Coronavirus - worldwide spread and control

Status
Not open for further replies.
Watching the PM’s address to the nation, I felt like I was losing the plot a bit.

Can we discuss the science behind the graph which kept being shown?

upload_2020-5-10_22-5-53.jpeg


agree and understand what others have said about warning levels needing to reflect both R together with how many people are currently infected including pre symptomatic people. So in this case why on earth is this graph being plotted along with R?

Number new infections or deaths along Y axis, what’s on the X axis? R? So the x axis is back to front..

Wait.. and did R ever “peak” at 1? I am so confused as to what on earth this graph is saying.

Something else was niggling at me and I realised what it is. The other thing is this graph seems to say that as we come down in number of cases/deaths, that means R is coming down (maybe?), but that also means we need to start easing restrictions at every step because that’s what the pictures show. But why? You could easily keep lockdown until cases have come down completely before easing restrictions. Whereas this graph could be interpreted as :you need to ease those restrictions at each step in order to keep coming down the graph. And again it’s all related to this “R”.

Also no mention of the fact that the very act of easing will mean increased spread.. so how does that fit into the graph at all?

Anyway it confused me. Either this is what the scientists on SAGE have been advising the govt about, or the govt have come up with it themselves, but since they’ve used this to explain their whole strategy including alert levels, it’s a bit disconcerting.
 
Oh, not to mention..
This “equation”...
click on it to read the rest of the thread. I’m not sure but to me this just seems like a way to keep the NHS from being overwhelmed and a return to herd immunity, rather than a proper elimination plan.

So the R swings from left to right and right to left. When R is higher, tougher measures. When R is lower, less tougher measures.





 
lunarainbows said:
So the R swings from left to right and right to left. When R is higher, tougher measures. When R is lower, less tougher measures.
Click to expand...

So we have an alert level of R + number of cases. Given R is small and the number of cases is quite large (even at a few thousand) then the contribution of R is minimal and the alert level = number of cases. But the alert level is between 1 and 5 so …
 
Adrian said:
So we have an alert level of R + number of cases. Given R is small and the number of cases is quite large (even at a few thousand) then the contribution of R is minimal and the alert level = number of cases. But the alert level is between 1 and 5 so …
Click to expand...

Yeah. the “equation” makes no sense whatsoever. Can’t quite believe it’s been shown all around the nation.
 
lunarainbows said:
Can we discuss the science behind the graph which kept being shown?
Click to expand...
To me, that graph is implausible. It suggests that easing restrictions (or adjusting measures) will have no effect on the effective reproduction rate R (or Re). In reality one would expect easing restrictions to increase Re. The angle of the gradient of the downward slope would therefore be expected to decrease with each adjustment in social restrictions, or start to go up (if Re goes above 1). Unless the number of new infections is close to zero, any increase in Re (or failure to reduce Re) will result in more avoidable deaths.

They won't say it but to me the UK strategy still appears to be mitigation (ie slowed down herd immunity) rather than suppression. I can't think of any other reason for allowing Re to increase before the number of new infections is very close to zero - which it could be in a few weeks if we stuck with lockdown and got an effective track, trace and quarantine system in place.


Adrian said:
So we have an alert level of R + number of cases. Given R is small and the number of cases is quite large (even at a few thousand) then the contribution of R is minimal and the alert level = number of cases. But the alert level is between 1 and 5 so …
Click to expand...
Yes, this would appear to be one of the more stupid calculations to have been presented - in a highly competitive field. I can only assume that the addition sign is supposed to be a multiplication sign. But it is perhaps unreasonable to expect the UK's crack team of experts to be aware of the difference between such complex mathematical concepts.
 
Adrian said:
So we have an alert level of R + number of cases. Given R is small and the number of cases is quite large (even at a few thousand) then the contribution of R is minimal and the alert level = number of cases. But the alert level is between 1 and 5 so …
Click to expand...
I imagine it's more of a function than an equation

Alert level = f(R, No. Of infections)
 
Robert 1973 said:
Yes, this would appear to be one of the more stupid calculations to have been presented - in a highly competitive field. I can only assume that the addition sign is supposed to be a multiplication sign. But it is perhaps unreasonable to expect the UK's crack team of experts to be aware of the difference between such complex mathematical concepts.
Click to expand...

I would say alert level = f(R, Number_of_cases)

But then the question is what exactly is the function and this leads to questions over detail which probably haven't been thought out. My guess is that is will be someone (or a committee of people) looking at an estimate of the value of R (and hopefully error bars on that) along with the numbers of cases (is that total cases or new cases) and put their finger in the air and say today I think the alert level should be x
 
Robert 1973 said:
They won't say it but to me the UK strategy still appears to be mitigation (ie slowed down herd immunity) rather than suppression. I can't think of any other reason for allowing Re to increase before the number of new infections is very close to zero - which it could be in a few weeks if we stuck with lockdown and got an effective track, trace and quarantine system in place.
Click to expand...

I think so too. It makes no sense otherwise.

But then why bother with contact tracing at all? They’ve been talking about “track and trace“. Unless that’s also going to be an attempt to just trace some people, enough to keep cases from overwhelming the NHS.

It’s all worrying. :(
 
lunarainbows said:
I think so too. It makes no sense otherwise.

But then why bother with contact tracing at all? They’ve been talking about “track and trace“. Unless that’s also going to be an attempt to just trace some people, enough to keep cases from overwhelming the NHS.

It’s all worrying. :(
Click to expand...

I hope when they talk of track and trace they mean track trace and isolate. But if that is isolate in your house then there is the question as to whether the whole family need to be isolated or further contract tracking happens with family contacts. I'm assuming contract tracking needs to be quick - they could use the nightingale hospitals for isolation but I suspect they won't.
 
As reality without perception is not knowledgable, so far some observations from euromomo in participating countries:

The upper half of the site shows absolute deaths,
The lower half excess deaths (which may so far be comparable in numbers to 2018).
Countries can be seen separately as well.

https://www.euromomo.eu/graphs-and-maps


For an assessment not included are of course:
  • long term conditions in survivors
  • side effects from measures, including deaths
  • deaths due to overwhelmed health care
 
Last edited:
lunarainbows said:
The other thing is this graph seems to say that as we come down in number of cases/deaths, that means R is coming down (maybe?), but that also means we need to start easing restrictions at every step because that’s what the pictures show. But why?
Click to expand...
What to me sticks out as well is that the graph doesn´t change after any of the easings. I guess this possibility should be taken into visible account, for clarity and further ideas.

I still ask though, what the impact and effect of measures is at all. As in the three R´s I had given above can be seen, in Austria, Switzerland and Germany no effect on R was seen after any of the stronger measures have been implemented. Only in Switzerland (I think it was) R went abruptly further down from about 1 to 0.7.

After having relaxed the measures in Germany numbers still declined last week (though R has risen from about 0.7 to 1.13 - ???), but might be a bit more to a plateau, maybe will rise??


Leila said:
It's up to every district/county now to keep infections at bay and go back to implementing stricter measures when there are >50 cases per 100.000 citizens per week. So more of a localized rather than nationwide approach.
Click to expand...
Who seeks will find (though maybe not everywhere).

It might be more wise to count the hospital admissions.
 
Last edited:
spinoza577 said:
What to me sticks out as well is that the graph doesn´t change after any of the easings. I guess this possibility should be taken into visible account, for clarity and further ideas.

I still ask though, what the impact and effect of measures is at all. As in the three R´s I had given above can be seen, in Austria, Switzerland and Germany no effect on R was seen after any of the stronger measures have been implemented. Only in Switzerland (I think it was) R went abruptly further down from about 1 to 0.7.

After having relaxed the measures in Germany numbers still declined last week (though R has risen from about 0.7 to 1.13 - ???), but might be a bit more to a plateau, maybe will rise??


Who seeks will find (though maybe not everywhere).

It might be more wise to count the hospital admissions.
Click to expand...

If having >50 cases per county means stricter measures again maybe nobody wants to seek and find anymore.

I totally get why localized approaches are necessary. Seems like they have their flaws though, too.
 
After yesterday’s change in rules, Sky news footage of the tube in London today..:( :(
It is terrifying, where this is all going.

Coronavirus: Commuters pack London Tube platforms after PM's lockdown announcement
A Tube driver says there is "no social distancing going on" and it is "only normally this busy on a normal peak day pre-COVID".

https://news.sky.com/story/coronavi...orms-after-pms-lockdown-announcement-11986344
 
Woolie said:
From the Guardian: "Scientists are concerned that the coronavirus is adaptingt o humans"

https://www.theguardian.com/society...cerned-that-coronavirus-is-adapting-to-humans

Ignore the shitty title, the article is a thing of great beauty. So simple, so clear. I see its by someone called Ian Sample - great science communicator!
Click to expand...
research paper mentioned (still going thro peer review)
Controlling the SARS-CoV-2 outbreak, insights from large scale whole genome sequences generated across the world
ABSTRACT
Background SARS-CoV-2 most likely evolved from a bat beta-coronavirus and started infecting humans in December 2019. Since then it has rapidly infected people around the world, with more than 3 million confirmed cases by the end of April 2020. Early genome sequencing of the virus has enabled the development of molecular diagnostics and the commencement of therapy and vaccine development. The analysis of the early sequences showed relatively few evolutionary selection pressures. However, with the rapid worldwide expansion into diverse human populations, significant genetic variations are becoming increasingly likely. The current limitations on social movement between countries also offers the opportunity for these viral variants to become distinct strains with potential implications for diagnostics, therapies and vaccines.

Methods We used the current sequencing archives (NCBI and GISAID) to investigate 5,349 whole genomes, looking for evidence of strain diversification and selective pressure.

Results We used 3,958 SNPs to build a phylogenetic tree of SARS-CoV-2 diversity and noted strong evidence for the existence of two major clades and six sub-clades, unevenly distributed across the world. We also noted that convergent evolution has potentially occurred across several locations in the genome, showing selection pressures, including on the spike glycoprotein where we noted a potentially critical mutation that could affect its binding to the ACE2 receptor. We also report on mutations that could prevent current molecular diagnostics from detecting some of the sub-clades.

Conclusions The worldwide whole genome sequencing effort is revealing the challenge of developing SARS-CoV-2 containment tools suitable for everyone and the need for data to be continually evaluated to ensure accuracy in outbreak estimations.
Click to expand...
https://www.biorxiv.org/content/10.1101/2020.04.28.066977v1
 
Status
Not open for further replies.
Back
Top Bottom