Coronavirus - worldwide spread and control

Status
Not open for further replies.
https://amp.theguardian.com/commentisfree/2020/may/11/boris-johnson-advice-coronavirus-spread-work?

If we follow Boris Johnson's advice, coronavirus will spread
David Hunter
A large-scale return to work without the ability to test, trace and isolate risks creating super-spreader events
  • David Hunter is a professor of epidemiology and medicine at the University of Oxford

“Those defending the government’s Covid-19 response have reasonably pointed out that policy mistakes are always clearer in retrospect. So let me make a prediction. If we take the prime minister’s advice and return to work in large numbers now – and without the ability to test, trace and isolate – then virus spread will increase, there will be super-spreader events and local or regional lockdowns will have to be reconsidered. The prime minister implied in his speech that relapse will somehow be our fault – we were not sufficiently “alert”. The responsibility will lie, however, with a government that has encouraged a premature return to work before the epidemiologic conditions and interventions were in place to make it safe to do so.”
 
I just got a very fine link, so that I thought I should share it,
a pretty well done video comment on mortalities as shown by euromomo.
Not what one may think anyway.

"NOT What You May Think! Viral Mortality Comparison Between 2018 and 2020"
 
Jonathan Edwards said:
They definitely still don't get it.
Where is the bit about employing 40,000 new pubic servants to actually do what needs to be done to rescue the economy (and in the meantime creating 40,00 new jobs paying tax and thereby substantially expanding the economy).
Click to expand...

What seems to me to be ludicrous is that, not unreasonably there are "n" times (10X?) 40,000 civil servants CURRENTLY working from home on their computers --- why not transfer some of them to contact tracing. OK, do what the insurance companies do i.e. monitor the calls "for training purposes" and re-train people if need be. They possibly do not need to employ additional people --- so why isn't it happening?
 
spinoza577 said:
I just got a very fine link, so that I thought I should share it,
a pretty well done video comment on mortalities as shown by euromomo.
Not what one may think anyway.

"NOT What You May Think! Viral Mortality Comparison Between 2018 and 2020"
Click to expand...


I have just watched that video. What he has done is compare the about 144,000 deaths attributed to Covid in 20 European countries so far over about 8 weeks or so since the pandemic took off, with the 160,000 deaths from flu in those countries over the 18 week flu season of winter 2018.

He then looks at the graphs of the deaths in each country and shows the death rate coming down since lockdown, and starts extrapolating wildly, assuming that lockdown reductions will continue basically to zero deaths in the next couple of months, and assuming it will magically die out over the summer. He fails to mention that now that lockdown is being lifted over the next weeks or months, more spikes of cases are likely, and the fact that it's a new virus not like flu, and we have no idea whether it will die out over the summer.

He uses this unfounded extrapolation to demonstrate his prediction of only about twice as many deaths overall from Covid than there have been so far. He is clearly doing this to pretend it will all be over in a couple of months and it's not really much worse than a flu season.
 
Trish said:
I have just watched that video. What he has done is compare the about 144,000 deaths attributed to Covid in 20 European countries so far over about 8 weeks or so since the pandemic took off, with the 160,000 deaths from flu in those countries over the 18 week flu season of winter 2018.

He then looks at the graphs of the deaths in each country and shows the death rate coming down since lockdown, and starts extrapolating wildly, assuming that lockdown reductions will continue basically to zero deaths in the next couple of months, and assuming it will magically die out over the summer. He fails to mention that now that lockdown is being lifted over the next weeks or months, more spikes of cases are likely, and the fact that it's a new virus not like flu, and we have no idea whether it will die out over the summer.

He uses this unfounded extrapolation to demonstrate his prediction of only about twice as many deaths overall from Covid than there have been so far. He is clearly doing this to pretend it will all be over in a couple of months and it's not really much worse than a flu season.
Click to expand...
He also does not acknowledge that testing has not been afforded to many and therefore deaths have not been attributed to COVID.

FT excess death stats are a better resource - they have averaged national / city / state death rates over previous 4 to 5 years and compared it to current rates. You are then comparing all cause mortality with all cause mortality. There will be glitches as there will be deaths due to treatments being suspended for other conditions, and people avoiding hospital, but the curve directions are telling .

https://www.ft.com/content/a26fbf7e-48f8-11ea-aeb3-955839e06441

There is also a feature now to pick and compare. When I did UK and Sweden , they were spookily similar.
 
This is the England government guidance issued to schools, educational settings (18 and under), early years and childcare providers:

https://www.gov.uk/government/publi...-measures-in-education-and-childcare-settings

If I was still working in schools as a Learning Support Assistant, I would be handing in my notice. The kids I was working with coughed and sneezed over me all day - I was constantly getting viral infections (unlike many PWME I am very prone to getting viral and bacterial infections).

My daughter brought over my shopping this afternoon after being in school with 12 kids. She commented to stay well away from her as none of the kids had been able to keep the 2 metre rule or stop themselves from coughing or speaking when close to her (these are primarily the well children of key workers who do not have any 'special needs'). My daughter has a reputation as a very strict teacher, so this wasn't a problem with her classroom behaviour management - the kids just can't adapt their behaviour for the duration of whole hour lessons and during their 'free time'.
_____________________

Moderator note: This post has been copied and replies moved to
Covid-19 - School closures and partial school closures
 
Last edited by a moderator:


Edit: too tired last night to say more. But very good article. They look at each of the 5 “tests” the U.K. govt has set out and what actually needs to be done in order to ease lockdown. And talk about R rate. They end with:

“The stakes in making the right decision cannot be higher. However, the data on covid-19 mortality show that it is not “just a flu.” Some previously healthy patients also face complicated recovery after acute infection. The UK has made some progress in controlling the pandemic since it entered lockdown on 23 March. However, a system of testing, tracing, and isolating is still not in place. Adequate provision of high quality PPE is not secure. Transmissions in care homes, hospitals, and households are far too high. Of course, the Government needs to balance health, wealth, and public acceptance of lockdown in making decisions. In this case, however, the health implications are so overwhelming that loosening the lockdown too much now will not do the economy and people’s livelihoods any good.

For those who are still not convinced that a cautious approach to easing lockdown is required, it is worth noting that the Rt in Wuhan at this stage of lockdown was below 0.2.“
 
Last edited:
spinoza577 said:
I just got a very fine link, so that I thought I should share it,
a pretty well done video comment on mortalities as shown by euromomo.
Not what one may think anyway.

"NOT What You May Think! Viral Mortality Comparison Between 2018 and 2020"
Click to expand...

Agree with @Trish, there are all kinds of problems with this guy's argument.

It doesn't start well: the presenter uses the term "excess death rates" to describe what are actually officially reported COVID-related death rates per million. The whole idea of excess death rates as a concept is that does not discriminate as to cause, so talking about excess death rates "due to COVID" makes no sense at all.

Another huge blunder is that he does not compare like with like: he compares all-cause mortality over a previous winter with deaths officially attributed to COVID. He needs to pick one or the other - either compare deaths officially attributed to flu with deaths officially attributed to COVID. Or alternatively, compare all-cause mortality for the reference winter season with all-cause mortality for the covid period. Those comparisons will not be as favourable to his conclusions of course - that the rate of COVID deaths is "not as bad as you think". Never let the data get in the way of a good story!

And finally, as @Trish says, he doesn't take into account the differences in the time periods he's comparing: an entire winter season for "flu" and six or so weeks for COVID. And the big whammy is that he seems to think its all going to just get better, but of course the more likely scenario is that as countries reopen, cases will rise. There's no evidence at all that this disease will just go away in the summer - it raged through Australia and NZ in our summer until the two countries put a lid on it.

I'm all for carefully examining the relevant data, and not overstating the case. But I'm utterly flummoxed as to why people would actively misconstrue data in a way that minimises the reality of this virus. Is it somehow too unbearable to face up to? Or is it that people just want to be able to do what they want, without having to feel guilty about other people dying?
 
Trish said:
I have just watched that video. What he has done is compare the about 144,000 deaths attributed to Covid in 20 European countries so far over about 8 weeks or so since the pandemic took off, with the 160,000 deaths from flu in those countries over the 18 week flu season of winter 2018.
Click to expand...
Yes, and so far.

He then looks at the graphs of the deaths in each country and shows the death rate coming down since lockdown, and starts extrapolating wildly, assuming that lockdown reductions will continue basically to zero deaths in the next couple of months, and assuming it will magically die out over the summer. He fails to mention that now that lockdown is being lifted over the next weeks or months, more spikes of cases are likely, and the fact that it's a new virus not like flu, and we have no idea whether it will die out over the summer.
Click to expand...
He says that the virus seems to behave like other respiratory viruses, influenza and other corona viruses.

Apart from being cautious, it´s not he who has to prove that this is the case, but it´s you who has to prove that this virus behaves in another manner.

You have also to show that the lockdown has an effect that would be worth to continue with it, or to reinduce it. This is - difficult as it is, admittedly - not a question of logic but a question of empirical investigation.
 
Last edited:
Woolie said:
It doesn't start well: the presenter uses the term "excess death rates" to describe what are actually officially reported COVID-related death rates per million. The whole idea of excess death rates as a concept is that does not discriminate as to cause, so talking about excess death rates "due to COVID" makes no sense at all.
Click to expand...
True, such an assumption is simplifying, saying that excess rates 2018 are due to flu and 2020 due to covid-19. This objective accounts for both viruses and years.

But do you have an alternative to figure out what goes on in a certain time period, say to investigate how sever a flu season was? Note that it is good luck that this year - as it seems to be the case - the flu didn´t cause problems.

Admittedly, as far as possible, you need to have a further look. But with covid-19 you have two additional impacts: A) overwhelmed health care, and B) effects from measures. So any deaths here will even subtract from excess deaths roughly interpreted as covid-19 deaths. Admittedly A) should be also true for some influenza deaths, and B) might not be high (at least not already).

Another huge blunder is that he does not compare like with like: he compares all-cause mortality over a previous winter with deaths officially attributed to COVID. He needs to pick one or the other - either compare deaths officially attributed to flu with deaths officially attributed to COVID. Or alternatively, compare all-cause mortality for the reference winter season with all-cause mortality for the covid period. Those comparisons will not be as favourable to his conclusions of course - that the rate of COVID deaths is "not as bad as you think". Never let the data get in the way of a good story!
Click to expand...
The official covid-19 deaths are biased in two ways: 1. not all deaths positive for covid-19 may have been counted 2. Many might have died anyway (as it is due to influenza). Because of 2. the concept of excess death is used (see point above). Saying it most directly, these excess deaths are the influenza deaths then, as the officially reported influenza deaths (death cases positive for influenza) are misleading.

Here he says now that official covid numbers may match up with excess deaths, which is obvious enough, for a rough comparison. It would be better, I agree, to have original euromomo numbers, but this sidestep is rather not in the favour of making covid-19 smaller.


I don´t see why you want to reject to get an idea of the situation (only because of the quality of a disease or whatsoever). Don´t you see that this allows capriciousness?
 
Last edited:
From my point of view a lot of this is sophistry.

If someone dies because they can't get shopping, or because they were too afraid to seek treatment for another condition, or because they had an accident and couldn't get the correct treatment in time, or because they were inappropriately kicked out of hospital onto the streets to clear beds for covid patients, or due to lack of care that they would normally have received, and a thousand other things.

If these deaths occured because covid-19 happened, if these deaths would probably not have occurred if covid-19 hadn't happened, then those are deaths due to covid-19 -- no matter what the actual recorded cause of death.
 
You can't take the number of excess deaths in 2018 and compare them with the number of excess deaths in 2020. The bare numbers do not take into account the precautions being taken this year to minimize infection with covid.

In 2018 we did not have any social distancing, pubs were packed, people had large weddings and thousands went to concerts. We did not wash our hands for 20 seconds every time we went out and came back in either or wear masks and use PPE in hospitals. There was no lockdown.

So the excess deaths this year so far have been despite all the precautions we took to prevent infection whereas in 2018 we carried on life as normal and did not give a thought to becoming ill.

I have been wary of this sort of calculation since I read that the Royal Free epidemic did not exist because there were no extra people becoming ill. It was just a random effect that they all became sick in a short period of time

You can prove anything with numbers out of context.
 
Wonko said:
If these deaths occured because covid-19 happened, if these deaths would probably not have occurred if covid-19 hadn't happened, then those are deaths due to covid-19 -- no matter what the actual recorded cause of death.
Click to expand...
In first instance I agree, but I principally disagree for an assessment of the whole situation, and insist to take possibilities into account.

If we would like to know how dangerous the virus is, we are in this knowledge not interested in how dangerous our inability to cope with a difficult situation is.

The latter point may not only account for non-virus caused deaths (which otherwise might not have occured), but also for virus caused deaths which could have been hindered. There was an investigation somewhere, China I think, that this latter point would have been important. So, this is again something that needs to be distinguished, and this is here not only academic play.
 
Mithriel said:
In 2018 we did not have any social distancing, pubs were packed, people had large weddings and thousands went to concerts. We did not wash our hands for 20 seconds every time we went out and came back in either or wear masks and use PPE in hospitals. There was no lockdown.

So the excess deaths this year so far have been despite all the precautions we took to prevent infection whereas in 2018 we carried on life as normal and did not give a thought to becoming ill.
Click to expand...
For flu there is precaution as well, in fact in 2018 the vaccine didn´t work well (so have heard). Nevertheless you would not come up to stop normal life (or is this changing now?).

Secondly, you need to show that the measures have an effect, this is not clear by itself. I said it already twice if not thrice, R went down in three countries already BEFORE stronger measures have been implemented or could excert their presumed effect. This raises of course a lot of questions about differences seen in countries.

A good argument in your favour may be that Sweden probably has more excess deaths than Norway or Finland, which can be - directly - attributed to covid-19. Here another question comes up, if this won´t level out in the next months, and this would be important to know because measures have side effects.


I have been wary of this sort of calculation since I read that the Royal Free epidemic did not exist because there were no extra people becoming ill. It was just a random effect that they all became sick in a short period of time

You can prove anything with numbers out of context.
Click to expand...
Statistically the Royal Free outbreak may be non-existent. But therefore there is also no discussion about society-wide measures justified (as it is justified with corona).

Even more important than argumentation is power of judgement, sadly there is no methodology on it possible, so that numbers are misused - as you say - can sadly not be hindered if not through a spirit that is curious and behaves also otherwise human.
 
Last edited:
Status
Not open for further replies.
Back
Top Bottom