Another View of the Statistics - Mine

[Revenant]

Originally Posted by friendly fred

In the USA, we have 34,273 deaths out of 3,723,634 tested.

That's a one percent death rate (well, actually less than that) out of symptomatic patients who met the testing criteria of fever and difficulty breathing - a credible candidate for a test in the judgement of medical professionals who have gotten a lot of experience by now.

Even if deaths double from this amount, which is what the IHME model essentially predicts - that will be a 2% death rate among symptomatic patients. However, some small antibody studies, such as in California, state this:

"These prevalence estimates represent a range between 48,000 and 81,000 people infected in Santa Clara County by early April, 50-85-fold more than the number of confirmed cases."

If true we are looking at a death rate about the same as the flu.

.02/50 = 0.0004

Expressed as a percentage 0.04%

Death rate per 100,000 people who get it: 40 people, probably old and sick already.

If everyone in Texas gets infected: 30,000,000 x 0.0004 = 12,000 deaths.

IHME is predicting about one thousand deaths in Texas.

So, have we temporarily forestalled 11,000 deaths of old and sick people but sacrificed herd immunity and sacrificed our way of life? Given up the freedoms our forefathers died to get for us?

https://www.medrxiv.org/content/10.1...Ccg_ohr2FE_hZU

Potential flaws in my estimates:

1. I don't know how many are asymptomatic that were tested.

2. I don't know who received multiple tests.

3. Assumes robust and accurate testing and reporting.

4. Poor selection criteria (i.e. improperly randomized and representative) in Santa Clara County.

I'd rather hire Stevie Wonder as my chauffeur than hire you as a statistician.

[friendly fred]

Originally Posted by Revenant

I'd rather hire Stevie Wonder as my chauffeur than hire you as a statistician.

You couldn't afford me and I don't work for gay people such as yourself.

[friendly fred]

Originally Posted by EagleEye

Those studies are interesting but have not been formally peer reviewed.


The other data point that I am curious about is the definition of fatalities. Are these studies only hospital based COVID-19 deaths or morgue based. Not all people who are dying of this are doing so in a hospital. I did not dive deep enough into these studies to determine if non-hospital fatalities are factored in, and if not, what are the numbers dying in a non-hospital setting.

Valid points.

[1blackman1]

They’ve only recently started trying to add non-hospital deaths.

[Revenant]

Originally Posted by friendly fred

In the USA, we have 34,273 deaths out of 3,723,634 tested.

That's a one percent death rate (well, actually less than that) out of symptomatic patients who met the testing criteria of fever and difficulty breathing - a credible candidate for a test in the judgement of medical professionals who have gotten a lot of experience by now.

Look at it this way instead.

As of April 21, we have 42,458 dead and 788,920 confirmed cases. That includes all of the over 4 million tested as of today.

That is a death rate of 5.4% of those infected. So the main question is what percentage of the country is going to ultimately get it - i.e., become confirmed cases.

Let's assume only 20 percent of the country gets it and they have a death rate of 5.4%. That is about 1.1% of the total population of 330M. So, that is about 3.5M dead.

Scale it back to only 10 percent of the country getting it. That is still about 1.75M dead.

How about 5% of the people? Still have about 900,000 dead. And let's be real - this is a highly transmissible disease. A lot more than 5% are going to catch it.

I don't know where that death toll of 68,000 went. Maybe that was only for the first wave. But it is WAY low. We are already at 42K dead. Does anyone think we are only going to lose 26K more? NY will get that many by itself in the next couple of months.

And one other thing. NEVER COMPARE THE WUHAN FLU TO THE SEASONAL FLU EVER AGAIN. They are an order of magnitude apart.

We have 42K+ dead in seven weeks. The seasonal flu kills about 20K-30K in an entire YEAR.

[friendly fred]

Originally Posted by Revenant

Look at it this way instead.

As of April 21, we have 42,458 dead and 788,920 confirmed cases. That includes all of the over 4 million tested as of today.

That is a death rate of 5.4% of those infected. So the main question is what percentage of the country is going to ultimately get it - i.e., become confirmed cases.

Let's assume only 20 percent of the country gets it and they have a death rate of 5.4%. That is about 1.1% of the total population of 330M. So, that is about 3.5M dead.

Scale it back to only 10 percent of the country getting it. That is still about 1.75M dead.

How about 5% of the people? Still have about 900,000 dead. And let's be real - this is a highly transmissible disease. A lot more than 5% are going to catch it.

I don't know where that death toll of 68,000 went. Maybe that was only for the first wave. But it is WAY low. We are already at 42K dead. Does anyone think we are only going to lose 26K more? NY will get that many by itself in the next couple of months.

And one other thing. NEVER COMPARE THE WUHAN FLU TO THE SEASONAL FLU EVER AGAIN. They are an order of magnitude apart.

We have 42K+ dead in seven weeks. The seasonal flu kills about 20K-30K in an entire YEAR.

Let's assume only 20 percent of the country gets it and they have a death rate of 5.4%. That is about 1.1% of the total population of 330M. So, that is about 3.5M dead.

Scale it back to only 10 percent of the country getting it. That is still about 1.75M dead.

Let's assume at 20% getting COVID-19 they missed by 55 times as referenced in the LA County report: 3.5 million/55 = about 64,000 dead

And one other thing. NEVER COMPARE THE WUHAN FLU TO THE SEASONAL FLU EVER AGAIN. They are an order of magnitude apart.

We have 42K+ dead in seven weeks. The seasonal flu kills about 20K-30K in an entire YEAR.

For it to be an order of magnitude different, it would need to be 200,000 to 300,000 dead from the seasonal flu.


"An order of magnitude is an approximation of the logarithm of a value relative to some contextually understood reference value, usually ten, interpreted as the base of the logarithm and the representative of values of magnitude one." -Wikipedia

However, we don't have enough right now to say for sure, but these two California reports are two good data points for those of us hoping the lethality is significantly less than currently and officially assumed.

[friendly fred]

Originally Posted by 1blackman1

They’ve only recently started trying to add non-hospital deaths.

Unfortunately that number could be easily fudged.

[HedonistForever]

There are a variety of tests out there. Are they all 100% accurate? I doubt it. As a matter of fact, reports say that some test's from China didn't work at all and had to be discarded. So how can we say with any degree of certainty, how many false positive or false negative tests are being used in what ever data is posited?


Until we know for sure and I doubt we ever will, the total number of people who actually had the virus and how many of those died, we will never have an accurate percentage of death rate to the number who had the virus. What I think we can safely say is that more people, far more people will survive this than will succumb to this but when there are people who insist the "one life is to many" and we have to keep 22 million people from working until there is no more death in the country because of COVID-19, we will never adequately recover from this from an economic standpoint.


https://www.health.com/condition/col...flu-every-year


This year's flu season is shaping up to be possibly less severe than the 2017-2018 season, when 61,000 deaths were linked to the virus. However, it could equal or surpass the 2018-2019 season's 34,200 flu-related deaths.


Overall, the CDC estimates that 12,000 and 61,000 deaths annually since 2010 can be blamed on the flu. Globally, the World Health Organization (WHO) estimates that the flu kills 290,000 to 650,000 people per year.


And yet we didn't feel the need to put 22 million people out of work to combat that number. What will we say if the COVID-19 kills 60,000 or slightly more than seasonal flu?

[oeb11]

R - at least you publish your basic assumptions.
Still - no basis in fact.

A lot we do not know - including the asymptomatic + minimally symptomatic never reported or tested.

this expands (likely) the denominator.

reference - read John Ioaniddis article from march - he is a Stanford epidemiologist - article was posted in this Forum.

he has a very good discussion about the errors of assumptions for the pandemic.

[oeb11]

IMHO - this Wuhan virus is behaving more and more like a case fatality rate similar to Influenza A.

Yes, I ignored your precious prohibition on that behavior.

Too BAD!!!
Unless you have credential as an epidemiologist - which i doubt from reading your post - you are welcome to your opinion - but not to try to control other people's freedom of opinion.

[Tiny]

Originally Posted by Revenant

Look at it this way instead.

As of April 21, we have 42,458 dead and 788,920 confirmed cases. That includes all of the over 4 million tested as of today.

That is a death rate of 5.4% of those infected. So the main question is what percentage of the country is going to ultimately get it - i.e., become confirmed cases.

Let's assume only 20 percent of the country gets it and they have a death rate of 5.4%. That is about 1.1% of the total population of 330M. So, that is about 3.5M dead.

Scale it back to only 10 percent of the country getting it. That is still about 1.75M dead.

How about 5% of the people? Still have about 900,000 dead. And let's be real - this is a highly transmissible disease. A lot more than 5% are going to catch it.

I don't know where that death toll of 68,000 went. Maybe that was only for the first wave. But it is WAY low. We are already at 42K dead. Does anyone think we are only going to lose 26K more? NY will get that many by itself in the next couple of months.

And one other thing. NEVER COMPARE THE WUHAN FLU TO THE SEASONAL FLU EVER AGAIN. They are an order of magnitude apart.

We have 42K+ dead in seven weeks. The seasonal flu kills about 20K-30K in an entire YEAR.

I can get to your results but question your assumptions. If you calculate the fatality rate as a % of the people actually infected, including those who are never tested, you have knowledgeable, smart epidemiologists predicting everything from 0.15% to 1.2%. Based on the papers that FF and Oeb linked too, you'd suspect the real number will be closer to the 0.15% than the 1.2%. But on the other hand based on what we're seeing in NYC you can say with some confidence that 0.15% is too low.

Assume 60% of the population is infected before you get herd immunity. At the lower end, 0.15%, you have 294,000 deaths. And at the upper end, 1.2%, you get 2.4 million deaths.

HOWEVER, will 60% of the population be infected before we have a vaccine? I don't think that will happen. Even without any involvement by governments, many people are going to social distance and wear masks on their own. Add in reasonable regulation by government, testing and tracing, and requirements for masks in environments like public transport, and it should be a lot less. This is why a middle way is a good idea, IMHO.

[oeb11]

Tiny - i think the overall population based fatality rate will vary with area of the country and lifestyle.

NY- in the city people live on top of each other with public transport and sidewalks all crowded.



Texas - we live in Homes with yards and use cars. Much less risk of virus transmission than the crowded warrens of the NE coast cities.



Plus the total incompetence of the DPST leadership encouraging people to be out and about infecting each other in March, 2020. something they will deny doing now.

The usual DPST conduct.



Your numbers make more sense , at this time, as data is coming in the true picture of the virus is evolving.

[Tiny]

We've got another data point. 14% of 3000 New Yorkers tested throughout the state had Covid 19 antibodies:

https://www.cnbc.com/2020/04/23/new-...uomo-says.html

There have been 15,302 deaths in New York state. The population of New York is 19.5 million.

If you believe the 14%, the number of people infected would be 2.7 million. 15,302 deaths to date implies at least a 0.57% fatality rate. Assume 60% infection rate to get to herd immunity and this would imply at least 1.1 million deaths in the USA with no testing, tracing, masks, vaccine, new treatment or social distancing.

[Tiny]

Originally Posted by oeb11

Tiny - i think the overall population based fatality rate will vary with area of the country and lifestyle.

NY- in the city people live on top of each other with public transport and sidewalks all crowded.



Texas - we live in Homes with yards and use cars. Much less risk of virus transmission than the crowded warrens of the NE coast cities.

Oeb, I'm not sure the fatality rate as % of the number of people infected will vary hugely. Yes, it should be higher if the area has more older people or more people with underlying conditions. Or if the health care system is poor or overloaded.

Yes, the infection rate, the percentage of people infected, should vary depending on population density and lifestyle.

[oeb11]

Tiny - I think the case mortality rate can vary - as populations differ - in NE slums like Baltimore - a high incidence of drug abuse, smoking, obesity, and poor diet - result in conditions presdisposing to a poor outcome with Wuhan virus infection.

Very different from Texas middle class suburbs with much less of those pre-disposing conditions.

No - it is not a "racist" allegation - it may be that certain racial groups may be more affected by these conditions in slum areas.

I do not believe anything inherent about the Wuhan virus affects individuals of any race differently - it is the underlying conditions that predispose.

Assuming the access to and quality of medical care is equal - which may not be a reasonable assumption in some areas. Agreed
!
IMHO!