Coronavirus

[clutch]

Also, from the Imperial College paper -- this is the prediction model in the full picture:

 

Once you lift the intervention in September (as in the picture), you get the same spike in the fall.

 

Hence, the paper recommends putting in the intervention in place for 18 months. Are we all ready for that?

 

I actually applaud them for this recommendation -- this means they are really thinking like health professionals and not as a blend of health pros + politicians/economists.

 

Their very last paragraph in the paper can be a bit chilling:

"... However, we emphasise that is not at all certain that suppression will succeed long term; no public health intervention with such disruptive effects on society has been previously attempted for such a long duration of time. How populations and societies will respond remains unclear."

 

 

[alpha]

 

They try to get herd immunity at an acceptable cost (in terms of lives). So far, by their own judgement, that is still the case and who are we to judge otherwise?

 

 

According to the latest WHO information there is no evidence getting infected with Covid will provide immunity to future infections. That's why they warned against "immunity passports".

[Spekulatius]

 

They try to get herd immunity at an acceptable cost (in terms of lives). So far, by their own judgement, that is still the case and who are we to judge otherwise?

 

 

According to the latest WHO information there is no evidence getting infected with Covid will provide immunity to future infections. That's why they warned against "immunity passports".

 

I agree, but they are gambling on this. If we can’t get immunity an vaccine very likely won’t work reliably either and we are in deep trouble with 2-3 million excess dead in the US alone after this has run its course. That is unless we found ways to reduce the IFR rate, which I think is somewhat likely, but by how much?

 

I also think we are gambling on this without explicitly stating so. At least in parts of the US like NYC we probably reach herd immunity before any vaccine is coming to the market one way or another.

[clutch]

 

...

No one is arguing the area of the curve is different. The area under the curve will be approximately the same, but flattening and delaying peak prevents healthcare overload (which would lead to more deaths and more area under the curve), allows to buy time for positive black swan (treatment, vaccine, warm/humid weather) to emerge.

Really?

 

Flattening the curve strategies do not incorporate in their projections a positive black swan emerging: a vaccine, effective treatment, or heat/humidity reducing spread/severity/mortality (positive black swans) because this is impossible to predict. They also do not necessarily incorporate extra deaths from healthcare overload (which would be very hard to estimate).

 

Hence our nonsense peddlers seize on this as "proof" that flattening has no real benefit. They are left out of models because it is impossible to model black swans (even positive ones) and extra deaths that result from overload. Just because they are left out does not mean these are not very real benefits of flattening (they are).

 

As the bank example I used--if everyone goes to the bank to ask for their deposits all at once vs over time, you will see a real, nonlinear difference as the bank will fail in the fmr scenario.

 

Again, I am not going back to square one (arguing about benefits of curve flattening which should have been settled 6 weeks ago) and coming down to the level of our great nonsense dwellers here...

 

So you are now arguing that we should make decisions based on the potential events that are left out of the model. Why make/use any models then? (sarcasm)

[Guest cherzeca]

the great Andrew McCarthy on the assault upon liberty by treating " science expert" advice as authority:  https://www.nationalreview.com/2020/04/coronavirus-restrictions-government-bears-burden-of-proof-before-denying-freedoms/ (I dont believe it is paywalled)

[Gregmal]

For the “models were wrong” crowd (i.e. perpetual nonsense peddlers):

 

 

A lot of limitations w the IHME model (for example would miss any second waves, dumb decisions by policy makers to reopen too soon), but it’s done a heckuva job so far.

 

Source:

 

A bit of cherry-picking.

 

In the same thread:

Nostromo26

"Why did you stop plotting deaths on the 20th? Daily deaths haven't gone down since then. Here's a chart updated through 4/25."

 

But regardless, this model predicts that the number of deaths would be close to 0 by July... I guess we will start counting deaths like China at that point.

 

This is whats known as straw grabbing from Dalal "I commit to nothing" Trump. Notice his primary information source has now shifted down "the curve", going from the almighty and highly reliable Twitter to Reddit!

 

Another interesting data point

 

Italy- population 60M, cases 195k, deaths 26k

NYC- population 8M, cases 155k, deaths 12k

Rest of US- population 320M, cases 800k, deaths 42k

 

So maybe the fear peddling Dalal at the least needs to revise his model in order to hope next time to "be like Italy" rather than the negative implications continuously drawn about "being the next Italy". As the facts show, NY is really the only massive, massive, negative outlier. Perhaps Dr. Dalal should spend more time doing this job and less time worrying about playing politics on message boards...

[Dalal.Holdings]

This is whats known as straw grabbing from Dalal "I commit to nothing" Trump. Notice his primary information source has now shifted down "the curve", going from the almighty and highly reliable Twitter to Reddit!

 

Another interesting data point

 

Italy- population 60M, cases 195k, deaths 26k

NYC- population 8M, cases 155k, deaths 12k

Rest of US- population 320M, cases 800k, deaths 42k

 

So maybe the fear peddling Dalal at the least needs to revise his model in order to hope next time to "be like Italy" rather than the negative implications continuously drawn about "being the next Italy". As the facts show, NY is really the only massive, massive, negative outlier. Perhaps Dr. Dalal should spend more time doing this job and less time worrying about playing politics on message boards...

 

Gregmal: Exhibit A in "Often Wrong, Never in Doubt"--fools of randomness. No wonder these guys love Trump--he resembles them to an uncanny degree! Loves attacking those who called it right and his memory problems border on critical dementia (helps himself feel better about being often wrong though).

 

On top of that, I love hearing his views on women, minorities, dying New Yorkers, etc. A real class act!

 

His posts continue to detract value from most of this forum--the "best of the best" of CoBF! Can't shake the sunk cost fallacy of his ardent support of Trump over the years.

 

I hope you & your kind are active participants in the market for sure.

[Cigarbutt]

 

...

No one is arguing the area of the curve is different. The area under the curve will be approximately the same, but flattening and delaying peak prevents healthcare overload (which would lead to more deaths and more area under the curve), allows to buy time for positive black swan (treatment, vaccine, warm/humid weather) to emerge.

Really?

Flattening the curve strategies do not incorporate in their projections a positive black swan emerging: a vaccine, effective treatment, or heat/humidity reducing spread/severity/mortality (positive black swans) because this is impossible to predict. They also do not necessarily incorporate extra deaths from healthcare overload (which would be very hard to estimate).

Hence our nonsense peddlers seize on this as "proof" that flattening has no real benefit. They are left out of models because it is impossible to model black swans (even positive ones) and extra deaths that result from overload. Just because they are left out does not mean these are not very real benefits of flattening (they are).

As the bank example I used--if everyone goes to the bank to ask for their deposits all at once vs over time, you will see a real, nonlinear difference as the bank will fail in the fmr scenario.

Again, I am not going back to square one (arguing about benefits of curve flattening which should have been settled 6 weeks ago) and coming down to the level of our great nonsense dwellers here...

OK.

Another question (or two) to help me understand then:

Have South Korea, China (+ or - adjusted numbers) and New Zealand etc succeeded (past, present and future) or not in reducing the area under the curve?

Staying away from sunk costs and how 'others' have been wrong before, then what's the implication for the US, in terms of go-forward policy?

[patience_and_focus]

 

They try to get herd immunity at an acceptable cost (in terms of lives). So far, by their own judgement, that is still the case and who are we to judge otherwise?

 

 

According to the latest WHO information there is no evidence getting infected with Covid will provide immunity to future infections. That's why they warned against "immunity passports".

 

I agree, but they are gambling on this. If we can’t get immunity an vaccine very likely won’t work reliably either and we are in deep trouble with 2-3 million excess dead in the US alone after this has run its course. That is unless we found ways to reduce the IFR rate, which I think is somewhat likely, but by how much?

 

I also think we are gambling on this without explicitly stating so. At least in parts of the US like NYC we probably reach herd immunity before any vaccine is coming to the market one way or another.

 

I think people are getting confused again. There are multiple related but distinct things going on here.

 

(a) Immunity is not strictly binary - there are grades of it for each person (strength of immune response - a quantitative measure). Some people elicit strong immune response, other have weak response resulting in poor immunity.

 

(b) There is also a time dependency depending on the virus, for example flu needs new immunity every year, whereas measles immunity lasts lifetime. We do not know yet where SARS-CoV2 falls in terms strength or time component.

 

© Having a positive antibody test without symptoms is even more complicated. If we assume the test is not a false positive (a lot of antibody tests have been shown to totally useless in terms of accuracy), then many of these tests give a binary answer. But from (a) we know that strength (quantitative measurement) is important.

 

(d) Herd immunity is highly dependent on (a) being high in strength in general population and (b) immunity lasting a long time.

 

We really do not know how this will play out in terms of biology. I think the economic opening should be planned accordingly with this uncertainty taken into account. There are multiple reasonable plans to do a roll out (for example - https://www.gatesnotes.com/media/assets/media/files/Pandemic-I-The-First-Modern-Pandemic.pdf), but it need co-ordination at the federal level with states following general strategy with some flexibility. States can have some leeway for sure, but it can't be the wild wild west it is right now.

[Spekulatius]

 

Staying away from sunk costs and how 'others' have been wrong before, then what's the implication for the US, in terms of go-forward policy?

 

This is one I am unsure about too. Th NY antibody study with a 21% positive rate in NYC let’s me believe, that there will be too many active cases to prevent further spreading regardless of what we do with test and trace. The testing is probably capacity is probably 2 order of magnitude too lower to test most people if we open the economy, which we have to do no matter what, before the vaccine is a factor in 18 month (best case).

 

So in opinion this means that we go down the path of heard immunity, at least in bigger cities, but most likely everywhere unless we constrain movement between states or even cities for 18 month.

 

Now heard immunity or vaccines may or may not even exist or be feasible, but no matter, virtually everyone just isn’t get the virus in this case sooner or later.

 

I would like hear different viewpoints on how we still contain this using test and trace from out current starting point of test capacity and the likely opening of the economy in May or early June.

 

Also, I would like to hear if anyone thinks that schools can be closed for 18 month. Opening up a school (which can be staggered into kindergartens, elementary school etc) will simultaneously expose a large number of people to the virus and most likely create  a significant spike in cases, no matter how we do it. Can we keep them closed? Should we? I don’t think we can, but others may have a different viewpoint. 

[Gregmal]

 

Staying away from sunk costs and how 'others' have been wrong before, then what's the implication for the US, in terms of go-forward policy?

 

This is one I am unsure about too. Th NY antibody study with a 21% positive rate in NYC let’s me believe, that there will be too many active cases to prevent further spreading regardless of what we do with test and trace. The testing is probably capacity is probably 2 order of magnitude too lower to test most people if we open the economy, which we have to do no matter what, before the vaccine is a factor in 18 month (best case).

 

So in opinion this means that we go down the path of heard immunity, at least in bigger cities, but most likely everywhere unless we constrain movement between states or even cities for 18 month.

 

Now heard immunity or vaccines may or may not even exist or be feasible, but no matter, virtually everyone just isn’t get the virus in this case sooner or later.

 

I would like hear different viewpoints on how we still contain this using test and trace from out current starting point of test capacity and the likely opening of the economy in May or early June.

 

Also, I would like to hear if anyone thinks that schools can be closed for 18 month. Opening up a school (which can be staggered into kindergartens, elementary school etc) will simultaneously expose a large number of people to the virus and most likely create  a significant spike in cases, no matter how we do it. Can we keep them closed? Should we? I don’t think we can, but others may have a different viewpoint.

 

If schools opened, would you send your kid to school? This is an interesting cog IMO. Many people need schools to be open in order to go back to work. They certainly, after weeks/months of financial drain, dont want to be paying for daycare. How many of those people dont feel comfortable sending their kids back to school even if it is an option? The number is greater than zero and could contribute to the slow reopening. Do you set up a nurse with a thermometer next to the metal detectors as kids enter?

[Investor20]

https://www.washingtonexaminer.com/news/nearly-a-third-of-coronavirus-antibody-test-participants-show-exposure-in-massachusetts

 

Nearly a third of coronavirus antibody test participants show exposure in Massachusetts

Researchers collected drops of blood from residents in Chelsea, Massachusetts, and determined that 64 people tested positive out of 200 for antibodies linked to COVID-19.

 

 

 

Getting closer to herd immunity?

At 30%, we can stop worrying about false positives.

[Guest cherzeca]

"If schools opened, would you send your kid to school? This is an interesting cog IMO. Many people need schools to be open in order to go back to work. They certainly, after weeks/months of financial drain, dont want to be paying for daycare. How many of those people dont feel comfortable sending their kids back to school even if it is an option? The number is greater than zero and could contribute to the slow reopening. Do you set up a nurse with a thermometer next to the metal detectors as kids enter?"

 

the NYC teachers union wont stand for it.  wont happen. they would come after de Blasio with lynching ropes

[RichardGibbons]

again, one can be scientific and create models regarding transmission or use a little common sense and distinguish between those in the population that would be at-risk in the event of transmission, and focus mitigation efforts on them, and let transmission among the not-at-risk proceed (as we do without massive shutdowns every other flu season)

 

Yeah, this is still stupidity and not common sense, as you were told a few days ago, a few days before that, a few days before that.... It is not possible to inflect a large part of society while keeping the at-risk population safe.

[RichardGibbons]

The other interesting black swan thing that people aren't talking about is that the more people who get this, the higher the chance of the disease mutating in a noteworthy way. That could be good (mutating into something that is even more infectious, not deadly, but gives immunity for all variants) or bad (same as now, but more deadly).

[Cigarbutt]

https://www.washingtonexaminer.com/news/nearly-a-third-of-coronavirus-antibody-test-participants-show-exposure-in-massachusetts

Nearly a third of coronavirus antibody test participants show exposure in Massachusetts

Researchers collected drops of blood from residents in Chelsea, Massachusetts, and determined that 64 people tested positive out of 200 for antibodies linked to COVID-19.

Getting closer to herd immunity?

At 30%, we can stop worrying about false positives.

This is interesting.

The debate here seems to be based on the following perspective:

Is Chelsea ahead or behind the curve?

 

-It seems that it's ahead of the curve in terms of spread.

By simple extrapolation (rule of three basically without modification or modelization and using reported statistics).

In a typical year, Chelsea should 'see' about 8 deaths from the flu (using state and national stats).

This year, it looks like Chelsea has 'seen' so far about 40 Covid-19 deaths (not a direct number, obtained indirectly but method reasonable).

This means (using a similar basic methodology) that, for the US at large to reach a similar 30% immunity, expected COVID-19 deaths would reach about 150-165k.

 

-So, how do you want to go from here for the rest of your country?

 

[Spekulatius]

https://www.washingtonexaminer.com/news/nearly-a-third-of-coronavirus-antibody-test-participants-show-exposure-in-massachusetts

 

Nearly a third of coronavirus antibody test participants show exposure in Massachusetts

Researchers collected drops of blood from residents in Chelsea, Massachusetts, and determined that 64 people tested positive out of 200 for antibodies linked to COVID-19.

 

 

 

Getting closer to herd immunity?

At 30%, we can stop worrying about false positives.

 

For context Chelsea is the hardest hit community in MA. ~1500 confirmed cases are 3.8% of the population. Sot this means that undetected cases are 8x confirmed cases. It’s similar NY antibody study actually.

But this also means that the rest  of MA is still Far behind and the rest of the US Even more so.

 

But you are correct, Chelsea may be on the way to herd immunity, if immunity exists.

[treasurehunt]

Interesting article on excess deaths in March and April in several countries that were hit hard by COVID-19: https://www.ft.com/content/6bd88b7d-3386-4543-b2e9-0d5c6fac846c This article is not behind a paywall.

 

The conclusion is that total fatalities from COVID-19 are likely much higher than reported deaths, even in developed countries that have good reporting systems.

[Dalal.Holdings]

Interesting article on excess deaths in March and April in several countries that were hit hard by COVID-19: https://www.ft.com/content/6bd88b7d-3386-4543-b2e9-0d5c6fac846c This article is not behind a paywall.

 

The conclusion is that total fatalities from COVID-19 are likely much higher than reported deaths, even in developed countries that have good reporting systems.

 

Good find. Those graphs are telling. Definitely not "like the Flu". And the sudden spikes reveal that this virus was not "spreading for months" in these regions, but rather there was a sudden surge in cases.

 

Many assume the denominator of CFR is underestimated, but the numerator is also underestimated in many measuring attempts (a lot of amateurs don't account in the delay from onset to death for example). This is another indicator we may be missing certain deaths from covid (eg. deaths at home).

 

Hard to parse through these things, but excess mortality is definitely occurring for all causes. For one, when hospitals in hard hit areas are full of covid patients, other patients will experience worse healthcare...which can increase mortality in those groups.

 

A lot of noise and uncertainty with a pandemic that looks an order of magnitude deadlier than the Flu. In my book, that means precaution is warranted.

[Guest cherzeca]

" total fatalities from COVID-19" is an undefined term where so many fatalities have co-morbidities and are associated with elderly

[patience_and_focus]

Interesting article on excess deaths in March and April in several countries that were hit hard by COVID-19: https://www.ft.com/content/6bd88b7d-3386-4543-b2e9-0d5c6fac846c This article is not behind a paywall.

 

The conclusion is that total fatalities from COVID-19 are likely much higher than reported deaths, even in developed countries that have good reporting systems.

 

Good find. Those graphs are telling. Definitely not "like the Flu". And the sudden spikes reveal that this virus was not "spreading for months" in these regions, but rather there was a sudden surge in cases.

 

Many assume the denominator of CFR is underestimated, but the numerator is also underestimated in many measuring attempts (a lot of amateurs don't account in the delay from onset to death for example). This is another indicator we may be missing certain deaths from covid (eg. deaths at home).

 

Hard to parse through these things, but excess mortality is definitely occurring for all causes. For one, when hospitals in hard hit areas are full of covid patients, other patients will experience worse healthcare...which can increase mortality in those groups.

 

A lot of noise and uncertainty with a pandemic that looks an order of magnitude deadlier than the Flu. In my book, that means precaution is warranted.

 

+1

 

That is the one of the important points of flattening the curve as much as possible. Because when hospitals are hit hard, the deaths due to other causes spike simply because lack of resources / fear of going to hospital, etc, etc. One can argue that even though these subjects did not get Covid, they died as a secondary effect of exponential rise in Covid infections. Other reasons of-course is to be able to ramp up testing, expand capacity for covid related supplies, develop plan and co-ordination etc when we start opening up.

 

On another note, Denmark has decide to pay ~ 75% of every citizen's income upto a large sum until shutdown is slowly lifted.

 

https://www.theatlantic.com/ideas/archive/2020/03/denmark-freezing-its-economy-should-us/608533/

 

https://www.forbes.com/sites/mortenjensen/2020/03/31/how-denmark-is-navigating-through-the-coronavirus/#346d6019fc7e

 

[Guest cherzeca]

"they died as a secondary effect of exponential rise in Covid infections"

 

no, they died as a result of a bureaucrat's mistaken policy.  hospitals are underutilized.  dont believe me...go visit one.  there was a short period fo time when hospitals in "hot spots" like NYC had high demand of ICU beds...that has passed. 

[Dalal.Holdings]

Interesting article on excess deaths in March and April in several countries that were hit hard by COVID-19: https://www.ft.com/content/6bd88b7d-3386-4543-b2e9-0d5c6fac846c This article is not behind a paywall.

 

The conclusion is that total fatalities from COVID-19 are likely much higher than reported deaths, even in developed countries that have good reporting systems.

 

Good find. Those graphs are telling. Definitely not "like the Flu". And the sudden spikes reveal that this virus was not "spreading for months" in these regions, but rather there was a sudden surge in cases.

 

Many assume the denominator of CFR is underestimated, but the numerator is also underestimated in many measuring attempts (a lot of amateurs don't account in the delay from onset to death for example). This is another indicator we may be missing certain deaths from covid (eg. deaths at home).

 

Hard to parse through these things, but excess mortality is definitely occurring for all causes. For one, when hospitals in hard hit areas are full of covid patients, other patients will experience worse healthcare...which can increase mortality in those groups.

 

A lot of noise and uncertainty with a pandemic that looks an order of magnitude deadlier than the Flu. In my book, that means precaution is warranted.

 

+1

 

That is the one of the important points of flattening the curve as much as possible. Because when hospitals are hit hard, the deaths due to other causes spike simply because lack of resources / fear of going to hospital, etc, etc. One can argue that even though these subjects did not get Covid, they died as a secondary effect of exponential rise in Covid infections. Other reasons of-course is to be able to ramp up testing, expand capacity for covid related supplies, develop plan and co-ordination etc when we start opening up.

 

On another note, Denmark has decide to pay ~ 75% of every citizen's income upto a large sum until shutdown is slowly lifted.

 

https://www.theatlantic.com/ideas/archive/2020/03/denmark-freezing-its-economy-should-us/608533/

 

https://www.forbes.com/sites/mortenjensen/2020/03/31/how-denmark-is-navigating-through-the-coronavirus/#346d6019fc7e

 

Yep--and in the flattening curve models, these deaths are not accounted for because they are very very hard to predict/model. Hence the flattened curve scenario seems to have a similar AUC in the models when in reality it will likely be smaller than the area under the curve of the sharp peaking scenario.

 

Another thing that flattening allows--people to prepare (i.e. deploy a navy hospital ship to NYC, manufacture and stock up on PPE/vent, etc) and change their behaviors--social distancing, masks, etc--these things have taken some time for westerners in particular to adopt.

 

Buying some time is generally good with something like this--as the FT article shows, there was literally a Tsunami in deaths that without significant intervention would have overwhelmed systems in place.

[treasurehunt]

" total fatalities from COVID-19" is an undefined term where so many fatalities have co-morbidities and are associated with elderly

 

Sure, but part of the point in looking at all causes mortality is to sidestep such semantic questions. If you see a lot more people dying in March and April of this year compared to the same period in the last five years, you can reasonably attribute most of the excess deaths to COVID-19 (assuming you don't find some other global phenomenon that could also be a cause). I mean this in the very basic sense that these deaths would not have occurred if not for  COVID-19. A few of the excess deaths -- such as a  higher number of suicides? -- might be due to the reaction to COVID-19 rather than to the disease itself; and some deaths could be due to folks with other conditions receiving worse care, as Dalal pointed out. Still, I find the excess all causes mortality data to be a decent way of estimating the impact of COVID-19.

 

[Liberty]

" total fatalities from COVID-19" is an undefined term where so many fatalities have co-morbidities and are associated with elderly

 

Pretty sure those who died and would be alive otherwise feel much better because of this observation.