Covid-19

  • You are viewing Orangepower as a Guest. To start new threads, reply to posts, or participate in polls or contests - you must register. Registration is free and easy. Click Here to register.
May 4, 2011
1,764
932
743
Charleston, SC
Do you mean 95% CI? Yes, that is pretty much standard. The interesting part is that the lower limit falls either above or just inside what we thought the R0 was. I may have to rethink the post where I said it couldn't have been engineered.
I think he means that the 95% CI is so wide. And I'm assuming you know and others may not, CI width depends in a host of factors (sample size, variance, etc.). We're still dealing with some small samples and large variance across observations.
 

Binman4OSU

Legendary Cowboy
Aug 31, 2007
32,876
16,370
1,743
Stupid about AGW!!
Do you mean 95% CI? Yes, that is pretty much standard. The interesting part is that the lower limit falls either above or just inside what we thought the R0 was. I may have to rethink the post where I said it couldn't have been engineered.
What else has an R0 like this that is naturally occurring ?

This would be even higher than Small Pox R0 distribution and challenging Rubella, Mumps and Pertussis

Still no where close to Chickenpox or Measles
 

Boomer.....

Territorial Marshal
Feb 15, 2007
6,792
6,026
1,743
OKC
Dr. Fauci...agrees with your first paragraph. Until we test everyone we can not know the mortality rate. At this point it is unlikely we could even if we had a reliable test test everyone. By the time an accurate test is widelyavailable the epidemic will be over. The public will lose interest in testing. Only retrospective antibody testing after the epidemic is over will we be able to access the number who were infected and lived.
I was thinking the antibody tests are or could be available soon. There is a company in Norman (IMMY) producing the tests now. With the tests, hopefully you could determine if you had already been infected and have immunities. Not sure about the possibility of contracting it a second time as has been mentioned previously.
 

Binman4OSU

Legendary Cowboy
Aug 31, 2007
32,876
16,370
1,743
Stupid about AGW!!
AHAHAHHAHAHA
Back to back tweets
So the WH task force (which includes) Trump and Pence, hospitals and the medical industry is using this to get more funding, gain more authority and make themselves famous

While the Left is using it to undermine and steal elections.

The radical Left and Right need to just go away

1586360055475.png
 

SLVRBK

Johnny 8ball's PR Manager
Staff
A/V Subscriber
Oct 16, 2003
14,892
5,491
1,743
Katy, TX
Looking at the numbers for the 5 largest European countries/economies published today (Germany, UK. France, Italy and Spain). Looking at this simply because the population and development of those 5 countries is equivalent to US. Early on Germany wasn't reporting all deaths as related to COVID-19, if underlying condition was present their laws required that as the reported cause of death. This has since been altered but I'm not sure if the numbers were adjusted.


Total tests: 2,520,123
Tests per million: 7,683
E5 reported Infections: 555,916
per million: 1,721
Deaths: 50,259
per million: 174


Total tests: 2,097,209
Tests per million: 6,336
US reported Infection: 402,471
per million: 1,216
Deaths: 12,914
per million: 39


Used info at Worldometers to compile: https://www.worldometers.info/coronavirus/#countries
 
Sep 18, 2006
1,136
1,489
1,743
SEC country :(
Do you mean 95% CI? Yes, that is pretty much standard. The interesting part is that the lower limit falls either above or just inside what we thought the R0 was. I may have to rethink the post where I said it couldn't have been engineered.
I think he means that the 95% CI is so wide. And I'm assuming you know and others may not, CI width depends in a host of factors (sample size, variance, etc.). We're still dealing with some small samples and large variance across observations.
Yes, sorry, I meant the actual interval, 3.8-8.9, not the 95%. There's a lot of fudge factor for the relative size of the numbers, so I didn't know if that was common in epidemiology.
 

State

Russian Bot
Mar 15, 2007
13,959
8,168
1,743
TX
HOLY SHIT.. 5.7!!!!

This would imply that with 400K known cases there are ~2 million unknown cases out there

That's nuts.

Per this model, with that R0 we're already well past the peak. If the CFR is 1% we can expect 2 million to die regardless of anything we do. Per this model, the CFR would have to be .1% or lower to match the actual numbers we're seeing.

If that R0 is accurate, in effect the worst is already over and we can go back to normal.

https://www.nytimes.com/interactive...wDztAqZ8I4WGIZBuU9ZijThMynA2oOgfgyATlvQN2hXbQ
 

RxCowboy

Has no Rx for his orange obsession.
A/V Subscriber
Nov 8, 2004
75,301
52,476
1,743
Wishing I was in Stillwater
Yes, sorry, I meant the actual interval, 3.8-8.9, not the 95%. There's a lot of fudge factor for the relative size of the numbers, so I didn't know if that was common in epidemiology.
It's impossible to know if that's "wide" or not. Thinking of it in those terms really doesn't make any sense, because 95%CI is 95%CI.
 
May 4, 2011
1,764
932
743
Charleston, SC
Yes, sorry, I meant the actual interval, 3.8-8.9, not the 95%. There's a lot of fudge factor for the relative size of the numbers, so I didn't know if that was common in epidemiology.
Depends on the area and disease. I primarily work with mental health and substance use epidemiology with some background in HIV. Some disorders have wide CIs because you have small numbers and inconsistent data, which is what we have here.
 

RxCowboy

Has no Rx for his orange obsession.
A/V Subscriber
Nov 8, 2004
75,301
52,476
1,743
Wishing I was in Stillwater
Depends on the area and disease. I primarily work with mental health and substance use epidemiology with some background in HIV. Some disorders have wide CIs because you have small numbers and inconsistent data, which is what we have here.
But 95%CI is 95%CI. It doesn't make sense to me to think of it as "wide" or "narrow". It is what it is.
 
May 4, 2011
1,764
932
743
Charleston, SC
It's impossible to know if that's "wide" or not. Thinking of it in those terms really doesn't make any sense, because 95%CI is 95%CI.
Not to derail the thread, but we do make interpretations based on how wide the CI is since it's an indicator of how certain we are about a particular parameter estimate. In this case though, it is what it is and both are informative. As I took it, the CI indicates the r0 very well could still be down around 4, but is almost certainly higher than original estimates from China (although I'd note, these data are also based on Wuhan and I'm not super confident in their reporting). The lower bound is close to prior estimates, but the upper bound is astronomically higher.
 

Binman4OSU

Legendary Cowboy
Aug 31, 2007
32,876
16,370
1,743
Stupid about AGW!!
Not to derail the thread, but we do make interpretations based on how wide the CI is since it's an indicator of how certain we are about a particular parameter estimate. In this case though, it is what it is and both are informative. As I took it, the CI indicates the r0 very well could still be down around 4, but is almost certainly higher than original estimates from China (although I'd note, these data are also based on Wuhan and I'm not super confident in their reporting). The lower bound is close to prior estimates, but the upper bound is astronomically higher.
Here is how the CDC looks at the CI..decision making on the CI is not done at the epidemiologists level
As a result, most epidemiologists take a common-sense approach rather than a strict statistical approach to the interpretation of a confidence interval, i.e., the confidence interval represents the range of values consistent with the data from a study, and is simply a guide to the variability in a study

https://www.cdc.gov/csels/dsepd/ss1978/lesson2/section7.html
 
May 4, 2011
1,764
932
743
Charleston, SC
Here is how the CDC looks at the CI..decision making on the CI is not done at the epidemiologists level
As a result, most epidemiologists take a common-sense approach rather than a strict statistical approach to the interpretation of a confidence interval, i.e., the confidence interval represents the range of values consistent with the data from a study, and is simply a guide to the variability in a study

https://www.cdc.gov/csels/dsepd/ss1978/lesson2/section7.html
Right, the debate isn't what a CI is, but whether we'd interpret it as "wide" in this case or at all. My reply is more meaning that as researchers, we do make interpretations about how wide a CI is for the very reasons that it's an indicator of the range of values that are likely. In this case, an r0 that may range from 4 to 8 or 9 is a big difference and importantly indicates that even in that broad range, data suggest that original r0 estimates were off.