Help the Fight Against COVID-19

Status
Not open for further replies.
lyricbread said:
Just for comparison, below is a picture of Heinz Field, the home of the Pittsburgh Steelers. At full capacity, it holds 68,400 people.
View attachment 65329
According to the CDC, there are currently 33,494 cases of COVID-19 in the United States, which is less than half of the capacity of Heinz Field.

These maps with the scary red blobs are very misleading.
View attachment 65330
Click to expand...
They need to be scary. As I pointed out before the number of infections is around 100 times the reported cases. This is because of the data lag. It takes about two weeks from infection to the case being reported. The growth rate is about 10x per week, or 100x in two weeks.

Reinforcing that is the fact that your numbers are only one day old and there are now over 42,000 reported cases in the US which means that there are close to 1/2 million infections that will end up being reported. In a week there will be 5 million infected. Of those up to 10% will die.

The actual number of people infected is then 10 times greater than the reported infections because those who don't get very sick will never go to the doctor and get tested. By my estimations there are already 5 million infected in the US. In a week it will be 50 million. Of those up to 1% will die.
 
stink said:
Wow!!! Talk about conflating words...whatever bro..
Click to expand...
My point being, words are being written and said all the time that have impact on you whether you have an emotional reaction to them or not. Holding the creators of those words to evolving, higher standards is not unreasonable.
 
FractalAudio said:
As I pointed out before the number of infections is around 100 times the reported cases. This is because of the data lag. It takes about two weeks from infection to the case being reported. The growth rate is about 10x per week, or 100x in two weeks.

Reinforcing that is the fact there are now over 42,000 reported cases in the US which means that there are close to 1/2 million infections that will end up being reported. In a week there will be 5 million infected. Of those up to 10% will die.

The actual number of people infected is then 10 times greater than the reported infections because those who don't get very sick will never go to the doctor and get tested. By my estimations there are already 5 million infected in the US. In a week it will be 50 million. Of those up to 1% will die.
Click to expand...
+ 1000
I can't believe that some forum members aren't taking Cliff's math seriously.
If you own a Fractal Audio product, yet you don't take Cliff's math seriously, why do you own a Fractal product?
(gets off soapbox)
 
BaronVonGrim said:
When Chinese are stealing intellectual property involving biological weapons... they might end up being given something they dont understand.. something mislabeled... a gift.

They think they've hired a double agent from Harvard... but ends up being a tripple agent.

They would behave the same way... if was a Chinese made bio weapon or a US bio weapon gone rogue...smuggled in by an espionage ring.
Click to expand...
Amazing !! Are you the guy who convinced Bush about the so called mass destruction weapons of Saddam in Iraq, that have never been found.... ? In fact a quadruple secret agent reexported them to a South American nazi conspiracy team; with a link to Being called OS117 the french connection. It's an old story..I'm sure you can discover who they are
 
solo-act said:
+ 1000
I can't believe that some forum members aren't taking Cliff's math seriously.
If you own a Fractal Audio product, yet you don't take Cliff's math seriously, why do you own a Fractal product?
(gets off soapbox)
Click to expand...
Well if that number of infected is true we could seriously revise the casualties ratio to a lower value. I own a Fractal unit, not even the latest, and am perfectly convinced it is the best the world actually offers for guitar players. Still this does not make FAS a biologic expert company; or at least permit me to hope such disastrous predictions will turn out to be wrong.
 
Wuhan death rate among people who had symptoms has been revised downward: https://www.wsj.com/articles/corona...initial-estimates-new-study-finds-11584663474. And that's with shitty, third world medicine. Going with Cliff's "10x infections as confirmed infections" heuristic, this brings us to 0.14% fatality ratio among those infected. Given that we already have numbers for the total number of infections in a totally unconstrained 2009 H1N1 epidemic, 60M, that boils down to 86K dead. And that's if no effective therapeutic is found in the next few weeks.
 
iaresee said:
I'll give you a perfect example of how words impact you whether you react to them or not: https://en.wikipedia.org/wiki/Constitution_of_the_United_States -- those words dictate an immense amount about your life whether you chose to respond.whatever bro..
Click to expand...
iaresee said:
My point being, words are being written and said all the time that have impact on you whether you have an emotional reaction to them or not. Holding the creators of those words to evolving, higher standards is not unreasonable.
Click to expand...

Nope.. but one can ALWAYS choose which words matter... you can't argue that fact... the older I get, the less I F**king care about "words"
 
plexi59 said:
Wuhan death rate among people who had symptoms has been revised downward: https://www.wsj.com/articles/corona...initial-estimates-new-study-finds-11584663474. And that's with shitty, third world medicine. Going with Cliff's "10x infections as confirmed infections" heuristic, this brings us to 0.14% fatality ratio among those infected. Given that we already have numbers for the total number of infections in a totally unconstrained 2009 H1N1 epidemic, 60M, that boils down to 86K dead. And that's if no effective therapeutic is found in the next few weeks.
Click to expand...
Just for reflection : the 2003 heat wave in france caused approximately 20000 casualties. Up to day we have 860 with the worst to come in the next 2 weeks.
 
lauke-lux said:
Well if that number of infected is true we could seriously revise the casualties ratio to a lower value. I own a Fractal unit, not even the latest, and am perfectly convinced it is the best the world actually offers for guitar players. Still this does not make FAS a biologic expert company; or at least permit me to hope such disastrous predictions will turn out to be wrong.
Click to expand...
Yes, the mortality rate is lower than the case fatality rate. Right now the CFR is around 14%. But that is inflated because people who don't get sick enough to warrant a trip to the doctor (or ER) aren't "cases".

If we assume that only 10% get sick enough to warrant a trip to the doctor then the mortality rate is only 1.4%. Still pretty terrible though.

But it doesn't matter because it just means the actual number of infections is 10 times the reported number of cases.

Let's assume only 10% get sick enough to warrant testing. Then that means there are actually around 25 million people in the world infected right now. Where does the 25 million come from?:
1. It takes around 5 days on average before you start displaying symptoms.
2. It takes another 3-4 days before you are ill enough to go to the doctor or call an ambulance.
3. It takes 2-3 days for the test results to come back.
4. It takes another day for the results to be registered.
That's around 12 days. The growth rate is around 1.2. So in 12 days there are about 10 times as many cases. The data lags by 12 days, meaning the actual number of cases is 10 times higher than reported.

Right now there are around 250K "active" cases reported. Multiplying by the data lag then we get 2.5 million. Multiply by another 10 for the number of cases that go unreported and you get 25 million.

Now, it could be that only 1% get sick enough to warrant testing. Or it could be 50%. There's not enough data yet. If we look at S. Korea, who it seems tests everybody regardless of whether they are sick or not, the CFR is 3%. That's troubling. If we use that number then 20% of cases warrant testing and the number of actual cases is only 12.5 million. But the CFR is twice as high so the same number of people die. Or it could be 1%. But then that means the actual number of cases is 250 million. Same number of people will die though.

At current active cases we're looking at a minimum of 300K deaths if no new cases were ever reported. Given exponential spread is still ongoing that will likely increase to somewhere between 3 and 30 million. Only until every government shuts down everything simultaneously will the spread slow. The piecemeal approach that the various governments are using is insufficient. As long as there is human contact the virus will spread, and spread rapidly.

Everything about this virus indicates it's among the most contagious we've ever seen. It's not as deadly as SARS or MERS or Ebola or polio or smallpox. But it seems to be far more contagious.
 
@lyricbread:

lyricbread said:
According to the CDC, there are currently 33,494 cases of COVID-19 in the United States, which is less than half of the capacity of Heinz Field.

These maps with the scary red blobs are very misleading.

In addition, the CDC reports that there have been about 400 deaths in the United States from COVID-19 as of today.

In comparison, there have been over 23,000 deaths from the flu this season.
Click to expand...
That (33,494 cases, 400 deaths) wouldn't be scary at all, if COVID-19 numbers doubled as slowly as the flu does.

But in the U.S., the case numbers for COVID-19 double around every 3 days.

For the H1N1 flu it was never less than 17 days.

Think about what 3 days to double the case-number means, over the course of one month:
Mar 20 - 15,477 cases (200 dead)
Mar 23 - 33,494 cases (400 dead)
Mar 26 - 66,988 cases (800 dead)
Mar 29 - 133,976 cases (1,600 dead)
Apr 02 - 267,952 cases (3,200 dead)
Apr 05 - 535,904 cases (6,400 dead)
Apr 08 - 1,071,808 cases (12,800 dead)
Apr 11 - 2,143,616 cases (25,600 dead)
Apr 14 - 4,287,232 cases (51,200 dead)
Apr 17 - 8,574,464 cases (102,400 dead)
Apr 20 - 17,148,928 cases (204,800 dead)

Considering that the U.S. has under a million hospital beds, total,
and considering that the U.S. normally has 600,000 of those beds already-occupied in an average winter flu season,
and considering that that leaves, at most, under 400,000 beds before we're full-up and run out of capacity,
and considering that, worldwide, more than 10% of all COVID-19 cases require hospitalization,
and considering that 10% of the Apr 20 number, above, is 1.7 million, which is a lot more than 400,000...,

it's easy to see that COVID-19 is not really comparable with the flu.

If we can take action to cause the doubling-rate to be reduced, so that the case-number takes 17+ days to double, then it becomes comparable with the flu. But not until then.

Even then, data is missing: How many days is the typical (hospital-requiring) flu case hospitalized, compared with the typical (hospital-requiring) COVID-19 case? That makes a difference. If it's 10 days' hospital stay for COVID-19 (a number I think I saw somewhere, but can no longer find), and less for the flu (can't find that number at all) then this would be another way COVID-19 was worse than flu. But I don't have that info.

And of course exponential doublings don't go on forever. Even if we don't do any social-distancing, they would eventually the hit a ceiling, equal to the U.S. population (350 million) minus the population that's already had the virus and recovered from it (???? million).

Still, there isn't really any danger, even in a bad flu season, of having 1.7 million people requiring hospitalization and fighting over 400,000 beds.

With COVID-19, there is...unless we slow down the doubling.

I should also add...
...the above focuses on the doubling rate for cases. The doubling rate for deaths is similar, so I put those numbers in, above.

Yikes. How many deaths (in addition to the usual ones) can the mortuaries of America process per month, before you start having to dump your friends' and family's bodies into unmarked mass graves? Do you know the answer to that? I don't.

(I think there's reason for hope that the number of deaths per thousand cases is dropping, and will continue to do so with widespread use of chloroquine. But I could be wrong. We'll see.)

So, again: We gotta slow this thing down.

Get the days-to-double number high enough (which is to say: the rate-of-transmission low enough), and we can ride this thing out without too many scenes of abject horror.

If it stays at 3-ish? Yikes.
 
Last edited:
Dr. Dipwad said:
@lyricbread:


That (33,494 cases, 400 deaths) wouldn't be scary at all, if COVID-19 numbers doubled as slowly as the flu does.

But in the U.S., the case numbers for COVID-19 double around every 3 days.

For the H1N1 flu it was never less than 17 days.

Think about what 3 days to double the case-number means, over the course of one month:
Mar 20 - 15,477 cases
Mar 23 - 33,494 cases
Mar 26 - 66,988 cases
Mar 29 - 133,976 cases
Apr 02 - 267,952 cases
Apr 05 - 535,904 cases
Apr 08 - 1,071,808 cases
Apr 11 - 2,143,616 cases
Apr 14 - 4,287,232 cases
Apr 17 - 8,574,464 cases
Apr 20 - 17,148,928 cases

Considering that the U.S. has under a million hospital beds, total,
and considering that the U.S. normally has 600,000 of those beds already-occupied in an average winter flu season,
and considering that that leaves, at most, under 400,000 beds before we're full-up and run out of capacity,
and considering that, worldwide, more than 10% of all COVID-19 cases require hospitalization,
and considering that 10% of the Apr 20 number, above, is 1.7 million, which is a lot more than 400,000...,

it's easy to see that COVID-19 is not really comparable with the flu.

If we can take action to cause the doubling-rate to be reduced, so that the case-number takes 17+ days to double, then it becomes comparable with the flu. But not until then.

Even then, data is missing: How many days is the typical (hospital-requiring) flu case hospitalized, compared with the typical (hospital-requiring) COVID-19 case? That makes a difference.

And of course exponential doublings don't go on forever. Even if we don't do any social-distancing, they would eventually the hit a ceiling, equal to the U.S. population (350 million) minus the population that's already had the virus and recovered from it (???? million).

Still, there isn't really any danger, even in a bad flu season, of having 1.7 million people requiring hospitalization and fighting over 400,000 beds.

With COVID-19, there is...unless we slow down the doubling.
Click to expand...
It's even worse than that because Covid-19 patients often require ICU beds and ventilators. The number of those is far less than 400,000.
 
Population of South Korea is about 1/7th of the US (if you include the illegals). Their total number of deaths so far is 111. Their total number of cases is 9K. Even if all of the remaining 5700 patients die (very unlikely), that's 5800 dead per 50M, or 113 people per million. If we are in the same conservative ball park, that's 40K dead. I don't see "millions" here.
 
plexi59 said:
Population of South Korea is about 1/7th of the US (if you include the illegals). Their total number of deaths so far is 111. Their total number of cases is 9K. Even if all of the remaining 5700 patients die (very unlikely), that's 5800 dead per 50M, or 113 people per million. If we are in the same conservative ball park, that's 40K dead. I don't see "millions" here.
Click to expand...
I hope you're correct,
but what we're trying to do is make projections based on insufficient (sometimes misleading) data.

In that situation, we're doing better than wild-guessing. We're partly-informedly guessing. (But it's still a guess; I get that.)

Hopefully the scenario I spelled out, above, is wrong. Based on existing data, though, it seems like the best available guess.

If you think a number will turn out to be misleading, it's important to give reasons why, and then we can adjust the projection accordingly.

For example: Do you have reason to think the days-to-double number is actually much greater than 3? If so, let's re-run the numbers with your number: 5 days, or 10 days, or whatever you think it should be.

Or, do you think there's some reason that the exponential curve will start to hit a "ceiling" that's much lower than 100 million? All exponential growth curves in a population eventually hit a "ceiling." There's a turning point that happens as they're approaching the "ceiling," after which they start to resemble a logarithmic curve, with the ceiling as their "upper limit" that they never quite reach.

The "ceiling" for seasonal flu is much lower than the size of the general population, precisely because so many people get the seasonal flu vaccine. (And also because people who don't get the vaccine, and do get the flu, remain immune to that same flu for the rest of the season, and can't be re-infected.)

For COVID-19, there's no vaccine, but there are a lot of people who've already had it, who were never tested for it, and who already recovered with only mild symptoms. I say "a lot of people," ...but how many is "a lot?"

I don't think anyone believes 100 million people have already had it and recovered. But let's say that they had, just for the sake of argument.

In that case, with a population of 350 million, minus the 100 million who are already immune, you still have 250 million people to go. My April 20 number in my earlier post wasn't anywhere near approaching that. And it's only when you start approaching the "ceiling" that the curve starts bending downwards on its own (that is, without "help" from social distancing).

So, I don't know of any reason that the exponential curve won't still be in its "climbing" phase by April 20th.

Do you?

Happy to hear it, if you do.

My best bet is that the growth will slow, and the 3-days-to-double number will increase, precisely because we're social-distancing, and quarantining, and hosing ourselves down with Purell, and all that.

My hope is that it'll do this soon enough, and by enough, that we don't have a 5-day waiting-period for the E.R.
 
Last edited:
Status
Not open for further replies.
Back
Top Bottom