only irresponsible comment is yours. |
There's nothing wrong with publishing statistics. But people on a basketball board should understand better than most that there are limits to the conclusions you can draw from certain statistics due to both sample size (eg. Layshia Clarendon was the best 3 point shooter in the WNBA last year), and the reliability of how they were obtained (eg. assist numbers is about the closest analogy I have)
People also don't tend to die the instant they are infected, so in the relatively early stages of an outbreak it seems inherently foolish to me to divide one number by the other and draw a conclusion from that, but maybe that's just me. I've been tracking numbers on this thing daily since the start of February and have drawn my own conclusions from them, but from those numbers, death rate is one thing I gave up on quite early for the reasons I previously mentioned.
But I will certainly follow your thread with interest.
_________________ Thanks for calling. I wait all night for calls like these.
Last edited by Luuuc on 03/12/20 7:42 pm; edited 1 time in total |
|
FrozenLVFan
Joined: 08 Jul 2014 Posts: 3519
Back to top |
Posted: 03/12/20 7:41 pm ::: |
Reply |
|
jammerbirdi wrote: |
Again. Not a scientist or a doctor and numerically illiterate but I just think there’s something going on in this country with the death rate. There are by now so many people in this country who have been exposed to coronavirus. So now a government body in Ohio has this estimation. I don’t doubt that for a minute. I would guesstimate that a million or more in my state have been exposed. Compared to what I believe and what Ohio is estimating is happening there... I’m asserting that people in this country are simply not dying of coronavirus. Maybe no more or less even than the flu.
Tell me how I’m wrong. |
These figures don't account for the lag between confirmation and deaths. Given the rapid uptick in cases (numbers have tripled in the US in the past three days), the confirmations are primarily people early in the course of the disease with minimal or no symptoms. The deaths are a population that was infected long enough to become seriously ill. To have a valid number, you need to follow the same group of patients over the course of their disease, not compare a just-diagnosed group with a group who's had time to develop complications.
Or what Luuuc said.
|
|
GlennMacGrady
Joined: 03 Jan 2005 Posts: 8289 Location: Heisenberg
Back to top |
Posted: 03/12/20 8:06 pm ::: |
Reply |
|
It's always been logically and statistically obvious that the number of confirmed cases will be less than the number of actual cases. But since the number of actual cases is not now and never will be known, the only way to work with the data, such as computing death rates, is to use the confirmed cases statistics.
Therefore, the death rate statistics that I (and others) are computing can be thought of as worst case death rates, since we know the actual number of cases is larger (by an unknown amount) than the confirmed cases.
One thing I'm trying to tease out in this thread are trends in the number of confirmed cases, deaths, and death rates -- using a uniform computation.
The most important current number is simply deaths. That's not going to change as a function of the number of tests. And everyone who doesn't die remains alive. So, to me, the number of deaths is what we should worry about. Or, more correctly for most countries including the U.S., what we shouldn't be currently worried about . . . because the numbers are so trivial compared to thousands of other lethal agents, such as other illnesses, accidents, natural disasters and criminal acts.
That last paragraph was my opinion. At midnight EDT, I'll post a factual update on the U.S. "death rate" (which I put in quotes just for Luuuc). |
|
jammerbirdi
Joined: 23 Sep 2004 Posts: 21046
Back to top |
Posted: 03/12/20 8:14 pm ::: |
Reply |
|
FrozenLVFan wrote: |
jammerbirdi wrote: |
Again. Not a scientist or a doctor and numerically illiterate but I just think there’s something going on in this country with the death rate. There are by now so many people in this country who have been exposed to coronavirus. So now a government body in Ohio has this estimation. I don’t doubt that for a minute. I would guesstimate that a million or more in my state have been exposed. Compared to what I believe and what Ohio is estimating is happening there... I’m asserting that people in this country are simply not dying of coronavirus. Maybe no more or less even than the flu.
Tell me how I’m wrong. |
These figures don't account for the lag between confirmation and deaths. Given the rapid uptick in cases (numbers have tripled in the US in the past three days), the confirmations are primarily people early in the course of the disease with minimal or no symptoms. The deaths are a population that was infected long enough to become seriously ill. To have a valid number, you need to follow the same group of patients over the course of their disease, not compare a just-diagnosed group with a group who's had time to develop complications.
Or what Luuuc said. |
Hmm. Okay, I don't understand. Confirmed cases in the US are 'primarily people in the early course of the disease with minimal or no symptoms?'
Okay. I'm not sure about that really. But whatever on that for now.
The 'long enough to become seriously ill' part. That's an assumption, I would contend. You (or that) is assuming that this disease hasn't been in the US as long or significantly as long as it has been in Italy. I can't believe that. Compared to Italy, absolutely hoards of human beings traveled here from China in the months of Dec-Jan. These are numbers that would dwarf the entirety of people entering Italy in that period. And the shipping entering the port at Long Beach with personnel from China.
I don't know, of course, but I'm suggesting that this bug has been here as long as it has been in Italy. Maybe I need to be shown a graph or something. But my gut tells me that California is the first stop in the west for anything and everything coming out of the far east. We'll see.
I'm starting to think that there is a possibility that this won't be as deadly as it has been in Italy and that the death rate might more resemble a seasonal flu. But that's just me._________________ Every woman who has ever been presented with a career/sex quid pro quo in the entertainment industry should come forward and simply say, “Me, too.” - jammer The New York Times 10/10/17 |
|
FrozenLVFan
Joined: 08 Jul 2014 Posts: 3519
Back to top |
Posted: 03/12/20 8:17 pm ::: |
Reply |
|
Your statistical method is flawed. You can't use numerators and denominators from different groups. We'll have to agree to disagree.
What we need to worry about are the subgroup of patients who become seriously ill and threaten to overwhelm our healthcare system. China reported 15-20% of cases were ill enough to require hospitalization. Who knows how valid their statistics are. China also reported a subgroup of younger patients (20's-50's) who died...physicians who were over-exposed and over-worked taking care of patients.
|
|
jammerbirdi
Joined: 23 Sep 2004 Posts: 21046
Back to top |
Posted: 03/12/20 8:18 pm ::: |
Reply |
|
So back to that first point. So why, with so few testing kits out there and so few tests being done, are so many confirmed cases among people aren't showing symptoms? Is it because they were on cruise ships or something? It just doesn't seem to follow that people without symptoms would be getting tested for coronavirus._________________ Every woman who has ever been presented with a career/sex quid pro quo in the entertainment industry should come forward and simply say, “Me, too.” - jammer The New York Times 10/10/17 |
|
jammerbirdi
Joined: 23 Sep 2004 Posts: 21046
Back to top |
Posted: 03/12/20 8:20 pm ::: |
Reply |
|
FrozenLVFan wrote: |
Your statistical method is flawed. You can't use numerators and denominators from different groups. We'll have to agree to disagree.
What we need to worry about are the subgroup of patients who become seriously ill and threaten to overwhelm our healthcare system. China reported 15-20% of cases were ill enough to require hospitalization. Who knows how valid their statistics are. China also reported a subgroup of younger patients (20's-50's) who died...physicians who were over-exposed and over-worked taking care of patients. |
I certainly agree with your first sentence in the second paragraph.
So is it possible to be over-exposed to the virus? I've never heard of that before._________________ Every woman who has ever been presented with a career/sex quid pro quo in the entertainment industry should come forward and simply say, “Me, too.” - jammer The New York Times 10/10/17 |
|
FrozenLVFan
Joined: 08 Jul 2014 Posts: 3519
Back to top |
Posted: 03/12/20 8:29 pm ::: |
Reply |
|
jammerbirdi wrote: |
FrozenLVFan wrote: |
Your statistical method is flawed. You can't use numerators and denominators from different groups. We'll have to agree to disagree.
What we need to worry about are the subgroup of patients who become seriously ill and threaten to overwhelm our healthcare system. China reported 15-20% of cases were ill enough to require hospitalization. Who knows how valid their statistics are. China also reported a subgroup of younger patients (20's-50's) who died...physicians who were over-exposed and over-worked taking care of patients. |
I certainly agree with your first sentence in the second paragraph.
So is it possible to be over-exposed to the virus? I've never heard of that before. |
Lack of gowns, masks, gloves, etc meaning those physicians were exposed to the disease almost continuously.
|
|
Luuuc #NATC
Joined: 10 Feb 2005 Posts: 21999
Back to top |
Posted: 03/12/20 8:44 pm ::: |
Reply |
|
Yeah, dosage matters. A small enough dosage of virus particles and due to probabilities, it might not even result in an illness. A high enough dosage and the immune system is immediately overwhelmed and doesn't have time to react and fight it off.
(Disclaimer - That's my viruses-for-dumb-people understanding of it. Not to be taken as medical advice!)
_________________ Thanks for calling. I wait all night for calls like these.
|
|
|
|