So, are you saying the death rate for Covid-19 cases (0.015%) is less than seasonal flu? I just want to be clear on this point. My methodology aside, you think everything going on is pointless or a hoax? Your number means 1500/0.015% or 10,000,000 in the US are infected, minimum.

Statistics != math.

 

Statistics != math.

How dare you use logic!

 

So, are you saying the death rate for Covid-19 cases (0.015%) is less than seasonal flu? I just want to be clear on this point. My methodology aside, you think everything going on is pointless or a hoax? Your number means 1500/0.015% or 10,000,000 in the US are ill, minimum.

Edit: Or, another example, about 23% of the population of New York state is ill? 46,000 confirmed cases times 100 out of a population of about 20,000,000?

Why don't you try actually reading the very next sentence where I very explicitly stated: "Unfortunately, it's not likely that rosy a picture because of the time lag involved."

But it would not surprise me at all if the total number of infected (including people that have already recovered) is well over a million -- in fact it would surprise me if it were less than that.

 

Just some data from Massachusetts, as of 3/25/2020.

There were 2,471 confirmed cases

There have been 25 confirmed deaths due to COVID-19

This is a projected death rate of 1%

The numbers change daily. Today, Massachusetts added another 679 confirmed cases. None of them have died yet.

So, again, the testing program that everyone has been decrying as being far too inadequate and not testing nearly enough people has somehow been able to detect every single case of COVID-19 in the state of Massachusetts?

Or is it that COVID-19 can only kill people that have been confirmed to have it via laboratory testing? If that's the case, then the solution is simple -- stop testing altogether so that no one else dies.

The death rate only has meaning or utility if it is the percentage of the total number of people that get infected by the disease who die. Limiting it to the percentage of people that are confirmed to have the disease renders it absolutely useless for anything. You can't even use it to make meaningful comparisons from one place to another because of the extreme degree of variability in who gets tested and how many tests are performed.

So you have to infer how many people have been infected but who never got tested based on other considerations, usually by determining the various rates in more closely monitorable groups, seeing how those rates compare to other rates that are applicable to the broader population, and then making assumptions about how to apply and adjust those rates (hence the huge range of uncertainty in the annual numbers for the seasonal flu).

 

So, again, the testing program that everyone has been decrying as being far too inadequate and not testing nearly enough people has somehow been able to detect every single case of COVID-19 in the state of Massachusetts?

Or is it that COVID-19 can only kill people that have been confirmed to have it via laboratory testing? If that's the case, then the solution is simple -- stop testing altogether so that no one else dies.

The death rate only has meaning or utility if it is the percentage of the total number of people that get infected by the disease who die. Limiting it to the percentage of people that are confirmed to have the disease renders it absolutely useless for anything. You can't even use it to make meaningful comparisons from one place to another because of the extreme degree of variability in who gets tested and how many tests are performed.

So you have to infer how many people have been infected but who never got tested based on other considerations, usually by determining the various rates in more closely monitorable groups, seeing how those rates compare to other rates that are applicable to the broader population, and then making assumptions about how to apply and adjust those rates (hence the huge range of uncertainty in the annual numbers for the seasonal flu).

I never stated a figure for total cases. As you pointed out. My figure was based on cases confirmed by testing.

I think where I erred was not qualifying the death rate as an estimate.

 

I never stated a figure for total cases. As you pointed out. My figure was based on cases confirmed by testing.

I think where I erred was not qualifying the death rate as an estimate.

But an estimate of what?

 

Nope. Not enough info.

The problem I see here, is not the total amount of fatalities, but rather how far easier it is for the CV19 to spread, and how it does so even before symptoms appear:

The new virus that causes Covid-19 disease appears to be a different beast: While it also can eventually lead to pneumonia, the virus does a great job of replicating in the upper respiratory tract, even when people don’t have any symptoms or just begin to feel sick.

https://www.vox.com/2020/2/20/21143785/coronavirus-covid-19-spread-transmission-how

What this means to me, is that even if the % of fatalities caused by CV19 were to be lower than ordinary flu, the total number of deaths caused by the former will quite possibly be much higher simply because much more people will more likely be infected by the former than by the latter. Also, who's to say that people can't be infected by both strains?

​

 

I have a hard time believing numbers for the flu as if it was this low, the flu outbrakes should be much easier to contain, no?

Regarding flu for example: https://www.cdc.gov/flu/about/keyfacts.htm

Persons can be asymptomatic and contagious. They can have very mild symptoms and be contagious. I like how the go and redefine asymptomatic flu and symptomatic flu. Well, if you have asymptomatic flu, you are still a carrier...

 

I have a hard time believing numbers for the flu as if it was this low, the flu outbrakes should be much easier to contain, no?

Regarding flu for example: https://www.cdc.gov/flu/about/keyfacts.htm

Persons can be asymptomatic and contagious. They can have very mild symptoms and be contagious. I like how the go and redefine asymptomatic flu and symptomatic flu. Well, if you have asymptomatic flu, you are still a carrier...

I'm not quite sure which numbers you are referring to. The flu mortality rate? The usual overall average estimate that has been thrown around for many years is about 0.1%. But it varies widely from year to year, as do the numbers for symptomatic illnesses, hospitalizations, and deaths. Good years to bad years span about a factor of six or so in each of these categories.

https://www.cdc.gov/flu/about/burden/index.html

The numbers in their table from 2010 through the estimates for the last two seasons put the average numbers at 28.6 million symptomatic illnesses and 24.7 thousand deaths and for an average mortality rate of 0.131%, which varies from a low 0.096% to a high of 0.176%, but half of the years are right in the 0.13% +/- 0.005%, so that's about as stable a number as we can hope for given the volatility.

The data for the 2013--2014 appear reasonably "typical" at 30 million symptomatic illnesses and 38 thousand deaths.

But here's where things get murkier. What COULD be a reasonably apples-to-apples comparison (but there's a bunch of complicating factors, too) would be the number laboratory-confirmed flu cases that resulted in death. The number of laboratory-confirmed hospitalizations is pretty easy to get -- the CDC's FluView database puts it at 35/100k by the end of the 2013--1014 season which, with a population of about 319 million back then, puts it at about 111 thousand (which is in the ballpark of the numbers I typically see of between 100k and 170k). However, the number of deaths in that group is a lot more problematic (plus, I haven't been able to track down and estimate) because it has to be inferred from the codes entered on the death certificate, few of which directly lay the blame on the flu. That's very different from the current CoVID-19 in which entities are going out of their way to identify deaths attributable to this virus. If we (somewhat naively) estimate the comparable rate we would end up with a death rate of 38k/111k or 34%, which really only underscores that we don't have adequately comparable data.

Switching gears a bit, I have frequently been coming across burden reports and vaccine effectiveness reports that generally claim that the vaccines are commonly able to prevent infection between 45% and 55% of the time, with some notable outliers when it wasn't a good match. But those same reports indicate that the estimated number of deaths avoided by the vaccine is usually in the 4000 to 6000 per year (with two notably higher years at about 10k and 12k). That doesn't seem to jive, unless the devils are in the details and the vaccine is predominantly good at preventing illness in younger people, who generally have pretty low death rates, but far less so in older adults who account for a considerably disproportionate fraction of the deaths.

 

But it would not surprise me at all if the total number of infected (including people that have already recovered) is well over a million -- in fact it would surprise me if it were less than that.

Ten million, not one million. Your death rate of 0.015% implies 10,000,000 since there are 1,500 deaths. With all respect, you are getting wrapped around the axle with the numbers and not looking at what is happening. The world would not be in lock down if the death rate was 0.015% of the number of people who are ill with Covid-19. If this were the case, the spread of the virus might not even be known. If would just be circulating with the seasonal flu and of no concern. By this logic, everything going on is a massive overreaction or a hoax.

 

(Some text removed for clarity) The world would not be in lock down if the death rate was 0.015% of the number of people who are ill with Covid-19.

(Some text removed for clarity) By this logic, everything going on is a massive overreaction or a hoax.

The numbers available now are not complete. The data are changing while people try to make sense of it. Maybe some months after the wave recedes the interesting parameters will be stable and available, but right now little can be said with certainty.

There are people and organizations in the world that can advance their causes by causing concern and panic and that may be happening now. Governments are sometimes more concerned about being thought to be doing the right thing actually doing the right thing which might be completely different from what is being done.

 

The numbers available now are not complete. The data are changing while people try to make sense of it. Maybe some months after the wave recedes the interesting parameters will be stable and available, but right now little can be said with certainty.

There are people and organizations in the world that can advance their causes by causing concern and panic and that may be happening now. Governments are sometimes more concerned about being thought to be doing the right thing actually doing the right thing which might be completely different from what is being done.

Very true. The biggest example of opportunism is the US Congress. However, this sort of thing is not of concern to the health care people dealing with the disease or the people trying to apply reasonable models to the spread and consequences of a disease. The emergency room and ICU people are not out to take advantage. Just the opposite. They are scared for themselves, their families and the population they serve. They are not being manipulated. Remember, these are hard boiled people who deal with serious illness and death for a living. We must always be careful to not mix real public health issues and policy with political/economic policy. A massive bailout of industry and finance and blatant pandering for votes is not health care policy.

Concerning "little can be said with certainty". Engineers deal with certainty or work like mad to make things as certain as possible before acting. This ideal does not exist during a pandemic. You pick numbers and go with it. All this blather now about death rates is, ultimately, pointless. It is a very, very nasty bug and not just seasonal flu. We have to deal with it the best we can.

 

We must always be careful to not mix real public health issues and policy with political/economic policy.

Yet, economic issues are health issues. It's a well documented fact that alcoholism and suicide increase during bad economic times. Depression (from loss of a job) has been shown to produce illness and death.

An uncertain number of deaths will be directly attributable to the attempted prevention of another set of uncertain deaths.

This is the danger of considering only the benefits of a particular action, and not the costs.

 

And then there are those with computer models making predictions...

 

https://www.sfgate.com/living-in-sf/article/Coyotes-are-being-seen-on-the-empty-streets-15159105.php

 

I can’t believe how quickly this Thread populates with everyone’s opinion. The truth is laying in front of us, no one knows for sure how deep this will cut, a flesh wound is a flesh wound you will die if it gets infected. But, this Virus is a living creature that can evolve it needs a host, secretly it waits for it’s next victim. If you can survive, you win, if you don’t you loose. I have been practicing Social distancing for years, I’m not a hugger, I don’t shake hands, I keep my distance, if your a close talker I move back, awkward you say, well that’s just me. I don’t want to share anything with you, however, what I don’t know is what I don’t know, and if someone wants to lick my spoon or cough into my food or drops it on the floor etc. You get the point only so much I can control and that is me. What you do is up to you, I can’t control you only the Universe can do that, the earth is about to shed a few viruses called humans and won’t be done till it’s done and the fat lady sings.

kv

 

You guys doing the math seem to be leaving out one thing. Until yesterday the test results of the Covid19 done by hospitals and private testing labs weren't even in the mix, it was only what was done at the CDC labs.

edited to remove hedges reference.

 

You guys doing the math seem to be leaving out one thing. Until yesterday the test results of the Covid19 done by hospitals and privet testing labs weren't even in the mix, it was only what was done at the CDC labs.

Note: Why are we testing hedges???

As WBahn has pointed out, that (the source of the numbers) is a flaw in many of the numbers/statistics we have posted.

My post, with the graphic was not intended to represent ‘real numbers’. I posted the information to demonstrate the mathematics of an exponential growth model, with and without a dampening factor. It did not appear that way because of the labels. No matter what the real numbers become (and several of you are correct in that we won’t know until some time after the pandemic passes) their patterns can be predicted mathematically.

 

RE: Post 1339

The epidemiologist who authored the CDC model said he got it wrong! The true result for the model is significantly LOWER!

 

Yet, economic issues are health issues. It's a well documented fact that alcoholism and suicide increase during bad economic times. Depression (from loss of a job) has been shown to produce illness and death.

An uncertain number of deaths will be directly attributable to the attempted prevention of another set of uncertain deaths.

This is the danger of considering only the benefits of a particular action, and not the costs.

+1

https://forum.allaboutcircuits.com/threads/coronavirus.166679/post-1487761