[evan_au]

In America 106 red cross workers from 9 different states.
https://nypost.com/2020/12/01/covid-19-was-likely-in-us-weeks-earlier-than-thought-study/

Thanks for the link. It looks like the original paper submission is here (free PDF, possibly prior to peer review): "Serologic testing of U.S. blood donations to identify SARS-CoV-2-reactive antibodies: December 2019-January 2020"
https://academic.oup.com/cid/advance-article/doi/10.1093/cid/ciaa1785/6012472

Some thoughts after reading it (as a non-expert):
- These are not Red Cross workers, they were donors to the Red Cross blood transfusion program. The researchers went back to archived samples to search for antibodies. The samples had been archived for future disease surveillance (which is what this was).
- They detected antibodies in 1.4% of the 7,389 samples tested
- These samples don't show that these people had SARS-COV2; it shows that they had antibodies to a coronavirus similar to SARS-COV2.  They note that some common coronaviruses trigger a reaction in COVID-19 antibody tests.

The blood samples are tested anonymously; it may be worthwhile "de-anonymizing" the data to conduct further research, as it may reveal more sources.
- CDC experts could interview these blood donors, to determine if they had ever suffered from SARS or MERS, or been in areas where it was common (you could expect that everyone has been exposed to the "common cold" coronaviruses).
- They should also check if these people had been in contact with bats

from 9 different states...
From Patient 0 to 106, takes a while.

These states are widely separated, so it is unlikely that we are talking about 106 patients who caught it from each other, and all just decided to go to the blood bank.

If 106/7,389 (1.4% infected) is a random sample of the population, then California, with a population of 40 million would have had around half a million cases of SARS-COV2 infection by January 2020. We know that the fatality rate was about 3% in the early days, so we are talking about 50,000 hospital admissions and 17,000 deaths from a mystery respiratory virus.
- That would not have been missed!
- In fact, the infection rate in the population would have been higher than 1.4%, as anyone with symptoms would have deferred their blood donation.

Reading the paper, they also tested samples going back as far as 10 years, as a control. 3 out of 519 also tested positive for antibodies reacting to SARS-COV2, or 0.6%.
- So clearly, there are a lot of people with coronavirus antibodies in their bloodstream, in the absence of a pandemic
- This represents almost half of the 1.4% found in December 19 and January 20.

none of the sera can be considered “true positives."

So they warn that a single snapshot in time cannot confirm an ongoing infection.
- But they do point to one donor in Connecticut on January 10, 2020 as very likely to have been exposed to SARS-COV2.

In reality, rather than 106 Red Cross workers with SARS-COV2 infection, we are talking about around 1 to 60 members of the general public who may have been exposed to SARS-COV2 (or some similar virus).

Past infections .  current infections

There is an additional line of evidence here, which is the number of variants seen in the early days of the pandemic.
- This genetic clock points to a "patient zero" perhaps around September-October.

[Bored chemist]

The early information from China suggested an R0 of 3. Actually 3.28

But that value of 3.28 is based on the whole of the outbreak, rather than just the early days which is (we all agree) the number that we should use.

So use the right number.
Or you will just make it increasingly clear that your intention here is to deceive.

[Bored chemist]

some of those athletes have attributed to COVID-19

They are athletes guessing at what caused their symptoms.
How did they rule out the idea that they just caught a cold?

[Petrochemicals]

At an R0 of 1.5 the doubling rate is 4 days. If the number of new cases in mid september were 1 per day then by early december there would be a million new cases per day. You need a different reason for the cases not to show. You also need to explain why Maryland wasn't an early center of disease.

The actual period for Corona virus transmission is 4 days from contact, with the standardised non controlled reproduction rate of 3 in wuhan in 2 months starting from one you would havenewly infected number of 4.75 million, and a total case count of around 9 million, another 4 days and that rises to around the 25 million mark.

Ok where is the data?

People can only transit once infected in the throat or mouth, someone with an infection on the eyes, will take time for the virus to present in the throat area. Same with someone who catches the virus by touching something and then rubbing their eyes.

"Currently, according to the Centers for Disease Control and Prevention (CDC)Trusted Source, the incubation period for the novel coronavirus is somewhere between 2 to 14 days after exposure"
https://www.healthline.com/health/coronavirus-incubation-period#incubation-period

Other studies have shown upto a month. So your claim of 4 days is misleading.

As.we see here
https://www.dailymail.co.uk/news/article-7986663/Incubation-period-new-coronavirus-long-24-DAYS-expert-claims.html

A 24 day incubation period.

Someone's changed their tune, damn let down by the daily fail.

Ok getting somewhere.


Weekly incubation period and an R.0 of 3.

Past infections .  current infections

Patient 0
Week 1.  1              ,   3
Week 2   3              ,   9
Week 3   9              ,  27
Week 4   27            ,  81
Week 5   81            ,  243
Week 6 243            ,  729
Week 7  729           , 2,187
Week 8 2,187         , 6,561
Week 9 6,561         , 19,683
Week 10 19,683     , 59,049
Week 11 59,049     , 177,147
Week 12 177,147   , 531,441
Week 13 531,441   , 1,594,323
Week 14 1,594,323 , 4,782,969

Ok so that's  over 4 million  infections in 3 and half months from patient 0.

Compared to over 4 million infections in 11 with the first data set.

And for the first 2 months basically  way under 2000 except for the last week

[Jolly2 (OP)]

In America 106 red cross workers from 9 different states.
https://nypost.com/2020/12/01/covid-19-was-likely-in-us-weeks-earlier-than-thought-study/

Thanks for the link. It looks like the original paper submission is here (free PDF, possibly prior to peer review): "Serologic testing of U.S. blood donations to identify SARS-CoV-2-reactive antibodies: December 2019-January 2020"
https://academic.oup.com/cid/advance-article/doi/10.1093/cid/ciaa1785/6012472

Some thoughts after reading it (as a non-expert):
- These are not Red Cross workers, they were donors to the Red Cross blood transfusion program. The researchers went back to archived samples to search for antibodies. The samples had been archived for future disease surveillance (which is what this was).
- They detected antibodies in 1.4% of the 7,389 samples tested
- These samples don't show that these people had SARS-COV2; it shows that they had antibodies to a coronavirus similar to SARS-COV2.  They note that some common coronaviruses trigger a reaction in COVID-19 antibody tests.

The blood samples are tested anonymously; it may be worthwhile "de-anonymizing" the data to conduct further research, as it may reveal more sources.
- CDC experts could interview these blood donors, to determine if they had ever suffered from SARS or MERS, or been in areas where it was common (you could expect that everyone has been exposed to the "common cold" coronaviruses).
- They should also check if these people had been in contact with bats

from 9 different states...
From Patient 0 to 106, takes a while.

These states are widely separated, so it is unlikely that we are talking about 106 patients who caught it from each other, and all just decided to go to the blood bank.

If 106/7,389 (1.4% infected) is a random sample of the population, then California, with a population of 40 million would have had around half a million cases of SARS-COV2 infection by January 2020. We know that the fatality rate was about 3% in the early days, so we are talking about 50,000 hospital admissions and 17,000 deaths from a mystery respiratory virus.
- That would not have been missed!
- In fact, the infection rate in the population would have been higher than 1.4%, as anyone with symptoms would have deferred their blood donation.

Reading the paper, they also tested samples going back as far as 10 years, as a control. 3 out of 519 also tested positive for antibodies reacting to SARS-COV2, or 0.6%.
- So clearly, there are a lot of people with coronavirus antibodies in their bloodstream, in the absence of a pandemic
- This represents almost half of the 1.4% found in December 19 and January 20.

none of the sera can be considered “true positives."

So they warn that a single snapshot in time cannot confirm an ongoing infection.
- But they do point to one donor in Connecticut on January 10, 2020 as very likely to have been exposed to SARS-COV2.

In reality, rather than 106 Red Cross workers with SARS-COV2 infection, we are talking about around 1 to 60 members of the general public who may have been exposed to SARS-COV2 (or some similar virus).

Past infections .  current infections

There is an additional line of evidence here, which is the number of variants seen in the early days of the pandemic.
- This genetic clock points to a "patient zero" perhaps around September-October.

Yeah my data suggests mid September.

Says to me we need a better antibody test. The report doesn't entirely rule of covid 19 for the others. But 1 infected person is  more certian instead of the 106.

Maybe an antibody test isn't the best marker. Are there any tests that could identify a past covid 19 infection better?

[Jolly2 (OP)]

At an R0 of 1.5 the doubling rate is 4 days. If the number of new cases in mid september were 1 per day then by early december there would be a million new cases per day. You need a different reason for the cases not to show. You also need to explain why Maryland wasn't an early center of disease.

The actual period for Corona virus transmission is 4 days from contact, with the standardised non controlled reproduction rate of 3 in wuhan in 2 months starting from one you would havenewly infected number of 4.75 million, and a total case count of around 9 million, another 4 days and that rises to around the 25 million mark.

Ok where is the data?

People can only transit once infected in the throat or mouth, someone with an infection on the eyes, will take time for the virus to present in the throat area. Same with someone who catches the virus by touching something and then rubbing their eyes.

"Currently, according to the Centers for Disease Control and Prevention (CDC)Trusted Source, the incubation period for the novel coronavirus is somewhere between 2 to 14 days after exposure"
https://www.healthline.com/health/coronavirus-incubation-period#incubation-period

Other studies have shown upto a month. So your claim of 4 days is misleading.

As.we see here
https://www.dailymail.co.uk/news/article-7986663/Incubation-period-new-coronavirus-long-24-DAYS-expert-claims.html

A 24 day incubation period.

Someone's changed their tune, damn let down by the daily fail.

Ok getting somewhere.


Weekly incubation period and an R.0 of 3.

Past infections .  current infections

Patient 0
Week 1.  1              ,   3
Week 2   3              ,   9
Week 3   9              ,  27
Week 4   27            ,  81
Week 5   81            ,  243
Week 6 243            ,  729
Week 7  729           , 2,187
Week 8 2,187         , 6,561
Week 9 6,561         , 19,683
Week 10 19,683     , 59,049
Week 11 59,049     , 177,147
Week 12 177,147   , 531,441
Week 13 531,441   , 1,594,323
Week 14 1,594,323 , 4,782,969

Ok so that's  over 4 million  infections in 3 and half months from patient 0.

Compared to over 4 million infections in 11 with the first data set.

And for the first 2 months basically  way under 2000 except for the last week

Not at all it's one data set with an R0 os 2 and a 2 week incubation,  compared to an R0 of 3 with the incubation of a week.

I.was actually thinking about combining them. As a means to compensate for long and shorter incubation periods, and the variance between 1.4 and 3 R0. Its not a perfect world. Models will never entirely predict the actual transmission that occurred. We can only use different techniques to build different models and look at the implications and match to data.

It's hard because we only have recorded cases, not all cases, we only have recorded deaths not all deaths.

[Bored chemist]

Not at all it's one data set with an R0 os 2 and a 2 week incubation,  compared to an R0 of 3 with the incubation of a week.

And what does it look like if you use the real values, rather than silly ones?
Or do you not want to do that because it makes it clear that your idea is absurd?

[Jolly2 (OP)]

Combine
Weekly incubation period and an R.0 of 3.
And
Fortnightly incubation with an R0 of 1.7, I'll round up.

Past infections .  current infections

Patient 0
Week 1.  1              ,   3
Week 3   3              ,   5
Week 4   5              ,  15
Week 6   15            ,  26
Week 7   26            ,  78
Week 9   78            ,  133
Week 10 133          ,  399
Week 12 399          ,  678
Week 13 678          ,  2,035
Week 15 2,035       ,  3,460
Week 16 3,460       ,  10,378
Week 18 10,378     ,   17,643
Week 19  17,643    ,  52,928
Week 21 52,928     ,  89,977
Week 22 89,977     , 269,932
Week 24 269,932   , 458,884
Week 25 458,884    , 1,376,651
Week 27 1,376,651 , 2,340,307
Week 28 2,340,307 , 7,020,919


Ok so that's  over 4 million  infections in 6 and half months from patient 0.

Compared to over 4 million infections in 11 months with the first data set.

And 4 million infections after 3 and half with the second.

[Jolly2 (OP)]

Not at all it's one data set with an R0 os 2 and a 2 week incubation,  compared to an R0 of 3 with the incubation of a week.

And what does it look like if you use the real values, rather than silly ones?
Or do you not want to do that because it makes it clear that your idea is absurd?

Yeah clearly the CDC and WHO all have silly numbers go take it up with them and stop trolling me. Better still do your own maths

[Petrochemicals]

Someone's changed their tune, damn let down by the daily fail.

Not at all it's one data set with an R0 os 2 and a 2 week incubation,  compared to an R0 of 3 with the incubation of a week.

That was what meant. We are not starting in September  anymore are we? If we did by your revised reckoning we would be finished by Christmas by around 0.5T infected

[Jolly2 (OP)]

Someone's changed their tune, damn let down by the daily fail.

Not at all it's one data set with an R0 os 2 and a 2 week incubation,  compared to an R0 of 3 with the incubation of a week.

That was what meant. We are not starting in September  anymore are we? If we did by your revised reckoning we would be finished by Christmas by around 0.5T infected

Based on 4 million infections.

Data set one puts patients 0 around March.

Based on the secound puts patient 0 in mid September.

Based on the combined data set.
Puts patient 0 mid July.

That's looking at the 20,000 deaths from covid we see around Feb-March 2020.

The combined data set matches with an out break from Fort Detrick which closed mid July.

The second matches with an outbreak at the wuhan games, and a possible zoological origin. In September.

The first matches with samples found in Spain.

Ofcourse different strains probably have slightly different R0 and incubation periods. There were at least 5 Strians of Covid at the begining of the pandemic recorded.

[Bored chemist]

Better still do your own maths

I did,

Kryptid the article suggests 10 day incubation not 5.

Actually:

On average, symptoms showed up in the newly infected person about 5 days after contact.

And it is known that they are infectious  a day or two before there are symptoms

Which mean that we are talking about R is about 7 and the time scale from one generation to the next is about 4 days.
That's 7.5 "generations" in a month (30 days/ 4 days ) and with R = 7 you get 7^7.5 cases in a month
2,178,889.9
And in Jolly's fairy tale world where nobody notices for 3 months you get (with an absurdly simple model)
10,344,414,165,316,372,973 cases.
which is massively  more than the population of the planet.
But somehow he thinks people might not notice.

Yeah clearly the CDC and WHO all have silly numbers go take it up with them

As far as I am aware, the CDC and WHO have not published the relevant data for R

If you believe that they have, please show me where it is.

However the Nature paper I quoted gives the value for R in the early stages of the Wuhan outbreak.
Since we all agree that's the number we need, why are you not using it?

[Jolly2 (OP)]

Better still do your own maths

I did,

Kryptid the article suggests 10 day incubation not 5.

Actually:

On average, symptoms showed up in the newly infected person about 5 days after contact.

And it is known that they are infectious  a day or two before there are symptoms

Which mean that we are talking about R is about 7 and the time scale from one generation to the next is about 4 days.
That's 7.5 "generations" in a month (30 days/ 4 days ) and with R = 7 you get 7^7.5 cases in a month
2,178,889.9
And in Jolly's fairy tale world where nobody notices for 3 months you get (with an absurdly simple model)
10,344,414,165,316,372,973 cases.
which is massively  more than the population of the planet.
But somehow he thinks people might not notice.

Yeah clearly the CDC and WHO all have silly numbers go take it up with them

As far as I am aware, the CDC and WHO have not published the relevant data for R

If you believe that they have, please show me where it is.

However the Nature paper I quoted gives the value for R in the early stages of the Wuhan outbreak.
Since we all agree that's the number we need, why are you not using it?

Your welcome to your dooms day scenario.  Both the WHO and CDC have taken official positions on the R0 and it isnt 7.

As already cited,  China, Germany and Sweden all conducted studies and concluded an R0 of 3.28 at the begining of the pandemic.
http://www.sci-news.com/medicine/covid-19-reproduction-number-08132.html


Besides with an R0 of 7 with a 5 day incubation,  you would go from  patient 0 to over 4 million infections in arround a month an a half.

As we dont see the 20,000 deaths worldwide from covid 19 until March 2019, patient 0 clearly wasn't in January 2019.

In no way do the numbers you suggest match reality.

[Bored chemist]

First question: do you think that R0 only has one value, or do you recognise that, for example in the UK it's near 1 (hopefully a little below) ?

Do you think that Germany and Sweden are part of Wuhan?
If not, why would you include them in an average of estimates of R0 when what you are trying to find is R0 for Wuhan and the surrounding area?
(Don't forget that the original outbreak is the ONLY place that tells you how the virus spreads in a population that were not expecting it)


The paper you cite refers to a bunch of studies.
Only one covers the Hubei province It gives R0 = 6.5.

And, it's silly to refer to it as a "Doomsday scenario" since it already happened.


So... once again.

As far as I am aware, the CDC and WHO have not published the relevant data for R

If you believe that they have, please show me where it is.

Please don't waste further time on studies conducted anywhere but Wuhan and the surrounding area- because they can not be relevant to the question of spread through a population which was unaware of the risk.

And that's the only value of R0 that matters if we are looking at how it would have spread if it had turned up "early" in Spain or wherever.

[Bored chemist]

In no way do the numbers you suggest match reality.

They are the numbers  calculated by observing reality; Of course they match it.

The problem with your analysis is that you refuse to consider that R0 changes because people change their behaviour.

[Petrochemicals]

From the bmj

Viral load kinetics could also explain some of the differences between SARS-CoV-2 and SARS-CoV-1. In the respiratory tract, peak SARS-CoV-2 load is observed at the time of symptom onset or in the first week of illness, with subsequent decline thereafter, which indicates the highest infectiousness potential just before or within the first five days of symptom onset (fig 2).7 In contrast, in SARS-CoV-1 the highest viral loads were detected in the upper respiratory tract in the second week of illness, which explains its minimal contagiousness in the first week after symptom onset, enabling early case detection in the community.

https://www.bmj.com/content/371/bmj.m3862

Symptoms emergence 4- 6 days after exposure

[Bored chemist]

From the bmj

Viral load kinetics could also explain some of the differences between SARS-CoV-2 and SARS-CoV-1. In the respiratory tract, peak SARS-CoV-2 load is observed at the time of symptom onset or in the first week of illness, with subsequent decline thereafter, which indicates the highest infectiousness potential just before or within the first five days of symptom onset (fig 2).7 In contrast, in SARS-CoV-1 the highest viral loads were detected in the upper respiratory tract in the second week of illness, which explains its minimal contagiousness in the first week after symptom onset, enabling early case detection in the community.

https://www.bmj.com/content/371/bmj.m3862

Symptoms emergence 4- 6 days after exposure

And with a couple of days of asymptomatic transmission, it means that people can be spreading it within 3 days.

[Jolly2 (OP)]

From the bmj

Viral load kinetics could also explain some of the differences between SARS-CoV-2 and SARS-CoV-1. In the respiratory tract, peak SARS-CoV-2 load is observed at the time of symptom onset or in the first week of illness, with subsequent decline thereafter, which indicates the highest infectiousness potential just before or within the first five days of symptom onset (fig 2).7 In contrast, in SARS-CoV-1 the highest viral loads were detected in the upper respiratory tract in the second week of illness, which explains its minimal contagiousness in the first week after symptom onset, enabling early case detection in the community.

https://www.bmj.com/content/371/bmj.m3862

Symptoms emergence 4- 6 days after exposure

And with a couple of days of asymptomatic transmission, it means that people can be spreading it within 3 days.

Ok make it even worse then. Under those number patient 0 would have to have been in February. 

We know the death rates.

To hit 20,000 at a minimum the numbers needed would put paintent 0  in February. So you are just proving the numbers you cite even more wrong.

[Bored chemist]

So you are just proving the numbers you cite even more wrong.

The problem with your analysis is that you refuse to consider that R0 changes because people change their behaviour.

[Jolly2 (OP)]

In no way do the numbers you suggest match reality.

They are the numbers  calculated by observing reality; Of course they match it.

The problem with your analysis is that you refuse to consider that R0 changes because people change their behaviour.

And you are failing to notice the lockdown measures in Wuhan came into effect way after the outbreak started.

December 31st wuhan 2019  quote "The Wuhan Municipal Health Commission released a briefing on its website about early signs of a pneumonia outbreak in the city."

On 2 January 2020 41 admitted hospital patients in Wuhan, China, were confirmed to have contracted (laboratory-confirmed) the 2019-nCoV (novel coronavirus);
 
January 7 2020 In a closed meeting of the Central Politburo of the Communist Party of China, Xi Jinping "made requests for the prevention and control work of the coronavirus outbreak" and issued instructions to similar ends. This meeting occurred 13 days before Xi's first public comments on the outbreak on 20 January

January 10 2020 The gene sequencing data of the isolated 2019-nCoV, a virus from the same family as the SARS coronavirus

A second death occurred in a 69-year-old man in China on 15 January

25th January 2020 "General Secretary of the Communist Party of China Xi Jinping called the "accelerating spread" of the coronavirus a "grave situation" in a Party Politburo meeting,[187] and that it was "mutating" as Beijing escalates measures to contain the illness"

They actually starts to escalate measures to control it. They start that's already atleast 2 months the virus is known to be free

Under the numbers you cite, you would have expected 20,000 deaths at the end of December. With patient 0 being at the end of November. But we know there were cases in October the Chinese government suppressed knowledge of.

So you are just proving the numbers you cite even more wrong.

The problem with your analysis is that you refuse to consider that R0 changes because people change their behaviour.

People dont notice until January a virus present from atleast October.  Again you are talking nonsense.

Go troll somewhere else and stop wasting everyone time.