COVID COVID-19 Coronavirus Containment Efforts

[atyy]

Re: cremation of cadaver of Covid victims:

It is understandable to prohibit conducting wake for dead loves ones due to Covid but i need to ask these questions---- can there be no effective safety protocol conceptualized to allow, at least, viewing and say a little prayer (depending of your belief), the process of cremation? When the Covid victim was brought to the hospital, presumably by relatives, there is no strict measures employed as stringent as when that Covid patient died..

Currently in Singapore during a partial lockdown (known here as a "circuit breaker"), for all funeral wakes and services (not only those who have died with COVID-19 infection), only 10 people or fewer are allowed in the room at anyone time. Earlier, this number was up to 250, provided safe distancing could be observed. If the deceased had COVID-19, then the body is doubly bagged and the coffin air-tight. Cremation is recommended, unless that conflicts with religious beliefs.

https://www.straitstimes.com/singap...e-followed-for-cremation-or-burial-of-victims

https://www.nea.gov.sg/our-services/public-cleanliness/environmental-cleaning-guidelines/circuit-breaker-measures/frequently-asked-questions
- Can funeral wakes be held for those who passed away from COVID-19 infection? How long can these wakes be held for?
- Yes. MOH has assessed that funeral wakes can be held for those who passed away from COVID-19 infection. Like all other funeral wakes held during the circuit breaker period, they should be held in accordance to the circuit breaker measures listed here.
Although measures have been put in place to ensure safe handling of the deceased infected with COVID-19 and that physical contact of their bodies is not permissible, funeral wakes should be kept within 3 days. This is to minimise any potential risk of transmission between visitors during the funeral wake.

 

[bhobba]

Interesting article on death rate when you have symptoms:
https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2020.00455

Of course the actual death rate depends on how many asymptomatics there are which many think is greater than the .16% of the flu. We need more antigen tests to know that one. Personally I think it's about .1-.2%

Thanks
Bill

 

[mfb]

Lombardy had 16,000 deaths in a population of 10 million - and that's only the recorded cases. To make that compatible with your estimate basically everyone in the whole region would have had to be infected.

The US lost 0.03% of its population so far (again only recorded cases). Did 20% get infected already?
100% in the state of New York (23,000 deaths in a population of 19 million)?
The Bronx recorded 3100 deaths in a population of 1.4 million (0.22% of the total population).

Antibody tests suggest a way lower fraction of people who got it, and correspondingly a much higher infection fatality risk.

 

[PeroK]

Interesting article on death rate when you have symptoms:
https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2020.00455

Of course the actual death rate depends on how many asymptomatics there are which many think is greater than the .16% of the flu. We need more antigen tests to know that one. Personally I think it's about .1-.2%

Thanks
Bill

Did you mean 1-2%? That's what I thought that study has concluded.

 

[bhobba]

Did you mean 1-2%? That's what I thought that study has concluded.

It didn't include asymptomatic or cases so mild you don't even know you are sick. Studies in places like California have shown there are a lot more of those than people previously thought:
https://www.medrxiv.org/content/10.1101/2020.04.14.20062463v2
'These prevalence point estimates imply that 54,000 (95CI 25,000 to 91,000 using weighted prevalence; 23,000 with 95CI 14,000-35,000 using unweighted prevalence) people were infected in Santa Clara County by early April, many more than the approximately 1,000 confirmed cases at the time of the survey.'

That has led a number of people to say they think it really is a lot less eg .1-.2%. But that's just a guess, we need to wait for the mass antibody tests to know. But as MFB points out that does not gel with other places like Lombardy.

Thanks
Bil

 

[PeroK]

It didn't include asymptomatic or cases so mild you don't even know you are sick. Studies in places like California have shown there are a lot more of those than people previously thought. That has led a number of people to say they think it really is a lot less ie about .1-.2%. But that's just a guess, we need to wait for the mass antibody tests to know. But as MFB points out that does not gel with other places like Lombardy.

Thanks
Bil

It doesn't gel with the UK either. We've had 36,000 deaths, which tends to be people who have died as a direct result of having the virus. Note that we have had about 55,000 excess deaths this year- i.e. more than expected. So, 36,000 is probably a lower limit for deaths directly from COVID-19.

I guess it's not impossible that 18-36 million people in the UK have had it. But, if it really spread that quickly, then lockdown has largely been a waste of time since half of us have had it anyway!

 

[bhobba]

I guess it's not impossible that 18-36 million people in the UK have had it. But, if it really spread that quickly, then lockdown has largely been a waste of time since half of us have had it anyway!

Yes looking at those examples does make it look like the California results are some kind of outlier. Thanks guys for giving it a reality check.

Thanks
Bill

 

[Jarvis323]

Yes looking at those examples does make it look like the California results are some kind of outlier. Thanks guys for giving it a reality check.

Thanks
Bill

The the Santa Clara antibody study has been pretty thoroughly rejected by the scientific community if I remember correctly.

 

[bob012345]

The the Santa Clara antibody study has been pretty thoroughly rejected by the scientific community if I remember correctly.

Part of the reason for rejection was some people just didn't like the views of someone who funded it claiming a potential conflict of interest. I didn't know science was only legitimate if its funded by the right people. I didn't know you can so easily buy Stanford researchers (of course you can't, it's a ridiculous charge to make.) Another actually serious issue is that the antibody test is inaccurate. The authors revised the conclusions from 2.5-4.2% to 1.3-4.7%. Still, a lot of infected people.

Update (May 18): A whistleblower complaint filed last week with Stanford University reveals that the Santa Clara study was partially funded by JetBlue Airways founder David Neeleman, who has spoken out against the use of lockdowns to slow the spread of COVID-19, BuzzFeed News reports. The information, which was not publicly disclosed, raises “concern that the authors were affected by a severe conflict of interest,” according to the complaint, which was filed by someone involved with the research. The complaint also suggests that the study’s authors disregarded warnings raised by Stanford professors about the accuracy of the antibody test used. In interviews with BuzzFeed, Neeleman and study coauthor Eran Bendavid denied that Neeleman or other funders had influenced the study.

Update (May 1): Bhattacharya and colleagues respond to criticisms of the Santa Clara study in a revised preprint posted yesterday. Using updated statistical analyses, the team now estimates that between 1.3 percent and 4.7 percent of the county’s population—the equivalent of 25,000–91,000 people—have been infected with SARS-CoV-2.

https://www.the-scientist.com/news-opinion/how-not-to-do-an-antibody-survey-for-sars-cov-2-67488

 

[Jarvis323]

The conflict of interest is something I wasn't aware of.

The study is flawed because the sample wasn't random or likely representative of the population, the test is inaccurate, and the number of positives relative to the inaccuracy of the test is low enough that most or all of the positive cases could just be testing errors. The authors also made statistical errors in the analysis. My understanding is that any conclusion from that data would be highly speculative. Any conflict of interest would just be a cherry on top.

It also was unfortunately used heavily to spread misinformation.

The NYC antibody study is a better one to look at since more people have been infected, the results are less sensitive to noise.

 

[bob012345]

The conflict of interest is something I wasn't aware of.

The study is flawed because the sample wasn't random or likely representative of the population, the test is inaccurate, and the number of positives relative to the inaccuracy of the test is low enough that most or all of the positive cases could just be testing errors. The authors also made statistical errors in the analysis. My understanding is that any conclusion from that data would be highly speculative. Any conflict of interest would just be a cherry on top.

It also was unfortunately used heavily to spread misinformation.

The NYC antibody study is a better one to look at since more people have been infected, the results are less sensitive to noise.

But it is already widely known the actual infection rates are likely much higher than the official case counts. My point was it was nonsense to claim conflict of interest in the first place. There was no conflict of interest. To claim so assumes all the researchers involved are dishonest. You can't discount studies even as "cherry on top" because you don't like someone who funded it. That's completely unscientific. But it is already widely known the actual infection rates are likely much higher than the official case counts.

 

[DennisN]

Various:

Oxford vaccine:
Oxford University Covid-19 vaccine trial has only 50 per cent chance of success (The Telegraph, 23 May 2020, sadly behind paywall)
"Exclusive: Project leader Prof Hill warns against 'over-promising', as vaccine success is far from guaranteed"

Brief news from around the world:
Factbox: Latest on the worldwide spread of the coronavirus (Reuters, May 23, 2020)

Edit: And an article in the Guardian about the Swedish policy and debate:
Sweden 'wrong' not to shut down, says former state epidemiologist (The Guardian, 24 May 2020)
"Scientist who oversaw the response to Sars says country has failed the vulnerable"

 

[bob012345]

Various:

Oxford vaccine:
Oxford University Covid-19 vaccine trial has only 50 per cent chance of success (The Telegraph, 23 May 2020, sadly behind paywall)
"Exclusive: Project leader Prof Hill warns against 'over-promising', as vaccine success is far from guaranteed"

Brief news from around the world:
Factbox: Latest on the worldwide spread of the coronavirus (Reuters, May 23, 2020)

Edit: And an article in the Guardian about the Swedish policy and debate:
Sweden 'wrong' not to shut down, says former state epidemiologist (The Guardian, 24 May 2020)
"Scientist who oversaw the response to Sars says country has failed the vulnerable"

In my view, Sweden hasn't failed even if some of their health officials now think so. Looking at all the countries blindly, you wouldn't single Sweden out as compared to all other European countries. Comparisons just to Scandinavian countries is misleading. Then knowing Sweden did not tank it's economy to get the results they achieved, seems to indicate they made a rational choice. Sure, they could have done some things better within that model as we all could have. When we add up all the costs of the shutdowns, I think Sweden was right. If there is a second, third or even more waves of the pandemic, there will likely not be an endless series of global shutdowns even without a vaccine.

 

[Jarvis323]

But it is already widely known the actual infection rates are likely much higher than the official case counts. My point was it was nonsense to claim conflict of interest in the first place. There was no conflict of interest. To claim so assumes all the researchers involved are dishonest. You can't discount studies even as "cherry on top" because you don't like someone who funded it. That's completely unscientific. But it is already widely known the actual infection rates are likely much higher than the official case counts.

What is non-sense about there being a conflict of interest? The research was funded by a company that had a financial stake in (and agenda to try) ending the lock down.

The fact the study was conducted in violation of scientific principles, is the reason the paper and analysis is not accepted. The complaint about a conflict of interest, is just that those conflicts are supposed to be reported. So the researches also had an ethical violation. The undisclosed conflict of interest in combination with the bad science, is just bad optics at least. It's also hard to imagine a team of so many experienced researchers would accidentally make those mistakes on such a high stakes project, so it looks suspicious. That's just a fact. Make of it what you will.

The true number of infection is expected to be higher than reported, but that doesn't mean you can just throw out any number for the infection rate that is lower than reported and expect it to fly.

 

[DennisN]

In my view, Sweden hasn't failed even if some of their health officials now think so.

I just want to say that I'm not posting about Sweden as an argument for or against lockdowns. As a Swede, I just wanted to give a view of the domestic debate here.

Then knowing Sweden did not tank it's economy to get the results they achieved, seems to indicate they made a rational choice.

I just want to point out that we are suffering financially due to the pandemic even though we have a more relaxed policy. And we are also quite dependent on the economies of other countries.

 

[kyphysics]

How long would it take to develop herd immunity for COVID-19?

What if we just let it spread and kill off the weak. Everyone else who survives gets anti-bodies to protect them.

Could we achieve that within two to three years? When people talk of herd immunity, what is the time-frame for getting there? Thanks!

 

[bob012345]

I just want to say that I'm not posting about Sweden as an argument for or against lockdowns. As a Swede, I just wanted to give a view of the domestic debate here. I just want to point out that we are suffering financially due to the pandemic even though we have a more relaxed policy. And we are also quite dependent on the economies of other countries.

But how much worse would it be with a forced shutdown? But the real issue I see with Sweden is not the number of infections, but why such a higher percentage die? That seems independent of the remaining open policy.

 

[atyy]

Oxford vaccine:
Oxford University Covid-19 vaccine trial has only 50 per cent chance of success (The Telegraph, 23 May 2020, sadly behind paywall)
"Exclusive: Project leader Prof Hill warns against 'over-promising', as vaccine success is far from guaranteed"

It seems community transmission rates are dropping which may make it hard to conclude whether a vaccine is effective or not.
https://news.sky.com/story/coronavi...accine-has-only-50-chance-of-working-11993739

Other grounds for caution are discussed in:
https://www.nature.com/articles/d41586-020-01092-3 19 May 2020
Coronavirus vaccine trials have delivered their first results — but their promise is still unclear
Scientists urge caution over hints of success emerging from small human and animal studies.
The Nature news item points to an interesting paper.
https://www.biorxiv.org/content/10.1101/2020.05.13.092619v2
Convergent Antibody Responses to SARS-CoV-2 Infection in Convalescent Individuals
Davide F. Robbiani et al

 

[kadiot]

Experts at Center for Infectious Disease Research and Policy (CIDRAP) propose a SARSCoV2 SMART testing framework to ensure that the right test is available to the right person at the right time, with timely test results guiding actions that minimize illness, deaths & COVID19 spread.

Key elements:

Right Infrastructure: Factors such as institutional support and supply chain availability must be in place.

Right Population: Testing must be targeted based on the goals of testing.

Right Test: Different types of tests (e.g., molecular, antigen, antibody) are appropriate in different settings.

Right Interpretation: The test sensitivity and specificity—and how well it performs at low versus higher levels of disease in the population—must be considered.

Right Action: Based on test results, what actions are needed to minimize illness, deaths, and disease spread?

 

[Vanadium 50]

California results are some kind of outlier

California statistics have always been an outlier: they are seeing an order of magnitude fewer cases than New York. This may be partially due to less public transportation.

If you are speaking about the Bhattacharya study, there are several studies that suggest Covid spreads faster and is less deadly than originally modeled. To my mind, the reason not to take it as seriously as when it first came out is that the subject recruitment is not as described: apparently an investigator's spouse did their own recruitment. Even if the conclusions are correct, the paper is useless since it does not describe what was done. It might describe what was intended to be done, but that's not the same thing.

The thought that a $5000 donation would somehow throw the results is ludicrous. $5000 buys about 6 days of a postdoc.

If I were a cynic, I might say the objection is (as stated in the the-scientist.com article) is that this result provides support for the Right. Personally, I think we should try and get the facts straight without worrying about who they help.

 

[Vanadium 50]

What if we just let it spread and kill off the weak.

Advocating killing off the weak is despicable.

 

[bhobba]

The the Santa Clara antibody study has been pretty thoroughly rejected by the scientific community if I remember correctly.

It certainly is rejected by me now. The consequences are now clearly and obviously absurd. Strange thing is a few interviews I saw used it to postulate the much lower death rate at about what I mentioned eg the interview with the now retired Swedish epidemiologist justifying their countries policies. Oh well back to the drawing board.

Thanks
Bill

 

[bhobba]

Advocating killing off the weak is despicable.

In Australia we now have about 100 deaths - most are the aged and vulnerable. Just one cost of what we are doing is called Jobkeep. Instead of being sacked, your employer gets unemployment benefits of $1500 per fortnight paid to them to keep them employed. This is at a minimum administratively simpler than being sacked, then applying for unemployment benefits etc, but also has advantages as far as self esteem is concerned of those that would otherwise be unemployed. An excellent idea. Originally thought to cost $130 Billion, but due to an estimation error now thought to cost a lot less, at $70 Billion. And that's just one cost - other costs and how badly the economy has been affected would make the figure much higher. Initially the prediction was 50,000 to 150,000 deaths in Australia.
https://www.smh.com.au/politics/fed...0-000-coronavirus-deaths-20200316-p54amn.html

Using the upper limit and just the $70 billion that's about 1/2 million per life saved, but likely a lot more. No country founded on free democratic principles, but especially in a county like Australia built on the spirit of mateship and the battler, would consider the cost - every life is precious beyond measure. Money can be repaid, the economy rebuilt, but life is irreplaceable.

Thanks
Bill

 

[atyy]

It certainly is rejected by me now. The consequences are now clearly and obviously absurd. Strange thing is a few interviews I saw used it to postulate the much lower death rate at about what I mentioned eg the interview with the now retired Swedish epidemiologist justifying their countries policies. Oh well back to the drawing board.

While the Santa Clara study was flawed, there are other reasons not to rule out a death rate near 0.3%. One should of course also not rule out higher death rates near 1% (rough calculation with NYC antibody testing suggests about 0.8%).

In Singapore, the current death rate is about 25 deaths in 30,000 infections (detected by PCR), which is a rate of 0.08%. This is almost certainly too low as a direct generalization to the overall population, as many infections are among workers whose median age is much younger than the median age of the population. It has however, not yet been ruled out that this low number is due to quite a high testing rate.

The action a region chooses to take depends on much more than just the death rate. For example, it would include uncertainties in the data, the number of cases, the rate of new cases, hospital and ICU capacity, resources of the government and population, which varies from place to place. Overall, Sweden seems not to have had an overwhelmed healthcare system (though there seem to be concerns about availability of care), which suggests that the estimates they used for planning were reasonable (though it could of course have been better, and the Swedish government has itself said some things did not go as planned).

 

[PeroK]

In Singapore, the current death rate is about 25 deaths in 30,000 infections (detected by PCR), which is a rate of 0.08%.

That number seems highly suspect, to say the least.

 

[atyy]

That number seems highly suspect, to say the least.

I gave a reason it should not generalize directly (atypical age distribution of those infected). Do you have other reasons?

 

[PeroK]

I gave a reason it should not generalize directly (atypical age distribution of those infected). Do you have other reasons?

It's just not credible.

 

[OmCheeto]

While the Santa Clara study was flawed, there are other reasons not to rule out a death rate near 0.3%. One should of course also not rule out higher death rates near 1% (rough calculation with NYC antibody testing suggests about 0.8%).

Today I found NYC data that is grouped by zip code.
They have 21 areas that have ≥ 0.3% "mortality" rates.
The highest is 0.62%, but is a bit of an outlier.
The rates start consistently at 0.45%, and of course, go down from there.

Populations ranged from 12,000 to 110,000 for the ≥ 0.3% rated areas.

The range of people tested who tested positive was between 25% and 45%.

[ref]

 

[atyy]

It's just not credible.

It's probably quite correct for the current situation in Singapore - again, I want to stress that without taking factors such as the age distribution into account, the current low death rate likely does not generalize to the situation in which most of the Singapore population gets infected. However, the current number is not due to a high false positive rate, since the confirmed cases are all by PCR, not antibody testing. There are also checks for false positives by PCR, and some initially positive cases by PCR have been shown to be false positives, so overall the denominator is not inflated. It is also unlikely to be due to misclassification of deaths. There have been some cases that were attributed to heart problems, and that tested positive after death. These cases were not added to the COVID-19 deaths, as they were thought not to be due to COVID-19 or to complications related to COVID-19. Possibly they were misclassified, but data released by the government suggests that the death rate due to non-COVID-19 heart problems is not higher than last year, suggesting the misclassification error is small.

 

[atyy]

Today I found NYC data that is grouped by zip code.
They have 21 areas that have ≥ 0.3% "mortality" rates.
The highest is 0.62%, but is a bit of an outlier.
The rates start consistently at 0.45%, and of course, go down from there.

I think that is consistent with what I called the 0.8% death rate, which I should have more properly called the infection fatality rate. The IFR is estimated with the denominator being the number of people infected, which is about 20% of the NYC population. If we take the denominator to be the whole population (both infected and uninfected), then the same 0.8% IFR would give about 0.15% COVID-19 population mortality.